Note: Descriptions are shown in the official language in which they were submitted.
DEMANDE OU BREVET VOLUMINEUX
LA PRRSENTE PARTIE DE CETTE DEMANDE OU CE BREVET COMPREND
PLUS D'UN TOME.
CECI EST LE TOME 1 DE 2
CONTENANT LES PAGES 1 A 391
NOTE : Pour les tomes additionels, veuillez contacter le Bureau canadien des
brevets
JUMBO APPLICATIONS/PATENTS
THIS SECTION OF THE APPLICATION/PATENT CONTAINS MORE THAN ONE
VOLUME
THIS IS VOLUME 1 OF 2
CONTAINING PAGES 1 TO 391
NOTE: For additional volumes, please contact the Canadian Patent Office
NOM DU FICHIER / FILE NAME:
NOTE POUR LE TOME / VOLUME NOTE:
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
IN THE UNITED STATES RECEIVING OFFICE
AUTOMATED SYSTEMS AND METHODS FOR
OBTAINING, STORINGi PROCESSING AND UTILIZING
IMMUNOLOGIC AND OTHER INFORMATION OF AN INDIVIDUAL OR
POPULATION FOR VARIOUS USES
CROSS-REFERENCE TO RELATED APPLICATIONS:
This application claims priority to and the benefit of United States
Provisional Patent
Application Serial No. 61/002,704, filed on November 8, 2007, the disclosure
of which is also
hereby incorporated herein by reference.
Applicant also refers to the disclosure of elated application United States
Utility Patent
Application Serial No. 11/796,727, filed on April 27, 2007, and which was
published as U.S.
Patent Application Pub. No. 2008-0091471 Al.
TECHNICAL FIELD:
The present invention relates to individualized health care, immunology and
medical informatics,
and more particularly to automated systems and methods for acquiring, storing,
processing and
utilizing immunologic and other information of individuals and populations for
decision making
in various public health, medical and health care related applications.
BACKGROUND OF THE INVENTION:
Personalized medicine is considered by many to be the wave of the future. A
personalized
medicine approach seeks to identify whether a given individual needs a given
treatment or
intervention prior to administering it, rather than relying on "standards"
representing an average
person in a group or population.
-1-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
This approach is based on the well known fact that some individuals in a
demographic
population have naturally low or naturally high values which are not best
measured against a
statistical mean for the demographic population, but against that individual's
own measured
history.
Determination of the immune status of individuals to, for example, vaccine-
preventable diseases
requires an assay system that can detect antibodies that may be present at
very low levels,
especially when natural or vaccine exposure may have been many years
previously. In addition,
such an assay system could be used more generally to assess an individual's
immune competence
at different stages of that individual's life, as well as to also measure the
vaccine status of
individuals with varying special needs and requirements (e.g., military
personnel or travelers).
What is thus needed in the art is a system and method for measuring and
processing
immunologic information of individuals and populations through various points
in time of their
lives so as to better track each individual's immune status and make
appropriate diagnostic,
prophylactic and therapeutic recommendations.
What is further needed in the art is a supporting structure to conveniently
store the results of such
screenings for easy access and processing, for data mining purposes as well as
for use in a
variety of commercial, research and governmental applications where a
knowledge of the
immunological indicia of customers, subjects and citizens can create
efficiencies and
optimizations, as well as allow for the exploitation of commercial
opportunities and improve the
quality of life.
SUMMARY OF THE INVENTION:
-2-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
A system and method for assessing the immunological status of one or more
individuals in a
patient population is presented. The method includes establishing a database
comprising a
plurality of records of information each representative of the immune status
of an individual in
the population, each of said records including (1) current information from
one or more assays
for the presence of a biochemical, and (2) individual specific information
comprising one or
more of said individual's medical history, said individual's doctors'
observations and historical,
demographic, lifestyle, and familial information relating to said individual.
The method further
includes processing the information in said database to find trends or
patterns relating to the
immune status of individuals in said patient population; and using the said
trends or patterns as
part of a health care related decision making process. In exemplary
embodiments of the present
invention, processing the information in the database includes generating a
list of correlations
between variables or fields in the database. The correlations in the list can
be further refined
automatically, and a set of hypotheses or literature ctations can be linked to
the final correlations.
The correlations, the processing, their associated hypotheses can then be
reported to a user or
automatically fed into another system component to generate a medical or
health related
decision. In exemplary embodiments of the present invention, a first assay
panel containing a
plurality of cytokine assays can be administered and the results processed.
Based on automatic
analyses of the cytokine data, a second tier or set of assays can be run on
the same individual.
The cytokine assay results being used to inform the contents of a second assay
panel.
BRIEF DESCRIPTION OF THE DRAWINGS:
Section I Figures
Fig. 1 depicts a generalized exemplary process flow according to exemplary
embodiments of the
present invention;
-3-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Fig. 2 depicts an exemplary system overview according to exemplary embodiments
of the
present invention;
Fig. 2A depicts an alternate exemplary system overview according to exemplary
embodiments of
the present invention;
Fig. 2B depicts yet another alternate exemplary system overview according to
exemplary
embodiments of the present invention;
Figs. 3 and 4 depict various exemplary configurations for assaying a patient
sample according to
an exemplary embodiment of the present invention;
Fig. 5 depicts a detailed system diagram according to an exemplary embodiment
of the present
invention;
Fig. 5A depicts a detailed system diagram according to an alternate exemplary
embodiment of
the present invention;
Fig. 5B depicts T helper cell commitment towards specific lineages;
Fig. 5C depicts an exemplary Thl/Th2 Paradigm model as it existed circa 2000;
Fig. 5D depicts an exemplary evolving Thl/Th2/Thl7/Treg paradigm, which now
includes arms
that recognize the importance of Th 17 and Treg cells;
Fig. 5E depicts an exemplary model illustrating how Treg-mediated control of
CD80/CD86
expression may control the threshold of antigen recognition, crucial for
preventing the activation
of low avidity self-reactive T cells that are below the cut-off imposed during
thymic selection;
-4-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Fig. 5F depicts a model of the development of the immune response in
schitosome infection;
Section II Figures
Fig. 6 depicts exemplary assay results in an exemplary database according to
the present
invention;
Fig. 7 depicts exemplary diagnostic module recommendation types according to
an exemplary
embodiment of the present invention;
Fig. 8 illustrates an exemplary perceptron network which implements a rule for
a normal
individual using as inputs the results of an exemplary menigicoccal diagnostic
panel;
Fig. 8A illustrates the exemplary perceptron network of Fig. 8 implementing a
similar rule for an
abnormal individual;
Fig. 9 depicts an XML representation of the exemplary perceptron networks of
Figs. 8 and 8A;
Fig. 10 depicts an exemplary symbology for diagnostic goals which can be used
to articulate
diagnostic goals in an exemplary embodiment of the present invention;
Fig. 11 illustrates exemplary diagnostic goals using the symbology of Fig. 10;
Fig. 12 illustrates an exemplary database schema for patient information
according to an
exemplary embodiment of the present invention;
Fig. 13 illustrates an exemplary database schema for visit information
according to an exemplary
embodiment of the present invention;
-5-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Fig. 14 illustrates an exemplary database schema for test results according to
an exemplary
embodiment of the present invention;
Fig. 15 depicts exemplary patient age intervals used in an exemplary database
according to an
exemplary embodiment of the present invention;
Fig. 16 is a plot of an exemplary female antibody comparison over a number of
years according
to an exemplary embodiment of the present invention.
Fig. 17 is a plot of an exemplary comparison of two individual females, one
vaccinated and one
not vaccinated, according to an exemplary embodiment of the present invention;
Fig. 18 is a plot of exemplary antibody levels in a compliment-deficient
individual according to
an exemplary embodiment of the present invention;
Fig. 19 is a plot of exemplary antibody levels in a healthy individual
according to an exemplary
embodiment of the present invention;
Fig. 19A is an example SQL query according to an exemplary embodiment of the
present
invention; and
Fig. 19B is a table illustrating the correlation among antibody levels in an
exemplary female
population according to an exemplary embodiment of the present invention;
Figs. 20 through 20F illustrate exemplary data mining results obtained from
operating on an
exemplary database according to an exemplary embodiment of the present
invention;
Fig. 21A illustrates an exemplary pattern detection process flow according to
an exemplary
embodiment of the present invention;
-6-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Fig. 21B illustrates an exemplary pattern detection process flow with
hypothesis generation
according to an exemplary embodiment of the present invention;
Fig. 21 C illustrates an exemplary automatic pattern detection process flow
according to an
exemplary embodiment of the present invention;
Figs. 21D-1 through 21D-37 illustrate automated data mining protocols
according to an
exemplary embodiment of the present invention;
Figs. 21D-38 thorugh 21D-40 respectively illustrate exemplary algorithms for
Hepatitis A Virus
(HAV), Hepatitis B Virus (HBV), and Hepatitis C Virus (HCV) Testing according
to an
exemplary embodiment of the present invention;
Figs. 21 E-1 through 21 E- 12 depict exemplary data analysis results obtained
using an exemplary
embodiment of the present invention;
Figs. 21F-1 through 21 F-6 depict the results of predictive models built using
cytokine data
according to an exemplary embodiment of the present invention;
Figs. 21G-1 through 21G-12 depict the results of running an exemplary patient
population rule
mining protocol according to an exemplary embodiment of the present invention;
Figs. 21H-1 thorugh 21H-10 depict the results of running an exemplary
individual patient
vaccine recommendation protocol according to an exemplary embodiment of the
present
invention;
Fig. 21I is an exemplary output from an exemplary automated data mining
protocol according to
an exemplary embodiment of the present invention, segmenting an exemplary
database by
Region of origin, Sex and the cytokine assay IFN-gamma;
-7-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Section III Figures
Fig. 22 is a process flow diagram for use in a healthcare management
embodiment according to
the present invention;
Fig. 23 is a subset of the process flow depicted in Fig. 22;
Fig. 24 is an alternative process flow chart for healthcare management
according to the
exemplary embodiment of the present invention;
Fig. 24A is a more detailed process flow chart similar to that of Fig 22;
Fig. 25 is an alternative process flow chart for managing healthcare according
the exemplary
embodiment of the present invention;
Fig. 25A is the process flow chart of Fig. 25 with an additional optional
element;
Fig. 26 is an alternative process flow chart for managing healthcare according
to the exemplary
embodiment of the present invention;
Fig. 26A is an alternative version of the process flow of Fig. 26 with greater
detail;
Fig. 27 is a process flow chart for cervical cancer prevention according to
the exemplary
embodiment of the present invention;
Fig. 28 is a process flow chart for managing the care of women of childbearing
age according to
the exemplary embodiment of the present invention;
-8-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Fig. 29 is a process flow chart for an exemplary "Vaccine-O-Mat" application
according to an
exemplary embodiment of the present invention;
Fig. 29A is a system diagram of entities involved in the vaccine distribution
application
according to an exemplary embodiment of the present invention;
Fig. 29B illustrates the necessary connectivity for the vaccine distribution
application illustrated
in Fig. 29A;
Figl 29C is the connectivity displayed in that Fig. 29B recast by use of an
interapplication
connectivity provider according to an exemplary embodiment of the present
invention;
Fig. 30 is an exemplary flow chart for use in a life insurance optimization
application according
to an exemplary embodiment of the present invention;
Fig. 31 is an exemplary process flow chart for use in an immunosenescence
management
application according to an exemplary embodiment of the present invention;
Fig. 32 is an exemplary process flow chart for a disaster management
application according to an
exemplary embodiment of the present invention;
Fig. 33 is an alternative process flow chart for the psychological aspects of
disaster response for
a disaster response application according to an exemplary embodiment of the
present invention;
Fig. 34 depicts exemplary process flow in an immunogenicity discovery
application according to
an exemplary embodiment of the present invention;
Fig. 35 illustrates components of an exemplary two-sided market application
according to an
exemplary embodiment of the present invention; and
-9-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Fig. 36 illustrates components of an exemplary drug hypersensitivity two-sided
market
application according to an exemplary embodiment of the present invention.
-10-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
TABLE OF CONTENTS
Page
CROSS-REFERENCE TO RELATED APPLICATIONS:
................................................ 1
TECHNICAL FIELD:
...............................................................1
........................................
SUMMARY OF THE INVENTION:
...............................................................................
... 2
BRIEF DESCRIPTION OF THE DRAWINGS:
................................................................ 3
DETAILED DESCRIPTION OF THE INVENTION:
....................................................... 16
EXEMPLARY ASSAY PANELS
...............................................................................
....24
A. COLLEGE STUDENT DIAGNOSTIC PANELS .......................................29
1. Meningococcal Diagnostic Panel
...................................................29
2. Sexually Transmitted Diseases Assay Panel ................................
30
3. Persistent Immunity Induced by Childhood Vaccines ....................31
B. ADULT DIAGNOSTIC PANELS
.............................................................. 31
1. Measurement of Immunity Induced By Vaccines for Military
Personnel
...............................................................................
....... 31
2. ImmunoScore Measurement of Vaccine-Induced Immunity for
Travelers
...............................................................................
........32
3. Cytokine Measurement in ImmunoScore .......................................
33
4. Quantitation
...............................................................................
.... 78
C. IMMUNOSCORE EXEMPLARY SUPERPANELS ..................................83
1. ImmunoScore Diagnostic Panel and Preventive Therapy for
Autoimmune Disease
.................................................................... 83
2. ImmunoScore Diagnostic Panel: Aging, Longevity, Cancer and
Human Cytomegalovirus
............................................................... 86
D. EXEMPLARY IMMUNOSCORE SUPERPANELS ..................................99
1. Middle School Student ImmunoPrint Super Diagnostic Panel ....... 99
-11-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Table of Contents
(continued)
Page
2. Exemplary ImmunoScore Diagnostic Panels for Women of Child-
Bearing Years
..............................................................................1
07
EXEMPLARY IMMUNOSCORE SYSTEM DATABASES
........................................... 113
A. General Overview
...............................................................................
..113
B. Exemplary Illustrative Database
......................................................... 122
1. Overall Description
......................................................................122
2. Impact of Data Mining
..................................................................127
3. Diagnostic Module
.......................................................................130
4. Data Mining Module
.....................................................................140
C. Exemplary Canadian Immigrant Project Database Used To
Illustrate Data Mining and Hypothesis Generation ...........................
146
D. Data Mining - Analyses and Conclusions
......................................... 154
1. Linear regression analysis - correlation coefficients ...................
155
2. Geometric mean values
............................................................... 156
3. Percent support between variables
............................................. 158
4. Possible Conclusions
..................................................................159
E. PATTERN DETECTION AND HYPOTHESIS GENERATION ............... 162
1. Initial Exemplary Analysis: Data Mining Steps.............................
172
F. AUTOMATED DATA MINING
............................................................... 174
1. Exemplary Software Development Environment ......................... 174
2. Client-Server Computing
............................................................. 176
3. Third-party Applications
............................................................... 176
4. Extending Pipeline Pilot
............................................................... 176
5. Integrating Protocols
....................................................................178
-12-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Table of Contents
(continued)
Page
6. Data Mining Tool
.........................................................................178
7. Single Patient Vaccine Recommendations ..................................
183
8. Patient Population Rule Mining
................................................... 186
9. Age Binned with Differences
....................................................... 189
10. Automated Data Mining
............................................................... 191
G. EXEMPLARY INTERNAL HYPOTHESIS DATABASE ......................... 199
H. EXPLANATION AND BASIS OF EXEMPLARY RULES CREATED
FOR PROCESSING cip DATABASE
....................................................211
1. INTERPRETATION OF CERTAIN RESULTS OF AUTOMATED
DATA MINING
...............................................................................
........226
J. EXTENSION OF DATABASE AND AUTOMATIC DATA MINING
FUNCTIONALITY
...............................................................................
...231
K. EXEMPLARY ANALYSES PERFORMED ON CIP DATABASE ..........233
L. EXEMPLARY RESULTS USING DATA MINING PROTOCOLS ON
CIP DATABASE
...............................................................................
.....242
USES OF IMMUNOSCORE INFORMATION AND AUTOMATED DATA MINING
RESULTS IN VARIOUS COMMERCIAL, RESEARCH AND GOVERNMENTAL
CONTEXTS
...............................................................................
..................................246
A. Health Insurance Underwriting and Management
.............................246
B. Health Care/Health Insurance Credit ExchanGe
................................258
C. Veterans Health Care Management (Variant of Health Care)............262
D. Socialized Medicine Management
......................................................263
E. Supplemental Insurance (AFLAC Model)
...........................................264
F. ImmunoScore and the Wellness Industry
..........................................266
G. Women of Childbearing Age/Screening of Pregnant Women ..........270
H. Vaccine-o-Mat/Vaccine Distribution Network
....................................272
-13-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Table of Contents
(continued)
Page
1. Consumer Accessibility to Immunologic Information ......................276
J. Immunoscore Connectivity Via Interapplication Translator/Data
Integrator
...............................................................................
...............277
K. Immunologic Informatics Based Life Insurance Underwriting.........278
L. Diagnosing and Managing Immunosenescence in the Elderly ........282
M. Frozen Storage of Naive Immune Cells (IRP Considerations)..........294
N. Vaccine Use Outcome/Design
.............................................................296
0. Research Services
...............................................................................
296
P. Immigration Consulting
....................................................................... 297
Q. Disaster Survivors: Immunizations, Recovery, Prognosis and
Treatment
...............................................................................
............... 301
R. Monitor Adoptive Immunotherapy/Transplants
.................................304
S. Elective Surgery
...............................................................................
.... 304
T. Services to Charitable Foundations Promoting Immunological
Well Being
...............................................................................
..............305
U. Discovery of Unwanted Immunogenicity of Therapeutics ................306
V. Two-Sided Market Applications
..........................................................309
W. Drug Hypersensitivity
..........................................................................322
X. Health Care Transparency and Competition
......................................332
1. Consistent high quality
................................................................ 332
2. Lower cost - follows from high quality
..........................................332
3. Available to all - for ethical, political, systemic, and business
reasons, health care must be available to everyone ....................333
4. Single model - every provider in the system must compete to offer
the best product at the best price
................................................333
-14-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Table of Contents
(continued)
Page
5. Shaped by market forces - the consumer market has the sustained
systemic power to bring consumers more for less ....................... 333
6. Practical - the solution must arise from present realities .............
333
7. Progressive - dramatic change can not occur all at once........... 333
8. Self-reinforcing - as any part of the health care system moves
toward a new reality, that movement must allow and encourage
other parts to move forward as well
............................................. 333
Y. User Access Via Data Networks and On-line Advertising ................343
Z. Prophilatic therapies during surgery
.................................................346
AA. Contraindications for biological active theraputics
..........................346
WHAT IS CLAIMED:
...............................................................................
.................... 360
-15-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
DETAILED DESCRIPTION OF THE INVENTION:
General Overview
In what follows, systems and methods of the present invention will be often
referred to as
the "ImmunoScore" system, method and/or database, as the case may be.
"ImmunoScore" is a trademark and/or service mark currently envisioned by the
assignee
hereof to be utilized in connection with exemplary embodiments of the present
invention.
The present invention is directed to the collection, processing, and use of
immunologic
information. Immunologic information is to be understood in a broad sense,
including
any information which may be useful as an indicator of any immunological
function of a
mammalian body. More specifically, the present invention includes acquiring
information that is indicative of the immune status of an individual,
processing that
information, storing the raw information as well as the outputs from the
processing stage,
and of that information at various times and in various ways to recommend
various
actions such as prophylactic or further diagnostic interventions, or
abstention from action,
for individuals or population. The present invention exploits a number of
advances in
technology as well as advances in how people think about medical treatment. In
exemplary embodiments of the present invention, a number of immunological or
-16-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
immunological related (in a broad sense) assays can be administered to an
individual.
Using modern technology such as, for example, the MIM Analyzer marketed by
BioVerisTM Corporation of Gaithersburg, Maryland, one can run a large number
of
assays, such as, for example 20, 40 or 60, and obtain results therefrom in a
relatively
short period of time. Moreover, these assay results can be stored in a memory,
either
locally or at one or more central servers or in associated databases, and can
be
operated upon by various algorithms or rules which can generate information as
to
that individual's immune status as well as recommendations for further
augmenting
that immune status or taking further action in response to the information
acquired,
from the assays and their processing. This information can be used in a
variety of
commercial, research, and healthcare contexts. Thus a variety of business
methods or
opportunities can be created or facilitated using the information obtained
according to
the methods of the present invention.
The present invention is described in three distinct sections. The first
section
describes the scientific background and motivation for creating various assay
panels
to be administered, singularly or in combination with other assay panels, to
different
individuals in different populations at different times in each individual's
life cycle.
This discussion culminates in suggested or exemplary assay super panels which
can
be administered in various contexts to various individuals.
A second section describes how information obtained from results of the
administered
assays can be stored, processed, and utilized. This discussion comprises,
inter alia, a
description of an exemplary database in which (i) results from numerous assays
can
be stored along with (ii) individual-specific information and (iii) the
outputs of
various algorithms which operate upon the assay results of that individual.
This
-17-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
section also presents an exemplary database upon which immunologic data mining
was performed according to the techniques of the present invention, and
summarizes
interesting and illustrative results form that exercise.
In a third and final section, a variety of business and commercial methods are
described in which information from the assay panels as stored in the database
and
further processed can be used to increase business efficiencies, create new
markets
and opportunities, and/or provide useful tools for research and development.
Before describing each of these three areas in detail, a brief overview of a
generalized
method and system according to exemplary embodiments of the present invention
is
presented with reference to Figs. 1, 2, 2A and 2B.
Fig. 1 depicts an exemplary process flow according to an exemplary embodiment
of
the present invention. Beginning at 101, an assay or panel of assays can be
conducted
on a biological sample, e.g., blood, urine, etc., which has been taken from an
individual. Such individual can simply be an individual or he or she can be a
member
of a population or sub-population whose immunologic informatics are of use to
some
entity or enterprise. For example, the individual could be an insured of a
health
insurance company that is using the techniques of the present invention to
efficiently
manage the healthcare of its insureds so as to minimize costs. Or,
alternatively, such
an individual could be an immigrant whose vaccination history is unknown but
whose
immune status is of interest to his new country's immigration service. Such
exemplary embodiments are described more fully below in Section 111.
In Fig. 1, at 102 the results of the assay or assays conducted at 101 can be
obtained,
and at 103 there can be an optional step of analyzing the assay results
locally. In
exemplary embodiments of the present invention assays can be conducted and
read in
-18-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
a variety of assay reading devices. There are many assays available using
known
techniques. Some of them are more sophisticated and some less sophisticated.
In
exemplary embodiments of the present invention, an assay reading device can,
for
example, obtain results at 102, store those results and analyze them locally,
for
example, in a processor communicably connected to the assay reading device.
Alternatively, if only raw assay results are obtained from a less
sophisticated
technology, those results can, for example, be sent over a data network and
stored in a
database record. This is illustrated at 104. At 105, the results can be
analyzed by
accessing the particular record associated with the particular individual to
whom the
assay panel or panels were administered at a given time. Such analysis can
involve a
variety of algorithms ranging from a simplistic look at quantity of antibodies
per
defined unit of blood or other bodily fluid, or it can also, for example,
include a
complex analysis where a variety of assay results are input and combined in
linear and
non-linear ways to produce some metric of immunologic significance. Such
algorithms are described more fully below in Section II. Finally, at 106,
based on the
results of the above described analysis, recommendations can be generated.
Such
recommendations can include, for example, that the individual obtain one or
more
vaccines, that the individual be administered prophylactic therapies to boost
his or her
immune system, or that the individual be administered gene therapy to correct
the
genetic defect which places him or her at risk of communicating a certain
disease or
condition, to name a few.
In general, in many exemplary embodiments according to the present invention
process flow will be equivalent to or substantially similar to the process
flow depicted
in Fig. 1. In each of those exemplary embodiments, one or more panels of
assays can
be conducted with respect to one or more individuals. Results can be obtained,
stored
-19-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
and analyzed, and based on such analysis, recommendations for action (or
inaction,
such as, for example, in cases of over-vaccination, as described above) can be
recommended.
Fig. 2 is an exemplary generalized system diagram which correlates to the
generalized
method depicted in Fig. 1. With reference to Fig. 2, there can be seen a
number of
assay devices 201. These assay devices include one or more assay panels which
have
been conducted with respect to an individual or individuals and for which
results have
been obtained. The results obtained from the assay devices can, as described
in
connection with the generalized method in Fig. 1, be locally analyzed at each
assay
device, provided that such assay device has a data processor and memory and
the
results can be stored locally at the assay device. Alternatively, the assay
device
results can, for example, be communicated over a data network 202 to a central
processor 204 and stored in a central database 203. The central processor 204
can
access the records which it has received and analyze them by implementing a
number
of analytic algorithms as described more fully below.
Central processor 204, based on its analysis, can generate recommendations
based on
decision trees and criteria embedded in the various analytic algorithms it
implements.
These recommendations can be displayed locally at the central processor at
display
205 and can there be printed in a tangible medium for distribution to
interested
persons. Alternatively, the central processor 204 can, for example, send the
results of
its analysis over a data network to various users who can access the results
at user
terminals 210.
Fig. 2A presents an alternative generalized system diagram similar to Fig. 2.
However, as can be seen in Fig. 2A, there is an additional database, the
business rules
-20-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
database 220, communicably connected to central processor 204. In such an
exemplary system the central processor can implement algorithms to operate on
stored assay data which can, for example, also take as inputs various business
rules in
generating a decision regarding a recommendation. For example, as described
more
fully below in Section III, an exemplary embodiment of the present invention
can be
utilized to help a health insurance underwriter manage its population of
insureds.
There can, for example, be an annual or semi-annual requirement of all
insureds to
have assays for various immunological components conducted on their blood or
other
bodily fluids. After analysis of the results of such assays, an insurance
company can
determine whether a particular insured is susceptible to one or more given
diseases or
other ailments which would result in increased expenditures for medical
treatment.
The insurance company could then decide if it was not more economical to
require the
insured to undergo certain prophylactic treatments, such as, for example,
vaccines or
immune system boosting therapies, etc., where the cost of such prophylactic
therapies
is less than, as determined by some user determined factor, the expected
exposure for
medical care if the insured contracts one or more of the diseases or ailments
to which
he or she is susceptible.
In such context, there would need to be a number of business rules where such
user
defined quantities, threshold levels, cost functions or metrics, figures of
merit,
expected risks, etc., can be input and articulated or incorporated in a number
of rules.
Such rules can then be taken into account by the central processor in
implementing
algorithms which take as inputs data from business rules database 220 as well
as a
primary ImmunoScore database 203.
-21-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Fig. 2B presents an alternative generalized system diagram similar to Figs. 2A
and
2B. However, as can be seen in Fig. 2B, there is shown yet another additional
database, a hypothesis and rules database 250, communicably connected to
central
processor 204. In such an exemplary system a central processor can, for
example,
implement data mining algorithms to operate on stored immunologic and
background
data to find a set of correlations. Such data mining algorithms can for
example, be
used to corroborate known or expected relationships, such as, for example, a
correlation in antibody levels for measles, mumps and rubella in persons born
in the
United States after 1960, where the three vaccines were given simultaneously.
In fact,
an interesting follow-up would be to track if the rates of antibody levels for
each of
these three diseases change in the individual at a similar or a different
rate, and if
different, determine why.
Alternatively, for example, such data mining algorithms can be used to find
counter-
intuitive, or generally unknowns connections between variables or fields in
the
database.
In either case, once a set of correlations is obtained, intelligence in an
exemplary
system can be used to automatically generate a set of hypotheses to explain
such
correlations (or, if known, any follow-up data related thereto, as described
above) and
proceed to test the viability of each hypothesis using the data in the
database. Or,
alternatively, such intelligence can inform a user that additional data is
needed to vet a
hypothesis.
This process is explained more fully in Section II below.
Further, using such correlations, an exemplary system can, for example, also
take as
inputs various business rules in generating a decision regarding a
recommendation.
-22-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
For example, as described more fully below in Section III, an exemplary
embodiment
of the present invention can be utilized to help a health insurance
underwriter manage
its population of insureds. There can, for example, be an annual or semi-
annual
requirement of all insureds to have assays for various immunological
components
conducted on their blood or other bodily fluids. After analysis of the results
of such
assays, an insurance company can determine whether a particular insured is
susceptible to one or more given diseases or other ailments which would result
in
increased expenditures for medical treatment. The insurance company could then
decide if it was more economical to require the insured to undergo certain
prophylactic treatments, such as, for example, vaccines or immune system
boosting
therapies, etc., where the cost of such prophylactic therapies is less than,
as
determined by some user determined factor, the expected exposure for medical
care if
the insured contracts one or more of the diseases or ailments to which he or
she is
susceptible.
In such context, there would need to be a number of business rules where such
user
defined quantities, threshold levels, cost functions or metrics, figures of
merit,
expected risks, etc., can be input and articulated or incorporated in a number
of rules.
Such rules can then be taken into account by the central processor in
implementing
algorithms which take as inputs data from business rules database 220 as well
as a
primary ImmunoScore database 203.
Given the generalized exemplary method of Fig. I and the generalized exemplary
systems of Figs. 2, 2A and 2B, what is next described are a number of
exemplary
assay panels which can be administered to an individual or members of a
population
according to exemplary embodiments of the present invention. The scientific
-23-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
background behind the various exemplary assay panel, as well as which segments
of
the general population such panels are best administered to, are also
described.
EXEMPLARY ASSAY PANELS
The present invention is, inter alia, concerned with assessing the "protective
immune
status" or "immunologic status" of an individual or population. A "protective
immune status" is understood to be represented by an array of detectable
components
(phenotypic and/or genotypic) of an immune system (adaptive and/or innate)
that
comprise its protective capacity against harmful substances and/or cells (such
as, for
example, microorganisms or cancer). Such components can, for example, consist
of
genes as well as gene products. Genes can include, for example, those which
encode
immunologic receptors (such as, for example, toll-like receptors ("TLR"s) and
chemoattractant receptors) as well as effector molecules (such as, for
example,
cytokines and chemokines) which may also, for example, exist as genetic
polymorphisms capable of deleterious and/or beneficial effects. Gene products
can
include, for example, antibodies, complements, cytokines, chemokines,
chemoattractant receptors, TLRs, lectins, and other immune-related ligands.
Harmful
substances can consist of, for example, chemicals and/or toxins originating
from the
environment, microorganisms, or one's self.
Once diagnostic information is acquired from an individual regarding his or
her
immune status, this information can be, for example, added to a system
database.
Such a database can contain, for example, not only the results of ImmunoScore
diagnostic testing but a wide variety of demographic data and patient history
information as well. Such a system database can, for example, be used to
record
adverse events occurring coincident with immunizations. Such information can
be
-24-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
invaluable to, for example, the ACIP for making recommendations regarding
immunization scheduling, as well as help discover unsuspected patterns and
correlations relevant to immune status and immune response.
ImmunoScore diagnostic testing can be, for example, tailored to meet an
individual's
specific immunization status needs. In addition, each individual can, for
example,
receive their own personal ImmunoScore card that they could carry with them to
health care office visits, and the database information can be easily
transferable in the
ever-increasingly likely event that they change physicians or other primary
health care
providers. Additionally, ImmunoScore data, analysis of such data and relevant
database information can, for example, be stored as part of a person's
totality of
health information and medical records, in electronic formats such as, for
example,
entries in electronic health information databases, or computer chips embedded
in, for
example, "smart" cards or "smart driver's licenses."
For economy of description, most of the references cited herein are provided
in full
citation in Appendix A to the Immunologic Informatics Patent. Throughout the
text
citations are made to author and year of publication alone.
One component of ImmunoScore data can be, for example, the raw as well as
processed results of diagnostic tests or assays relating to immune status, as
described
below. ImmunoScore diagnostic testing is envisioned to be done on a small
assay
device or testing instrument that can be located, for example, in a doctor's
office. The
testing can be done, for example, with a sample of an individual's whole
blood,
plasma, serum, saliva, milk, semen, tears, or urine. In the case of blood, for
example,
the sample can be obtained by a finger prick, heel stick, ear stick, other
skin prick,
capillary draw, venous draw, or an arterial draw. The instrument can, for
example,
-25-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
take assay panels and the patient sample. Patient information can also be
input. The
resulting information can be, for example, displayed to a user, printed,
stored in a
removal medium, stored in the instrument, and/or transmitted (wired or
wireless) to
other devices such as via an intranet, a VPN or the Internet, for example.
Numerous systems and methods have been developed for the detection and
quantitation of analytes of interest in biochemical and biological substances
that can
be used, for example, in such an instrument. Such methods and systems which
are
capable of measuring trace amounts of microorganisms, pharmaceuticals,
hormones,
viruses, antibodies, nucleic acids and other proteins can be of great value to
researchers and clinicians.
A substantial body of art has been developed based upon well known binding
reactions, such as, for example, antigen-antibody reactions, nucleic acid
hybridization
techniques, and protein-ligand systems. The high degree of specificity in many
biochemical and biological binding systems has led to many assay methods and
systems of value in research and diagnostics. Typically, the existence of an
analyte of
interest is indicated by the presence or absence of an observable "label"
attached to
one or more of the binding materials. Of particular interest are labels which
can be
made to luminesce through photochemical, chemical, and/or electrochemical
means.
"Photoluminescence" is the process whereby a material is induced to luminesce
when
it absorbs electromagnetic radiation. Fluorescence and phosphorescence are
types of
photoluminescence. "Chemiluminescent" processes entail the creation of
luminescent
species by chemical transfer of energy. "Electrochemiluminescence" entails
creation
of luminescent species electrochemically.
-26-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Electrochemiluminescent (ECL) assay techniques are an improvement over
chemiluminescent techniques. They can, for example, provide a sensitive and
precise
measurement of the presence and concentration of an analyte of interest. In
such
techniques, the incubated sample is exposed to a voltammetric working
electrode in
order to trigger luminescence. In the proper chemical environment, such
electrochemiluminescence is triggered by a voltage impressed on the working
electrode at a particular time and in a particular manner. The light produced
by the
label is measured and indicates the presence or quantity of the analyte. For a
fuller
description of such ECL techniques, exemplary reference is made to US patents
5,221,605; 5,705,402; 6,140,138; 6,325,973; and 6,451,225. The disclosures of
the
aforesaid patents are hereby incorporated herein by reference.
Amplification techniques for nucleic acids may be combined with the above
assay
techniques. For example, US patent 6,048,687 discloses how NASBA can be
combined with an ECL technique; and US patent 6,174,709 discloses how PCR can
be combined with an ECL technique. The disclosures of the aforesaid patents
are also
hereby incorporated herein by reference.
An assay instrument can, for example, be, or be similar to, the BioVeris
Corporation
M 1 R or M 1 M instruments with an added sample processing front end [ Roche
products] . Aspects of these instruments are disclosed in pending US patent
application numbers 10/600,165 and 10/841,569, each under common assignment
herewith. The disclosures of these patent applications are hereby incorporated
herein
by reference.
In exemplary embodiments of the present invention, an assay instrument can
include,
for example, amplification techniques such as PCR or NASBA. In exemplary
-27-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
embodiments of the present invention, the instrument can use fluorescence,
chemiluminescence, or ECL assay techniques. In exemplary embodiments, multiple
measurements can be done simultaneously; in other exemplary embodiments of the
present invention, multiple measurements can be done sequentially. In
exemplary
embodiments of the present invention, an assay instrument can, for example,
contain
self-test and/or self-calibration components.
In exemplary embodiments of the present invention, a sample can be added to an
assay panel, and the combination then inserted into the test instrument, as
shown in
Fig. 3. In alternate exemplary embodiments, the sample and assay panel can be
separately inserted into the test instrument, as shown, for example, in Fig.
4.
As described below, entries to an exemplary master ImmunoScore database can
be,
for example, coded so as to protect patient confidentiality. A patient could,
however,
be able to learn from their physician in real time, for example, which
vaccines he or
she might need to ensure protection from vaccine-preventable illnesses. The
physician can, for example, offer the vaccine, or other therapy, during the
same visit,
or shortly thereafter. Any possible adverse effects from any delivered
vaccinations
could be subsequently entered into an ImmunoScore database and that
information
could be shared with the ACIP or other agencies or bodies, as described more
fully
below.
The actual assays can be performed, for example, based upon the needs of the
individual or individuals being examined. Age, occupation, travel plans,
immigration
status, military status, and previous health status can all be considered
prior to
initiation of ImmunoScore diagnostic analyses in exemplary embodiments. In
-28-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
exemplary embodiments of the present invention, the following exemplary broad
categories can, for example, be utilized as focal points for test panels:
1. Entry to primary school.
2. College entry.
3. Age 19-49 years.
4. Age 50-64 years.
5. Age > 65 years.
6. Health-care professionals.
7. Military personnel:
recruits and officer accessions;
alert forces;
individualized according to occupational or personal needs; and
veterans.
8. Travelers.
9. Immigrants.
10. Individuals with identifiable health risks (not necessarily
exclusively):
a. Complement-deficient individuals (e.g. meningococcal
disease susceptibility);
b. Genetically identified (e.g. HLA haplotype, sepsis
susceptibility) disease-susceptible individuals;
c. Mannose-binding lectin-deficient individuals;
d. Hepatitis B vaccine poor/non-responders; and
e. Ethnic groups and others known to respond poorly to
polysaccharide, conjugate, or other vaccines.
A. COLLEGE STUDENT DIAGNOSTIC PANELS
1. Meningococcal Diagnostic Panel
-29-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
In exemplary embodiments of the present invention, the following tests can be
included in a meningococcal diagnostic panel:
1. Antibody (Ig) to (4 tests):
Group A Meningococcal Polysaccharide (GAMP)
Group C Meningococcal Polysaccharide (GCMP)
Group Y Meningococcal Polysaccharide (GYMP)
Group W-135 Meningococcal Polysaccharide (GWMP)
2. Antibody (IgM) to Group B Meningococcal Polysaccharide (GBMP) (1 test)
3. Serum levels of complement components (7 tests):
C5
C6
C7
C8
C9
Properdin
MBL
4. Measurement of genetic polymorphisms (5 tests):
FcyRIIa receptor
IL-1
ILA R
IL-6
IL-10
2. Sexually Transmitted Diseases Assay Panel
In exemplary embodiments of the present invention, the following tests can,
for
example, be used for ImmunoScore measurement of immunity to STDs:
-30-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Antibodies to Chlamydia - IgG, IgA, and IgM (3)
Antibodies to HSV - IgG to HSV-1 and HSV-2 (2)
DNA analyses of HPV types - particular emphasis on high-risk
Antibody to N. gonorrhoeae (1)
Antibody to T. pallidum (1)
T-cell related response to T. pallidum
Antibody to HIV
T-cell related response to HIV
Antibodies to GBS serotypes (at least 3)
Measurement of Thl/Th2 cytokines (many as current evolving
definitions)
3. Persistent Immunity Induced by Childhood Vaccines
In exemplary embodiments according to the present invention, the following
tests for
measurement of immunity to childhood vaccines can be included in an exemplary
ImmunoScore panel directed to college students, or in other exemplary
embodiments,
to adults in general:
Antibody to HBs (1)
Antibody to diphtheria toxin (1)
Antibody to tetanus toxin (1)
Pertusis antibodies (4):
Antibody to pertussis toxin (PT)
Antibody to pertactin (PRN)
Antibody to filamentous hemagglutinin (FHA)
Antibody to fimbriae
Antibody to PRP (Hib) (1)
Antibodies to poliovirus serotypes P1, P2, and P3 (3)
Antibody to measles (1)
Antibody to mumps (1)
Antibody to rubella (1)
Antibody to varicella (1)
Antibody to pneumococcal serotypes (7)
B. ADULT DIAGNOSTIC PANELS
1. Measurement of Immunity Induced By Vaccines for Military Personnel
In exemplary embodiments of the present invention military personnel can be
administered the following diagnostic panels:
1. College Student ImmunoScore Panels consisting of:
= Meningococcal Diagnostic Panel;
-31-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= Sexually Transmitted Disease Diagnostic Panel;
= Persistent Immunity Induced by Childhood Vaccine Diagnostic Panel; and
as described above; and. in addition
2. Military personnel can have specific vaccination needs as outlined in Table
3
below depending on their assignments and type of deployment. Specific
branches of the service may also have specific vaccination needs and
permutations of the basic diagnostic panels. Thus, in exemplary
embodiments, military personnel can be administered one or more of the
following tests:
TABLE 3
Vaccine Diagnostic Panels Exclusive to the Military:
Vaccine Antibody Marker
Adenovirus 4 & 7 Neutralizing antibody
Anthrax PA
Cholera LPS IgG
Plague Fraction I Capsular Antigen
Smallpox Neutralizing antibody
Lyme disease OspA
2. ImmunoScore Measurement of Vaccine-Induced Immunity for Travelers
In exemplary embodiments of the present invention, an ImmunoScore traveler's
assay
panel can, for example, include the following:
Antibody to HAV (1)
Antibody to HBs (1)
Antibody to Japanese Encephalitis (1)
Antibody to rabies (1)
other rabies related cytokine assays (as necessary)
Antibody to Typhoid fever (1)
Antibody to yellow fever (1)
Antibody to diphtheria toxin (1)
Antibody to tetanus toxin (1)
Pertusis antibodies (4):
Antibody to pertussis toxin (PT)
Antibody to pertactin (PRN)
Antibody to filamentous hemagglutinin (FHA)
Antibody to fimbriae
Antibodies to poliovirus serotypes P1, P2, and P3 (3)
-32-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Antibody to measles (1)
Antibody to mumps (1)
Antibody to rubella (1)
3. Cytokine Measurement in ImmunoScore
Introduction
An individual's immune system functions as an informational system that is
shaped
during that person's life after exposure to pathogens. Immune interventions,
such as
vaccines, that manipulate the "knowledge" of the immune system are among the
most
cost effective in modem medicine. Currently, globally immunotherapy for non-
communicable diseases is not showing the same success achieved in fighting
infection. Despite considerable experimental advances in understanding immune
tolerance, autoimmune diseases continue to be treated by non-specific
immunosuppression. The substantial experimental data generated with animal
models
remain limited in their capacity to allow predictions and guide clinical
interventions
(Lage, 2008).
The immune system should be considered a complex network, given that it
consists of
more than 200 cytokines and chemokines and contains millions of lymphocyte
clones
and its macroscopic activity is dictated by the interactions of all these
components.
How complexity influences immunology is demonstrated by the almost universal
failure to predict the outcome of gene-inactivation experiments, the absence
of
effective vaccines for malaria and other parasites, tuberculosis or HIV, and
the
context dependent effects of some immunotherapy interventions (which induce
either
tolerance or immunity).
Over recent decades the immune system has been subject to a great deal of
investigation. Growing complexity has often been a major byproduct of the
-33-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
discoveries reported, and subsequently models such as the Thl/Th2 paradigm,
were
developed to cope with such complexity. Regarding autoimmune diseases,
verifying
and expanding such models is desirable, because it has proven difficult to
extrapolate
findings to existing models that were often developed in different contexts
(Delaleu,
et at. 2008). Recent technological advances have greatly increased the amount
of
information and the number of proteins that can be investigated in any given
system
and put into a scientific context simultaneously.
By studying the immune system through the application of reductioninst
principles, its
mediators have been thoroughly analyzed over recent decades. This has yielded
tremendous scientific advances. However, studying the properties of the immune
system's isolated components is limited in terms of elucidating how system
properties
emerge, because they may strongly rely on and arise from interactions between
numerous system components. The complexity of the immune system should not
paralyze immunology research. The realization that the immune system is a
complex
network has led to wider use of mathematical models for simulating its
activity and
testing hypotheses in silico. ImmunoScore technology represents a novel way to
analyze the implications of multiple molecules in a specific condition and
provide
insight into the inter-relationships that define a specific immune system
status.
Cytokines are a large and diverse group of plasma-membrane associated or
secreted
proteins that bind cell-surface receptors and thereby regulate many important
biological processes. These processes include development, hematopoesis,
inflammation, immune responses, and tissue repair. Whether in a healthy
individual
or in an acute or chronic disease situation, cytokines act in concert rather
than in
isolation, and no single cytokine in a cross-sectional model is adequate to
serve as an
-34-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
absolute screening marker. It is essential to understand the regulation of
cytokine
production in healthy individuals as well as individuals with distinct disease
states.
The application of ImmunoScore technology to cytokine analyses will help to
establish the viability and merits of a multi-marker approach for clinical
risk
stratification. ImmunoScore technology will examine expressed levels of
cytokines
as serum markers, as correlation between mRNA levels and protein expression
has
previously been demonstrated to be poor in a model of autoimmune disease (Hu,
et al.
2007).
Introduction: T-helper Cell Subsets
Uncommitted CD4+ T helper cells can be induced to differentiate towards T
helper I
(ThI), Th2, Th17, and regulatory (Treg) phenotypes according to the local
cytokine
milieu (Figure 1). Thl cells secrete (among others) IFN-y and TNF-a, which
allow
these cells to be particularly effective in protecting against intracellular
infections by
viruses and bacteria that grow in macrophages, as well as eliminating
cancerous cells
(Kidd, 2003). Th2 cells secrete IL-4, IL-5, IL-10, and IL-13 which upregulate
antibody production and target parasitic organisms. Th2 cells activate B
cells, which
are adapted for defense against parasites that are vulnerable to IL-4 switched
IgE
production, IL-5 induced eosinophilia, and IL-3 and IL-4 stimulated mast cell
proliferation and degranulation (Kaiko, et al. 2007). Th17 cells secrete IL-
17, IL-17F,
IL-6, IL-22, and TNF-a and appear to play a role in both tissue inflammation
and
activation of neutrophils to combat extracellular bacteria. Treg cells secrete
IL-I 0
and TGF-(3, which modify helper T cell activity and suppress some of their
functions,
inducing tolerance to antigens.
Anomalous T cell responses bolster a range of diseases, including asthma,
allergy,
and autoimmune disease. Fundamental immune elements of these diseases are the
-35-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
development of antigen-specific T-helper cells. Th 1, Th2, and Th17 cells are
associated with the clinical features and disease progression. The phenotypes
of these
polarized T cells that differentiate from naive precursors is determined by
the
complex interaction of antigen presenting cells with naive T cells and
involves myriad
factors, including the dominant cytokine environment, co-stimulatory
molecules, the
type and amount of antigen presented, and a wide variety of signaling
cascades. The
decision to take the immune response in a certain direction is not made by one
signal
alone, but rather through many different elements acting synergistically,
antagonistically, and through positive and negative feedback loops to activate
a Th 1,
Th2, or Th17 immune response, or combination thereof (Kaiko, et al. 2007).
Cytokines are the most influential factors that modulate T cell phenotype, and
their
mechanism of action involves intracellular signals transmitted through
cytokine
receptors expressed on the surface of T cells. In essence, any cell that
differentially
secretes or consumes key cytokines can regulate the function of other effector
cells
that are activated in close proximity (Sojka, et al. 2008).
Evolution of Th1/Th2 Paradigm to Include Th17 and Treg Cells
The initial concept of the Thl/Th2 paradigm is depicted in Figure 2, where the
T
helper cell immune response was balanced on opposite sides of a teeter-totter.
Cytokines produced during one type of response were imagined to be counter-
productive to the other type of response in this model, expression of the Th 1
response
would cause a dampening of the Th2 response, and vice versa. Chronic over-
expression of either type response would be undesirable to the individual,
with a
chronic Th I response seen to cause autoimmunity and graft rejection, among
others,
-36-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
and a chronic over-expression of the Th2 response to be the cause of atopic
diseases
and allergies.
With the discovery of Th17 and the re-discovery of Treg cells, it became
apparent that
the teeter totter model depicted in Figure 2 was too simplistic, and the newer
models
now include the Th 17 and Treg arms to accommodate these cell types (Figure
3). In
the recent past, inflammatory responses were assigned as an over-expression of
Th 1
cells, while at the same time allowing that inflammation could occur in the
absence of
the signature cytokine of Thl cells, IFN-y. Now, the Th17 cell pool is seen as
having
a significant contribution to inflammation. Th17 responses in the presence of
a ThI
response can presumably lead to autoimmune disease, while Th17 responses in
the
presence of a Th2 response can lead to allergic or atopic disease. Treg cells
are
envisioned as the cells that dampen the immune response to avoid autoimmune
and
allergic reactions, however, over-expression of the Treg cell population is
also not
desirable as this can lead to chronic infection, or more strikingly, acute,
fatal
infection.
Thl Cytokine Signals
Th1 cell development begins with the secretion of IL-12 and type I IFNs (IFN-a
and
IFN-P). These cytokines are released by macrophages and dendritic cells (DCs)
upon
activation by intracellular pathogens (Farrar, et al. 2002). IL-12 induces the
production of IFN-y from the Thl cells, which then acts in an autocrine manner
to
generate a positive feedback loop, producing more IL-12. IFN-y acts as an
inhibitor
of the Th2 pathway by preventing Th2 cell proliferation. Once the IL-12
receptor is
expressed, IL-12 is then able to bind its receptor and further reinforce the
differentiation of Thl cells. IL-12 signaling activates the transcription
factors STAT-
3, STAT-4 and nuclear factor-KB to promote the production of cytokines
associated
-37-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
with the Th 1 phenotype (Kaiko, et al. 2007). The IFN-y secreted by Th 1 cells
as they
develop stimulates surrounding naive Th cells to begin polarization into more
Th I
cells, in a self-renewing paracrine loop (Kidd, 2003). Other proposed Th 1
polarizing
factors include IL-27, and the intercellular adhesion molecule-1 (ICAM-1)
binding its
receptor (Salomon and Bluestone, 1998).
Thl effector cell signature:
Induced by: IL-12
Produce: IFN-y, TNF-a, IL-2
Suppressed by: IL-10, TGF-(3, IL-23
Th2 Cytokine Signals
The production of Th2 effector cells primarily involves the action of
cytokines IL-4,
IL-6, IL-l0, and ILA 1. IL-4 induces the production of STAT-6 in naive T
cells,
which in turn activates the expression of the zinc finger transcription factor
GATA-3
(Ouyang, et al. 1998). GATA-3 augments promoter activity or reverses chromatin
structure based suppression of regions that are responsible for controlling
Th2
cytokine gene expression. This results in the release of cytokines
characteristic of the
Th2 phenotype: IL-4, IL-5, IL-9, IL-10, and IL-13. Another result is the
inhibition of
expression of IL-12 receptor and therefore Thl development (Farrar, et al.
2002). As
Th2 cells mature, they produce increasing levels of IL-4, which generates a
paracrine
loop and induces neighboring naive T cells to develop to Th2 cells (Kidd,
2003). IL-6
is also released early in Th2 cell development, and up-regulates IL-4 and
inhibits
STAT-1 phosphorylation, thereby preventing IFN-y synthesis (Dodge, et al.
2003).
IL-6 also plays an integral role in Th17 differentiation. IL-1I released by
myeloid
cells acts directly on T cells to stimulate IL-4 and IL-5 synthesis and also
to inhibit
IFN-y production.
-38-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The induction of mast cell degranulation and the release of histamine have
been
demonstrated to polarize the function of DCs and Th cells towards a Th2
phenotype
(Mazzoni, et al. 2006). Degranulation reduced the capacity of DCs to induce
ThI
cells and instead promoted the development of increased numbers of IL-4
secreting T
cells. This indicates that mast cells may have a critical function in the
development of
the antigen specific Th2 cell phenotype in mast cell-mediated diseases, such
as
asthma.
It seems likely that the inducible co-stimulator (ICOS) is capable of co-
stimulating
distinct effector functions, depending on the density of surface expression
and tissue
localization of the immune response. There appears to be a relationship
between
ICOS cell-surface density and the type of cytokines produced (Kaiko, et al.
2007).
There is a strong association between intermediate expression of ICOS and
secretion
of Th2 cytokines, and high levels of ICOS expression and release of the
regulatory
cytokine IL-10 (Lohning, et al. 2003).
Th2 effector cell signature:
Induced by: IL-4
Produce: IL-4, IL-5, IL-13, IL-10
Suppressed by: IL-10, TGF-(3
Th17 Cytokine Signals
Th 17 cells represent a subset of CD4+ cells that is both distinct from and
antagonized
by cells of the Thl and Th2 lineages. Although found throughout the body, Th l
7
cells are predominantly found in the lung and digestive mucosa suggesting a
homeostatic role in those tissues (Kryczek, et al. 2007). The generation of Th
17 cells
is inhibited by IL-4 and IFN-y potentially by down-regulation of the IL-23
receptor
(Harrington, et al. 2005). IL-23 appears to be essential for the production of
a robust
Th 17 response, but is not responsible for the initial induction of the Th 17
phenotype.
-39-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Rather, Th 17 cells appear to be induced by a combination of IL-6 and TGF-(3.
The
combination of these cytokines induces the predominant generation of Th 17
cells with
minimal numbers of Tregs in a mutually exclusive pattern (Veldhoen, et al.
2006;
Bettelli, et at. 2006). As TGF-(3 is involved in the development of both Tregs
and
Th 17 cells, which may occur through the inhibition of IL-4- and IFN-y-
dependent
pathways, it appears that IL-6, a known inhibitor of Treg development, plays
an
integral role in switching between these inflammatory and suppressive cell
types
(Kaiko, 2007). Neutralizing IL-17 in cultures of Th17 cells alters the balance
in favor
of Tregs, suggesting an important inhibitory action of IL-17 on Treg cells
(Nardelli, et
al. 2004).
Th17 effector cells signature:
Induced by: TGF-0, IL-6
Produce: IL-17, IL-21, IL-22
Maintained by: IL-23
Suppressed by: IL-4, IFN-y, IL-2, IFN-a
Treg Cytokine Signals
Every adaptive immune response involves recruitment and activation of not only
effector T and B cells but also Tregs, and that the balance between the two
populations is critical for the proper control of the quality and magnitude of
adaptive
immune responses and for establishing or breaching tolerance to self- and non-
self
antigens (Sakaguchi, et al, 2008). The exact mechanism by which the Tregs
exert
their effect is currently unknown, although it is believed that their
suppressive
function may be contact-dependent (Afzali, et al. 2007). Other studies show an
important role for TGF-0 and IL-10 production as mediators of Treg activity
that is
contact-independent (Dieckmann, et al. 2002; Longhi, et al. 2006). Both TGF-(3
and
IL-2 are important for the development of Tregs (Afzali, et al. 2007; Malek
and
Bayer, 2004). Foxp3 expression as a complex leads to Treg cell-mediated
-40-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
suppression in a cell-cell contact-dependent or -independent manner (Li and
Greene,
2008). In humans, disruption of Foxp3 function leads to an immune dysfunction,
polyendocrinopathy, enteropathy, X-linked (IPEX) syndrome characterized by
autoimmune disease, allergy, and inflammatory bowel disease (Bennett, et al.
2001).
In the absence of cell-cell contact, Tregs can suppress T cell activity by
either directly
secreting IL-10, TGF-J3, and IL-35, or competing for cytokines via receptors
that
contain the common y-chain, which binds to IL-2, IL-4, and IL-7 (Sojka, et al.
2008).
Given the diverse array of suppressive mechanisms, Treg activity needs to be
attenuated to mount effective immune responses to infection. Treg function can
be
modulated by a variety of pro-inflammatory signals including Toll-like
receptor
triggering and direct inhibition by tumor necrosis factor-a (TNF-a) (Liu and
Zhao,
2007; Valencia, et al. 2006). The up-regulation of B7 expression (CD80 and
CD86)
by antigen-presenting cells represents a central event in the activation of
naive T cells
and may serve as a mechanism to disrupt regulatory T cell tolerance by
rendering
effector T cells unresponsive to suppression (Sojka, et al. 2008). B cells
with their
lower expression of CD80/CD86 appear to be more efficient antigen-presenting
cells
than dendritic cells for inducing effector Treg cells (Benson, et al. 2007).
Treg cell signature:
Induced by: TGF-(3, IL-2
Produce: IL-10, TGF-0, IL-35
Suppressed by: IL-6, IL-17, IL-31
Dendritic Cells
Dendritic cells are recognized as one of the most important cell types for
initiating the
priming of naive CD4+ helper T (Th) cells and for inducing CD8+ cell
differentiation
into killer cells (Banchereau, et al. 2000). Immature dendritic cells are
found at
strategic anatomical sites throughout the body, thereby allowing them to
respond
rapidly to microbial invasion (Pashine, et al. 2005). Activation of lymphoid
dendritic
-41-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
cells, because of their preponderance to secrete IL-12, may be important for
priming
Th1-like responses, while early activation of myeloid dendritic cells may lead
to Th2-
like responses (Pulendran, 2004). It has also been suggested that the
production of II-
6 by dendritic cells may be reponsible for inhibiting the suppressor activity
of Treg
cells (Pasare and Medzhitov, 2004). This production of IL-6 might also
reasonably be
assumed to enable the activation of Th 17 cells.
B cells and success of vaccinations
The most dramatic health problem of the aged immune system is the increaing
rates of
morbidity and mortality from recurrent and invasive infections of the
respiratory tract
caused by encapsulated bacteria such as Streptococcus pneumoniae (Sankilampi,
et al.
1997). It has been reported that increased susceptibility to secondary
pneumococcal
pneumonia is at least in part caused by excessive I1-10 production and reduced
neutrophil function in the lung (van der Sluijs, et al. 2004).Two populations
of B cells
have been identified in human peripheral blood: mature and memory B cells. IgD-
CD27+ memory B cells can produce IgG, IgM, and IgA, while IgD+CD27+ IgM
memory B cells predominantly produce IgM (Shi, et al. 2003).The presence of
IgM
memory B cells in the blood correlates with protection from pnuemococcal
infection
(Kruetzmann, et al. 2003). Natural antibodies make up most of the IgM in the
serum
and have the function to limit the growth and dissemination of pathogens
during the
early phases of infection and potentiate the immune response (Ochsenbein, et
al.
1999). Physiological and transient disposition to pneumococcal infection of
young
children (under 2 years of age) is associated with the lack of circulating IgM
memory
cells and of serum anti-polysaccharide IgM (Kruetzmann, et al. 2003). Decline
of
splenic functions may reflect diminished numbers of aged IgM memory B cells.
Effectiveness of pneumococcal polysaccharide vaccine in older adults on
protection
-42-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
against pneumococcal infections may be associated with the increase and
activation of
circulating IgM memory B cells, resulting in rapid synthesis of anti-
polysaccharide
IgM antibodies (Shi, et al. 2005).
Regulatory B Cells
The maintenance of tolerance is an absolute requirement of a sophisticated
regulatory
apparatus to prevent or dampen overzealous immune responses. In addition to
the
ability of B cells to prime and activate the immune system, B cells with
regulatory
function (Bregs) have been identified in experimental models of autoimmunity,
infections, and cancer, supporting the notion that, similar to Tregs, Breg-
mediated
suppression is an important means for the maintenance of peripheral tolerance.
This
regulatory funntion appears to be directly mediated by the production of IL-10
and
TGF-(3 and by the ability of the B cells to interact with pathogenic T cells
to inhibit
harmful immune responses (Mauri and Ehrenstein, 2008). B cells are typically
characterized by their ability to produce antibodies. However, B cells possess
additional immune functions, including the production of cytokines and the
ability to
function as secondary antigen presenting cells. As with T cells, the B cell
population
contains functionally distinct subsets capable of performing both pathogenic
and
regulatory functions (Mizoguchi and Bhan, 2006). B cells can play a pathogenic
role
in acquired immune responses by producing autoantibodies that contribute to
the
development of autoimmune diseases (Murakami and Honjo, 1997; Korganow, et al.
1999;Fields, et al. 2003). The existence of an immunoregulatory B cell subset
that
plays a role in immune regulation resulting in complete recovery from
experimental
autoimmune encephalomyelitis was reported in a murine model of that disease
(Wolf,
et al. 1996).
-43-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
IL-10 from regulatory B cells can repress the production of IL-6 and IL-12 by
DCs,
thereby inhibiting the differentiation of Th 17 and Th 1 cells, respectively
(Lampropoulou, et al. 2008).
Like their T cell counterparts, B cells can be divided into functionally
distinct
regulatory subsets capable of inhibiting inflammatory responses and inducing
immune
tolerance (Mizoguchi and Bhan, 2006).
Role of Basophils
Basophils activated by IL-3 or antibody to FcERI induce B cell proliferation
and the
production of IgM and IgGl in the presence of activated CD4+ T cells; this B
cell
proliferation and immunoglobulin production requires IL-6, IL-4 and cell
contact
(Kawakami, 2008). Activated basophils enhance the humoral memory response by
secreting IL-6 and by altering the phenotype of CD4+ cells (that is, by
inducing CD4+
T cell up-regulation of IL-4, IL-5, IL-10, IL-13, and the transcription factor
GATA-3
and down-regulation of IFN-y and IL-2) (Denzel, et al. 2008).
Trauma and Cytokines
The immune system undergoes numerous changes after traumatic injuries,
including a
down-regulation of the Th l response and up-regulation of the Th2 response
(Miller, et
al. 2007). They Thl response is suppressed as illustrated by diminished IL-2,
IFN-y,
and IL-12 levels after major injury. The enhancement of the Th2 profile is
marked by
elevated IL-10 and IL-4. Certain cytokine profiles, ratios, and polymorphisms
may
help identify patients at increased risk of systemic inflammatory response
syndrome
(SIRS), sepsis, multiple organ failure (MOF), and deep venous thrombosis. Some
provocative indications for individuals more susceptible to complications
include
(Miller, et al. 2007):
= decreased IL-12
-44-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= increased IL-10
= increased sIL-2Ra
= increased IL-18
= IL-18 promoter genetic polymorphisms
= IL-6 : IL-l0 ratio
Identification of those Th l/Th2 cytokine profiles associated with worse
prognosis
may one day allow clinicians to risk stratify injured patients and identify
those at
increased risk of developing SIRS, sepsis, MOF, and deep venous thrombosis.
Stress and Cytokines
Recent evidence indicates that the major stress hormones, glucocorticoids and
catecholamines, systemically inhibit IL-12, TNF-a, and IFN-y, while
simultaneously
upregulating IL-10, IL-4, and TGF-P production indicating a generic ThI to Th2
shift
(Calcagni and Elenkov, 2006). However, in certain local responses and under
certain
conditions, stress hormones may actually facilitate inflammation through
induction of
IL-I, IL-6, IL-8, IL-18, TNF-a, and CRP production. Autoimmunity, chronic
infections, major depression, and atherosclerosis are characterized by a
dysregulation
of the pro/anti-inflammatory and Thl/Th2 cytokine balance (Calcagni and
Elenkov,
2006). These authors stated that conditions that are associated with
significant
changes in stress system activity, such as acute or chronic stress, cessation
of chronic
stress, pregnancy and the postpartum period, or rheumatoid arthritis through
modulation of the systemic or local pro-anti-inflammatory and Thl/Th2 cytokine
balance, may suppress or potentiate disease activity and/or progression.
Stress-
hormones induced inhibition or up-regulation of innate and Th cytokine
production
may represent an important mechanism by which stress affects disease
susceptibility,
activity, and outcome of various immune-related diseases.
-45-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Inflammatory Bowel Disease (IBD) and Cytokines
Traditional dogma has had different viewpoints of inflammatory bowel disease
and
cytokines: Crohn's disease (CD) has been thought to have a Thl motif, while
ulcerative colitis (UC) was thought to have given rise to a Th2 expression
(Mudter
and Neurath, 2007). CD has been associated with elevated expression of IFN-y,
TNF-
a, and IL-12. In UC, the pattern is less clear; there is a modified Th2
response
associated with cytokines such as IL-15 and IL-10 (Torres and Rios, 2008).
Other
publications have reported that the IL-17/IL-23 pathway may have a pivotal
role in
intestinal inflammation (Hue, et al. 2006; Kullberg, et al. 2006).
Atherosclerosis
Atherosclerosis historically was considered to be mainly a degenerative
disease, but it
is now well ascertained that its pathogenesis is inflammatory (Jawien, 2008).
Serum
levels of the IL-1 family of cytokines (including IL-18 and IL-33) have been
correlated with various aspects of cardiovascular disease and their outcomes
(Apostolakis, et al. 2008). IL-IRa, a natural antagonist of IL-1, possesses
anti-
inflammatory properties, mainly through the endogenous inhibition of IL-I
signaling
(Apostalakis, et al. 2008).
Oxidized low density lipoprotein (OxLDL) is not only pro-inflammatory and pro-
atherogenic, but several of the neoepitopes generated during oxidation are
highly
immunogenic and result in the generation of auto-antibodies. The overall
evidence
supports the notion that IgG auto-antibodies to OxLDL are associated with pro-
atherogenic properties, and that IgM auto-antibodies to OxLDL are associated
with
atheroprotective properties (Gounopoulos, et al. 2007). ImmunoScore would
track
trends in anti-OxLDL antibody levels in patients over time and add this
information to
-46-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
the ImmunoScore database. Similarly, antibody levels to Hsp 60 would be
tracked to
determine if these antibodies are beneficial or detrimental to patients.
One researcher proposed that in general, it is believed that the ThI response
and its
mediators: IFN-y, TNF-a, IL-1, IL-12, and IL-18 enhance atherogenesis, while a
Th2
response and its mediators: IL-4, IL-5, IL-10 and IL-13 inhibit the
development of
atherosclerosis (Jawien, 2008). Another group put the atherosclerosis profile
onto the
ThI7/Treg axis by stating that acute coronary syndrome was associated with an
increase in Th17 cytokines (IL-17, IL-6, and IL-23) and a decrease in Treg
cytokines
(TGF-(I and IL-10) (Cheng, et al. 2008). By measuring all the associated
cytokines
over time and various demographics, the ImmunoScore technology would be able
to
specifically enumerate and assign significance to these assays and relate the
results to
the individuals being tested.
Another study found that in patients with an inflammatory response, as
demonstrated
by elevated levels of IL-6 in serum, CMV seropositivity was a strong and
independent
predictor for cardiac death (Blankenberg, et al. 2001). High CMV antibody
titers may
be associated with a chronic inflammatory response resulting in increased IL-6
levels.
This in turn, can lead to an increase in CRP levels and a poor prognosis for
coronary
artery disease outcome. It has been shown that statins can reduce the
inflammatory
response. Periodic ImmunoScore measurements in a treated patient population
would
assist physicians regarding course and effectiveness of statin treatments, as
serum
CMV positivity without inflammatory response is not indicative of fatal
cardiovascular events (Blankenberg, et al. 2001).
Another group proposed classifying cytokines related to both atherosclerosis
and
diabetes in the following categories: "noxious" comprising IL-1, IL-2, IL-6,
IL-7, IL-
-47-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
8, IL-15, IL-17, and IL-18; "protective" comprising IL-4, IL-10, IL-11, IL-12,
and IL-
13; and "aloof' comprising IL-5, IL-9, IL-14, IL-16, and IL-19 through IL-29
(Fisman, et al. 2008).
Proposed ImmunoScore Atherosclerotic Disease Panel
Indicators for Poor Prognosis (Thl/Th17 axis)
= IL-6
= IL-12
= IL-18
= IL-33
= IFN-y
= TNF-a
= CRP
= Antibody to CMV
Indicators for Improved Prognosis (Th2/Treg axis)
= IL-10 (with exception of transplant patients)
= TGF-(3
= IL-5
Indeterminate Prognostic Value (Th2)
= IL-4
= Antibody to OxLDL
= Antibody to Hsp 60
Th17 autoimmune pathogenesis
Th17 may play an essential role against certain extracellular pathogens.
However,
Th17 cells with specificity for self-antigens are highly pathogenic and lead
to the
development of inflammation and severe autoimmunity. Interleukin (IL)-17 was
originally named cytotoxic T lymphocyte-associated antigen-8 (CTLA-8)
(Paradowska, et at. 2007). There are six members of the IL-17 protein family
(IL-
17A through IL-17F). The IL-17 family plays a key role in the regulation of
immune
-48-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
and inflammatory response, in the homeostasis of several tissues, and the
progression
of autoimmune diseases (Paradowska, et at. 2007).
IL-23 and IL-17 are associated with a number of human autoimmune disorders.
Th17
cells are likely to be highly pathogenic in rheumatoid arthritis (McGeachy and
Cua,
2008). IL-17+, CD4+ and CD8+ T cells have been identified in active lesions in
the
brain of multiple sclerosis patients (Tzartos, et al. 2008). Psoriasis has
also been
linked to inappropriate Th17 cell responses. IL-17, IL-23, and 11-22 are all
elevated in
psoriatic lesional skin (Lee, et al. 2004; Wilson, et al. 2007; Wolk, et at.
2004).
As the Thl and Th2 cell subsets cross-regulate the differentiation of the
other cell
type, they also appear to negatively regulate the differentiation of Thl7
cells
(McGeach and Cua, 2008). Addition of IL-2, IFN-y, or IL-4 to cultures inhibits
either
IL23- or TGF-(3 plus IL-6-stimulated differentiation of mouse and human Th17
cells
(Annunziato, et al. 2007; Harrington, et al. 2005; Murphy et al. 2003; Park,
et al.
2005; Wilson, et al. 2007).
Foxp3+ regulatory T cells (Treg) are necessary and sufficient to prevent
autoimmunity throughout the lifespan of an individual. TGF-(3 induces Foxp3 in
naive T cells, but TGF-0 and IL-6 together drive the generation of Thl7 cells
(Korn,
et al. 2007). A group found that in humans, IL-23 and IL-10 are able to drive
naive
CD4+ T cells toward the Th17 phenotype (Wilson, et al. 2007).
Th17 cells probably have a specific role in normal immune function through the
coordinated action of their effector cytokines and chemokines, similar to the
well
established functions of Thl and Th2 cells in regulating cellular immunity and
antibody production. The signature cytokine and chemokine profile of Th l 7
cells
-49-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
suggests that these cells regulate the immune function of epithelial cells
rather than
cells of the classical immune system (Wilson, et al. 2007).
When mucosal immunity is not countered by anti-inflammatory mediators,
excessive
pro-inflammatory responses result in chronic inflammatory bowel disease (IBD)
(Braegger, 1994). One study demonstrated that IL-23 was essential for the
manifestation of chronic intestinal inflammation, whereas IL-12 was not. A
critical
target of IL-23 was shown to be a unique subset of tissue-homing memory T
cells,
that were specifically activated by IL-23 to produce IL- 17 and IL-6. They
concluded
that this pathway might be responsible for chronic intestinal inflammation as
well as
other chronic autoimmune inflammatory diseases (Yen, et al. 2006).
= Th17 differentiation
o TGF-t3 (may inhibit in humans? - Chen and O'Shea, 2008)
o IL-10 (Toh and Miossec, 2007; Chen and O'Shea, 2008)
o IL-6
o IL-23
= Th 17 amplification
o IL-21 (produced by Th17 cells)
= Th17 stabilization
o IL-23
Thymus-produced self-reactive T cells, which become activated in the periphery
by
recognition of major histocompatibility complex/self-peptide complexes,
stimulate
antigen presenting cells (APCs) to secrete IL-6. APC-derived IL-6, together
with T
cell-derived IL-6, drives naive self-reactive T cells to differentiate into
arthritogenic
Th17 cells. In mice, deficiency of either IL-17 or IL-6 completely inhibits
the
development of arthritis, while IFN-y deficiency exacerbates the development
of
-50-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
arthritis (Hirota, et al. 2007). In humans, it is not yet clear whether
rheumatoid
arthritis is a Th 1 or a Th 17 mediated disease (Lubberts, 2008).
The IL-17 family of cytokines has been implicated in the pathogenesis of
rheumatoid
arthritis (RA) and juvenile idiopathic arthritis (JIA) (Nistala, et al. 2008).
In JIA, IL-
17 is increased in patients with active disease as compared to those in
remission (de
Jager, et al. 2007). Conversely, Treg cells are present at significantly
higher numbers
in patients with a milder clinical phenotype than in those with a more severe
form of
arthritis (De Kleer, et al. 2004).
= Th17 cell cytokine production (Lubberts, 2008):
o IL-17A
o IL-17F
o IL-6
o TNF-a
o GM-CSF
o IL-21
o IL-22
o IL-26
IL-6
The pleiotropic cytokine IL-6, previously called B cell stimulatory factor-2
(Bcl-2 or
BSF-2) or IFN-02, has emerged in recent years as a key regulator of the
transition
from innate to adaptive immunity through its ability to modulate leukocyte
recruitment at inflammatory sites (Kishimoto, 2006; Ohsugi, 2007). It has been
found
that there is a thermally-sensitive alert system utilizing IL-6 signaling that
promotes
immune surveillance, thus shedding light on the benefits of mounting a febrile
reaction during inflammation (Vardam, et al. 2007).
-51-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
One of the most important systemic actions of IL-6 is induction of the acute
phase
response. Acute phase proteins are produced primarily by the liver and include
proteins that promote the immune response through activation of complement,
induction of pro-inflammatory cytokines, and stimulation of neutrophil
chemotaxis
(Cronstein, 2007). In humans, tow of the most prominent acute phase proteins
are
CRP and serum amyloid A (Van Snick, 1990).
IL-6 exerts a significant influence on the course of inflammation in humans.
There is
evidence that IL-6 is capable of mediating both pro-inflammatory effects,
including
the induction of intercellular adhesion molecules and the recruitment of
leukocytes,
and anti-inflammatory effects, such as the suppression of the pro-inflammatory
cytokines, TNF-a and IL-I (Wong, et al. 2003).
IL-6 in combination with its soluble receptor, sIL-6Ra, influences the
transition from
acute to chronic inflammation (Gabay, 2006). Prospective studies have shown
that
long-term IL-6 levels are associated with coronary heart disease (CHD) risk as
strongly as are some major established risk factors (Danesh, et al. 2008).
This group
pointed out the pressing need for paired studies of individuals, in that owing
to
fluctuations in IL-6 values over time, comparisons using only baseline values
may
yield biased estimates of the true association between 11-6 and CHD, which can
be
corrected, for the most part, by using data from paired measurements. In
addition,
given the central role of IL-6 levels in inflammatory pathways and its
continuous
association with CHD risk, it warrants further investigation as a plausible
potential
therapeutic target. The ImmunoScore database will serve to hold an
individual's
paired measurements of various cytokines and enable future prospective studies
for
many diseases, as well as enable the study of CHD. In addition, ImmunoScore
-52-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
diagnostic applications could track fundamental immune parameters in
individuals
undergoing IL-6 directed therapy.
Cardiovascular disease is a leading cause of mortality in rheumatoid arthritis
(RA).
Endothelial dysfunction often precedes manifest atherosclerosis. Among
immunological and metabolic laboratory markers, anticyclic citrullinated
peptide
antibodies, IgM rheumatoid factor, circulating immune complexes, pro-
inflammatory
cytokines including TNF-a and IL-6, ThO/ThI cells, homocysteine, dyslipidemia,
decreased folate and vitamin B production, and paraoxonase activity may all be
involved in the development of vascular disease in RA (Szekanecz, et al.
2007). The
early diagnosis of endothelial dysfunction and atherosclerosis, active
immunosuppressive treatment, the use of drugs that control atherosclerosis,
changes
in sedentary lifestyle, and the close follow-up of RA patients may help to
minimize
cardiovascular risk in these individuals.
High serum levels of IL-6 have been linked to risks for several conditions,
such as
cardiovascular disease, type 2 diabetes, mental health complications, and some
cancers. Stress-induced immune dysregulation has been shown to be significant
enough to result in health consequences, including reducing the immune
response to
vaccines, slowing wound healing, reactivating latent viruses, and enhancing
the risk
for more severe infectious disease. There is evidence that psychological
stress
promotes immune dysfunction that negatively impacts human health (Godbout and
Glaser, 2006).
Local cellular environmental factors and an individual's genetic
susceptibility play a
role in the transduction of IL-6 signals. Depending on the expression of CD45
on
multiple myeloma cells, IL-6 can either result in proliferation or apoptosis
of CD45+
-53-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
cells depending on circumstantial stimuli (Ishikawa, et al. 2006). Chronic
obstructive
pulmonary disease (COPD) is a multicomponent disease characterized by abnormal
inflammatory response of the lungs to noxious particles, accompanied by
systemic
effects like weight loss, muscle wasting, reduced functional capacity, and
impaired
health status. A persistent low-grade systemic inflammatory response,
determined in
part by genetic components, is present in a portion of the COPD population
(Yanbaeva, et al. 2006).
IL-22
Elevated serum and plasma levels of IL-22 are indicative of Crohn's disease.
Normal
population mean level are approximately 2 pg/mL, while mean levels in Crohn's
disease patients reach 24 pg/mL, and are higher with flares of the disease
(Wolk, et al.
2007).
IL-23
Interleukin-23 is composed of the IL-12p40 subunit and a novel p19 subunit. It
can
enhance the proliferation of memory T cells and the production of IFN-y, IL-
12, and
TNF-a from activated T cells. IL-23 can also act directly on dendritic cells
and
possesses potent anti-tumor and anti-metastatic activity in murine models of
cancer
(Hao and Shan, 2006).
IL-23 possesses unique roles in the differentiation and expansion of memory T
cells.
IL-23 is also associated with Th 17 responses and the cytokine produced by the
antigen presenting cells (i.e. IL-12 vs. IL-23) determines in part if a
response is Thl
or Th17 (Tan, et al. 2008).
IL-23 is an inflammatory cytokine that plays a key role in the pathogenesis of
several
autoimmune and inflammatory diseases. It orchestrates innate and T cell
mediated
inflammatory pathways and can promote Th17 cell responses (Izcue, et al.
2008). IL-
-54-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
23 has been associated with several inflammatory disease including rheumatoid
arthritis, inflammatory bowel disease, and Helicobacter pylori associated
gastritis.
The immune response in the intestine is typically a delicate balance between
effector
and regulatory T cell responses, and IL-23 plays a key role in this balance.
Factors
may promote inflammation not only by direct effects on inflammatory mediators,
but
also indirectly by impeding regulatory mechanisms (Izcue, et al. 2008). IL-6
has been
identified as an inflammatory mediator that desensitizes T cells to Treg cell
mediated
suppression (Pasare and Medzhitov, 2003). IL-23, via its ability to impede
Treg cell
responses in the intestine, may promote host protective immunity at this site.
Cytokines and Autoimmunity
The presence of IL-17 mRNA or IL-17 protein in tissues and biological fluids
of
patients has been associated with rheumatoid arthritis (Kotake, et al. 1999;
Honorati,
et al. 2001), multiple sclerosis (Matusevicius, et al. 1999; Kurasawa, et al.
2000),
systemic lupus erythematosus (Wong, et al. 2000), inflammatory bowel disease
(Nielsen, et al. 2003; Fujino, et al. 2003), atopic dermatitis (Koga, et al.
2008), Lyme
arthritis (Infante-Duarte, et al. 2000), and psoriasis (Albanesi, et al.
2000).
In autoimmunity, IFN-y does not appear to be pathogenic, but rather
protective, as
inhibition of IFN-y signaling enhances the development of pathogenic Th 17 and
exacerbates autoimmunity (Harrington, et al. 2005). Also, the neutralization
of IL-4,
produced by Th2 cells is critical to in neutralizing the development of IL-
17;
however, neither IFN-y nor IL-4 seem to be effective on already established Th
17
pathogenesis (Harrington, et al. 2005).
In organ-specific autoimmunity, the balance of cytokines is a key determinant
of
resistance or susceptibility. Animal models of experimental autoimmune
-55-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
encephalomyelitis (EAE) are considered to mirror conditions of multiple
sclerosis
(MS) in humans. In EAE, disease susceptibility is thought to correlate with
the
expression of pro-inflammatory cytokines such as IL-17, IFN-y, TNF-a, IL-6,
and IL-
1 R. On the other hand, Th2 cytokines such as IL-4 and IL-13 have been shown
to be
important for preventing or easing disease symptoms (Cash, et al. 1994;
Olsson,
1995). IL-25 is expressed in organ systems where regulation of inflammation is
of
critical importance (Kleinschek, et al. 2007). In healthy digestive and
respiratory
tracts, an anti-inflammatory environment must be maintained due to constant
exposure to commensal microbes.
Although IL-25 and IL-17 are members of the same cytokine family, they play
opposing roles in the regulation of organ-specific autoimmunity. The type 2
responses promoted by IL-25 drive a novel regulatory mechanism for controlling
Th17 responses. This regulation relies on IL-13 and not IL-4, suggesting that
IL-13
may be secreted at higher levels in the target organs during autoimmune
inflammation
(Kleinschek, et al. 2007).
In mice, IL-25 is expressed by lung epithelial cells as a result of innate
immune
responses to allergens. IL-25 promotes Th2 cell differentiation in an IL-4-
dependent
manner and has been shown to be a critical factor regulating the initiation of
innate
and adaptive pro-allergic responses (Angkasekwinai, et al. 2007).
Human patients with IBD have elevated IL-17 and IL-22 in affected colonic
tissue
and serum, depending on disease activity and severity (Fujino, et al. 2003;
Nielsen, et
al. 2003; to Velde, et al. 2007). Patients with rheumatoid arthritis have
elevated 11-17
and IL-22 in synovial fluid (Kotake, et al. 1999; Ikeuchi, et al. 2005). 11-22
is
-56-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
increased in psoriatric serum and high levels of IL-23 have been detected in
psoriasis
lesions (Wolk, et al. 2006; Piskin, et at. 2006).
Etanercept is a TNFR-Ig fusion protein that has been used clinically to block
TNF at
molecular and cellular levels. A group studying the effect of this drug on
psoriasis
patients found that improvement in psoriasis disease correlated with the rapid
down-
modulation of DC and Th17 cell products and downstream effector molecules.
Final
disease resolution correlated with later down-modulation of Thl cells (Zaba,
et al.
2007).
Although many patients have been treated with immunomodulatory drugs, there
are
surprisingly limited data on therapeutic mechanisms in human inflammatory
disease.
Consistent monitoring of serum cytokines during therapies for autoimmune
disease
with ImmunoScore technology will provide benefit to physicians, patients, and
pharmaceutical researchers. In the case of etanercept treatment of psoriasis
described
above, it was found that first DC and Th17 effects in lesions were lessened,
but
disease was not completely resolved until Thi effects were also ameliorated.
For
other inhibitors of TNF (of which there are three currently in clinical use),
there may
be different mechanisms as yet undiscovered. Careful monitoring of the
progress of
autoimmune diseases and treatments to ease disease symptoms are to be a
hallmark of
ImmunoScore implementation. Based upon current understanding of autoimmune
disease flares, the following cytokines are to be monitored in patients
suffering from
autoimmune disease symptoms. Progressive spikes of each of these cytokine
groupings are to be expected. Effective therapeutic treatment should be
indicated by
faster cycling of each stage.
-57-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
1. When expressed together, these cytokines indicate a "highly aggressive"
Th17 profile:
o IL-17
o IL-22
2. For Th17 induction and stability:
o IL-23
o IL-6
o TGF-13
3. Cytokines indicative of impending flare resolution:
o IFN-y (ThI cytokine - interferes with long term Th17 cell survival)
o IL-4 (Th2 cytokine - interferes with long term Th 17 survival)
o IL-27
4. Treg cytokines:
o IL-10
o TGF-f3
It has been proposed that there is a reciprocal relationship betweem
pahtogenic Th17
cells and Foxp3+ Treg cells, in which IL-6, an acute phase protein induced
during
inflammation, acts as a pivot to determine whether the immune response is
dominated
by the highly inflammatory Th17 cells or protective Treg cells (Bettelli, et
al. 2007).
Cytokines and Cancer
Immunosuppressive networks mediated by IL-10 and TGF-(3 seem to inhibit cell-
mediated immune responses against cancer cells (Zou, 2005). Clinical data show
a
decreased ratio of circulating Thl cells to circulating Th2 cells and their
associated
cytokines in different cancer types and also in chronic inflammatory
conditions that
are associated with increased risk of cancer (Tan and Coussens, 2007).
Increased levels of circulating cytokines and their receptors (most often of
the pro-
inflammatory cytokine IL-6) have been found in observational studies of
patients with
-58-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
various types of cancer, both at the diagnosis of the primary disease and in
those with
metastases, compared with healthy people and people with benign tumors
(Seruga, et
al. 2008)
Various specific cancer treatments can stimulate the immune system to produce
pro-
inflammatory cytokines that are associated with toxic effects of treatment
such as
cancer-related fatigue, flu-like systemic effects and bone loss. They can lead
to
impaired quality of life of patients with cancer and poor compliance with
treatments
(Seruga, et al. 2008). Stimulation of the immune system by specific cancer
treatments
might also have a substantial role in producing anti-cancer effects. Cancer
drugs
might differentially effect the secretion of cytokines in humans with cancer,
and this
secretion might be a tool with which to monitor the therapeutic indices of
drugs.
Periodic ImmunoScore diagnostic measurements of cytokines relevant to specific
cancer types would be invaluable to doctors and their patients.
Cancer patients frequently suffer from fatigue and some suffer from cognitive
impairment during and after treatment for cancer (Lawrence, et al. 2004; Vardy
and
Tannock, 2007). An animal study has indicated the likely importance of IL-6 in
the
development of cognitive impairment (Sparkman, et al. 2006). IL-10 has been
shown
to counteract the production of IL-6 in microglial cells (Heyen, et al. 2000).
In
another study, patients with myeloid leukemia or myelodysplastic syndrome who
had
higher serum IL-6 levels were found to have poorer executive function, whereas
higher levels of IL-8 were associated with better memory (Myers, et al. 2005).
Many cancer patients, during both treatment and long-term follow-up,
experience
psychological distress including anxiety and depression (Zabora, et al. 2001).
Studies
of chronic and acute stress showed increased circulating levels of 11-6 and
TNF-a
-59-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
compared with controls (Kuecolt-Glaser, et al. 2003; Graham, et at. 2006).
Cancer-
related fatigue is strongly associated with a depressed mood (Bower, et at.
2006).
Clinical studies have also shown an association between circulating levels of
IL-6 and
resistance to chemotherapy (De Vita, et al. 1998; Zhang and Adachi, 1999). 11-
6 is
one of the most ubiquitously deregulated cytokines in cancer patients and high
levels
of circulating IL-6 most commonly predicted poor outcome in observational
studies
(Hong, et al. 2007). Stage is an important prognostic factor in every cancer
type and
in observational studies there is a consistent trend of higher levels of
circulating
cytokines in more advanced stages of various cancers than in early stages,
which
further supports an association with the outcome of cancer (Seruga, et at.
2008). A
study of oral squamous cell carcinoma found significant contributions of IL-6
and
TNF-a to disease (Vairaktaris, et al. 2008).
Some immunomodulatory cytokines have been demonstrated to have anti-tumor
activity. These include TNF-a, IFN-y, IFN-a, IL-2, IL-12, IL-15, and IL-18.
Others
with promise of anti-tumor activity include IL-21, IL-23, and IL-27 (Weiss, et
at.
2007).
Pro-inflammatory cytokines are involved in the development and progression of
cancer and are also associated with fatigue, depression, cognitive impairment,
anorexia, and pain, which all affect the quality of life of the patient.
Sustained
production of some cytokines may also be associated with cancer recurrence and
progression. Strategies to monitor and inhibit the effects of such cytokines
might
therefore have profound effects on quality of life and survival.
Proposed ImmunoScore Cancer Cytokine Panel
= IL I -RA
-60-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= sIL2R
= IL-6
= IL-8
= IL-10
= TNF-a
= M-CSF
= VEGF
Treg Manipulation
Depletion of naturally arising Tregs not only elicits autoimmunity, but also
augments
immune responses to non-self antigens (Sakaguchi, et al. 2008). Treg depletion
produces inflammatory bowel disease, which likely results from excessive
immune
responses to commensal bacteria in the intestine (Singh, et al. 2001). Removal
or
reduction of CD4+CD25+ Tregs also provokes effective tumor immunity in
otherwise
non-responding animals and augments microbial immunity in chronic infection,
leading to eradication of tumors or microbes (Wang and Wang, 2007; Belkaid and
Rouse, 2005). Conversely, CD4+CD25+ T cells enriched from normal mice suppress
allergy, establish tolerance to organ grafts, prevent graft versus host
disease after bone
marrow transplantation, and promote fetal-maternal tolerance (Sakaguchi,
2005).
Several key concepts have been formulated regarding dominant self-tolerance
and
immune regulation (Sakaguchi, et al. 2008). First, the normal immune system
generates Tregs that are engaged in suppressing immune responses towards self,
quasi-self (such as autologous tumor cells), and non-self (such as microbes
and
allografts). Second, the normal thymus produces potentially pathogenic self-
reactive
T cells as well as functionally mature Tregs; mature Tregs persist in the
periphery and
exert dominant control over the self-reactive T cells. Third, Treg deficiency
in the
-61-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
periphery is sufficient to evoke chronic T cell-mediated autoimmunity and
immunopathology.
In vitro, Tregs suppress the proliferation and cytokine production
(particularly IL-2)
of responder T cells in the presence of antigen presenting cells (Takahashi,
et al.
2000). Several mechanisms of Treg mediated suppression have been proposed, and
these include secretion by the Treg of immunosuppressive cytokines, cell
contact
dependent suppression, and functional modification or killing of antigen
presenting
cells. IL-10 and TGF-f3 contribute to suppression of inflammatory bowel
disease by
Treg depletion (Read, et al. 2000). TGF-(3 may act as a direct mediator of
suppression, and/or maintain Foxp3 expression and suppressive activity in
Tregs (von
Boehmer, 2005). A recent study has shown that Foxp3 natural Tregs
predominantly
produce immunosuppressive IL-35: ectopic expression of IL-35 confers
regulatory
activity on naive T cells, whereas recombinant IL-35 suppresses T cell
proliferation
(Collison, et al. 2007).
Foxp3 Tregs are abundant in tumors. Natural Tregs that promote self tolerance
may
act to impede immune surveillance against cancers in normal individuals and
suppress
potential responsiveness to autologous tumors in cancer patients. Targeting
Tregs is a
promising approach for cancer immunotherapy. Such approaches could include
local
depletion of Tregs in the tumor mass, attenuation of Treg function at the time
of
therapeutic vaccination with tumor antigen, and ex vivo expansion of tumor
infiltrating lymphocytes after the depletion of Tregs (Sakaguchi, et al.
2008).
Role(s) of TGF-13
TGF-(i is well known as an important cytokine that promotes the
differentiation of
anti-inflammatory Foxp3+ Treg cells. The finding that TGF-(3 is also required
for the
-62-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
differentiation of pro-inflammatory Th17 cells was unexpected (Veldhoen, et
al.
2006; Bettelli, et al. 2006). A model has been proposed wherein Th17 and Treg
cell
subsets may work together to either elicit or restrain tissue inflammation
(Cua and
Kastelein, 2006). The gut mucosa has high concentrations of TGF-(3, which may
induce a local population of Foxp3+ Treg cells to maintain homeostasis in an
environment filled with commensal bacteria. The gut lamia propria also
naturally
contains a considerable number of IL-17 producing cells that may help to
maintain the
mucosal barrier (Mangan, et al. 2006). When there is a breach of the
protective
mucosa, dendritic cells secrete copious amounts of IL-6 and IL-23. This then,
likely
activates Th17 cells to release IL-17, TNF-a, and GM-CSF which recruit
neutrophils
to protect the host from invading pathogens (Cuan and Kastelein, 2006). In the
final
phase of infection, this model predicts that microbe-specific Thl and Th2
cells would
enter the inflamed mucosa. These cells secrete pro-inflammatory factors, which
activate macrophage killing functions and promote anti-microbial antibody
responses,
respectively. They also repress the differentiation and function of Th 17
cells. As the
infection subsides, IL-6 and IL-23 production is reduced and the balance
swings
toward favoring the development of TGF-(3-dependent Treg cells, essential for
maintaining mucosal homeostasis.
Role of IL-2
IL-2 has multiple targets. It facilitates differentiation of CD4+ T cells to
Th I and Th2
cells and expands CD8+ memory T cells and natural killer cells. On the other
hand,
IL-2 promotes apoptosis in antigen activated T cells. IL-2 also maintains
Foxp3+
natural Tregs, expands them at high doses and facilitates TGF-(3-dependent
differentiation of naive T cells to inflammatory Th 17 cells (Laurence, et al.
2007).
Thus, assuming that the main source of IL-2 is activated T cells, there is a
negative
-63-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
feedback control of immune responses via IL-2; that is, IL-2 produced by
activated
non-regulatory T cells contributes to the maintenance, expansion and
activation of
natural Tregs, which in turn limits the expansion of non-regulatory T cells
(Sakaguchi, et al. 2008). Disruption of this IL-2 mediated feedback loop at
any step
promotes the development of autoimmune/inflammatory disease. Further,
manipulation of this feedback loop is instrumental in tuning the intensity of
Treg
mediated suppression, hence the strength of a variety of immune responses.
IL-2 is an essential cytokine for both generating and then limiting T cell-
dependent
immune responses. IL-2 has potent T cell growth factor activity (Smith, 1980).
The
strongest support for a crucial role for IL-2 in the generation of the immune
response
is that T cell proliferation and function in vitro can be substantially
inhibited using
monoclonal antibodies specific for either IL-2 or IL-2R (Malek and Bayer,
2004).
The third key activity of IL-2 is its ability to sensitize activated T cells
to undergo
apoptosis by a tumor necrosis factor (TNF) dependent pathway (Lenardo, 1991).
Impaired production of Treg cells is sufficient to account for the lethal
autoimmunity
that is associated with IL-2 and IL-2R deficient mice. The main function of IL-
2
seems to be the production of Treg cells and the maintenance of peripheral T
cell
tolerance (Malek and Bayer, 2004). In therapeutic settings, manipulation of
Treg cell
number or function might be accomplished by targeting IL-2 or IL-2R.
Relevant to Canadian Immigrant Population
As described below, an exemplary database comprised of records of bioassay and
patient history information was obtained and used to illustrate various
algorithms
according to exemplary embodiments of the present invention. The individuals
whose
-64-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
records were obtained were all immigrants to Canada. Accordingly, that
database is
sometimes referred to herein as the "CIP Database."
Treg cells appear to play a crucial role in controlling the threshold for T-
cell
activation via dendritic cell co-stimulation. When Treg cell activity is high,
as is
postulated to occur in response to infection in the developing world,
dendritic cell
expression of co-stimulatory molecule is low (like that on B cells) and
production of
potentially harmful but relatively low affinity self-reactive effector T cells
is
inhibited. Conversely, as occurs in the developed world where Treg activity is
lower,
T cells are more readily activated by dendritic cells expressing high levels
of co-
stimulatory molecules in addition to otherwise harmless antigens, like self
determinants and allergens, resulting in the generation of Thl and Th2
effector T
cells, respectively. This way of thinking is attractive because it provides a
rational
explanation for the inverse relationship between the incidence of autoimmune
and
allergic diseases on the one hand and infectious diseases on the other hand in
the
developed world (Figure 4 - Basten and de St Groth, 2008).
Infections relevant to the Canadian Immigrant Population Database
There is increasing awareness that helminth infections can ameliorate pro-
inflammatory conditions. The outcome of shistosomal infection in mice depends
on
Th2 polarization (Belkaid and Rouse, 2005). The inhibitory effects of natural
Treg
cells on the ThI response have been shown to promote Th2 polarization and to
protect
the host from lethal inflammatory pathology ((McKee and Pearce, 2004).
Natural Treg cells also seem to be important in the disease caused by
hepatitis C
virus. A chief complication of this chronic infection is massive liver damage
that
often requires organ transplant (Belkaid and Rouse, 2005). Liver biopsies
obtained at
-65-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
the time of transplants show an inverse correlation between the number of
natural
Treg cells in the periphery and the histological inflammatory score. These
Treg cells
were actively secreting IL-10 and TGF-f3 (Cabrera, et al. 2004). People
chronically
infected with hepatitis C virus have more circulating natural Treg cells than
do
uninfected people, and depletion of Treg cells enhances antigen-specific CD8+
T cell
responses in vitro (Sugimoto, et al. 2003).
Schistosomiasis and the Canadian Immigrant Population Database
The estimated mortality owing to Schistosoma mansoni and Schistosoma
haematobium in sub-Saharan Africa is 280,000 each year (van der Werf, et al.
in
press). Schistosomiasis causes a range of morbidities, the development of
which
seems to be influenced to a large extent by the nature of the induced immune
response
and its effects on granuloma formation and associated pathologies in target
organs
(Pearce and MacDonald, 2002). The development of the immune response in
infection is shown in Figure 5. In the course of the infection, the immune
response
progresses through at least three phases. In the first three to five weeks,
during which
the host is exposed to migrating immature parasites, the dominant response is
Thl-
like. As the parasites mature, mate, and begin to produce eggs at weeks 5-6,
the
response changes; the Thl component decreases and this is associated with the
emergence of a strong Th2 response. This response is induced primarily by egg
antigens. During the chronic phase of infection, the Th2 response is modulated
and
granulomas that form around newly deposited eggs are smaller than at earlier
times
during infection. During acute illness, there is a measurable level of TNF-a
in the
plasma, and PBMCs produce large quantities of IL-I and IL-6 (de Jesus, et al.
2002).
At the end of the ThI phase, the production of IL-10 is likely at least partly
responsible for the down-regulation of the inflammatory functions (Montenegro,
et al.
-66-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
1999). In the chronic phase, prolonged Th2 responses contribute to the
development
of hepatic fibrosis and chronic morbidity (Cheever, et al. 2000). The main Th2
cytokine responsible for fibrosis is IL- 13 (Pearce and MacDonald, 2002).
Mediators
associated with Th I responses, such as IFN-y, TNF-a, and IL-12 can prevent IL-
13
mediated fibrosis (Hesse, et al. 2001).
Expression of 11-25 (IL-17E) is critical for immunity against helmith
infections
(Owyang, et al. 2006). 11-25 protein administration in an animal model results
in
elevated expression of Th2 cytokines, IL-4, IL-5, and IL-13 (Fort, et al.
2001). IL-25
also regulates the development of autoimmune inflammation mediated by IL-17-
producing T cells (Kleinschek, et al. 2007).
Helminth infections in patients with multiple sclerosis (MS) created a B cell
population producing high levels of IL-10, dampening harmful autoimmune
responses. One group concluded that increased production of B cell-derived IL-
10
and of neurotrophic factors are part of the parasite's regulation of host
immunity and
can alter the course of MS, potentially explaining environmental-related MS
suppression observed in areas of low disease prevalence (Correale, et al.
2008).
The burden and chronicity of helminth infections is an important variable that
may
determine whether helminths act as a risk factor for, or confer protection
against,
allergic diseases (Yazdanbakhsh, et al. 2002). The over-riding view is that
heavy
helminth infections protect against allergy (Lynch, et al. 1997). Moreover, it
has been
proposed that alterations of commensal bacteria influences intestinal immune
homeostasis by direct regulation of the IL-25/IL-23/IL-17 axis (Zaph, et al.
2008).
Additional Cytokine Assays - CIP Database
IL-1
-67-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
IL-I is a pleiotropic cytokine the primarily affects inflammatory responses,
immune
reactivity, and hematopoiesis (Dinarello, 2005; Apte and Voronov, 2002). Its
potency
stems from inducing cytokine, chemokine, pro-inflammatory molecule secretion,
adn
adhesion molecule expression in diverse cells, thereby amplifying and
sustaining the
response. Membrane-associated IL-la is immunostimulatory. Low level secreted
IL-
1(3 induces limited inflammatory responses followed by T cell activation. High
levels
of 11-1(3 are accompanied by broad inflammation with tissue damage (Dinarello,
2005;
Apte and Voronov, 2002; Mariathasan and Monack. 2007).
Interleukin-I includes a family of closely related genes; the two major
agonistic
proteins, IL-la and IL-1(3, are pleiotropic and affect mainly inflammation,
immunity
and hemopoiesis (Apte, et at. 2006). These IL-I molecules bind to the same
receptors
and induce the same biological functions. As such, they have been considered
identical in normal homeostasis and disease. However, the IL-1 molecules
differ in
their compartmentalization within the cell. IL-1(3 is active in its secreted
form, while
IL-la is active in cell-associated forms - either as the intracellular
precursor or as
membrane-bound IL-la. It has been proposed that membrane associated IL-la
expressed on malignant cells stimulates anti-tumor immunity, while secretable
IL-1(3,
derived from the microenvironment or malignant cells, activates inflammation
that
promotes invasiveness and also induces tumor-mediated suppression (Apte, et
al.
2006). Both sarcoma cell-derived IL-la and IL-10 promote tumor growth.
However,
IL-la exerts regulatory authority on the tumor cell-matrix cross-talk, and
only IL-10
initiates systemic inflammation (Nazarenko, et al. 2008).
IL-1 is also an important mediator of inflammation and a major cause of tissue
damage in rheumatoid arthritis (RA). In a mouse model of vaccination to
prevent RA,
-68-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
it was found that immunization with IL-1(3 was strongly protective against the
development of arthritis, while immunization with a similarly constructed IL-1
a
vaccine had no effect (Spohn, et al. 2008). Another group examining genetic
polymorphisms in genes coding for IL-la and IL-1(3 in RA found that in a
majority of
cases, genetic polymorphisms in these genes were not a major contributor to
genetic
susceptibility (Johnsen, et al. 2008), strengthening the argument for
ImmunoScore
based analyses which measures phenotypic expression of protein components.
IL-8
Interleukin-8 (IL-8 or CXCL8) is a chemokine known to possess tumorigenic and
pro-
angiogenic properties as well as leukocyte chemotactic activity (Brat, et al.
2005). Il-
8 has been found to play an important role in autoimmune, inflammatory, and
infectious diseases (Harada, et al 1994; Koch, et al. 1992; Smyth, et al.
1991).
Because of its potent pro-inflammatory properties, IL-8 is tightly regulated,
and its
expression is low or undetectable in normal tissues. Expression of IL-8 can be
induced by IL-l, TNF-a, IL-6, and IFN-y (Baggiolini, et al. 1994; DeForge, et
al.
1993). Potent inhibitors of IL-8 production include IL-4 and IL-10 (Mukaida,
et al.
1994; Xie, 2001).
There is evidence that IL-8 is involved in tumor formation and malignant
progression
(Brat, et al. 2005). Mast cell mediators including fibroblast growth factor-2
and IL-8
are mitogenic to melanoma cells. Current evidence supports an accessory role
for
mast cells in the development and progression of cutaneous malignancies, but
it is
currently unclear whether the mast cells have promoting or inhibitory effects
on
tumors (Ch'ng, et al. 2006).
-69-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
IL-8 is expressed in working muscles. The small transient release of IL-8 by
working
muscle is likely used for local angiogenesis, whereas systemic increase of
plasma IL-
8 is likely indicative of a tumor disease state (Akerstrom, et al. 2005).
Other
cytokines, including 11-6 and IL- 15 are released locally by working muscle
(Nielsen
and Pedersen, 2007). Interestingly, anti-oxidant vitamins, C and E, have been
shown
to inhibit the release of pro-inflammatory cytokines from human skeletal
muscle
(Fischer, et al. 2004).
The use of biochemical markers in neonatal infection has remained an important
area
of research. Many infection markers are components of the inflammatory cascade
and IL-6 and IL-8 have been demonstrated to have good diagnostic utility as
early
phase markers, while CRP and procalcitonin have superior diagnostic properties
during the later phases (Lam and Ng, 2008).
IL-17
Interleukin- 17, described extensively above, is a pro-inflammatory cytokine
which
induces differentiation and migration of neutrophils through induction of
cytokines
and chemokines including granulocyte-colony stimulating factor and CXCL8/IL-8.
IL-17 producing T cells have a pivotal role in the pathogenesis of autoimmune
diseases. IL-17 is also involved in protective immunity against extracellular
bacterial
or fungal pathogens such as Klebsiella pneumoniae and Candida albicans
(Matsuzaki
and Umemura, 2007).
IL-13
Interleukin- 13 plays a major role in various inflammatory diseases including
cancer,
asthma, and allergy. It mediates a variety of different effects on various
cell types
including B cells, monocytes, natural killer cells, and fibroblasts (Joshi, et
al. 2006).
-70-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Bronchial asthma is a complex disorder that is thought to arise as a result of
aberrant
T cell responses to non-infectious environmental antigens. In particular,
asthma
symptoms are closely associated with the presence of activated Th2 cytokine-
producing cells making IL-4, IL-5, and IL-13 in the airway wall (Nakajima and
Takatsu, 2007). Animal models of disease have provided compelling evidence
that
IL-13, independent of the other Th2 cytokines, is both necessary and
sufficient to
induce all features of allergic asthma (Wills-Karp, 2004). IL-13 has been
described as
a target for therapeutics, as it is involved in the pathogenesis of bronchial
asthma and
therapeutic agents have been described that block IL-13 signals (Izuhara, et
al. 2006).
Biological agents directed against the IL-13 pathway and new immunoregulatory
agents that modulate functions of Treg and Th17 cells are likely to be
successful
against asthma (Adcock, et al. 2008).
Th2 cells, producing IL-4 and IL- 13 have also been implicated in systemic
sclerosis
which is characterized by extensive fibrosis, microvascular stenosis, and
autoantibody
production (Sakkas, et al. 2006).
Receptors for IL-4 and IL-13 are overexpressed on malignant cells from brain
tumors.
These cells have been experimentally targeted by using a chimeric IL-13
constructed
with a mutated form of pseudomonas exotoxin (Shimamura, et al. 2006).
ImmunoScore technology would be useful monitoring clinical therapies and
overall
health of the immune system during the course of such treatments.
In tuberculosis, IL-4 and IL- 13 can undermine effective Th1-mediated immunity
and
make an individual more susceptible to TB infection (Rook, 2007). It has been
postulated that for a TB vaccine to be effective, not only must the Thi axis
be
promoted, the Th2 axis must be suppressed. Understanding the balance between
II-
-71-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
12, IL-13, 11-23, and IL-27 is crucial to the development of immune
intervention in
tuberculosis (Cooper, et at. 2007).
IL-13 levels have been shown to be elevated in patients with gastrointestinal
nematode and helminth infections of the liver (Grencis and Bancroft, 2004;
Hirayam,
2004). In patients with schistosomiasis, both IL-13 and IFN-7 were shown to be
elevated, suggesting a compartmentalization of the anti-schistosome immune
response
(Dessein, et al. 2004).
IL-15
Interleukin-15 is a pleiotropic cytokine that plays an important role in both
the innate
and adaptive immune system. IL-15 promotes the activation of neutrophils and
macrophages, and is critical to DC function. In addition, IL-15 is essential
to the
development, homeostasis, function, and survival of natural killer and CD8+ T
cells
(Diab, et al. 2005). Abnormalities of IL-15 expression have been described in
patients with rheumatoid arthritis or inflammatory bowel disease (Waldmann,
2002).
In contrast to the role of IL-2 which is in the elimination of self-reactive T
cells to
prevent autoimmunity, IL-15 is dedicated to the prolonged maintenance of
memory T
cell responses to invading pathogens (Waldmann, 2006). IL-15 has been proposed
as having anti-cancer properties, in addition to triggering innate immunity
(Shanmugham, et al. 2006).
TNFD (Lymphotoxin E)
Lymphotoxin (3 is implicated in lymphoid follicle development, production of
pro-
inflammartory cytokines, and can enhance the production of fibroblasts and
synoviocytes. The expression of lymphotoxin 0 is significantly increased in RA
patients (O'Rourke, et al. 2008). It was speculated that lymphotoxin (3 may
play a
-72-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
role in RA disease pathogenesis by contributing to a more intense inflammatory
reaction in the synovium.
Important to the concept of immunosenescence, in a mouse model, over-
expression of
lymphotoxin was shown to induce fulminant thymic involution (Heikenwalder, et
al.
2008). Host responses to cytomegalovirus (CMV) infections include early
initial
production of interferons. New data indicate that, preceding the induction of
type I
interferons, an earlier critical type I interferon elicited in primary
infected stromal
cells via the lymphotoxin P receptor system and mediated by B cells is
necessary to
kick-start an efficient antiviral response (Fodil-Cornu and Vidal, 2008).
In addition, signaling through the lymphotoxin pathway is a crucial element in
the
maintenance of the organized microenvironment. Inhibitors of the lymphotoxin
pathway have been shown to reduce disease in a wide range of autoimmune models
(Gommerman and Browning, 2003).
-73-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
ImmunoScore Total Immunoglobulin Assays
IgG (g/L)
= IgGI (normal range = 4.9-11.4 g/L)
= IgG2 (normal range = 1.5-6.4 g/L)
= IgG3 (normal range = 0.2-1.10 g/L)
= IgG4 (normal range = 0.08-1.4 g/L)
IgM (g/L)
IgA (g/L)
IgE (g/L)
ImmunoScore Cytokine Assay Panel
Thl.
= IL- 12 (induction)
= IL-27 (induction)
= IFN-y (produced by Th 1)
= TNF-a (produced by Th 1)
= IL-2 (produced by Thl late in cycle?)
= IL- 10 (Th 1 suppressed by)
= TGF-(3 (Th 1 suppressed by)
Th2
= IL-4 (induction and production by Th2)
= IL-5 (produced by Th2)
= IL- 13 (produced by Th2)
= IL- 10 (produced by Th2 to dampen Th I response?)
= IL-10 (suppressed by)
= TGF-(3 (suppressed by)
Th17
= TGF-0 (induced by - with IL-6)
= IL-6 (induced by - with TGF-(3)
= IL-1(3 (induced by)
= IL-17A (produced by Th 17)
-74-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= IL-17F (produced by Th17)
= IL-21 (produced by Th 17)
= IL-22 (produced by Th17 - seen as both inflammatory and anti-inflammatory)
= IL-23 (Th 17 response maintained by presence of IL-23)
= IL-1 (Thl7 response maintained by IL-1 which is produced by Thl cells)
= IL-4 (suppressed by IL-4 which is produced by Th2 cells)
= IFN-y (suppressed by IFN-y which is produced by ThI cells)
= IFN-a (suppressed by)
= IL-2 (suppressed by)
= IL-27 (suppressed by)
Treg
= TGF-0 (induced by - in the absence of IL-6)
= IL-10 (produced by - suppress Thi and Th2 responses)
= TGF-P (produced by - suppress Th I and Th2, expand Th17 together with IL-
6)
= IL-6 (suppressed by)
= IL-21 (suppressed by)
= IL-31 (suppressed by)
B cell differentiation and regulation
= IL-2
= 1L-4
= IL-7 (important in aging immune system)
= IL-9
= IL-10
= IL-15
= IL-21
Canadian Immigrant Population Database Cytokine Assays
-75-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= IL-la- inflammatory; likely stimulates Th17 response locally
= IL-1(3 - inflammatory; likely stimulates Th 17 response more systemically
= IL-2 - produced by Th 1 - induce Treg
= IL-4 - produced by Th2
= IL-5 - produced by Th2
= IL-6 - pro-inflammatory - together with TGF-13 induce Th 17
= IL-8 - pro-inflammatory
= IL- 10 - anti-inflammatory; produced by Treg
= IL-12 - induce Th 1
= IL- 13 - produced by Th2
= IL- 15 - trigger of innate immunity; anti-tumor role
= IL- 17 - pro-inflammatory; produced by Th 17
= IL-23 - maintenance of Th 17
= IFN-'y - produced by Th I
= TNF-a - pro-inflammatory; produced by Th 17 and Th 1
= TNF-(3 - pro-inflammatory; increased levels signal autoimmune disease flares
-76-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Fig. 5B depicts T helper cell commitment towards specific lineages. Depending
on
local cytokine milieu, naive CD4+ cells can differentiate to one of three
types of
CD4+ T effector cells (Th 1, Th2, or Thl7) or CD4+ immunosuppressive Treg
cells.
Green arrows indicate positive cytokine signals for differentiation, while red
arrows
indicate suppressive effects of cytokines on particular cell types. This
simplified
diagram captures only the cytokines of major influence as the state of the art
currently
stands.
Fig. 5C depicts Thl/Th2 Paradigm. Model of paradigm as it existed circa 2000.
Thl
cells known for important role in cell mediated immunity, whilst Th2 cells
acknowledged to be important for humoral immunity. At this time, it was
thought
that Thl over-response was solely responsible for autoimmune disease. The
story has
proven to be more complicated with the current understanding of the role of Th
17
cells.
Fig. 5D depicts an evolving Thl/Th2/Th17/Treg paradigm. The Thl/Th2 paradigm
now includes arms that recognize the importance of Th 17 and Treg cells.
Rather than
balance on one fulcrum between two opposite sides, the model now encompasses
more complicated interactions whilst tilting on at least two axes. Over-
expression of
any one of the four arms of the T cell immune response without response of the
opposite functions can lead to undesirable complications and over-reaction of
the
immune system. Th17 responses coupled with Thl responses can lead to
autoimmune
reactions, while Th17 coupled to Th2 responses can lead to allergic reactions.
Over-
expression of regulatory responses with Thl reactions can lead to chronic
microbial or
viral infection, while coupled with Th2 responses can lead to chronic
parasitic
infections.
-77-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Fig. 5E depicts an exemplary model illustrating how Treg-mediated control of
CD80/CD86 expression may control the threshold of antigen recognition, crucial
for
preventing the activation of low avidity self-reactive T cells that are below
the cut-off
imposed during thymic selection. Treg cells stimulated during high affinity
responses
to microbes would increase the threshold (indicated by green line) by reducing
dendritic cell expression of co-stimulatory molecules. Conversely, in the
absence of
strong Treg cell activity, the threshold of self antigen recognition may drop
below the
thymic cut-off (indicated by black line), allowing activation of low avidity
anti-self T
cells (Basten, et al. 2008).
Fig. 5F depicts the development of the immune response in schitosome infection
(Pearce and MacDonald, 2002).
4. Quantitation
As part of the ImmunoScore technology, immune responses can be quantitated.
For
example, ThI, Th2, Th17, and Treg responses can be quantitated. These
quantitative
values can be used in at one point in time, or can be trended over time to
help
determine a person's immune status. These values and/or their time responses
can
also be used with other factors such as antibody concentrations to help
determine a
person's immune status.
Immune responses such as Thl, Th2, Th17, Treg, and the like can be quantitated
using a collection of blood measurements; for example, a collection of
cytokine
concentrations. Some of cytokines suppress an immune response; suppression can
be,
for example, quantitatively characterized by negative coefficients in a
function
relating concentrations to response. Because cytokine concentrations can vary
by
-78-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
many orders of magnitude, immune response quantitation may be related to the
logarithm of the concentrations rather than the concentration. Because basal
concentrations of cytokines are not always the same, the ratio of the cytokine
concentration to its basal concentration (or the logarithm of the ratio) can
be used in
the quantitation.
As more particular examples, the Th l quantitation can be a function of
cytokine
concentrations; the Th I quantitation can be a function of the concentrations
of IFN-y,
TNF-a, IL-2, IL-12, IL-10, TGF-(3, and IL-23; the ThI quantitation can be a
polynomial function of the concentrations (or the logarithm of the
concentrations) of
IFN-y, TNF-a, IL-2, IL-12, IL-10, TGF-(3, and IL-23; the Thl quantitation can
be a
linear function of the concentrations (or the logarithm of the concentrations)
of IFN-y,
TNF-a, IL-2, IL-12, IL-10, TGF-(3, and IL-23. As a second set of examples, the
Th2
quantitation can be a function of cytokine concentrations; the Th2
quantitation can be
a function of the concentrations of IL-4, IL5, IL-13, IL-10, and TGF-(3; the
Th2
quantitation can be a polynomial function of the concentrations (or the
logarithm of
the concentrations) of IL-4, IL5, IL-13, IL-10, and TGF-(3; the Th2
quantitation can
be a linear function of the concentrations (or the logarithm of the
concentrations) of
IL-4, IL5, IL-13, IL-10, and TGF-13. Asa third set of examples, the Th17
quantitation
can be a function of cytokine concentrations; the Th 17 quantitation can be a
function
of the concentrations of TGF-(3, IL-2, IL-4, IL-6, IL-17, IL-21, IL-22, IL-23,
IFN-a,
and IFN-y; the Th17 quantitation can be a polynomial function of the
concentrations
(or the logarithm of the concentrations) of TGF-(3, IL-2, IL-4, IL-6, IL-17,
IL-21, IL-
22, IL-23, IFN-a, and IFN-y; the Th 17 quantitation can be a linear function
of the
concentrations (or the logarithm of the concentrations) of TGF-(3, IL-2, IL-4,
IL-6, IL-
17, IL-21, IL-22, IL-23, IFN-a, and IFN-y. As a fourth set of examples, Treg
-79-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
quantitation can be a function of cytokine concentrations; the Treg
quantitation can be
a function of the concentrations of IL-2, IL-6, IL-10, IL-31, IL-35, and TGF-
(3; the
Treg quantitation can be a polynomial function of the concentrations (or the
logarithm
of the concentrations) of IL-2, IL-6, IL-10, IL-31, IL-35, and TGF-13; the
Treg
quantitation can be a linear function of the concentrations (or the logarithm
of the
concentrations) of IL-2, IL-6, IL- 10, IL-3 1, IL-3 5, and TGF-(3.
Combining the quantitative measurements for Th1 and Th2 responses to form a
ThIITh2 response is another aspect of ImmunoScore technology. In some
embodiments, Th 1 1Th2 is simply the difference between the Th2 quantitation
and the
Th1 quantitation. For example, a positive Thl ITh2 value can be indicative of
a Th2
response, while a negative Th1jTh2 value can be indicative of a ThI response.
Combining the quantitative measurements for Th17 and Treg responses to form a
Th 17ITreg response is another aspect of ImmunoScore technology. In some
embodiments, Th171Treg is simply the difference between the Th 17 quantitation
and
the Treg quantitation. For example, a positive Thl7ITreg value can be
indicative of a
Th17 response, while a negative Th171Treg value can be indicative of a Treg
response.
Further distillation of immune response information can be done, for example,
by
combining the ThljTh2 response and the Thl7ITreg response. By considering
Th I ITh2 and Th 171Treg as two dimensions of an immune response, a magnitude
and
direction can be computed. A small magnitude can be interpreted as the immune
system being in balance. For large magnitudes, the direction can indicate the
type of
immune response. For example, if 0 represents a Th2 response; 90 a Th 17
response;
180 , a Th 1 response; and 270 , a Treg response; then a direction of 45 can
be
-80-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
indicative of allergies and/or fibrosis. Continuing this example, 135 can be
indicative of autoimmunity and/or an acute bacterial infection; 225 can be
indicative
of chronic protozoan & mycobacterial infections; 315 can be indicative of
Helminth
infections. Trend tracking magnitude and direction over time can be used to
differentiate between acute and chronic immune issues, such as the 135
direction that
may represent autoimmunity or an acute bacterial infection.
As additional knowledge about the immune system is learned, additional
dimensions
can be added to the two dimensional Thl ITh2, Th17jTreg example above. Small
magnitudes can still represent a balanced immune system, and for large
magnitudes
the direction can indicate the type of imbalance.
Example 1
Let Th,1 be the quantitative measurement of the Th 1 response.
Thl =ao+ a, In c;
In the above equation, c represents the concentration of the jth cytokine, a=
is the
coefficient relating the magnitude and sign of the amount the 1`h cytokine
affects the
Thl response, and ao is a weighted sum of the logarithm of the basal
concentrations.
Cytokines that do not affect the Th 1 response can either be excluded from the
summation or use a coefficient of 0. To simplify notation when combining
responses,
the coefficient of 0 method will be used. This equation is a linear function
of
logarithm of the ratio of the cytokine concentration to the basal
concentration. To
change the equation to a linear function of logarithm of the cytokine
concentration, set
ao=0.
-81-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Similar equations can be generated for the Th2 response, Th17 response, and
the Treg
response:
Th2 = Qo + Ii In ci
T h17 = yo + y, Inc,
Treg = 50+6,Inci
The Th 1 Th2 response can then be computed as
Th2 - Th.1 = 60 - ao + Zi (fl, -ai)Inci. The Th171Treg response can be
computed
as Th.17 - Treg = yo - So + i(yi - ap)In ci. The magnitude of the combined
response can be computed as
,a z
Qa-a0+ -2f)1nci + y0 -So+Y(y,- 6j)Inc,
i
The direction of the combined response can be computed as
tan-1 yo - Sa +: (yi - Si) In ci
p0 - ua t Ei(Ii-ut)1iit,
When tracking the direction over time, the solution to the arctangent that
minimizes
the magnitude of the direction change can be selected: for example, if the
first
direction is 1 and the direction changes by 2 in the clockwise direction, -1
should
be chosen rather than 359 .
-82-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
C. IMMUNOSCORE EXEMPLARY SUPERPANELS
1. ImmunoScore Diagnostic Panel and Preventive Therapy for Autoimmune
Disease
In exemplary embodiments of the present invention, some or all of the
following
assays can be included in an ImmunoScore Autoimmune Screening/Diagnostic
Panel:
1. Antibody Assays
= anti-myelin oligodendrocyte glycoprotein (MOG) antibody
= anti-measles virus antibodies
= anti-2 I -hydroxylase antibody
= anti-adrenal cortex antibody
= anti-Klebsiella antibodies
= anti-cardiolipin antibody
= anti-lupus anticoagulant antibody
= anti-beta-2-glycoprotein antibody
= anti-hematopoietic precursor cell antibodies
= anti-soluble liver antigen antibody
= anti-RO/SSA antibody
= anti-endomysial antibody (AEA)
= anti-tissue transglutaminase (anti-tTG)
= anti-Saccharomyces cerevisiae antibody (ASCA)
= anti-neutrophil antibody (pANCA)
= anti-porin protein C of E. coli antibody (anti-OmpC)
= anti-glutamic acid decarboxylase antibody (GADA)
o particularly anti-65 kDa isoform
= anti-protein tyrosine phosphatase-like molecule antibody (IA-2A)
= anti-glomerular basement membrane (GBM) antibody
= anti-neutrophil cytoplasmic antigens (ANCA)
= anti-GD 1 a/GD 1 b complex antibody
= anti-LMI antibody
= anti-GM 1 antibody
= anti-thyroglobulin antibody
= anti-nuclear antibodies (ANA)
o lupus anticoagulant (LA) antibody
o anti-phospholipid (aPL)
o anti-SS/A antibody
o anti-SS/B antibody
o anti-Sm antibody
o anti-RNP antibody
o anti-Jo 1 antibody
o anti-Scl-70 antibody
o anti-dsDNA antibody
o anti-Centromere B antibody
-83-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
o anti-Histone antibody
= anti-alphaIlbbeta 3 IgM
= anti-acetylcholine receptor (anti-AChR) antibody
= anti-muscle-specific tyrosine kinase (MuSK) antibody
= anti-neuroleukin antibody
= anti-gliadin antibody
= anti-CV 2 antibody
= anti-GQlb IgG
= anti-GQ 1 b IgM
= anti-thyroid peroxidase antibody
= keratinocyte cell-surface antibodies
o anti-BP 180 (bullous pemphigoid antigen 2)
o anti-BP 230 (bullous pemphigoid antigen 1)
= anti-intrinsic factor antibody
= anti-parietal cell antibodies
= anti-mitochondrial antibodies
o in particular, anti-E2 component of pyruvate dehydrogenase complex
(PDC) antibody
= anti-cyclic citrullinated peptide (CCP) antibody
= anti-heat shock protein (HSP) 65 antibody
= anti-HSP 90 antibody
= anti-DnaJ antibody
= anti-BiP antibody
= anti-heterogeneous nuclear RNP A2/B 1 antibody
= anti-heterogeneous nuclear RNP D antibody
= anti-annexin V antibody
= anti-calpastatin antibody
= anti-type II collagen antibody
= anti-glucose-6-phosphate (GPI) antibody
= anti-elongation factor
= anti-human cartilage gp39 antibody
= anti-Chlamydia antibodies
= anti-La/SSB antibody
= anti-fodrine antibody
= anti-salivary duct antibodies
= anti-Red Blood Cell (RBC) IgM
= anti-neutrophil cytoplasmic antibodies
= anti-thyroid microsomal antibody (ATMA)
= anti-smooth muscle antibody (SMA)
= anti -mitochondrial antibody (AMA)
= anti-extractable nuclear antigens (ENA) antibody
= anti-actin antibody (AAA)
= anti-hair follicle antibodies
o anti-anagen matrix antibody
o anti-cuticle antibody
o anti-cortex keratinocytes antibody
o anti-melanocyte nuclear antigen
-84-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= anti-human dermal microvascular endothelial cells (HDMEC) antibodies
o anti-81 kDa HDMEC antigen, in particular
= anti-Trypanosoma cruzi antibodies
= anti-oleic acid IgM
= anti-palmitic acid IgM
= anti-myristic acid IgM
= anti-azelaic acid IgM
= anti-malondialdehyde IgM
= anti-aceylcholine IgM
= anti-S-farnesyl-L-cysteine IgM
= anti-ganglionic nicotinic acetylcholine receptor antibody
= anti-follicle-stimulating hormone (FSH) IgA
= anti-V 14D IgA
= anti-V1413 IgG
= anti-cytoskeleton-associated protein 4/p63 (CKA4/p63)-specific antibody
= anti-cytokeratin 10 antibody
= anti-Voltage-Gated Potassium Channels (VGKCs) antibodies
= anti-Chlamydia pneumoniae antibodies
= anti-human cytomegalovirus (CMV) antibodies
= anti-Toxoplasma gondii antibodies
= anti-CENP-A antibody
= anti-CENP-B antibody
2. Cytokine Assays
= Interleukin-1 a (IL-la)
= IL-lei
= IL-2
= IL-4
= IL-5
= IL-6
= IL-7
= IL-8
= IL-10
= IL-12
= IL-13
= IL-15
= IL-18
= Interferon a (IFN-a)
= IFN-y
= TNF-a
= G-CSF
= MCP-1
= MIP- I a
= MIP-1(3
= MIP-3a
= MIP-3p
-85-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= EGF
= VEGF
= TNFRII
= EGFR
3. Toll-like Receptor (TLR) genetic variants
= TLR 2
= TLR 3
= TLR 4
= TLR 7
= TLR 8
= TLR 9
4. HLA Haplotype screening
= HLA A24
= HLA B8
= HLA B 18
= HLA B27
= HLA B51
= HLA B60
= HLA B62
= HLA DR2
= HLA DR3
= HLA DR4
= HLA DR5
= HLA DR7
5. Protein isoforms/ genetic polymorphisms/ serum protein levels
= Apolipoprotein E isoforms
o apo E2
o apo E3
o apo E4
= Serum Apolipoprotein A-IV level
= Mannose-binding lectin (MBL) polymorphism
= Serum Haptoglobin level
= SerumTransthyretin level
= Serum Fibrinogen level
= Serum Vitamin B12 level
= Serum Folic acid level
2. ImmunoScore Diagnostic Panel: Aging, Longevity, Cancer and Human
Cytomegalovirus
-86-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Old age is accompanied by an increased incidence of infection and poorer
responses
to vaccination. A progressive decline in the integrity of the immune system is
one of
the physiologic changes during mammalian aging. Perhaps the most profound
clinical
impact of age on the immune system concerns the response of the elderly to
vaccination (Pawelec, 2005). An immune risk phenotype (IRP) was described
wherein individuals possessed high CD8 and low CD4 numbers and poor
proliferative
response (Wikby, et al. 2005). Characteristics of the IRP are listed in Table
I (Vasto,
et al. 2007). The IRP consists of a cluster of these parameters, not each
parameter
individually. Which are the most important and which additional factors are
involved
remains to be determined.
Lifelong and chronic antigenic load may represent the major driving force for
immunosenescence, which impacts on human lifespan by reducing the number of
virgin antigen-non experienced T cells, and results in their replacement by
expanded
clones of antigen-experienced effector and memory T cells which display a late
differentiation phenotype. Gradually, the T cell population shifts to a lower
ratio of
naive cells to memory cells, the thymus releases fewer naive T cells with age
and
those T cells remaining, especially the CD8+ subset, also show increased
oligoclonality with age. Presumably, the repertoire of cells available to
respond to
antigenic challenge from previously encountered pathogens shrinks. In
addition,
older organisms often are overrun by memory cells that carry a single type of
T cell
receptor, i.e. the clonal expansion referred to above. Therefore, the memory
cells
from old individuals might recognize a limited set of antigens despite being
plentiful
in number, and in addition, are likely to show various degrees of
dysfunctionality.
Many of the clonal expansions filling the individual's immune system seem to
result
from previous infections by persistent viruses, especially CMV (Ouyang, et al.
-87-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
2003b), but also, to a lesser extent EBV (Ouyang, et al. 2003a) and possibly
other
herpes viruses (Vasto, et at. 2007). A high number of CD8+ cells are found to
be
specific for a single CMV epitope (Pawelec, et at. 2005; Pawelec, et at.
2004). In
humans, the accumulation of CMV-specific T cells has been observed to reduce T
cell
immunity toward EBV infection (Khan, et al. 2004) and influenza vaccination
(Trzonkowski, et al. 2003). Functional analyses performed with T cells from
nonagenarians demonstrated that they were characterized by decreased
functional
capacity when compared with similar cells isolated from middle aged
individuals
(Hadrup, et at. 2006). This suggests that increased numbers of CMV-specific T
cells
could be the result of a compensatory mechanism enabling control of CMV
despite
lower functional capacity (Hadrup, et at. 2006). The biology of CMV infection
in
humans can be conceptualized as an evolutionary "negotiated" balance between
viral
mechanisms of pathogenesis, persistence, and immune evasion and the host
cellular
immune response (Sylwester, et at. 2005).
One of the immunodominant viral antigens recognized by CMV-specific CD8+ T
cells
is derived from the 65-kDa phosphoprotein (pp65). Samples from octogenarian
and
nonagenarian populations revealed that a large number of CD8+CD28- cells were
specific for the pp65 antigen. These findings imply a co-dominant role of CMV
as a
cause for a compromised immunity in old age (Vasto, et al. 2007). A second
immunodominant antigen is the IE-1 antigen. Epitope specificity and
immunodominance of CD8 T cells against IE-1 and pp65 are comparable in acute
infection and long-term memory often with marked focusing of responses that
are
probably established very early on. However, the kinetics of CD8 T cell
responses for
these antigens expressed at opposite ends of the replicative cycle of the
virus reflect
the different modes of antigen presentation, which probably depend on levels
of viral
-88-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
activity occurring over the lifetime of the host (Khan, et al. 2007). Other
studies have
suggested an extraordinary complexity of CMV-specific T cell responses to
chronic
infection (Sylwester, et al. 2005). This complexity complicates efforts to
understand
the basis of the CMV immune balance and, in clinical practice, to determine
the
thresholds that define the boundary between controlled vs. progressive CMV
infection
in immunocompromised subjects and between normal and excessive CMV-specific
immunity in the elderly (Sylwester, et al. 2005).
There are a suggested sequence of stages for IRP individuals that begin with
the
acquisition of CMV infection in earlier life, followed by generation of
CD8+CD28-
cells to control persistent CMV infection, and eventually the development of
an IRP.
Recently, a group of rare individuals was discovered who moved out of the IRP
category by a process of immune suppression, including increases in IL-6 and
IL- 10
and decreases in the number of CD3+CD8+CD28" cells (Wikby, et al. 2006).
There are two theories regarding the evolution of senescence - mutation
accumulation
and antagonistic pleiotropy. The mutation accumulation theory of senescence
postulates that there are numerous loci subject to mutation to deleterious
alleles,
whose effects on survival or other components of fitness are restricted to
narrow
bands of ages (Rose, 1991). The equilibrium frequencies of such deleterious
alleles
will be higher the later in life in which they act (Charlesworth, 1994). The
alternative
path involves antagonistic pleiotropy, according to which genes that increase
early
performance are likely to become established in a population even if they have
adverse effects on later performance (Williams, 1957; Rose, 1991).
Antagonistic
pleiotropy was originally defined as meaning opposite effects of the same
allele at
different ages (Williams, 1957). Antagonistic pleiotropy in evolutionary
theory
-89-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
usually refers to opposite effects of a genotype on fecundity and survival.
The
existence of trade-offs between these two components of Darwinian fitness was
proposed to explain the evolution of senescence and the maintenance, via the
creation
of the heterozygous advantage, of polymorphism at loci involved in the
determination
of both traits (Kirkwood and Rose, 1991). In a later model, antagonistic
pleiotropy
involved, instead, relative survival values of a genotype at different ages
(Toupance,
et al. 1998). The two theories are not mutually exclusive, and modeling
exercises
have examined the validity of each (Charlesworth and Hughes, 1996).
An example of antagonistic pleiotropy would be the high expression of
testosterone in
a male gorilla, which could lead to increased aggression and strength that
would allow
the male to become dominant and mate more frequently, but may eventually lead
to a
shortened lifespan due to increased atherosclerosis. Recent studies at the
molecular
level have suggested that cellular senescence may be antagonistically
pleiotropic
because it prevents tumorigenesis, but also contributes to organismic aging
(Troen,
2003).
In one study, it was suggested that cellular senescence was antagonistically
pleiotropic, protecting from cancer early in life, but promoting
carcinogenesis in aged
organisms (Krtolica, et at. 2001). Another study (Hughes, et al. 2002) found
the AP
(antagonistic pleiotropy) model is consistent with the existence of a few
genes with
individually large effects on late-life fitness, whereas the MA (mutation
accumulation) process should lead to the maintenance of may deleterious
alleles at
intermediate frequencies within populations and these alleles can have
individually
small effects on late-life performance and health. Current methods of
identifying
aging genes (such as mutation studies and quantitative trait locus-mapping
-90-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
experiments) are most effective in finding alleles of large effect, and even
well
designed studies will probably miss genes with small effects. Novel approaches
are
needed to find such genes.
Cancer rates also increase sharply with age in both sexes, and the majority of
cases of
cancer occur in patients over the age of 65. Tumor progression is a complex
process
that depends on interactions between tumor cells and host cells. The
inflammatory
aspect of the host response is of particular interest because it includes the
release of
pro-inflammatory cytokines, some of which may promote tumor growth and hence
influence survival. Some kinds of solid tumors are likely affected by
regulatory
cytokine genotypes. In particular, pro-inflammatory genotypes characterized by
a low
IL-10 or a high IL-6 producer seem to be associated with a worse clinical
outcome
(Caruso, et al. 2004). On the other hand, recent evidence has linked IL-10 and
IL-6
cytokine polymorphisms to longevity. In fact, individuals who are genetically
predisposed to produce high levels of IL-6 have a reduced capacity to reach
the
extreme limits of human life, whereas the high IL-10 producer genotype is
increased
among centenarians (Caruso, et al. 2004). The opposite effect of IL-6 and IL-
10 in
cancer and longevity is intriguing. Inflammatory genotypes may be both friends
and
enemies. The immune system has evolved to control pathogens, therefore pro-
inflammatory responses are likely to be evolutionarily programmed to resist
fatal
infections, and a high IL-6 or a low IL-10 production is associated with
increased
resistance to pathogens. However, decreased level of IL-6 or increased level
of IL-10
might better control inflammatory responses and cancer development. These
conditions might result in an increased chance of long life survival in an
environment
with reduced pathogen loads (Caruso, et al. 2004).
-91-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Most tumor supressor genes can be classified as either caretakers or
gatekeepers
(Kinzler and Voglestein, 1997). Caretaker tumor suppressor genes prevent
cancer by
protecting the genome from mutations. They generally act by preventing DNA
damage or optimizing DNA repair. In addition to preventing cancer, genes that
help
maintain genomic integrity also prevent or retard the development of other
aging
phenotypes and age-related pathologies (Hasty, et al. 2003). Gatekeeper tumor
suppressors, by contrast, prevent cancer by acting on intact cells -
specifically,
mitotic cells that are at risk for neoplastic transformation. Gatekeepers can
virtually
eliminate potential cancer cells by inducing programmed cell death
(apoptosis).
Alternatively, they can prevent potential cancer cells from proliferating by
inducing
permanent withdrawal from the cell cycle (cellular senescence). Although
little is
known about how cells choose between apoptotic and senescence responses, there
is
little doubt that both responses are crucial for suppressing cancer (Campisi,
2001;
Green and Evan, 2002).
Increasing evidence suggests that the rise in cancer with age results from a
synergy
between the accumulation of mutations and age-related, pro-oncogenic changes
in the
tissue milieu. Most age related cancers derive from epithelial cells.
Epithelial tissues
are supported by a stroma, which is composed of extracellular matrix and
several cell
types. One age-related change that occurs in epithelial tissues is the
accumulation of
senescent cells. Cellular senescence is a potent tumor suppressive mechanism
that
irreversibly arrests proliferation in response to damage or stimuli that put
cells at risk
for neoplastic transformation. Senescent cells secrete factors that can
disrupt tissue
architecture and stimulate neighboring cells to proliferate. The suggestion
has been
made that senescent cells can create a tissue environment that synergizes with
oncogenic mutations to promote the progression of age-related cancers
(Krtolica and
-92-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Campisi, 2003). The recent evidence indicates that cellular senescence may be
an
example of evolutionary antagonistic pleiotropy.
A major difference between microbial pathogens and tumors as potential vaccine
targets is that cancer cells are derived from the host, and most of their
macromolecules are normal self-antigens present in normal cells. To take
advantage
of the immune system's specificity, antigens must be found that clearly mark
the
cancer cells as different from host cells. An area generating much interest is
the
possibility of overcoming mechanisms that downregulate or attenuate the immune
response, as is depicted in Fig. 5D (Berzofsky, et al. 2004b). With refernce
thereto,
Fig. 5D illustrates negative regulation of tumor immunosurveillance and
antitumor
immune responses. Fig. 5D(A) depicts CD4+CD25+ T regulatory cells, induced by
peptide presented by class II MHC molecules in the presence of IL-2, may
inhibit
induction of effector CD4+ or CD8+ T cells by a contact-dependent mechanism,
possibly involving cell surface and/or secreted TGF-(3, and Fig. 5D(B)
illustrates how
CD4+ NKT cells may be induced by tumor glycolipid presented by CDId to secrete
IL-13, which stimulates Gr-I+CDI lb+ myeloid cells to produce TGF-(3, which
inhibits
induction of CD8+ CTLs mediating tumor immunosurveillance. TGF-(3 may also
inhibit CD4+ T cells (not shown). Blockade of other mechanisms can improve
immunosurveillance and the response to vaccines. Other suppressor or negative
regulatory cells have been described in other contexts, but not as well study
in the
context of cancer (Berzofsky, et al.). Such mechanisms may have evolved to
reduce
inflammation and immunopathology or to prevent autoimmunity. Tumors have co-
opted these mechanisms to evade immunosurveillance.
-93-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Thus, it has been postulated that the excess of dysfunctional CD8 T cells is
indirectly
immunosuppresive by filling the "immunologic space" and shrinking the T-cell
repertoire for new antigens, as well as directly suppresive via cytokine
secretion. It is
associated with the IRP predicting two and four year mortality in longitudinal
studies
of very old people. It is hypothesized that deletion of such accumulations of
dysfunctional cells would be beneficial to the individual. It may be possible
to
distinguish functional CMV-specific cells (which are essential to maintain
immunosurveillance) from dysfunctional ones by their expression of certain
surface
molecules. This, coupled with methods directed at reinvigorating the thymus
(such
as, for example, the use of interleukin 7), and targeting CMV by pharmacologic
and
immunotherapeutic interventions might result in the immunorejuvenation
sufficient to
take elderly individuals out of the risk category and thereby extend healthy
longevity
(Pawelec, et al. 2006). Animal models suggest that IL-7 improves immune
reconstitution through increasing thymic output and, perhaps more importantly,
through antigen-independent homeostatic driven proliferation in the periphery
(Sasson, et al. 2006). A study in old Rhesus macaques showed that treatment of
the
elderly with IL-7 may provide an effective therapy to improve the immune
system
(Aspinall, et al. 2007).
In rural Gambians, the season of birth strongly predicts adult mortality.
Those born
during the harvest season have longer life spans than do those born during the
hungry
season, and the deaths associated with infectious diseases suggest permanent
early-
life influences on immunity (Ngom, et al. 2004). One group studied thymic size
and
output in Gambian infants born in either the hungry or the harvest season by
measuring signal joint T cell receptor-rearrangement circles (sjTRECs) at
birth and at
8 weeks of age. They found that by 8 weeks of age, those born in the hungry
season
-94-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
had significantly lower sjTREC counts (indicating poor immune function) than
did
those born in the harvest season. These results correlated directly with lower
ELISA
measurements of IL-7 in mothers' breast milk (Ngom, et al. 2004). This
research
group speculated that these data show a plausible pathway linking external
season
insults to mothers with thymic development in their infants, which suggests
possible
implications for long-term programming of immunity.
ImmunoScore Measurements and Applications. Thus, there is a balance between
viral mechanisms of pathogenesis, persistence, and immune evasion and the host
cellular immune response. The immunologic basis of this balance has not been
completely characterized. The nature and threshold of CMV-specific T cell
responses
required for long-term CMV containment yet remain to be defined. This
information
would facilitate identification of highly susceptible individuals and provide
a specific
target for immunotherapeutic approaches designed to establish, maintain, or
restore
immunologic protection (Sylwester, et al. 2005). There seem to be clinical
consequences to an overly robust CMV-specific T cell response. An obvious
prerequisite for a better understanding of what constitutes insufficient or
excessive
CMV-specific T cell immunity is the ability to evaluate the overall CMV-
specific T
cell response in infected individuals. Future longitudinal studies would
benefit from
combining data on viral reactivation and primary infection with immunological
monitoring (Hadrup, et al. 2006). The ImmunoScore diagnostic and database
systems
would provide just such an opportunity for data collection and monitoring
longitudinal data collection.
Although CMV seropositivity appears to be one of the driving forces for
induction of
CD8 T cell clonality, this is not currently detectable in the middle-age
population
-95-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
(Hadrup, et al. 2006). The influence of CMV on clonality only becomes relevant
at a
detectable level in the elderly. Superior detection capabilities available
through the
ImmunoScore technology might lead to earlier detection of possible immune
depletion as individuals pass through middle age.
ImmunoScore technology by its nature of compiling individual patient data
would
offer the opportunity for longitudinal design of research studies. The
longitudinal
design is a superior alternative to the cross-sectional method for conducting
ageing
research, but it has seldom been used due to extensive costs as studies are
currently
conducted. The ImmunoScore system would naturally build a longitudinal
component into patient care at no increased initial cost. The database would
yield
important insights into ageing and all its implications at a lower cost and
dramatically
improve healthcare.
Questions have been raised concerning CMV infection and its relationship to
the IRP
(Vasto, et al. 2007). Uncertainties that require clarification are: Is there
an
immunogenetic component influencing the IRP phenotype that might explain the
different degree of CMV clonal expansion vs. non-IRP phenotype? May this
difference depend on social and/or environmental factors? Might the genetic or
environmental component affect the degree of clonal expansion of CMV in IRP
individuals? What can be the main cause of death in IRP? Can IRP selection be
predictive in young as well as in old individuals? Is it possible to
revert/prevent
accumulation of CMV-specific cells?
These are all questions that can, in exemplary embodiments of the present
invention,
be addressed by the application of ImmunoScore diagnostic and database
technologies.
-96-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Immunogenetic components can, for example, be monitored using unique
technology
designed to investigate single nucleotide polymorphisms (SNPs) rapidly and
those
data could be stored in the ImmunoScore central database. Additionally, social
and
environmental factors can be part of the ImmunoScore demographic data
collected at
routine patient visits to their physicians. The accumulation of these data on
the
ImmunoScore database would yield potential relationships regarding
environmental
and social factors to the IRP.
Careful monitoring of the ImmunoScore database would shed more light onto
environmental and/or genetic factors contributing to the clonal expansion of
CMV T
cells in IRP individuals and the non-IRP individuals.
As the ImmunoScore data collection system is envisaged as a cradle-to-grave
system
of healthcare, the cause of death in IRP individuals can be collected and
collated.
Preliminary indications are that IRP selection is likely to be predictive in
the young as
well as in the very elderly. The ImmunoScore cradle-to-grave philosophy of
patient
data tracking can be invaluable in assessing these issues. Additionaly,
prevention/reversion of the accumulation of CMV-specific T cells would seem an
issue of paramount importance. Preliminary studies in animal models regarding
judicious use of IL-7 have been promising. ImmunoScore can, for example, track
treatments and even shed light on when such treatments should commence in the
life
of the afflicted individuals.
CMV Vaccine and Vaccines Against Chronic Viral Infections and Cancer. In a
recent
review of priorities for vaccine development, CMV was ranked in the highest of
five
tiers by the Institute of Medicine in the United States as a potentially cost-
saving
vaccine target (Stratton, et al. 2000). In general, CMV is acquired earlier in
life in
-97-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
developing countries and among the lower socioeconomic strata of the developed
countries (Stagno and Cloud, 1990). Recently, the seroepidemiology of CMV was
examined in Australia (Seale, et al. 2006). The pattern of age-specific
seroprevalence
of CMV antibody, as provided in Fig. 5C, closely matched the pattern found
from
analysis of the exemplary CIP database described in Section II, below. Indeed,
a
review of CMV seroprevalence studies conducted around the world revealed that
residents of developing countries have higher rates of CMV seropositivity than
those
of developed countries (Enright and Prober, 2004). The potential benefits of a
CMV
vaccine would include reduced transmision to pregnant women and less CMV
disease
due to primary infection or reactivation in organ transplant recipients and
the
immunosuppressed (Griffiths, et al. 2000).
It is possible that the development of a vaccine that is effective against
viruses that
cause chronic infection may require consideration of a paradigm different than
those
previously used for organisms causing acute infection (Berzofsky et al. 2004).
In
most cases of chronic viral infection, the immune response to the natural
infection is
not sufficient to eradicate that infection. The challenge for the 21st century
is to apply
the latest fundamental knowledge in molecular biology, virology, and
immunology to
developing vaccines that are more effective at eliciting immunity than the
natural
infections and consequently, effective against chronic viral and other
infectious
diseases in addition to cancer, which do not fit the classic paradigm.
ImmunoScore
diagnostic and database tracking would be invaluable in analyzing the efficacy
of a
CMV vaccine, as well as vaccines developed against HIV, hepatitis C virus
(HCV),
human papilloma virus (HPV) and Epstein-Barr virus (EBV), among others.
-98-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
As prophylaxis against acute infectious diseases, vaccines have been among the
most
cost-effective agents, saving many millions of lives. However, for treatment
of
chronic infections and cancer, vaccines have yet to achieve widespread
success.
Increased understanding of the immune system has raised new hope of harnessing
the
exquisite specificity of the immune system to attack cancer (Berzofsky, et al.
2004b).
In exemplary embodiments of the present invention exemplary ImmunoScore
diagnostic panels and database systems can add considerably to this knowledge
base
andcan, for example, assist in intelligent vaccine design and monitoring of
the
efficacy of the vaccines as they are developed.
Table 1: Characteristics of the Immune Risk Phenotype (IRP)
CD4:CD8 ratio < 1
Poor T cell proliferative responses to mitogens
Increased CD8+CD28" and CD8+CD57+cells
Low B cell count
CMV seropositivity
Clonal expansion of CD8 cells carrying receptors for.CMV
High proportion of dysfunctional cells amongst the CMV-specific CD8 cells
D. EXEMPLARY IMMUNOSCORE SUPERPANELS
1. Middle School Student ImmunoPrint Super Diagnostic Panel
In exemplary embodiments of the present invention, a middle school superpanel
can,
for example, comprise the following exemplary panels:
1.1. Persistent Immunity Induced by Childhood Vaccines
This panel is described above in section A3.
1.2. Sexually Transmitted Disease (STD) Diagnostic Panel
For children entering middle school (grades six through eight) a baseline
determination for antibody levels to STDs is advisable. Recommended tests for
ImmunoPrint measurement of immunity to STDs:
-99-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= Antibodies to Chlamydia - IgG, IgA, and IgM (3)
= Antibodies to HSV - IgG to HSV-1 and HSV-2 (2)
= DNA analyses of HPV types - particular emphasis on high-risk
= Antibody to N. gonorrhoeae (1)
= Antibody to T pallidum (1)
= T-cell related response to T pallidum
= Antibody to HIV
= T-cell related response to HIV
= Antibodies to GBS serotypes (at least 3)
= Measurement of Th l/Th2 cytokines (many as current evolving definitions)
= Antibodies to organisms that cause Urinary Tract Infection (UTIs)
o Escherichia coli
o Staphylococcus saprophyticus
o Proteus mirabilis
o Klebsiella pneumoniae
o Enterococcus species
o Pseudomonas aeruginosa
Currently, there are no vaccines available for any of these STDs, with the
exception of
the Merck HPV vaccine. Until this situation is ameliorated as to a particular
vaccine
preventable disease, an ImmunoScore STD diagnostic panel would thus be to
recommend treatments, track immunological response or provide other analyses,
and
not be used to recommend a vaccine or track the persistence of immunity
conferred by
it. Thus, in exemplary embodiments of the present invention an exemplary
ImmunoScore database can, for example, generate correlates of protection
information for all disease-causing organisms. As vaccines are developed,
ImmunoScore diagnoses could, for example, be designed to examine antibody and
other related immune responses to vaccine components.
= Chlamydia trachomatis infection is the most commonly reported sexually
transmitted disease in the United States, with the highest rates among
adolescent females and young women. Because up to 70% of chlamydial
infections in women are asymptomatic, routine screening and treatment of
infected persons is essential to prevent pelvic inflammatory disease,
infertility,
-100-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
ectopic pregnancy, and perinatal infections. The third U.S. Preventive
Services Task Force (USPSTF) recommends that primary care physicians
routinely screen all women whether or not they are pregnant if they:
o Are sexually active and aged 25 or younger.
o Have more than one sexual partner, regardless of age.
o Have had an STD in the past, regardless of age.
o Do not use condoms consistently and correctly, regardless of age.
According to studies reviewed by the third USPSTF:
o The cost of screening women who are not pregnant and who are at risk
for chlamydial infection may be less than the cost of treating
Chlamydia and its complications.
o Screening patients at greatest risk is more cost effective than screening
all patients.
o DNA or RNA amplification tests are more sensitive than culture.
A low cost diagnostic test for Chlamydia infection or immune response to a
Chlamydia vaccine would be a welcome addition to immune status determination
by
ImmunoPrint diagnostic testing.
= Herpes simplex virus type 2 (HSV-2) is the primary cause of genital herpes,
a
common sexually transmitted disease with at least 40 to 60 million infected
individuals in the U. S. Medically serious complications of HSV are rare but
constitute a significant burden, given the high rates of HSV seropositivity in
the population. Many prophylactic and therapeutic vaccination approaches
have been explored for the prevention or treatment of HSV infection.
Infection induces both humoral and T-cell immunity. Vaccine candidates for
-101-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
HSV-2 infection include subunit vaccines, killed and live attenuated virus
vaccines, and viral DNA vaccines.
= Human papillomaviruses (HPV) are small double-stranded DNA viruses that
are responsible for pathological conditions ranging from benign skin warts to
invasive cervical carcinomas. Cervical cancer is the second leading cause of
cancer death among women worldwide, and more than 99% of cervical
cancers contain HPV, particularly the high-risk HRP type 16 (HPV-16). Two
HPV oncoproteins, E6 and E7, are consistently expressed in HPV-associated
cancer cells and are responsible for their malignant transformation. These
oncogenic proteins represent ideal target antigens for developing vaccines and
immunotherapeutic strategies against HPV-associated neoplasms. More than
10,000 American women a year are diagnosed with cancer or precancerous
cells caused by HPV, and 3,700 of them will die. Eighty times that number
will die worldwide. An effective vaccine could prevent nearly all of those
deaths. The CDC is currently considering an HPV vaccine for all children
aged 12 years. A positive recommendation by the ACIP could start states
thinking of requiring the vaccine for entry into middle school.
= Neisseria gonorrhoeae, the causative agent or gonorrhea, is one of the most
common sexually transmitted pathogens worldwide. Although a robust
inflammatory response ensues during symptomatic infection, no apparent
protective immunity is developed following infection, as shown in a male
human challenge study and by the high incidence of recidivism among
patients attending sexually transmitted disease clinics. The search for a
vaccine against gonorrhea has been largely disappointing. In human vaccine
-102-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
trials, partially lysed gonococci, purified pilin, and purified porin were
shown
to be immunogenic, but all failed to elicit protection upon subsequent natural
exposure. The lack of protective immunity is likely due, in part, to the
capacity of many gonococcal surface antigens to undergo high-frequency
phase and antigenic variation.
= Individuals infected with Treponemapallidum subsp.pallidum develop
specific immune responses that are able to clear millions of treponemes from
sites of primary and secondary syphilis. Despite the fact that humans develop
robust immune responses against T. pallidum, they can be infected multiple
times. The response is a T-cell mediated delayed-type hypersensitivity
response in which T cells infiltrate syphilitic lesions and activate
macrophages
to phagocytose antibody-opsonized treponemes. How treponemes from
heterologous isolates can evade the recall response of a previously infected
individual is unknown. Data from animal studies suggest that both antibodies
and T cells play a role in protection but neither alone prevents infection. It
is
possible that antigenic diversity of T. pallidum accounts for the lack of
heterologous protection. The T. pallidum repeat protein K (TprK) is a strong
candidate for a treponemal factor involved in immune evasion. Epitope
mapping studies revealed that, during experimental infection, T cells are
directed to the conserved regions of TprK, while the antibodies are directed
to
the variable regions.
= A safe, effective prophylactic human immunodeficiency virus (HIV) vaccine
is urgently needed to curb the current AIDS epidemic. There are currently 40
million individuals in the world infected with HIV, and nearly 16,000 new
-103-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
infections occur worldwide each day. Effective HIV-1 vaccines must be
capable of protecting immunized individuals from infection with a broad array
of diverse viral variants. Attempts to develop a safe and effective AIDS
vaccine have been slowed, in part, by the difficulty in clearly defining
specific
immune responses that can prevent infection and limit disease progression.
This is in part due to the poor immunogenicity of the envelope glycoprotein,
the tremendous variability of the virus, its ability to evade and impair the
host's immune system, and its ability to persist by integrating into the
host's
immune system, and its ability to persist by integrating into the host's
genome
of a number of different cell types. It is generally believed that an
effective
HIV-1 vaccine must be capable of inducing neutralizing antibodies as well as
strong cell-mediated immune responses in outbred populations.
= Group B Streptococci (GBS) emerged dramatically in the 1970s as the leading
cause of neonatal infection and as an important cause of maternal uterine
infection. The burden from GBS disease in elderly persons has also increased.
In 1996, the first national consensus guidelines were released. Since then,
there has been a 70% reduction in early-onset neonatal GBS infection. In
2002, new national guidelines were released recommending:
o solely a screen-based prevention strategy
o a new algorithm for patients with penicillin allergy
o more specific practices in certain clinical scenarios
Yet clinical issues remain, including implementation of new diagnostic
techniques, management of preterm rupture of membranes, use of alternative
-104-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
antibiotic approaches, improvement of compliance, prevention of low birth
weight infants, emergence of resistant organisms, and vaccine development.
= Urinary tract infections (UTIs) are a leading cause of morbidity and
mortality
and health care expenditures in persons of all ages. Sexually active young
women are disproportionately affected, but several other populations,
including elderly persons and those undergoing genitourinary instrumentation
and catheterization, are also at risk. UTIs are the leading cause of gram-
negative bacteremia (Orenstein and Wong, 1999).
= Lymphocytes are the effector cells of acquired immunity. Two T helper
subsets are Thl and Th2, based on two distinct cytokine profiles that resulted
in the overall regulation of the immune response. The ThI cell (with its
associated cytokines: INF-y, TNF-a, IL-2, IL-12) is biased towards the cell-
mediated side of immunity, effective against intracellular parasites, and its
down regulation of Th2 can provide relief from allergic reactions due to IgE;
but detrimental effects may result in autoimmunity and graft rejection. On the
other hand, the Th2 cell (with its associated cytokines IL-4, IL-5, IL-6, IL-
10,
IL-13) favors humoral immunity, providing an effective correlate of protection
for most vaccines, and its down regulation ofThl can result in some benefit of
tolerance to prevent cellular autoimmune reactions; but certain harmful
characteristics related to IgE-based allergies and autoimmunity may result. In
order to diagnose or predict an immunologic disease and/or provide therapy or
prophylaxis, the Th polarization status must be determined; this should also
be
applied to measure susceptibility to infectious and neoplastic diseases. Th
status is measurable in terms of cytokine profiles, chemokine/chemoattractant
-105-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
receptors, specific effector cell products, or gene expression profiles. An
exemplary diagnostic panel is described in the table below:
Th1 Th2
Cytokines Receptors Cytokines Receptors
INF-y CCR5 IL-4 CCR3
TNF-a CXCR3 IL-5 CCR4
IL-2 CCR1 IL-6 CCR8
IL-12 IL-10 CRTh2
IL-13
-106-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
2. Exemplary ImmunoScore Diagnostic Panels for Women of Child-Bearing
Years
Adult immunization rates have fallen short of national goals partly because of
misconceptions about the safety and benefits of current vaccines. The danger
of
misconceptions is magnified during pregnancy when concerned physicians are
hesitant to administer vaccines and patients are reluctant to receive them.
Routine
vaccines that are generally safe to administer during pregnancy include
diphtheria,
tetanus, influenza, and hepatitis B. Other vaccines, such as meningococcal and
rabies,
may be considered. Vaccines that are contraindicated, because of the
theoretical risk
of fetal transmission, include measles, mumps and rubella; varicella; and BCG.
A
number of other vaccines have not yet been adequately studied; therefore,
theoretic
risks of vaccination must be weighed against the risks of disease to mother
and fetus.
The administration of vaccines during pregnancy poses a number of concerns to
physicians and patients about the risk of transmitting a virus to a developing
fetus.
This risk is primarily theoretical. No evidence exists of risk from
vaccinating
pregnant women with inactivated virus or bacterial vaccines or toxoids (CDC,
2002).
Physicians should consider vaccinating pregnant women on the basis of the
risks of
vaccination versus the benefits of protection in each particular situation,
regardless of
whether live or inactivated vaccines are used (Sur, et al. 2003). Generally,
live-virus
vaccines are contraindicated for pregnant women because of the theoretical
risk of
transmission of vaccine virus to the fetus. The following table summarizes
recommendations for vaccines commonly administered and their indication for
use
during pregnancy.
Table 11: Immunizations During Pregnancy
-107-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Considered safe if otherwise Contraindicated Special recommendations
indicated during pertain
pregnancy or safety
not established
Tetanus and diphtheria toxoids BCG* Anthrax
(Td)
Hepatitis B Measles* Hepatitis A
Influenza Mumps* Japanese encephalitis
Meningococcal Rubella* Pneumococcal
Rabies Varicella* Polio (IPV)
Typhoid
Vaccinia*
Yellow fever*
* = Live, attenuated vaccine
Women in their second and third trimesters of pregnancy have an increased risk
of
influenza-related complications including pneumonia and a four-fold risk of
hospitalization (Neuzil, et al. 1998). The CDC has recommended that women who
will be in the second or third trimester during influenza season and all
pregnant
women with additional high-risk medical conditions should receive vaccination
in the
fall. Despite publication of these guidelines, rates of vaccination among high-
risk
patients remain low (Silverman and Greif, 2001; Schrag, et al. 2003). Many
possible
explanations exist for this discrepancy, including vaccine unavailability,
logistical
concerns, poor reimbursement, fear of side effects, and lack of adequate
patient or
physician education (Wallis, et al. 2004).
-108-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
A number of maternal conditions were perceived as potential contraindications
to
influenza vaccination during pregnancy. The most common of these were the
first
trimester, history of preterm labor, history of intrauterine fetal demise, and
pregnancy
induced hypertension; none of these are listed by the CDC as contraindications
(Wallis, et al. 2004). According to this group, another potentially
significant obstacle
to influenza vaccination during pregnancy was physician reimbursement. Several
responders remarked that reimbursement from insurance companies played a part
in
whether they stocked the vaccine in their offices and whether it was
administered to
pregnant patients. Although they acknowledged the indications for the vaccine,
some
obstetricians stated that insurance plans have refused reimbursement for
vaccination
because they were not the patient's primary care provider for this
"preventive"
service. Although patients may still be instructed to obtain vaccination
elsewhere,
this additional obstacle to recommended obstetrical care may result in lower
immunization rates. These authors concluded by stating that further research
is
needed to determine effective methods of increasing vaccination rates in this
high-risk
population.
Cytomegalovirus (CMV) is found universally throughout all geographic locations
and
socioeconomic groups, and infects between 50-80% of adults in the United
States by
40 years of age. CMV is also the virus most frequently transmitted to a
developing
child before birth. The incidence of primary CMV infection in pregnant women
in
the U.S. varies from 1-3%. Healthy pregnant women are not at special risk for
disease from CMV infection. When infected with CMV, most women have no
symptoms and very few have a disease resembling mononucleosis. It is their
unborn
babies that may be at risk for congenital CMV disease. CMV remains the most
-109-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
important cause of congenital viral infection in the U.S. For infants who are
infected
by their mothers before birth, two potential problems exist:
1. Generalized infection may occur in the infant, and symptoms may
range from moderate enlargement of the liver and spleen (with
jaundice) to fatal illness. With supportive treatment most infants with
CMV disease usually survive. However, from 80-90% will have
complications within the first few years of life that may include
hearing loss, vision impairment, and varying degrees of mental
retardation.
2. Another 5-10% of infants who are infected but without symptoms at
birth will subsequently have varying degrees of hearing and mental or
coordination problems.
However, these risks appear to be almost exclusively associated with women who
previously have not been infected with CMV and who are having their first
infection
during pregnancy. There appears to be little risk of CMV-related complications
for
women who have been infected at least six months prior to conception. The
current
recommendations from the CDC for pregnant women with regard to CMV infection
are:
1. Throughout the pregnancy, practice good personal hygiene,
especially hand washing with soap and water, after contact with
diapers or oral secretions (particularly with a child who is in
day care).
-110-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
2. Women who develop a mononucleosis-like illness during
pregnancy should be evaluated for CMV infection and
counseled about the possible risks to the unborn child.
3. Laboratory testing for antibody to CMV can be performed to
determine if a woman already had a CMV infection.
4. Recovery of CMV from the cervix or urine of women at or
before the time of delivery does not warrant a cesarean section.
5. The demonstrated benefits of breast-feeding outweigh the
minimal risk of acquiring CMV infection from the breast-
feeding mother.
6. There is no need to either screen for CMV or exclude CMV-
excreting children from schools or institutions because the virus
is frequently found in many healthy children and adults.
Recently, it was found that hyperimmune globulin therapy in pregnant women was
associated with a significantly lower risk of congenital CMV disease (Nigro,
et al.
2005). This group concluded that treatment of pregnant women with CMV-specific
hyperimmune globulin is sage, and their findings suggested that it may be
effective in
the treatment and prevention of congenital CMV infection.
Specific ImmunoScore diagnostic panel recommendations must take into account
the
woman of child-bearing years status with regard to pregnancy. Ideally, an
ImmunoScore screening of a young women prior to child-bearing years would give
an
appropriate "baseline" reading of that individual. In this instance, for
example, a
positive serologic test for CMV would bean indication that CMV-Iike illness
during
-111-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
pregnancy would not be a cause of concern regarding transmission to that
mother's
infant during a pregnancy later in that woman's life.
Clearly, women of child-bearing years that are not pregnant, or not planning
to get
pregnant in the six months following ImmunoScore screening would have
different
recommendations than pregnant women. An ideal location and time for
ImmunoScore diagnostic screening women of child-bearing years would be during
their annual recommended visit to the OB/GYN. An early baseline could be
achieved
for each patient and the Specialist could make use of the specific
recommendations
without confusion as to which immunizations would be appropriate. It is very
important to assure immunity to the components of the measles-mumps-rubella
vaccine prior to pregnancy and the ImmunoScore service would enable that
assurance.
Accordingly, in exemplary embodiments of the present invention a Women of
Child-
Bearing Years ImmunoScore superpanel can be defined as follows.
2.1. Recommended tests for ImmunoScore Measurement of Immunity:
= Antibody to Cytomegalovirus (1)
o History of CMV infection needs to be captured to complete
ImmunoScore database and add relevance to pregnancy.
= Pregnancy test (1)
A pregnancy test is critical to making the correct decisions regarding
administration
of vaccines to women of this age group. There are, of course, other
considerations
here, but the status of the woman in question regarding pregnancy must be
resolved in
order to make accurate therapeutic decisions. In addition to CMV antibody, the
physician(s) of women of child bearing years need to be aware of the
recommendations of the CDC regarding immunizing pregnant women and the risks
of
immunization vs. the risks of foregoing immunizations. In addition, physicians
-112-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
should be aware that following appropriate immunization protocols and assuring
a
competent immune status is extremely important for women of child-bearing
years.
2.2. Persistent Immunity Induced by Childhood Vaccines Diagnostic Panel
Described above.
2.3. Sexually Transmitted Disease (STD) Diagnostic Panel
Described above.
EXEMPLARY IMMUNOSCORE SYSTEM DATABASES
A. GENERAL OVERVIEW
In exemplary embodiments of the present invention the results of immunologic
and
other assays of an individual together with additional medical, lifestyle,
environmental and other demographic information can be collected at the same
time
as, or derived from, the collected data, and can, for example, be stored in a
system
database. Such a database can, for example, serve as an electronic record of
immune
status and other data over a period of time, both for individuals as well as
for
populations or sub-populations, as described below. Additionally, for example,
such
a database can be augmented with information regarding diagnoses received,
treatments administered, pharmaceuticals prescribed, costs of medical services
perfomed, insurance re-imbursements, metrics as to the efficacy of treatments
and/or
pharmaceuticals administered, as well other relevant information to facilitate
evaluation of the efficacy and efficiency of medical services rendered, as
described
more fully below.
Thus, for example, for each run of an exemplary ImmunoScore assay within an
exemplary system, various categories of data can be collected. Data can, for
example,
-113-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
be stored in an electronic database using standard techniques as are known in
the art.
An example of data which can be stored and the manner in which it can be
stored is
next described. It is understood that this example is not intended to preclude
the
storage of additional collected or derived data as may prove useful for the
purposes of
trending, data mining, evaluation or diagnostic improvement, as described
below, or
as may be needed in or useful to any of the exemplary applications described
in
Section III below.
For each assay an exemplary system can record a unique assay ID, which can
incorporate, for example, among other information, an identifier for the assay
instrument. This ID can be unique over the universe of instruments, ensuring
that
when data is aggregated into a central system no two assay result records will
have
the same identifier. A possible implementation of this functionality is given,
for
example, by Microsoft's use of the GUID (Globally Unique Identifier), a 16
byte
identifier generated by a computer and guaranteed to be unique across all
computers.
Each record can include the time and date that the assay was performed, stored
to a
time resolution of, for example, one second. As is known, there are a variety
of
standard means of storing time and date information in a database. One simple
means
is, for example, to record the number of seconds from an arbitrary start time,
such as,
for example, January 1s`, 1900 at midnight.
Each record can, for example, also include an indication of the location where
the
sample was processed. This can include, for example, an identifier of the
instrument
used, as well as real-world location information, such as, for example, the
name and
address of the facility where the instrument has been installed.
-114-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The aforementioned exemplary fields comprise identification information which
is
important to maintain for all samples. In addition, information about the
sample and
patient can be stored in the database as well. Patient information can, for
example, be
stored in a form which is separate from the bulk of the data, and referenced
by a data
link. Patient information, which can include, for example, name, social
security
number, birth date or other information (such as is described below in
detail), can be
maintained with emphasis on security standards are known in the art. The
storage of
identifiable individual patient information in a separate virtual location
from the
remaining data can help to maintain such a high level of security.
In exemplary embodiments of the present invention, a system can, for each
assay
result, also store an identifier indicating exactly which assay was performed
on the
sample. This can indicate not only the analyte to be determined, but also
information
regarding the production of the reagents used in the assay. This information
can be
used to distinguish between, and compensate for, for example, lot-to-lot
variations in
assay manufacture. It can also allow for converting different assays for the
same
analyte into a normalized value, so that trends across geography as well as
time can be
obtained.
The measurement of an immune response to a particular disease or other analyte
can
involve the collection of a large quantity of low level data generated by an
instrument.
For an ECL instrument, for example, an instrument can measure the light
emitted
from the electrochemiluminescence over some time period as well as other
information such as voltages and currents used to induce the
electrochemiluminescence and the temperature near the electrodes through which
the
electrical energy is delivered to drive the electrochemiluminescent reaction.
From
-115-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
this "raw data" and possibly instrument calibration information, a single
number, for
example, can be computed to represent an ECL signal for that measurement.
Additional information can be computed from the raw data and instrument
calibration
information that indicates the quality of the ECL signal, for example, whether
the
instrument was operating in an appropriate environmental condition, whether
sample
was present, or whether the instrument was operating as expected. The raw data
and
such derived data can, for example, be stored in an exemplary ImmunoScore
system
database. In general the size of the storage required for this raw data can
vary
depending upon the resolution at which the data is captured. It is possible
that a finer-
grained resolution, resulting in a larger data storage requirement, will yield
more
useful analysis for some assays rather than others. Storage of both the raw
data and
the derived values can be done, for example, using industry-standard methods
for the
persistence of floating point numbers. For example, four (4) bytes of storage,
yielding
approximately six (6) significant digits, can be used for each stored value.
The quantity of greatest interest in an assay is the concentration of the
analyte under
evaluation. This concentration can be determined by converting a computed ECL
signal to a concentration. This conversion can be done, for example, by
backfitting
the ECL signal through a calibration curve that relates ECL signal to analyte
concentration. In general, such a calibration curve can vary from assay to
assay, and
can change over time for a given assay as that assay is refined.
Calibration curves enable both interpolation and extrapolation of ECL signal
measurements for samples with known analyte concentrations for ECL signal
measurements of samples of unknown amounts of analyte. The form of the
mathematical functions used in a curve fit can, for example, make assumptions
-116-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
regarding the continuity and/or smoothness of the underlying relation such as
through
interpolating the measurements with functions such as piecewise constant,
piecewise
linear, cubic spline, or for example, by throughfitting all the data with
linear,
quadratic, cubic, or quartic polynomials. For overconstrained systems,
parameters
can be computed by minimizing an error function such as, for example, least
squares
(e.g., Press et al. 1992) or total least squares (e.g., Van Huffel et at.
1991). The form
of the mathematical function may make assumptions about the assay mechanism,
such
as a one site saturation, two site saturation, one site saturation with
nonspecific
binding, two site saturations with nonspecific binding, a sigmoidal dose
response
curve with or without a variable slope, one-site competition, two-site
competition, or
a four-parameter logistic. Generation of a calibration curve entails selecting
the form
of the mathematical function and then fitting the parameters of the function
with
measurements. The measurements can, for example, be done on the test
instrument or
can be done in whole or in part elsewhere (e.g., at the place the assay is
manufactured). The measurements can either perfectly constrain or over-
constrain the
mathematical function. As noted, for overconstrained systems, model parameters
can
be computed by minimizing an error function such as least squares.
In exemplary embodiments of the present invention, for each analyte the form
of the
mathematical function or model (stored, for example, as an index into a table
of
known models), the computed model parameters, as well as the data used to
compute
the model parameters, can be associated with each measurement of the analyte.
To
reduce the amount of redundant information stored in the database, the
association for
each measurement can be a link to the calibration data rather than the
calibration data
itself. Instruments can be re-calibrated at any time, such as, for example, on
a weekly
basis or with every measurement. The quality of the calibration can also be
assessed,
-117-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
for example, through the running of controls or by computing the residual
error from
an overconstrained curve fit.
Thus, a calculated concentration can be stored by the system. This can be, in
exemplary embodiments of the present invention, the primary input to analysis
recommendation algorithms employed by the remainder of the system. It is noted
that
not all assays will result in a quantitative concentration. For example, some
assays,
due to the shape of their calibration curve, may yield two different
concentrations for
the same measured signal. Such assays are said to "hook." In such cases the
most an
exemplary system can store is an indicator that the measured concentration is
above a
certain level, the lower of the two returned calculated values. Other assays,
for various
reasons, may return only qualitative results rather than true quantitative
results. In all
cases, a system database can be capable of storing and retrieving the result.
For this
reason, in exemplary embodiments of the present invention, the result of an
assay can
be stored not as a simple floating point number, but as a complex object which
can
take into account the various scenarios described above. Such an object can
have, for
example, several fields of its own.
A compressed version of the database can, in exemplary embodiments of the
present
invention, consist of only the initial ID information, patient ID information,
test ID
information, and the calculated concentration of analyte. This is a minimal
set of data
which can prove productive for data mining and trending analysis, as detailed
below.
The additional data described herein can, for example, be used to enhance the
value of
this analysis.
Algorithms encoded or implemented or implemented in an exemplary system can be
used, for example, to determine a recommendation for action. This
recommendation
-118-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
can be based upon a calculated concentration of, for example, antibody
response.
Other information can also be considered, including, for example, the results
of other
assays upon the same sample within a given assay panel.
Regardless of the means of determining the recommended action, as described
above,
a final recommendation can be stored in the database. A system database can,
for
example, also store the "reasoning" behind the recommendation, allowing a
human to
later query the database to determine why a given course of action was
recommended.
Given that the number of recommended courses of action can be broad, these
actions
can be categorized and encoded. For example, a recommendation to administer a
particular vaccination may be encoded with one byte to indicate "give
vaccination"
and two additional bytes to indicate the particular vaccination that is
warranted. A
field for comments can also be included, to allow the capture of the system's
reasoning - in this case, an explanation of how algorithms and rules were
applied to
determine the stated conclusion.
A system database according to an exemplary embodiment of the present
invention
can be implemented, for example, as a shared resource spread over multiple
computer
platforms. For purposes of trending and analysis, it may be necessary to
accumulate
the data from a large number of systems into a central repository as depicted
in Figs.
2, or, for example, in the case of having only decentralized information, by
using a
mechanism or process to locate and query the distributed sources. The
individual
databases can therefore require the capability to link up with a defined
central
database and upload their contents. This can occur on a periodic basis, or as
may be
triggered by a user of the system. Additionally, there can be multiple central
servers,
so that a given enterprise may choose to aggregate their data at any level.
Unique IDs
-119-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
associated with sample and panel records can serve to allow for the
combination of
data from disparate sources without data "collision."
The linkage between local databases and a central database can be implemented,
for
example, across a local area network (LAN), a private data network, a VPN, an
intranet or across the Internet. It is also possible to link databases on a
periodic basis
using physical media, such as CD-ROMs. Similarly, various users such as, for
example, health care providers, individuals, insurance executives, consumers
of
research services, health care management personnel, etc., can access an
exemplary
system via a web based interface across a local area network (LAN), a private
data
network, a VPN, an intranet or across the Internet.
Once data has been accumulated into a central repository, a separate system
can be
used to perform data mining and data trending analysis upon the stored data.
There
are many valuable sorts of analyses which can be performed on the accumulated
data
in an exemplary system according to the present invention.
Given that each data record can, for example, be identified with a particular
individual
or patient and a particular time and date, it becomes possible to perform
trending
analysis of a patient's (or a population's) ImmunoScore profile over time. In
many
cases an individual's absolute measured value of an analyte is not as
important as the
trending of that value over a time. Some individuals may have naturally low or
naturally high values which are not best measured against a statistical mean
for their
demographic population, but rather against that individual's own measured
history.
As described above, each patient can, for example, also be placed within
certain
demographic categories. It can be useful to compare a patient's measured
ImmunoScore profile against the corresponding profile for the demographic
groups to
-120-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
which he or she belongs. Deviation from the measured means for a demographic
slice
of the population can prove more meaningful than can a comparison to a total
threshold. Thus, in exemplary embodiments of the present invention, collected
data
can be used to continually modify the demographic profile averages known to
the
system, taking care to not pollute the system with outlying data points. For
example,
it may prove useful to produce separate ImmunoScore demographic profiles for
patients who are known to have experienced vaccinations versus those for whom
there
is no known immunization record. Alternatively, as is described below in
Section III,
such an immunization record can be inferred and reconstructed, as in the
provision of
ImmunoScore services to national immigration services or authorities or bodies
dealing with such concerns.
Trending information in a demographic profile, for example, can also be
useful. For
example, tracking an indication of a typical person (e.g., mean, median, or
mode), or
an indication of the spread amongst people (e.g., standard deviation,
interquartile
range, or range) over time can enable an exemplary system to assess the
relationship
between immune status indicia and external factors, such as, for example,
seasonal
effects. Eating habits, sleeping habits, time aboard ship, etc. can be found
to affect
immune status in groups where these external factors are partially
controllable (such
as, for example, in military personnel). Comparing immune status indicators of
differing demographic profiles can have important epidemiological
significance.
Finally, it is expected that the collection of ImmunoScore data from a large
number of
individuals and/or populations can eventually lead to the improvement of
diagnostic
tests, thus forming a feedback loop. These improved diagnostic tests can then,
for
example, be deployed to field instruments, resulting in more accurate
measurements
-121-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
and diagnoses. Such exemplary embodiments having feedback loops can be
implemented, for example, with respect to particular populations or
demographic
groups, such as, for example, the military, college students, immigrants or
any other
group or combination thereof as described above.
B. EXEMPLARY ILLUSTRATIVE DATABASE
1. Overall Description
To illustrate the systems and methods of the present invention, a database
system was
constructed to serve as a testbed for the exercise of the business models
described
below. Such an exemplary database system was used to demonstrate the tools and
techniques that might be used in a full scale system according to the present
invention. Accordingly, a large data set was constructed using statistical
techniques.
The data was produced according to match existing knowledge about the
distribution
of immune response values among the general population.
The exemplary database system has two primary components. These two components
represent the algorithmically interesting sections that can be, for example,
present in a
full-scale operational system according to an embodiment of the present
invention.
Such a full system could, for example, contain other modules as well, along
the lines
of industry standard large scale database systems. Such an exemplary system is
depicted in Fig. 5 and is next generally described.
With reference to Fig. 5, an exemplary system architecture can be constructed.
The
exemplary system architecture can be, for example, divided into two sub-
systems, one
relatively local to "point of care" or locations where the individuals or
patients whose
immune status is to be analyzed are located. The other subsystem can be in a
central
location where complex data mining and analysis can occur. Thus, with
reference to
-122-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Fig. 5, an upper portion of the figure contains components which can be
located at the
point of care and a lower portion of the figure contains components which can
be, for
example, located at a system central location. The point of care is divided
from the
central location in the figure by a double dotted and dashed line for ease of
identification.
With reference to the point of care sub-system, there can be one or more
Instruments
505 which are devices which can read immunologic assays. Instruments 505 yield
Assay Results 506. Assay Results 506, along with Doctor's Observations 503,
Patient
History 502 and Demographic Information 501 regarding the individual or
patient can
all be stored in Local PatientEvent Database 510. Database 510 can be, for
example,
an online transaction processing database. Because the point of care sub-
system is
generally directed to generating a recommendation in a relatively short time,
there are
two pathways to Diagnostic Module 515. Diagnostic Module 515 applies
algorithmic
rules to the assay results to determine a proper course of treatment or action
based on
current readings and optionally on past history. Thus, there is a flow of
information
from Assay Results 506 to Diagnostic Module 515. Alternatively, Diagnostic
Module
515 can implement algorithms having other inputs besides the current Assay
Results
506, such as, for example, Demographic Information 501, Patient History 502,
and
Doctor's Observations 503 (understood to include any observations by any
health care
provider, or the like, in a general sense) which can be stored in Local
PatientEvent
Database 510. Thus, in Fig. 5, there is an arrow labeled "optional" running
from
Local PatientEvent Database 510 to Diagnostic Module 515. Regardless of which
source of information Diagnostic Module 515 draws upon, it can, for example,
output
the patient action recommendation 516 as indicated.
-123-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Returning to the central location sub-system of Fig. 5, a connection exists
between
Local PatientEvent Database 510 and a Central PatientEvent Database 520. This
connects the two sub-systems. It is contemplated that at regular intervals
data from
Local PatientEvent Database 510 can be uploaded to Central PatientEvent
Database
520. Moreover, although the central location sub-system could be mirrored in a
number of distributed central location subsystems, the point of care sub-
system is
contemplated to take data from numerous instruments and in fact have numerous
local
patient event databases in those locales. In short, the point of care sub-
system is
found wherever potential customers or patients are found. It is noted that
there can be
a myriad of such locations, given the various and sundry applications and
business
models that exemplary embodiments of the present invention contemplate.
Examples
of such applications are described more fully in Section III, below.
Therefore, there
could be a great number of local patient event databases all of which feed
into Central
PatientEvent Database 520. None of these additional point of care sub-systems
are
shown in Fig. 5, for reasons of ease of illustration.
Returning again to Central PatientEvent Database 520, it is noted that this
database
can, for example, also be an online transaction processing database or OLTP.
It is
contemplated that this database can, for example, periodically load data to an
online
analytic processing database, or OLAP in the form of PatientEvent Database
530.
PatientEvent Database 530 can be, for example, adapted to provide inputs to
complicated algorithms dealing with data mining and pattern detection, as next
described.
PatientEvent Database 530 can, for example, reside on a central server and
utilize a
data warehouse approach. There can be a variety of connections to PatientEvent
-124-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Database 530 such as, for example, a Query Module 531, a Data Mining Module
532
and a Pattern Detection Module 533. Query Module 531 can be, for example, an
interface by which a user can interactively search for information in database
530.
Query Module 531 can also access Central PatientEvent Database 520 implement a
variety of operations on the data there as well. Data Mining Module 532 can be
an
interface by which a user can interactively use OLAP tools to finds trends and
summaries in the stored data. Finally, Pattern Detection Module 533 can be a
program module which can be used to automatically search for patterns or other
"hidden" correlations between various data points in a database. It is
contemplated
that in exemplary embodiments of the present invention Pattern Detection
Module
533 can regularly sort through all of the stored data looking for patterns
using various
algorithms. Some of such algorithms can, for example, articulate some hunch or
a
correlative assumption provided by a panel of immunological experts for which
they
do not have hard data. Pattern Detection Module 533 is thus an important
feature in
exemplary embodiments of the present invention. Additional exemplary databases
which Patter Detection Module 533 can utilize are described below in
connection with
Fig. 5A.
The exemplary system depicted in Fig. 5 will next be described in greater
detail. A
first module of interest is termed Diagnostic Module 515. The function of this
software module is to input a set of assay results 506 obtained through
measurements
by instruments 505, and to make one or more recommendations 516 based upon the
analysis of assay results 506. Diagnostic Module 515 can be designed in such a
way
that additional assay panels can be slotted into an existing system as they
are
developed. Some exemplary algorithms used to make recommendations as a
function
of assay results are described in more detail below, including descriptions
both of
-125-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
algorithms used in the exemplary database as well as additional algorithms
that could
be implemented in various exemplary embodiments of the present invention.
Diagnostic Module 515 can rest upon a Local Database 510 containing Assay
Results
506 obtained from Instruments 505. These results are pertinent to an
individual
patient. Local Database 510 can also, for example, contain background medical
history 502 for that patient, demographic information 501 pertinent to the
patient, and
a summary of other medical observations 503 made by medical professionals or
persons fulfilling a similar function. Local Database 510 can also, for
example,
contain statistical information obtained from a larger central database, as
described
below.
A second exemplary module of interest is Data Mining Module 532. Whereas
Diagnostic Module 515 is intended for the analysis of a particular
individual's data at
a particular point in time, Data Mining module 532 can, for example, look at a
broader range of data collected from many individuals over a range, or
interval, of
time. Through analysis of this collected data a system can, for example, be
used to
support various business methods and other applications by deducing trends and
patterns within an immunological landscape. A particular result could be fed
back into
the Diagnostic Module's algorithms, improving their effectiveness by providing
additional specificity with regard to an individual's background, possibly in
terms of
background or demographic information such as, for example, gender, racial
background, geographic origin, lifestyle, economic circumstances social
circumstances, or age.
As can be seen from Fig. 5, while the Diagnostic Module's functionalities are
primarily local in nature and patient-specific, the Data Mining Module's
-126-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
functionalities are primarily central, and system-wide. As noted, this
structure is
reflected in the division of Fig. 5 into two zones, the "Point of Care" zone,
shown at
the top of the figure, and the "Central Location" zone, shown at the bottom of
the
figure.
Data Mining Module 532 depends upon the existence of a large central database
containing records from a wide variety of individuals over a long span of
time. Thus,
the local databases described above can, for example, exist in a federated
state with
the central database, uploading their information on a regular basis, where
this
information can, for example, be integrated into the full system.
2. Impact of Data Mining
Patterns can be detected within the data in an exemplary database which are
related to
demographic and other non-immunologic information such as, for example,
gender,
age, ethnicity, geographic origin, employment, etc. These patterns may not be
obvious until large numbers of individuals are assessed, using a computer that
can be
by nature much more efficient, unbiased, and precise in pattern recognition.
From such patterns, new correlates can, for example, can be established, and
old
correlates can be changed. For example, in immunization related applications,
it may
be proposed, based on previous data, that a serum antibody concentration of 2
micrograms per ml should be used to represent a threshold of protection
against
meningococcal disease, so that anyone with less antibody would be recommended
for
immunization. Subsequent and continued analysis, however, may show that this
threshold value should be reduced or raised for given individuals, depending
on, for
example, age or ethnic background, or some other undefined parameter. In turn,
an
ethnicity evaluation could lead to the discovery of a specific biological or
genetic
-127-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
marker. For example, the functional activity of Haemophilus influenzae type b
(Hib)
antibodies may vary with different individuals, where the same antibody
concentration may not possess the same level of bacteriocidal activity due to
differences in antibody avidity. For example, regarding age, Hib
polysaccharides
have been shown to be poorly immunogenic in children less than 2 years of age
(Granoff DM, 1985, J Pediatr 107:330-36). Similarly, regarding ethnicity it
has been
shown from previous studies that Eskimos and Apaches are more susceptible to
Hib
meningitis because they possess a less effective antibody repertoire to the
Hib
polysaccharide capsule, based on the presence or absence of certain variable
region
genes used in the production of the polysaccharide-specific antibodies.
Additionally, variations in host factors can lead to significant differences
in the
immune response to vaccines, which can also be discerned by data mining. For
example, late-stage complement deficiency may have no impact on antibody
production, but would certainly reduce the effectiveness of those antibodies
in killing
bacteria, thereby lowering their activity. In such case, the antibody
threshold for
protection may need to be raised in order to achieve the same level of
protection in
this subpopulation.
As previously described for Hib, the capacity for protective antibody
production is the
direct result of variable region gene haplotypes. In this case, ethnic
differences were
first observed as a gross marker, but the presence of specific genes was later
determined to be responsible. In a similar but different manner, HLA
haplotypes have
also been correlated with the susceptibility to certain infections, as well as
the
unresponsiveness to certain vaccines. For example, certain HLA antigens appear
to
be correlated with chronic hepatitis B virus (HBV) infections and HBV vaccine
-128-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
nonresponsiveness. In such cases, in exemplary embodiments, of the present
invention, subpopulations can be identified, initially by ethnicity, then
later by
genetics, to evolve a more specific and appropriate diagnostic outcome.
Another example of the influence of ethnicity on responsiveness to treatment
is the
case of NitroMed's BiDiITM, which was approved by the U.S. FDA in 2005 for the
treatment of heart failure in African Americans. BiDiITM is an orally
administered,
nitric oxide-enhancing drug that was shown to have clearly different effects
on blacks
versus whites in clinical trials, where the "differences may be related to
environmental, social, lifestyle, or genetic factors or to interactions among
all of
these." (see http://www.fda.gov/fdac/features/2005/505_BiDil.html). In
exemplary
embodiments of the present invention, data mining can, for example, be used to
observe and identify these kinds of effects and correlations, and then be
later used to
determine the specific underlying mechanisms.
Data mining can also be used, for example, to change or reverse previously
held
dogma(s) concerning long-term protection from vaccination. For example,
immunity
resulting from the smallpox vaccine, used extensively during the previous
century,
was originally thought to last for less than a decade. Recent analyses
however, have
shown that "more than 90% of volunteers vaccinated 25-75 years ago still
maintain
substantial humoral or cellular immunity (or both) against vaccinia, the virus
used to
vaccinate against smallpox." (Hammarlund E et al., 2003, Nature Medicine
9:1131-
37). The same study further showed that "Antiviral antibody responses remained
stable between 1-75 years after vaccination, whereas antiviral T-cell
responses
declined slowly, with a half-life of 8-15 years." While it is not clear what
level and
combination of responses is required for protection, the authors concluded
that "the
-129-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
morbidity and mortality associated with an intentional smallpox outbreak would
be
substantially reduced because of pre-existing immunity in a large number of
previously vaccinated individuals." This is exactly the type of information
that could
be obtained through data mining over time on large populations, as
contemplated in
exemplary embodiments of the present invention.
As noted above, an exemplary system similar to that depicted in Fig. 5 was
built using
standard software development tools and packages. The algorithms were encoded
using the XML data description language. The engine for executing the
algorithms
was built using the Java programming language. An Oracle database was used for
data storage and data mining querying. Excel spreadsheets were used for data
construction and analysis. Details of the construction are given below.
3. Diagnostic Module
3.1. Overview
Diagnostic Module 515 forms the heart of an exemplary ImmunoScore decision
system. At a basic level, the diagnostic module exists to provide relevant
information
and/or to suggest courses of recommended action (for various purposes,
depending
upon the application; see Section III below) based upon an individual's immune
status, as measured by instrumentation or obtained from elsewhere, in
combination
with other supporting data. There are many different ways that such a
determination
could be made. Next described are some exemplary algorithms that were used in
the
example system as well as other exemplary decision support algorithms which
could
be implemented using the same techniques.
One essential function of a diagnostic module can be, for example, to assist a
medical
or other professional in making decisions regarding which actions to take with
a
-130-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
specific individual, making use of data regarding that person's immune status.
As
noted, in exemplary embodiments of the present invention, an individual's
immune
status can be determined by conducting a panel of assays, each of which assays
can
produce an element of data. For purposes of the example database, information
presumed to be obtainable through such assays is summarized in Fig. 6. It
should be
noted that in practice some of this information may not yet be obtainable,
although it
is expected that assays could be developed along the lines of existing tests
in order to
complete this spectrum.
In addition to immune status information obtained from assays, a diagnostic
module
can make use of other information specific to the patient being examined. This
information falls into two principal categories: demographic information, such
as, for
example, age and gender, and patient medical history. Most demographic
information
can be simply expressed in a database. Patient medical history is more
problematic,
although there are many existing healthcare database systems which do this
adequately. The difficulty with patient medical history, however, is in
devising
algorithms which can make use of this qualitative data. It is expected that
particular
care can be taken to use algorithmic techniques which have proven adept in
dealing
with inconsistent or unreliable data, such as, for example, neural networks,
described
in greater detail below. This is due to the inherent unreliability of self-
reported
medical history data, along with the historic problems found in the transfer
of medical
records. If a system with built-in reliability checks is implemented, then it
can be
possible to rely more strongly upon historical data.
Thus, the exemplary system described below can store both demographic and past
medical history information for individual patients, but does not make use of
these
-131-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
factors in performing diagnostic assessments or recommendations of courses of
action. However, the algorithms implemented can easily be extended into these
realms once more information becomes available.
The output of Diagnostic Module 515 can be, for example, a series of
recommendations. A recommendation is simply defined as any discernible bit of
data
which might be of interest to a medical professional, health care or life
insurer,
medical services analyst, researcher or other user of the present invention in
determining a given course of action. In the case of a patient's immune
status, a
common recommendation could be, for example, to recommend a particular
vaccination, to conclude whether the individual is in an overall sense
healthy, to
conclude that certain potential hypotheses need further data to be fully
explored, to
tag the individual as being potentially immunosenescent, or to grant a health
insurance credit or debit relative to a health insurance policy or HMO
membership
fee. Or, for example, a recommendation not to vaccinate, to reduce the over-
vaccination of the populace. A summary of some exemplary types recommendations
that can be offered by an exemplary Diagnostic Module are provided in Fig. 7.
In exemplary embodiments of the present invention a Diagnostic Module can be
capable of producing a set of recommendations for each analysis. For example,
it
might recommend that both vaccine V be administered and that the individual be
retested in three weeks to monitor his or her response to such vaccine. For
each
recommendation, an exemplary Diagnostic Module can, for example, also provide
a
confidence level, which is a measure of the system's support for any given
conclusion. A user can take this confidence level into account when deciding
upon a
course of action. A course of action with a low confidence level but a high
financial
-132-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
cost, for example, could be delayed until additional data could be gathered to
more
strongly support the course of action.
In exemplary embodiments of the present invention a Diagnostic Module can, for
example, be constructed in a manner to allow the deployment of many different
algorithms within its basic shell. For the exemplary system, an algorithmic
approach
based upon perceptrons was used. This approach is detailed below. Additionally
described are alternative algorithmic approaches, each of which has different
strengths and weaknesses. It is noted that some of these approaches are
realistically
infeasible until such time as large-scale data collection of immune status
informatics
becomes available.
3.2. Perceptron algorithms
A perceptron is a simple neural network, a computer science representation
based
upon an analogy with the operation of human neurons. Perceptrons were invented
by
Frank Rosenblatt in 1957, and have been used in artificial intelligence
research since
that time. A perceptron is simplistic, but adequate for the computation of
algorithmic
diagnostic results within the exemplary system of the invention. More
importantly,
there is a clear progression between perceptrons and more sophisticated
artificial
intelligence techniques, which may be of use in more complex embodiments of
the
invention.
An example of a perceptron is given in Figs. 8 and 8A. These networks encode
the
decision making process for the running of a Meningococcal Diagnostic Panel,
as
described above. There are seventeen inputs to the algorithm, one for each of
the
measurements that can be taken in an exemplary meningococcal assay panel. Five
inputs are for the meningococcal serogroups, seven for the complement
components,
-133-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
and five for the genetic poymorphisms. There are two output recommendations
from
this panel R1 810 (or in Fig. 8A, R2 810) and R3 840. R1/R2 is a
recommendation to
vaccinate an individual with a meningococcal vaccine. R3 840 is a
recommendation
to monitor the individual on a stricter interval schedule than normal, because
the
individual may be more susceptible to this condition than the average
individual in the
populace. Figs. 8 and 8A depict the same prerceptron, with different values
for the
various nodes upon firing.
With reference to Fig. 8, serum IgG levels for vaccine-preventable serogroups
(A, C,
W-135, and Y) of Neisseria Meningitis can be assessed. As seen in the fifth
input to
RI, the panel also has a built-in facility to measure and consider serogroup
B, but
there is no currently available vaccine or clearly known threshold of
protection for
this serogroup, so it was left blank. A serum IgG level exceeding 2.0 ug/mL
for all
four serogroups would be presumptive of protection in an otherwise healthy
individual, i.e., an individual (i) found not deficient in serum levels of
measured
complement components, and (ii) having no deleterious genetic polymorphisms as
indicated in the CC Test 820 and Genetic Polymorphism Test 830. There would be
no immediate recommendation for meningococcal vaccination for these
individuals.
The following is a description of rule execution flow for the exemplary
perceptron of
Figs. 8 and 8A.
RI - Recommend Vaccination. With reference to Fig. 8,If the CC Test 820 and
the
Genetic Poly Test 830 show the person is normal, both of them will fire,
giving a
minimal total of 2.0 at R3. Then no contribution at RI from R3, and if any of
the
serogourps is deficient, RI will be at least =1.0 and RI will fire. If the CC
Test 820
or the Genetic Poly Test 830 show that the person is not normal, R3 840 will
fire,
-134-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
giving a base total of -4Ø Nothing will be contributed from the R3
conclusion as
even if the inputs to R1 810 from the four serogroup assays are all 1.0 (all
deficient),
this added to -4.0 = 0, which is <1.0, and R1 needs to be >= 1.0 to fire. Thus
Fig. 8
only operates as to normal individuals vis-a-vis the CC and Genetic Poly
tests.
R3 - Recommend Flagging. If the total at R3 840 is less than 2.0, the
individual is
not normal, R3 fires and the recommendation will be to flag this individual
for
monitoring.
Fig. 8A is similar to Fig. 8, except that it applies a different recommend
vaccination
rule, R2 at 810, for a different immunological context. The perceptron is
modified as
to values, but the nodes are identical.
R2 - Recommend Vaccination. With reference to Fig. 8A, if deficiencies were to
be
revealed in any of an individual's complement components, or if any
unfavorable
genetic polymorphisms were shown to exist, then it is likely that a serum IgG
level of
> 5.0 ug/mL (not the > 2.0 UG level as in the rule of Fig. 8) for the vaccine-
preventable serogroups would be desirable in these individuals. If these
individuals
had IgG levels exceeding 5.0 ug/mL for all four serogroups, no vaccination
would be
recommended. If the level of antibody to any of the four serogroups were to be
below
5.0 ug/mL, then a vaccination would be recommended. If the CC Test or the
Genetic
Poly Test show the person is not normal, one of them will fire, giving a
minimal total
of 10 at R2. Then, all that is required is for one of the serogroups to be
deficient (i.e.,
< 5.0 ug/ml) in order for the recommendation at R2 to evaluate to true.
R3 - Recommend Flagging. If the CC Test and the Genetic Poly Test show the
person is normal, both of them will fire, giving a minimal total of 2Ø If
the total is
-135-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
less than 2.0, R3 fires, they are not normal and the recommendation will be to
flag
this individual for monitoring.
Because al perceptrons operate on the data in parallel, an abnormal individual
can, for
example, be captured in the perceptron of Fig. 8A and can thus receive no
vaccination
recommendation from the perceptron of Fig. 8.
A perceptron operates through software by simulating the "firing" of nodes
based
upon numerical conditions being met. As each node fires, it can contribute to
the
firing of other nodes, in some cases positively and in some cases in an
inhibitory
fashion. The network as a whole has completed execution when the rightmost
nodes,
representing diagnostic recommendations, have either fired or have come to
rest.
The perceptrons in the exemplary system were encoded manually based upon
existing
knowledge of diagnostic recommendations in use today. Each perceptron can be
represented either graphically, as in Fig. 8, or textually, as in Fig. 9. Fig.
9 is thus a
textual representation of the perceptron network using a language called XML,
or
eXtensible Markup Language. In the exemplary these XML files can be deployed
to
the diagnostic module as discrete packets. An exemplary Diagnostic Module
connected to an instrument, or bank of instruments, could, for example, be
configured
with only those perceptron algorithms required for that site.
In addition, updated versions of these algorithms could be deployed as the
algorithms
are improved over time in a continuous process of system learning or
iteration. Thus,
in exemplary embodiments of the present invention, knowledge gained through
use of
the data mining module, detailed below, can be fed back into the individual
diagnostic
modules, thus improving the accuracy of the entire system. For example, it may
be
deduced through data mining of an exemplary database that the level of
antibody
-136-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
activity which is a strong indication of the need for vaccination is lower in
men than
in women. A new perceptron algorithm could then be deployed, for example,
including the gender of the patient as a new input node, with a link to the
vaccination
recommendation node.
More subtly, a perceptron can include within it a series of weights which can,
for
example, correspond to the importance of each bit of evidence to the
recommendation
procedure. Over time these weights can be continually adjusted and redeployed
to
reflect increased understanding of the role of each of the immunological
factors being
measured.
3.3. Alternate algorithmic approaches
There are a number of alternate algorithmic approaches which can be used
within a
Diagnostic Module. Each has varying strengths and weaknesses. An exemplary
system can, for example, include a combination of these approaches in order to
come
up with the most complete recommendation for a course of action.
The process of evaluating algorithmic approaches involves a consideration of
the
goals which are to be met. A Diagnostic Module can, for example, be configured
to
optimize for any one of a number of different criteria. Possible goals can
include, for
example, optimizing the welfare of the patient, minimizing costs for the
patient
related to the disease in question, minimizing overall patient healthcare
costs, and
minimizing life insurance costs. The decision algorithm used in the diagnostic
module
can thus vary depending on how these goals are prioritized.
A key difference between a system according to the present invention and
existing
systems is the use of an individual's immune status information and associated
data as
inputs to the decision procedure. This allows the system to provide more
tailored and
-137-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
individualized recommendations instead of relying upon aggregate statistical
measures. A second key difference is the introduction of historical patient
immune
status and other data. It is possible, for example, that a given individual's
antibody
level is below some computed norm, but is in fact high in relation to that
individual's
past results. This might conventionally be, for example, a contraindication
for
vaccination, a recommendation which would not be made if the individual's
immune
status were only to be compared to the population standards.
Using the exemplary symbology laid out in Fig. 10, various diagnostic goals as
shown
in Fig.1 I can be summarized.
3.4. Additional input data
This section describes additional data which could be incorporated into a
diagnostic
module in exemplary embodiemtns of the present invention.
As noted above, historical immune status information can be a useful addition.
Basing
a recommendation solely upon an individual's status at the current point in
time is an
adequate approach, but it risks making incorrect recommendations for those
patients
who do not fall within the average range of the population at large. A simple
extension to the system would be to move away from absolute measures of, for
example, antibody level and antibody activity level, and to substitute instead
relative
measures based upon the percent change in these values since the last
historical
measurement, or in comparison to the individual's historical averages. The
same
decision procedures could be applied, but retooled so that a decision rule
such as "the
level is greater than 30" becomes "the level is greater than 15% above the
patient's
baseline". In order for this to occur, an exemplary system can either maintain
a central
record of the patient's immune status over time, or provide means to allow the
-138-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
portable storage and transfer of this historical record, perhaps under the
patient's
control. Various forms of "smartcard" or electronic storage technologies as
are known
could be used for this purpose.
A second type of additional input data relates to demographic information.
Current
decision procedures do little to distinguish treatment recommendations based
upon an
individual's age, gender or racial background, although it is known that these
factors
have a considerable effect on the interpretation of immune status information.
Thus,
an exemplary system could make use of such demographic information,
customizing
the diagnostic algorithms to take into account observed patterns. Additional
research
would be required to deduce these patterns in the population as a whole in
order to
make reasonable modifications to the decision procedures.
3.5. Decision rule algorithms
A clear successor to the perceptron approach could be to extend the system to
full
neural networks. The distinction between perceptrons and more complex neural
networks is the incorporation into the latter of feedback links from later
nodes to
earlier nodes in the network. This not only increases the complexity of the
algorithms
which can be implemented, but allows for algorithms which improve over time
through a learning mechanism. Neural networks are a well-established domain of
artificial research. The primary impediment to neural networks is that they
are
difficult to construct by hand. A typical neural network is instead evolved
through the
use of training algorithms. These training algorithms require as input a set
of training
data. In an exemplary embodiment of the present invention, the training data
could
consist of immune status data from a large population of people coupled with
data
about the eventual onset of diseases in that population. Were such a database
to exist,
-139-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
neural networks could be constructed which could predict the onset of disease
based
upon features in an individual's immune status information. An advantage to
using
neural networks is that they could be a simple drop-in replacement to the
current
Diagnostic Module in terms of inputs and outputs.
4. Data Mining Module
4.1. Overview
The Data Mining Module is the large-scale component of exemplary systems
according to the present invention. As noted above, while the Diagnostic
Module
focuses upon obtaining results specific to a particular individual, the Data
Mining
Module can be, for example, designed to examine trends in large data sets
assembled
for many individuals and with many readings per individual. This capability is
necessary to support business models in which information is deduced about
immune
status patterns, as well as to improve the functionality of the Diagnostic
Module over
time.
As noted, an exemplary system was constructed using an Oracle database server.
The
schema for the database system is given in Figs. 12 through 14. The schema
used is
termed a `star schema', which is a database layout optimized for online
analytical
processing. This is a standard concept in data mining. More information about
the
data storage is given below.
4.2. Sample Data
A sample database was intended to represent actual immune status information
which
could be collected from a large population over a large span of time. The test
measurements contained within the database are randomly generated within the
constraints detailed below.
-140-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The exemplary database contains three distinct sorts of information.
The first block of information is individual immune status information. As an
example, the individual is assumed to be a patient in some healthcare context.
The
schema for the patient information table is given in Fig. 12. To summarize,
the
database contains information on the patient's birthdate, gender, racial
background
and geographic location. All of this information can potentially be used for
data
mining efforts related to immune status. The database also contains other
information
strictly for identification purposes, such as name and ID.
In the exemplary database, patient information was randomly generated. Gender
was
split evenly, and geographic placement was divided among four test cities.
Racial
backgrounds were assigned to match latest U.S. census figures available.
The second block of information is patient visit information. A schema for the
patient
visit information table is given in Fig. 13. To summarize, this information
covers data
that could, for example, be collected by a physician at the time of a
patient's visit.
There can be multiple visit information records for each patient. The majority
of this
information covers various symptoms present in the patient at the time of the
visit.
This information can be used within the Diagnostic Module, above, as part of
an
algorithm which takes into account diagnostic information other than the
immune
status assay results. This information can also be used in data mining to
discover
correlations between physical symptoms, immune status indicator levels, and
subsequent onset of disease. The visit information section of the database is
also used
to store recommendations from the Diagnostic Module.
In the exemplary database, symptomatic information was assigned randomly. The
example Diagnostic Module did not make use of symptomatic information.
-141-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The third block of information is the actual results of immune status assays.
In the
exemplary database there are 48 distinct simulated measured quantities,
although this
can be expanded, for example, to any reasonable number in a straightforward
manner.
The schema for this data block is given in Fig. 14.
In the exemplary database, assay test results are generated with care. The
distribution
of antibody levels are randomly generated based upon a log-normal distribution
with
an average of 50 micrograms per milliliter, as is consistent with measured
antibody
levels in practice. These values are used as initial baseline levels for the
patients in
the database. New values are then entered to simulate readings taken at set
time
intervals in the exemplary patients' lives, as indicated in Fig. 15. At each
age, the
antibody levels were perturbed using a small normal distribution, to simulate
variation
in the population over time. Results are biased to match the observed behavior
of
antibody activity in populations as they age, as shown in Fig. 16. All data in
Fig. 16
is from simulated vaccinated patients.
Half of the sample population was treated as if they had received a standard
vaccination schedule at age 5; the other half was left untreated. Antibody
levels were
adjusted to suit, as shown in Fig. 17. In addition, a subset of patients were
given
artificially lowered complement levels and antibody activity levels with no
change to
the measured antibody levels, simulating the effect of complement-deficient
patients
on the data mining procedure. This is shown in Figs. 18 and 19.
The intent behind this production of sample data was to produce a population
with
interesting characteristics that could be highlighted in the data mining
module.
Although the exact features used may not be strictly representative of the
population
as a whole, they represent the type of correlation that a system such as this
could
-142-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
detect within real patient data. It could easily be imagined, for example,
that
individuals of a particular racial background might naturally have elevated
levels of a
particular antibody. The system being described could be used to deduce that
fact,
which may have implications for the immunological care that such individuals
would
receive.
It is noted that all assay results, such as antibody levels, such as, for
example, "Gcmp
AVG" in Fig. 16, may be measured and quantified as units (U) per volume (e.g.,
ml),
where U may be defined as some arbitrary unit of a particular assay for the
purpose of
relative comparisons. In addition, U may be replaced by a more precise
measurement
of mass, such as micrograms, where possible and appropriate. Antibody
activity,
such as, for example, "Gcmp AVG" in Fig. 16, refers to the functional activity
of an
antibody, which may consist of, but not necessarily be restricted to,
bactericidal or
bacterial killing properties. In these specific examples, assay results from
individuals
may be processed for statistical purposes in the evaluation of a population,
as in Fig.
16, where individuals may be averaged (AVG) by appropriate statistical
formulas.
Where statistical processing assumes a normal distribution, geometric means
may be
used to average the results from different individuals, thereby requiring a
log
transformation of data sets, since it is generally found that only the log
values of
immune responses will follow a normal distribution.
4.3. Exemplary Use of the Patent Event Database
In exemplary embodiments of the present invention, a database used for data
mining
can, for example, be accessed in three different modes, as indicated in Fig.
5.
A first mode can be, for example, an interactive query mode. A user can
interactively
search for results in the database. Typically queries might include the
retrieval of a
-143-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
single individual's immune status over time, or the comparison of two such
individuals, as shown in Fig. 19. Queries can be submitted, for example, using
either
a graphical query tool or through the use of Structured Query Language (SQL),
a
computer language for the querying of databases. An exemplary SQL query is
shown
in Fig. 19A. Both of these methods of access are well-known in the industry.
With
reference to Fig. 5, a user can use the query mode via Query Module 531.
A second exemplary mode is the use of Online Analytical Processing tools, or
OLAP
tools, to find patterns within the database. A simple example of this is the
production
of aggregate statistics for subpopulations within the whole. In Fig. 19B, for
example,
a query for correlation coefficients to GCMP levels is restricted to female
patients. A
similar query might look at only patients from a distinct geographical area or
racial
background. Correlation statistics can also be generated, to test hypotheses
about
possible causal links among measured antibodies, between antibody measurements
and physical symptoms, or correlations between any of these and demographic
information. The utility of such a tool depends directly on the quantity and
quality of
data that is input into the system. For the exemplary system, trends that were
deliberately introduced into the sample data can be "discovered", but other
correlations are simply a function of random noise. In a real system, a
variety of
interesting patterns can be deduced. For the exemplary database, standard OLAP
tools
were used. With reference to Fig. 5, a user can use the data mining mode via
Data
Mining Module 532.
A third exemplary mode that is anticipated is the construction of a pattern
detection
module. This can, for example, comprise software programmed to sift through
the
accumulated immune status and other data and search for patterns that might
not be
-144-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
evident to a human observer. It is generally true that there are statistically
significant
patterns in the underlying data which are too subtle or too complex for simple
detection schemes. Such an automated detection system can, for example, rely
upon
one or more of the artificial intelligence pattern recognition techniques as
described
above and in the standard literature. In exemplary embodiments of the present
invention both neural networks and genetic algorithms can, for example, be
used to
perform this task. With reference to Fig. 5, a user can use the pattern
detection mode
via Pattern Detection Module 533.
Fig. 5A illustrates an alternative exemplary system architecture to that of
Fig. 5. Fig.
5A has a few additions, namely, Hypothesis Database 560 and Rules Database
565.
Each of these databases can be used, for example, when pattern detection
module 533
discovers a correlation between database variables. When that occurs, a list
of such
correlations can, for example, be reported to a human expert or group of
experts for
review. Or, for example, an intelligent system can attempt to recognize the
characteristics of such a correlation and associate possible hypotheses to
explain it.
These can be generated, for example, from a Hypothesis Database 560, and the
rules
by which a given correlation can, for example, be mapped to one or more
hypotheses
can be stored, for example, in a Rules Database 565. In such exemplary
embodiments, once a set of hypotheses is generated, an exemplary system itself
can
go back and mine the data to either rule out, corroborate, or confirm that
there is
insufficient data to either confirm or rule out, each hypothesis in the set.
In the latter
case the system can recommend that further information be collected, such as,
for
example, via additional assay panels known to the system, lab tests,
additional patient
history items, etc. This process is described in greater detail below.
-145-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
C. EXEMPLARY CANADIAN IMMIGRANT PROJECT DATABASE
USED TO ILLUSTRATE DATA MINING AND HYPOTHESIS
GENERATION
Appendix A contains selections (i.e. an initial set of records) from an
exemplary
database which was used to illustrate various data mining functionalities
according to
exemplary embodiments of the present invention. The database was created from
data obtained in interviews with and by performing tests on blood obtained
from a
number of newly arrived immigrants to Canada under the auspices of Dr. Chris
Greenaway (Assistant Professor in the Department of Medicine, McGill
University,
and a staff physician in the Departments of Microbiology and Internal
Medicine, Sir
Mortimer B. Davis Jewish General Hospital, Montreal, Quebec). As can be seen
from
the initial pages of the database, there are entries for assay results for
each of measles,
mumps, varicella, rubella, hepatitis A, hepatitis B, hepatitis C, tetanus,
diphtheria,
cytomegalovirus, strongoloides, schistosoma, filarial, and sixteen cytokines
(IL-la ,
IL-1(3, IL-2, , IL-4, IL-5, IL-6, IL-8, IL-10, IL-12p70, IL-13, IL-15, IL-17,
IL-23,
IFN-y, TNF-a, TNF-(3), as well as other factors such as age, gender,
region/country
of origin, socioeconomic status, etc.. In the descriptions that follow, this
database will
sometimes be referred to as the CIP database (for Canadian Immigration
Project).
The entire CIP database has approximately 1500 records. The database contains,
specifically, the following data:
Immunological Tests:
Hepatitis A
Measles (two different manufacturers for diagnostic testing)
Mumps
Rubella
Varicella
Tetanus
-146-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Diphtheria
Cytomegalovirus
Hepatitis B
Hepatitis C
CMV
Strongoloides
Schistosoma
Filaria
IL- l a
IL-1(3
IL-2
IL-4
IL-5
IL-6
IL-8
IL-10
IL-12p70
IL-13
IL-15
IL-17
IL-23
IFN-y
TNF-a
TNF-(3
Historical/Demographic Data:
Region of Origin, being one of-
Sub-Saharan Africa
Latin America and South America
Caribbean
Europe
Eastern Europe
South Asia
-147-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Southeast Asia
North Africa/Middle East
Demographic information:
Date recruited
Gender
Age (all participants were adults > 18 years of age)
Whether the interview was taken through an interpreter
Country of origin, being one of:
India
Bangladesh
Sri Lanka
Pakistan
Morocco
Vietnam
Congo
Other
Date moved to Canada
Citizenship status, being one of
Refugee claimant
Refugee
Immigrant
Other
Pregnancy
History of vaccine-preventable diseases
Participant had written vaccination record?
Participant's residence in home country had indoor toilet/no indoor toilet
If indoor toilet,
-148-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Flush?
Other?
Participant's residence in home country had outdoor toilet/no outdoor toilet
If outdoor toilet,
Outhouse?
Covered pit latrine?
Other?
Participant's residence in home country water supply
Tap inside?
Tap outside?
Closed well?
Public stand pipe?
Bottle?
Pump?
River?
Pump earth system?
Tap inside and closed well?
Other?
University education?
Participant's residence in home country degree of crowding (number of
individuals/room)?
Participant's residence in home country had electricity/no electricity?
The C1P database could be augmented to facilitate a broader scope of data
mining. In
such embodiments results of the following assays could be added: Tuberculosis,
Avian (H5N1) flu, Pandemic flu (not necessarily H5NI), Chronic infectious
diseases
other than CMV
-149-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
EBV, Herpes/Type?, Zoster outbreak/varicella antibody level following
outbreak?,
HPV, HIV
HTLV 1, Helicobacter pylori, Lyme disease, Tularemia, Parasite infections,
Malaria,
Strongyloides, Hantavirus, Leishmaniasis, Toxoplasmosis (particularly among
pregnant women), Antibody levels to other infectious diseases currently on
vaccination schedules: Hib,
Pneumococcal (conjugate vs. PS vaccines); Meningococcal (conjugate vs. PS
vaccines); Poliovirus; Traveler's vaccines; Japanese encephalitis; Cholera;
Yellow
fever; Military-specific vaccines: Anthrax, Smallpox, Plague, Rabies; Other
infectious agents not currently vaccinated for, including: Staphylococcus
aureus,
Moraxella catarrhalis.
In exemplary embodiments of the present invention, the following non-
immunologic
data can, for example, also be obtained and stored in an individual's
exemplary
Immunoscore database record:
Environmental considerations
Zip/Postal code
Rural/Urban home environment
Working environment
many interactions with many people
few interactions with few people
Interactions with types of people at home/work:
adults
children/age of children
interactions with local travelers
-150-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
interactions with global travelers
Commute to work
public transport/drive?
duration?
crowded/stressful?
Power source
proximity to power lines
type of fuel
proximity to power plant
Water
source
well
city
nature of treatment
Nutrition
Diet
high/low fat
meat/vegetable intake
Common food infections
Salmonella
Cholera
Hepatitis A
Typhoid
Alcohol consumption/volume
Vitamin supplements
Fitness
Regular exercise/sedentary
-151-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Cardiac/blood pressure assessment
History of smoking
Second hand smoking
School(s) attended
Day school/boarding school
Crowding at school?
Work environment
high/low/intermediate stress
job satisfaction
occupation
work described as physical/mental/combination?
safety considerations at work?
infectious organisms present
nosocomial infections a concern?
chemical agents?
Air quality
home
work
Animal exposure
pets
work
farm
lab
leisure
wooded environment?
horseback riding?
Family/personal history:
-152-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Chronic disease/nature?
Cancer/type?
Heart disease
Diabetes
Known immunodeficiency
Asthma
Kidney disease
Liver disease
Lung disease
Allergies/type?
Mental illness
Back problems
Joint pain/injury?
Chronic fatigue
Osteoporosis
Arthritis
Epilepsy
Education level
highest grade achieved
Education type
public/private?
education environment
crowding?
stress
quality of school (measured objectively, of course)
Military service?
Nature of deployment
-153-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Service branch?
Rank
It is understood that an exemplary database according to the present invention
can
contain records for various individuals from different countries and locales,
being
managed under various health care systems, and that various types of assays
can be
used to obtain assay results. Thus, in exemplary embodiments of the present
invention, the data stored in the database can, for example, be normalized to
some
database wide standard defined for each data field used in the database, or,
for
example, can be stored in its original form and any algorithm that seeks to
access data
first performs normalizing of the various records which are input to that
algorithm. It
is for the purposes of such normalizing that information regarding assay
manufacturer, type, and curve that maps an OD or other assay raw result to lUs
of an
antibody or other measured biochemical needs, in general, to be stored in the
database.
Next described are the results of data mining and hypothesis generation
studies
performed on the exemplary CIP database. These examples illustrate methods and
techniques that can be used in exemplary embodiments of the present invention.
D. DATA MINING - ANALYSES AND CONCLUSIONS
In exemplary embodiments of the present invention, immunologic information
stored
in an exemplary database (such as, for example, the CIP database, described
above)
can, for example, be analyzed in various ways and related to other variables
in the
database. Three useful examples of such analysis can, for example, include:
(1) linear
regression analysis on two variables to determine whether a positive or a
negative
correlation exists; (2) comparison of geometric mean immune values (obtained,
for
example, as antibody concentration, optical density, etc.) for both genders by
-154-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
geographical regions; and (3) percentage of positive or negative support
within a
population for one variable with respect to another. Examples of such analyses
are
described below using data from the CIP database.
1. Linear regression analysis - correlation coefficients
In exemplary embodiments of the present invention, tables of correlation
coefficients
(r) can, for example, be generated when comparing one particular immunologic
variable (such as, for example, varicella antibody optical density) against
other
disease-related immune measurements, either for both genders together or
separately.
For example, Fig. 20 presents the correlation coefficients between Varicella
OD and
various other variables in the CIP database. Fig. 20 presents three tables.
The top
table is the correlation of Varicella OD with each of nine other variables
from the CIP
database for all persons in the CIP database. The second and third tables
present this
information segregating males and females. For example, in Fig. 20, r values
have
been highlighted by shading when they are either > 0.05 or < -0.05, as a means
of
readily identifying patterns of relatedness (where I r 0.05 is considered as
"related").
With reference to Fig. 20, Varicella optical density is obviously highly
positively
correlated with Varicella titration dilution, inasmuch as one is calculated
from the
other, but other relationships also appear, although somewhat less pronounced.
If the
genders are separated, then the relationships appear even more strongly, as
expected,
since scatter is thus reduced. From the tables presented in Fig. 20, Measles
and
Mumps immunity (Dade assay) appear to be slightly correlated with Varicella
immunity.
-155-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The r values in the tables can also, for example, be graphed in such a way so
as to
better visualize any condition patterns, as is shown, for example, in Fig.
20A. Again,
the Measles and Mumps relationship to Varicella stands out above the others
(not
considering the Varicella titration dilution data, which is obviously
correlated to
Varicella OD).
2. Geometric mean values
In exemplary embodiments of the present invention, immune data can, for
example,
be statistically analyzed for the purpose of characterizing populations of
different
geographical regions, as well as for comparing results across genders. Such
mean
values can thus be graphically compared by gender and region to visualize
population
dynamics. For example, the geometric means of Rubella antibody concentrations
for
different regions can be graphically analyzed by gender, as shown in Fig. 20B.
With
reference thereto, a trend can be seen where males have higher antibody levels
than
females across all populations in the database. It is also apparent that
persons from
Southeast Asia show a lower antibody level relative to the other regions in
this study.
To help facilitate this assessment, dotted lines were drawn on Fig. 20B to
indicate the
mean of the means (geometric) from all of the populations (excluding Southeast
Asia)
separately for males and females. The arrows above the bars for the Southeast
Asia
data show the difference between the mean values for Southeast Asia compared
with
such mean of the means for all other regions. It would thus appear that
Southeast
Asia has a lower immune profile for Rubella. This can, for example, be
explained as
the effect of (i) no specific Rubella vaccine program; and (ii) a possibly a
lower
exposure rate compared with the rest of the world, making Southeast Asians
more
susceptible to this disease when traveling to other geographic regions.
-156-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
This finding highlights one of the many potential uses of the present
invention. As
described below in Section III, exemplary embodiments of the present invention
can
be directed to health insurance underwriting. Here, for example, knowledge of
the
fact that Southeast Asians tend to be vulnerable to Rubella would indicate
that such
persons, as a condition of maintaining insured status under a health plan or
HMO,
could be required to obtain Rubella vaccination.
In a similar manner, the geometric means of Hep A units in the CIP database
(which
are inversely proportional to antibody concentrations and derived from
immunoassays), were plotted in Fig. 20C for different geographical regions,
again
separately for each gender. In this case, it appears that there is no
significant
difference between males and females across all populations except one,
Eastern
Europe. Also, once again, Southeast Asia appears to be different from the
other
regions, where the Hep A antibodies are lower, as shown by higher assay units
which,
as noted, are inversely related to antibody concentration. In addition,
persons from
Eastern Europe are also seen as being generally lower in antibodies, and the
Eastern
European females (dotted bars) are seen as having particularly lower
antibodies than
the males. Again, in Fig. 20C a dotted line has been drawn to represent the
mean of
the means (geometric) from all of the populations, excluding Southeast Asia
and East
Europe, but combining males and females. Another dotted line has been drawn to
represent the mean of the means from the excluded populations, except for the
Eastern
European females, which are noticeably higher in units (and thus lower in
antibodies).
The arrows in Fig. 20C highlight the differences between (i) the overall
population
mean of means and the mean for Southease Asians and Eastern European males;
and
(ii) the overall population mean of means and the unit levels for Eastern
European
females. Overall, there appears to be less Hep A reactivity for Southeast Asia
and
-157-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Eastern Europe when compared with other regions; this is especially so among
East
European females. This may, for example, indicate no vaccination and possibly
less
exposure, with greater disease susceptibility. Thus, from a health
insurance/health
management perspective, an adult female from Eastern Europe should have a Hep
A
vaccination.
3. Percent support between variables
In exemplary embodiments of the present invention, the percentage of a
population
that demonstrates a positive or negative relationship for one variable with
respect to
another variable can, for example, be determined and graphically analyzed. For
example, using data from the CIP database, Rubella antibody levels were
measured in
females from China, and the results were grouped according to immune status:
immune support (protective high antibody level), low level support (equivocal
antibody level), or susceptible support (non-immune antibody level). The
percentage
of each of these groups that supports an association with another immune
variable,
either positively or negatively, for various different diseases was then
plotted in Fig.
20D. It is apparent that there is no significant difference in support for
Rubella with
Hep A (non-reactive or reactive) or with Varicella (positive); once again, in
Fig. 20D
dotted lines have been drawn to help visualize that the Rubella immune levels
show
no clear trend from immune to lower immunity to susceptible in these specific
cases
of other diseases. However, as regards Mumps, there is a clear trend for
Rubella
immune support when compared with Mumps. The arrow shows that there is a
greater percentage of Rubella immune support for positive mumps, i.e., immune
response for Rubella is correlated with that for Mumps.
-158-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The immune support of Mumps for other immune variables can, for example, also
be
used to compare different geographical regions, as is shown, for example, in
Fig. 20E.
In Fig. 20E, only the positive and negative Mumps support groups are plotted
(leaving
out the equivocal "low level support" group) for each of Eastern Europe and
Sub
Saharan Africa with respect to Hep A = non-reactive, measles = positive, and
Rubella
= immune. Dotted lines have been drawn to illustrate that there is no
difference in
Mumps immune support for Hep A = non-reactive in both regions, but the arrows
show that there is a difference for Measles = positive in Eastern Europe only,
and a
difference for Rubella = immune in both regions. Thus, a higher percentage of
Mumps immunity is seen with Measles immunity in East Europe, and with Rubella
immunity in both East Europe and Sub Saharan Africa.
In exemplary embodiments of the present invention, immune support can also be
related to other variables that do not measure immune status, such as, for
example,
education. An example of such a correlation analysis is shown in Fig. 20F. In
this
example, the positive and negative Mumps support groups are plotted for
Southeast
Asia and East Europe with respect to university attendance. From these results
it
appears that for Southeast Asia, a higher percentage of negative Mumps immune
support occurs when there is less university attendance, and in the expected
reciprocal
way, a higher percentage of positive Mumps immune support occurs with
university
attendance. For Eastern Europe, however, there is no relationship seen between
Mumps immune support and university attendance.
4. Possible Conclusions
The data mining examples described above demonstrate the usefulness, in
exemplary
embodiments of the present invention, of an analysis of relationships among
different
-159-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
variables, both immunologic and otherwise in an unbiased mathematical manner.
Regression analysis can, for example, be performed to just look for
correlations at
random, but the relationships may be weak and difficult to see. Additionally,
for
example, population means can be used to detect broad population differences
or
similarities. Also, percentage support analysis between different variables
can, for
example, allow for a greater focus on specific relationships between different
immune
status results and other factors that may affect them.
The examples described above point towards interesting correlations, some of
which
can be explained based on known immunization practices, and others which may,
for
example, indicate previously unforeseen relationships involving exposure to
disease.
For example, in countries where MMR (Measles, Mumps, Rubella) vaccines are
administered, one might expect to see a clear correlation of immunity for all
three
diseases; but this would usually occur only in developed countries such as the
U.S.,
Canada, and parts of Europe. Also, in some cases, there may only be single
immunizations for Measles. The immigrant populations used in the examples
discussed above, however, were most likely not immunized for the diseases
under
analysis, and thus most of the observed immunity would be due to environmental
exposure to the infectious agents of disease, or possibly some other agents or
substances that cross-react with these disease agents.
Due to socioeconomic conditions in these regions, it is possible that exposure
to one
disease might also indicate exposure to others, particularly in crowded areas,
or areas
where diseases are known to be endemic. It is therefore not surprising to see
positive
correlations between Mumps and Measles or Rubella, as seen in the China and
Eastern Europe data. In certain circumstances, however, the disease exposure
may be
-160-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
so prevalent (>90% of population) that there would be no way to establish
correlations to other factors since everyone has it; this might be the case,
for example,
for Mumps support with Measles in Sub Saharan Africa as shown in Fig. 20E.
Increased immunity for Mumps in Southeast Asia for those who attend a
university
could be the result of these more fortunate people being allowed greater
access to
vaccines, or, for example, it could be due to greater disease exposure in
crowded
dormitories. No difference in Mumps for university attendance in Eastern
Europe
might mean that there is greater disease incidence, or, for example, that
there is
greater university attendance, since both positive and negative support
percentages are
high. No difference in Rubella support for Hep A reactivity or positive
Varicella in
China may be, for example, the result of higher disease prevalence and
exposure
overall. A trend towards higher Rubella antibodies in males for all regions
might
indicate an unforeseen gender preference that could warrant further
epidemiological
studies in relation genetic polymorphism if this is not the result of broad
cultural
practices regarding vaccinations or disease exposure. The significantly lower
Hep A
antibody levels (higher assay units) only for females in Eastern Europe may,
for
example, might indicate a cultural phenomenon for further study.
These examples merely scratch the surface of what can be explored in terms of
epidemiology, immunity, socioeconomics, and genetic polymorphism in exemplary
embodiments of the present invention. Such exemplary analyses, can be used,
for
example, to design more focused studies on specific areas of interest or, for
example,
to test specific relationships that are only hinted at in the beginning. It is
also useful
to remember that the data in these examples only represent immigrants entering
Canada; it may therefore be important, in exemplary embodiments of the present
invention, to collect more samples and expand the database to other population
-161-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
segments, and/or to follow the same persons through time taking samples of
each
participant annually for an extended period of time.
As can be seen from the above description, in exemplary embodiments of the
present
invention, once a list of correlations has been obtained by analysis of a
given set of
records in an exemplary database, either humans or intelligent systems are
needed to
postulate explanatory hypothesis, which can then be verified, or at least can
be
attempted to be verified, excluded or determined as inconclusive.
E. PATTERN DETECTION AND HYPOTHESIS GENERATION
Fig. 21A illustrates an exemplary process flow for pattern detection according
to
exemplary embodiments of the present invention. With reference thereto, at 21
A01
patient information attributes can be collected and then grouped together into
separate
logical groupings. The following table illustrates such an exemplary grouping.
Logical group Attributes Example
Patient Information Patient's information that never changes
e.g. Gender, Birth Date
Current medical Visit date, female patient is pregnant or not at the
information time of visit, patient is taking medication or not etc.
Geography Patient's country of origin, region of origin
Immune Status Optical density of various diseases like HepA,
Rubella etc and also the immune interpretation i.e.
Positive, negative or susceptible for a disease.
Environmental Patients education level, Type of toilet, water supply,
conditions average people in house hold, number of rooms in
house hold, type of water supply etc.
Patients medical history Has patient been hospitalized before? If the patient
had diseases like measles, mumps etc and at what
a e, patient has vaccine record.
Miscellaneous Information that does not fall into any of the above
Next, at 21A05, the logical groups can be prioritized in an order in which
they are to
be correlated. For example, one could choose the highest priority logical
groups that
-162-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
you want to find correlations between (e.g. Immune Status v. Geography of the
patient). This can be done, for example, at 21A15, for all the logical groups.
At
21A20, correlations can be sought. This can be done, for example, as follows:
1. obtain the percentage of people in the same geographical regions that are
immune,
not immune or susceptible to diseases.
2. try to find a region where the patient population has variation in immunity
status
towards a disease. The reason for this is that if 89 % of the people are
Immune to
Mumps in, say, N. America, this means that there is not enough data for people
who are not immune to Mumps for evaluation. Whereas in South East Asia 67%
are immune to Rubella, therefore there is a large percentage of the population
(33%) that are either susceptible to Rubella or not immune. Thus when there is
difference in immune status in population in the same region the remaining
data
can be explored to attempt to determine the cause.
3. Try to evaluate the above results by next logical group (i.e. patient
information -
does the immunity status of a region differ by gender?). Obtain the percentage
of
population by region and gender that are immune, not immune or susceptible to
the disease. If there is a major variation in the percentage of male - female
population for same region that are immune or not immune, then there is a
discrepancy and the other data can then be explored to attempt to determine a
cause.
4. Use a data mining tool to find the correlations of the next logical group
(i.e., for
example environmental conditions on the patients within the same region and
gender and same immune status).
5. Obtain the geometric means of the optical density of the various diseases
by
geography and gender. This can determine if there is a difference in the
antibody
-163-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
level between genders living in the same geographical regions. After seeking
correlations at 21A20, if a correlation is found at 21A25, it can be reporte
at
21A27. The process can continue until all groups have been searched, and
process flow ends, at 21A50.
In exemplary embodiments of the present invention, Oracle Data Miner can be
used,
for example, as a tool for finding patterns in a database.
Using this tool, for example, there are different ways of finding correlations
in the
data.
= Association Rules :
Oracle data miner uses Apriori Algorithm to find these association rules.
Apriori Algorithm Details
Oracle Data Miner calculates the following two properties of association
rules:
> Support: S upport of an associating pattern is the percentage of task-
relevant data transactions for which the data is true.
If A => B
Support (A => B) = Number of tuples containing both A and B
Total number of tuples
> Confidence: Confidence is defined as the measure of certainty or
trustworthiness associated with each discovered pattern.
IfA=>B
Confidence (A => B) = Number of tuples containing both A and B
-164-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Number of tuples containing A
Associations can be calculated in 3 steps:
1. Find all combinations of items, called frequent itemsets, whose support
is greater than minimum support.
2. Decide the minimum support and minimum confidence required for
choosing the rules. As the data set under consideration was small we
kept the minimum support = 0.1 and minimum confidence as 0.1 so
that we do not miss any data that might have any inverse co relation or
strong co relation.
3. Use the frequent itemsets to generate the desired rules. Rules that
satisfy both minimum support threshold and minimum confidence
threshold are called strong rules. Reading the confidence and support
get the rules that are correlated.
For example, association rules generated for Chinese females who are
immune to rubella.
Some exemplary rules that can be generated can be, for example:
Rules Confidence Support
If Hep A Non-Reactive then Rubella
Immune 1.00000 0.25532
If Hep A Reactive then Rubella Immune 1.00000 074468
If Measles = Negative then Rubella Immune 1.00000 0:17021
If Measles = Positive then Rubella Immune 1.00000 0180851
If Varicella = Positive then Rubella Immune 1.00000 3=< `' 0:97872
Conclusions can be derived, for example, from the rules generated by
data miner.
Thus, Rule 1 means that 25 % of the Chinese females who are immune
to Rubella are Hep A non reactive. The trustworthiness of this statement
is 100%.
-165-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= Regression
Regression creates predictive models. The difference between regression and
classification is that regression deals with numerical/continuous target
attributes, whereas classification deals with discrete/categorical target
attributes. In other words, if the target attribute contains continuous
(floating-
point) values, a regression technique is required. If the target attribute
contains
categorical (string or discrete integer) values, a classification technique is
called
for.
The most common form of regression is linear regression, in which a line that
best fits the data is calculated, that is, the line that minimizes the average
distance of all the points from the line.
This line becomes a predictive model when the value of the dependent variable
is not known; its value is predicted by the point on the line that corresponds
to
the values of the independent variables for that record. Oracle Data Mining
provides both linear and non-linear regression models.
Algorithm options: Support Vector Machines (SVM)
Support Vector Machine (SVM) is a classification and regression prediction
tool that uses machine learning theory to maximize predictive accuracy while
automatically avoiding over fit of the data.
= Geometric Mean:
y - ~Y1Y2Ya=..Yõ
=
-166-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The geometric mean of a set of positive data is defined as the n`h root of the
product of all the members of the set, where n is the number of members.
Another way to calculate the geometric mean, which may aid in statistical
analyses, is to define it as the antilog of the mean of the log values for a
set of
numbers.
Exemplary Data Mining Algorithm
Using the CIP database, the following exemplary algorithm was performed:
1. The logical groups were prioritized so that the immune status (immune
assay results) could be shown according to geography, followed by
Gender. This is shown in all examples of the data mining (regression
analysis, geometric means, and percent support). All other logical
groups could be examined later for possible relationships that might
help explain the observed correlations.
2. Regression analyses were performed between all immune variables at
each geographic location, and by gender. Varying cut-offs could be set
to detect patterns of correlations from tabulated correlation
coefficients. For example, r values were highlighted in the table where
they were >0.05 or <-0.05, is described above in connection with Fig.
20.. This resulted a possible association of Varicella with Measles and
Mumps. These r values were also graphed to facilitate any
visualization, as demonstrated.
3. Geometric means of immune assay results were calculated for all
geographic regions, and by gender. Graphic analyses were performed,
-167-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
as demonstrated, to detect differences or similarities between regions,
as well as gender. For example, there appeared to be a gender
difference globally for Rubella immunity in favor of males, and a
lower immunity overall in Southeast Asia. Hep A showed this gender
difference only for East Europe, with lower overall reactivity in both
Southeast Asia and East Europe.
4. Setting the confidence at 100% for different immune subsets of a
disease, different geographical regions were examined for the percent
support of the association with other variables. For example, in each
immune subset of Rubella (immune, low level, susceptible) for
Chinese females, the percent support was determined for each of the
other disease immune variables. The graph (Fig. 20D) shows that
there is a greater association of Rubella immune support for positive
Mumps, when compared with Rubella low level or susceptible support.
In another graph (Fig. 20E), positive and negative Mumps support was
associated with other diseases in different geographical locations. In
this case, there was a greater percent positive Mumps support for
Measles and Rubella in both East Europe and Sub Saharan Africa.
5. To enhance the chances of seeing meaningful associations, regions
where there was a lower incidence of immune status result (e.g.,
<80%) were looked at, so that associations were not just based on the
fact that everyone has a particular status. For example, if 95% of a
population has a particular status, then that status could likely be
associated with anything; however, as noted above, since there is 67%
-168-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
immunity for Rubella in Southeast Asia, then there was enough non-
immunity to allow some possibility of detecting meaningful
associations.
6. Other logical groups were then now be examined for other possible
associations and explanations of previous associations. For example,
an association was graphically demonstrated between positive Mumps
support and university attendance in Southeast Asia (Fig. 20F).
Fig. 21B depicts the exemplary pattern detection process flow of Fig. 21A with
additional expert system functionalities. Thus, at 21B60, for each
correlation, the
hypothesis database can be searched for possible explanations of the given
correlation. In general, this can be done, for example, by using a Rules
Database and
Hypothesis as shown in Fig. 5A (and Fig. 2B) to map correlations to hypotheses
according to defined rules. Once a set of hypotheses is generated, for
example, at
21B65 each hypothesis can be tested, to the extent possible, authomatically,
using
data in the system database. Finally, at 21B67, a report can be generated
which lists
the generated hypothesis and states, based on system data, if that hypothesis
is
corroborated, ruled out, or inconclusive, as next described.
Thus, in exemplary embodiments of the present invention, a Hypothesis Database
(and associated Rules Database) can function as a repository for expert
knowledge.
When correlations are discovered by the system, these databases can be
consulted to
provide possible explanations as to why certain correlations may exist. A
Rules
Database can, for example, map - as a function of its conditions on
attributes, such as,
for example, the database variables involved in the correlation - correlations
to
-169-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
hypothesis already stored in a Hypothesis Database. For example, a possible
sequence may occur as follows:
1. Database Searched.
2. Correlation found between Rubella and Varicella where Antibody
levels are directly proportional.
3. Consult Hypotheses Database for possible explanations.
4. Possible explanations:
a. Cross Reactivity (when exposed to one disease, build up
resistance to the other)
b. Multiple disease vaccinations; and
c. Patient living in an area where risk of exposure is great.
5. System can automatically seek to verify whether each hypothesis
generated by the system, using the Rules Database and Hypothesis Database,
as above, is valid.
For example, the database records for the individuals involved in the
correlation can be checked for (i) vaccinations for either or both of Rubella
and Varicella, and for (ii) living and/or socioeconomic conditions conducive
to exposure.
Next, the hypothesis and the support/nonsupport/non-conclusiveness of
each hypothesis can be reported to humans, as shown in 21 B67 of Fig. 21 B.
6. After receiving a report, each correlation can be analyzed by a human
to determine if a new hypothesis should be added and fed back into the
Hypotheses Database; or if an existing hypothesis is operative in the given
context.
An example of this process can be illustrated with reference to Fig. 20B,
which
presents levels of Rubella antibody concentration across various regions using
data
form the Cl? database. First, the data was grouped by gender and region to
determine
if any trends were discovered.
As noted above, it was discovered that the females in Southeast Asia had
especially
low levels of Rubella antibodies. Upon further, lower level, geographic
analysis it was
found that the individuals from China were the ones with low levels.
Possible hypotheses for this occurrence are, for example:
-170-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
1. The females tested were never vaccinated for Rubella.
2. The females tested were not exposed to Rubella via the general populace.
As above, data already in the system can be used to examine the validity of
each of
these hypotheses.
In this way future correlations can, for example, be analyzed by the system
itself to
suggest possible reasons as to why trends or patterns have emerged.
Exemplary Automatic Pattern Detection Module
Fig. 21C depicts exemplary process flow for an exemplary automated pattern
detection module according to an exemplary embodiment of the present
invention.
With general reference thereto, the following exemplary process can be
implemented
in exemplary embodiments of the present invention for such a module:
1. Prepare data for data mining. Most data mining algorithms require data to
be
suitably transformed in order to produce good results. Some common data
transformations are: binning, normalization, missing value imputation, and
outlier removal. In exemplary embodiments of the present invention,
techniques used for transforming the data can be, for example, selected based
on attribute data type, attribute value range, attribute cardinality, and
percentage of missing values for an attribute or a record. (21CO1, 21C03,
21C05)
2. Group the attributes of the data into different logical groups like patient
current immune status, patient history, environmental conditions they lived
in,
geography, etc. (21 C07)
3. As this is a data centric data mining system for diseases, a focal point is
to get
the disease immune status relativity. The attribute importance of each
attribute
can be found to rank them in an ascending order to determine which attributes
effect patient's immune status to a particular disease. Attribute Importance
ranks the predictive attributes by eliminating redundant, irrelevant, or
uninformative attributes and identifying those predictor attributes that may
have the most influence in making predictions. (21C07)
Example For rubella interpretation attribute the following was found:
Attribute Importance Order
-171-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
RUBELLA ANTIBODY LEVEL 1
MEASLES OPD ZEUS 2
ELECTRICITY 3
PT GENDER 4
4. For each disease immune status (21C15) find a correlation with all possible
combinations of identified set of attributes found in Step 3. (21CIB), (21C20)
For example, a correlation between Rubella interpretation and Rubella
antibody level and gender was found.
5. For each correlation a threshold can be with the help of a (human) domain
expert (21C30)
6. Compare the correlation found by the data miner with the thresholds set.
Verify the combination of attributes resulting correlation with the disease
immunity status with acceptable threshold with the hypotheses database. If
such relation already exists remove this combination from further
investigation.
7. If the correlation can not be explained by existing hypotheses, analyze
this
attribute combination further for each attribute's contribution to the
correlation
of the whole set of attributes with disease immune status.
8. Derive association rules for the correlated attribute set found from Step
7.
9. Check rules with the discovered set of rules for its existence.
10. Analyze the Rules for determining the patterns in the data set using line
or
curve fitting.
11. Report discovered pattern and verify with existing hypotheses database.
For an initial exemplary analyses of the CIP database described above, where
data
was received in the form of an Excel spreadsheet, and data mining was
accomplished
using Oracle software, the following processes were utilized. Similar
processes can be
implemented in exemplary embodiments of the present invention. As described
below in detail, in a follow-up series of analyses on the CIP database, fully
automated
protocols using the PipelinePilotTM software environment were created.
1. Initial Exemplary Analysis: Data Mining Steps
1.1. Data Preparation:
1. The data is received in xls format.
-172-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
2. The data then needs to be scrutinized for each column and modified.
Example: some columns have data like ">250". That needs to be changed to
some number greater than 250 (Scientist discretion) since the data needs to be
imported into the database as a number and ">250" is not a number.
3. All the data is checked for valid values in the xls sheet before importing
into
the database.
4. Save the data.xls file as data.csv file (Comma separated file).
5. Create the table using the data received using the script createtable.sql.
6. Import the data from the xls sheet into the new table using the
Immnoscore.ctl
file.
1.2. Association Rules:
Association Rules provide the ability to show relationships that exist in the
data.
To find the association rules between the attributes of the data use the
Oracle Data
Miner (ODM). For example, obtain association rules for everyone that has a
Measles
Interpretation which is positive.
1. Create a view of records that have the attribute value "measles
interpretation"
as "positive".
2. Next go to Oracle Data Miner.
3. Click on Models > Association Rules > Build.
4. Name your Model and Click Next to continue.
5. Specify the location of the data used to build the model.
a. Schema: Select the schema containing the input table.
b. Input table: Select the table or view to use.
c. Records per Case: Select Single Record per case. (As each patient
record is I record in your view).
Click Next to continue.
6. ODM supports Apriori Algorithm to build Association Rules. You can change
the defaults.
Minimum Support: A real number between 0-1.Ask the scientist for details.
Minimum Confidence: A real number between 0 - 1.Ask the scientist for
details.
Limit Number of Attributes in Each Rule: Number between 2-100.
After specifying the values Click Next to continue.
7. Select data preparation if any is required. Click Next to continue.
8. Choose the attribute to include in your model. Click Next to continue
9. Click Finish to queue your mining activity.
10. Once the mining activity is executed without error, get the rules based on
Rule
Length Ascending, Support Descending and Confidence Descending.
11. Export the rules to an xis sheets.
1.3. Regression:
-173-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Regression Models provide the ability to predict numerical attributes about
data
entities.
Steps:
1. Create an Oracle View from the main table with all the numerical fields
that
you want in your regression model.
2. Click on ODM.
3. Click Model > Regression >Build
4. Name your Model. Click Next to continue.
5. Specify the location of the data used to build the model.
Schema: Select the schema containing the input table.
Input table: Select the table or view to use.
Records per Case: Select Single Record per case. (As each patient
record is 1 record in your view).
Click Next to continue.
6. ODM uses the Support Vector Machine algorithm for regression. Change the
values of the defaults by asking the scientists. Currently defaults given by
ODM are used.
7. Click Next to Continue.
8. Select Automatic Preparation option for your Model. Click Next to continue.
9. Select the attribute you want to predict. Click Next to continue.
10. Select all the attributes that must be in your model. Click Next to
continue.
11. Click Finish to queue the mining task on the server.
12. Once the task is done export the results to an xls sheet.
F. AUTOMATED DATA MINING
In exemplary embodiments of the present invention, partially or fully
automated
processing of the data in an exemplary database can be implemented in various
ways.
The following describes five exemplary software tools that can be used in
exemplary
embodiments of the present invention. These tools were created and then used
to
analyze an expanded version of the CIP database. As noted above, approximately
20% of the records of the CIP database (i.e., the first 330 records) are
provided in
Appendix A hereto for easy reference.
1. Exemplary Software Development Environment
The below described software tools have been implemented using Scitegic
Software's
Pipeline PilotTM programming tool. The Scitegic approach, known as data
pipelining,
uses a data flow framework to describe the processing of data. Figs. 21 D-1
through
21D-37 illustrate the following discussion.
-174-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Data pipelining is the rapid, independent processing of data records through a
branching network of computational steps. It has several advantages over
conventional technologies, including:
= Flexibility: Each data record is processed independently, allowing
processing
to be tailored to each record.
= Speed: Highly optimized methods allow rapid analysis of thousands (or
millions) of data records.
= Efficiency: Individual processing of data records limits memory use so that
many protocols can be executed simultaneously.
= Ease of use: Protocols are easy to construct, with visualization that
exposes
key data processing steps.
= Integration: Data pipelining is a powerful tool for connecting the different
data sources, databases, and applications required in the drug-discovery
enterprise.
Thus, Pipeline Pilot provides environments to design, test, and deploy data
processing
procedures called protocols. A protocol is made up of a set of components that
perform operations such as data reading, calculation, merging, filtering, and
viewing.
The connections between components define the sequence in which data is
processed.
Data from files, databases, user input, and the Internet is merged, compared,
and
processed, according to the logic of the protocol.
Protocols are constructed with a graphical drag-and-drop interface. The work
environment is divided into windows. On the left, the Explorer shows the
contents of
the database of available components and prebuilt protocols. Additionally, a
user can
-175-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
save new protocols in the database of components and publish them for
enterprise-
wide sharing and reuse. On the right, the workspace provides a way to create
new
protocols by dropping and connecting components.
The visual representation makes it simple to understand critical data
processing steps
in a potentially complicated procedure. Components are displayed as function-
specific
icons clearly identified with descriptive labels. Data records are passed
between
components through pipes represented by gray lines.
2. Client-Server Computing
Pipeline Pilot employs the client-server model of computing Figure 21D-1. The
professional client provides a way to create and edit components and
protocols, which
are stored on the server. All protocols are executed on the server. The server
can also
connect to resources on other machines, including files, databases and third
party
applications. This architecture provides a convenient way to integrate
Pipeline Pilot
and distribute resources efficiently across different locations.
3. Third-party Applications
Many Pipeline Pilot components integrate with third-party applications such
as, for
example, Microsoft Word and Excel, Spotfire Decision Site, and Accelrys DS
ViewerPro. A user can use these applications to read from, write to, and view
data.
4. Extending Pipeline Pilot
Pipeline Pilot includes the following types of components that extend the
program's
functionality:
-176-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= Open Database Connectivity (ODBC): A user can access databases such as
Oracle, SQL Server, and MS Access that reside anywhere on the user's network.
ODBC components allow a user to select, insert, delete, and update data.
= Run Program Components: A user can execute an operating system
command on the server or on a user's client and extend the functionality of a
protocol to include any operation that a user can invoke from a command line.
For
example, a user can write out one or more data files, invoke the command line
program to work on these files, and then read the results back into the
protocol
when the command execution is completed.
= Simple Object Access Protocol (SOAP) services: A user can make requests
to a calculator or service that resides on Unix, Linux, and other remote
machines.
The SOAP component sends the necessary data to the SOAP server, collects the
results, and adds the data to the current record.
= Scripting: A user can write scripts in one of the supported syntaxes
(VBScript, Perl, Java and Python) and use them in protocols. The script has
access
to the contents of each data record passed to the script component and to the
relevant protocol properties. This allows a user to write components such as
data
readers, data writers, data filters, and calculators.
= Visualization tools: Standard visualization tools such as Internet Explorer,
Excel, Spotfire, and Accelrys' viewer software are integrated with Pipeline
Pilot.
Additional third party viewers can be created by exporting the appropriate
data file
and executing an operating system command to start up the visualization
software
and read in the data file.
-177-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
5. Integrating Protocols
Pipeline Pilot works with data and computer services that commonly exist on
user
networks. After writing and publishing a protocol, a user can access the
functionality
of the protocol from environments other than Pipeline Pilot. For example, a
user can
run protocols on the Pipeline Pilot server from Internet Explorer, integrate a
protocol
into a third-party application such as Excel or Spotfire Decision Site, run
protocols
from SOAP clients or from the command line on any computer where a Pipeline
Pilot
client or server is installed.
Next described are five exemplary automated data mining tools that were
implemented in Pipeline Pilot, and that were used to analyze the CIP database
described above.
6. Data Mining Tool
This tool is designed to find correlations between variables, or fields in a
database,
across populations.
Run Application From:
Webport is the out of the box web-based client for Pipeline Pilot. Protocols
can be
deployed to Webport by saving them in the Protocols\Web Services folder within
Pipeline Pilot. Webport users simply go to a web page and from this location
protocols can be run. Users do not even need to know that Pipeline Pilot is
being
used; to the user it is simply a web page where information can be obtained.
This paradigm is used for all protocols deployed via Webport.
Protocols
-178-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= ProtocolsiWeb Services\ Wellstat_v3\Data Mining Tool
o ProtocolsiWeb Services\ Wellstat_v3\Utilites\Filter Form with Binning
o ProtocolslWeb Services\ Wellstat v3\Utilites\Filtered Graph with
Binning
From Webport, all protocols are visible, except those protocols stored in a
folder
called Utilities. Because of this, all of the protocols that are performing
queries and
calculations are "hidden" in the Utilities folder. The Data Mining Tool
protocol is
outside of the Utilities folder and thus it is visible from Webport. This
protocol is
merely a shell that points to the first protocol that needs to be run. On the
protocol's
Implementation tab, the Protocol Form parameter points to the protocol which
will
actually be run when the user clicks on the Data Mining Tool in Webport. For
the
Data Mining Tool, this protocol is the Filter Form with Binning protocol. The
data
entered into the form created by Filter Form with Binning is passed to
Filtered Graph
with Binning.
A protocol is composed of components, which are the building blocks within
Pipeline
Pilot. In general a component performs a particular task. Multiple components
can
be collapsed into a Subprotocol. In this case, multiple components appear to
be a
single component as one looks at the protocol. When a user double-clicks on a
subprotocol, it opens to show the components that make up the subprotocol. In
this
way, a subprotocol can hide more complex logic.
Input:
Below is a screen shot of an exemplary embodiment of a data mining tool and
its user
interface. It is via this interface that a user can input information that
will be passed
-179-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
to the data mining tool. As can be seen from the figure, this particular
version of the
data mining tool is called "Data Munger With Binning."
As can be seen from the exemplary screenshot in Figure 21D-2, a user can
select
which properties to group by, including the bin size for any groups for which
binning
is desired. It is often convenient to bin by age in 5 or 10 year increments.
An upper
and lower correlation threshold can also be set. Options regarding output
format can
include, for example, selecting to show the Value Table for each heat map
and/or to
create a PDF version of the output.
Lower and upper values for the standard deviations across groups can also be
specified. This is used to determine which attribute pairs appear in the
Difference
Between Groups output tab. This allows the user to set the cutoff values for
which
attribute pairs are "interesting" based on how much variation there is between
the
groups.
The protocol is set up to point to the necessary data files. The parameter
that points to
the full data file is at the top level of the protocol, so that the
administrator of this
application can easily set the path to the full data source. The Full Data
Source
parameter is accessed by clicking on the white space of the protocol, as
illustrated in
Figure 21D-3.
Output:
There are up to 4 output tabs for this application:
= Filters
= PDF of Filters (optional)
-180-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= Differences Between Groups
= Whole Data Set Pie Charts for Each Group
The first output tab includes a Pie Chart showing the percentage of the
population that
makes up each group for which a Heat Map was created. One correlation matrix
and
heat map are created for each data group, including only the selected
attributes (such
selected attributes are selected from the interface at the "Attributes to
Include in Heat
Map" field). Only correlation values within the specified range will be shown.
If the
Show Value Table option is selected, the values for the heat map are also
displayed in
a table. If the Create PDF option is selected, a second tab is created for a
PDF version
of the report.
Figure 21D-4 depicts an exemplary data set distribution pie chart, an
exemplary
correlation matrix visualized as a heat map corresponding to the segment of
the
database represented by the pink upper left piece of the pie, and a list of
all the cells in
the heat map and their actual correlation values.
In addition, a tab displaying the differences between groups for each pair of
attributes
can be created as illustrated in Figure 21 D-5. Only attribute pairs that have
a standard
deviation within the upper and lower standard deviation thresholds will be
displayed.
The table can, for example, be sorted by decreasing standard deviation values,
as
shown below. This report allows the user to assess which attribute pairs may
be
interesting based on the differences between the groups. For example,
attribute pairs
which have a larger standard deviation can be used to create different patient
suggestion rules for different groups within the populations.
-181-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
A final results tab can show the distribution of the data set for each of the
properties
the data has been grouped by, as separate pie charts. Taking both properties
together
results in the two property pie chart shown in Figure 21D-6.
Application Design:
As shown in Figure 21D-7, there are three major steps involved in this
application:
= Data preparation and grouping (red)
= Creation of Correlation Matrices and Heat Maps (purple)
= Creation of Reports (green)
Figure 21D-8 illustrates the Data Mining Tool protocol, which includes
subprotocols,
illustrated in Figure 21D-9, for Data Prep, Filtered Heat Maps and Difference
Table
creation and the creation of the Distribution Pie Charts. The components shown
are
actually subprotocols, which are composed of multiple components to carry out
the
described tasks.
The data set is grouped so that each group (i.e., each segment of the database
that
resulted from the "Group By" choices made via the interface) of data can be
acted on
independently. For each group, a correlation matrix is calculated and a Heat
Map is
used to display this data.
The correlation value for each attribute pair is then compared across all of
the groups
and the standard deviation is calculated. This can be used to determine
whether the
correlation is "universal" across the database, or only seen within certain
defined
segments.
-182-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The various pieces of information that are created are then placed into the
appropriate
report outputs.
7. Single Patient Vaccine Recommendations
Run Application From:
Webport
Protocols
= ProtocolslWeb Services\Single Patient Vaccine Recommendations
o ProtocolslWeb Services\Utilites\Single Patient Form
o ProtocolslWeb Services\Utilites\Single Patient Treatment Suggestions
o ProtocolsiWeb Services\Utilites\Create Learn Models (to be run only
once prior to these protocols)
The Single Patient Vaccine Recommendations protocol is outside of the
Utilities
folder and thus it is visible from Webport. This protocol is merely a shell
that points
to the protocols that do the work. On the protocol's Implementation tab, the
Protocol
Form parameter points to the protocol which will actually be run when the user
clicks
on Single Patient Vaccine Recommendations in Webport. For the Single Patient
Vaccine Recommendations, this protocol is the Single Patient Form. The data
entered
into the form created by Single Patient Form is passed to Single Patient
Treatment
Suggestions. Create Learn Models must be run prior to using the Single Patient
protocols.
Input:
A user browses to locate a file with a patient's data. An exemplary Webport
user
interface is shown in Figure 21D-10.
-183-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The protocol is set up to point to the necessary data files. These parameters
are at the
top level of the protocol, so that the administrator of this application can
easily set the
paths to the specified sources. These parameters are accessed by clicking on
the white
space of the protocol.
The Rules database is comprised of several files. Rules Source contains the
information about the Rule-ID and the Suggested Action. The Rules Definition
Source contains information about which conditions must be met for each rule.
And
Conditions Source contains information about conditions that are used to
describe the
rules, for example, Pregnancy = 1.
There are also files for internal and external references. The Keyword Source
file
documents the keywords for a each rule.The Literature Source file contains
information about published documents, including a field containing Keywords
that
are used to link the document to rules via the Keyword Source file.
Output:
The result page has two sections. The Results table shows which rules the
patient
satisfies, including the Conditions, Suggested Action, Internal References and
External References.
The Patient Data table shows the results for each assay, including predicted
values.
The percentiles for the entire data set, the patient's sex, the patient's age
group and
the patient's region of origin are also included for all OD and TITRE
properties.
-184-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Below is an image of an exemplary output for an exemplary individual in the
database, the first record of the CIP database provided in Appendix A below.
In this
exemplary embodiment, each rule in the Rule Database is compared to the
patient's
test values. If the patient matches all of the conditions for the rule, the
rule is
considered "satisfied" and will appear in the Results Table, illustrated in
Figure 21D-
12, including the conditions, suggested action, internal and external
references for
that rule.
Application Design:
As can be seen in Figure 21D-13, there are three major steps involved in this
application:
= Data prep (red)
= Determining Satisfied Rules (purple)
= Create Tables for Report (green)
As part of the data preparation, the patient's data is read in and any
properties that are
missing can be predicted using learn models created using the Create Learn
Models
protocol (which must be run once prior to running this protocol). This
protocol,
illustrated in Figure 21D-14, creates a Learn Model for each property that is
specified
as a property that should be predicted. These Learn Models can be called from
other
protocols in order to calculate values for properties for patients that are
missing
values.
The percentiles for the patient are calculated relative to the total
population, his or her
age bin, his or her sex and his or her region of origin.
-185-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The patient's results are then used to determine which of the rules in the
Rules
Database are satisfied, as illustrated in Figure 21D-15. The patient data
table includes
all values in the original patient data file, any predicted values, and the
various
percentile calculations. For any rule that is satisfied, the information about
the
suggested action, internal and external references are joined in from the
appropriate
files. This information is displayed in the Results table.
8. Patient Population Rule Mining
Run Application From:
Webport
Protocols
= Protocolsi Web Services\Wellstat_v3\Patient Population Rule Mining
o ProtocolslWeb Services\ Wellstat v3\Utilites\Multiple Patient Form
o ProtocolslWeb Services\ Wellstat_v3\Utilites\Multiple Patient
Treatment
o ProtocolslWeb Services\ Wellstat_v3\Utilites\MuItiple Patient Data
Link Table
The Patient Population Rule Mining protocol is outside of the Utilities folder
and thus
it is visible from Webport. This protocol is merely a shell that points to the
protocols
that do the work. On the protocol's Implementation tab, the Protocol Form
parameter
points to the protocol which will actually be run when the user clicks on
Patient
Population Rule Mining in Webport. For the Patient Population Rule Mining,
this
protocol is the Multiple Patient Form. The data entered into the form created
by
-186-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Multiple Patient Form is passed to Multiple Patient Treatment. The Multiple
Patient
Data Link protocol is run when the user clicks on a bar in any of the output
graphs.
Input:
A user browses to locate file with patient's data and selects what property to
bin on.
For properties that require binning, the bin size is also entered. An
exemplary
interface appears below, which allows a user to select a file to read, and
select the
variable to graph by (with or without binning). In the example shown in Figure
21D-
16, the graphs are drawn according to age, in bins of 20 years.
The protocol is set up to point to the necessary data files for the Rules
Database.
parameters are at the top level of the protocol, so that the administrator of
this
application can easily set the paths to the specified sources. These
parameters are
accessed by clicking on the white space of the protocol, as illustrated in
Figure 21D-
17.
A description of the files that make up the Rules database can be found in the
section
describing the Single Patient Vaccine Recommendations.
Output:
As shown in Figure 21D-18, the output of the Patient Population Rule Mining
protocol consists of bar charts and a summary table with information about the
standard deviation for each rule across the groups.
The Bar Chart in Figure 21 D-18 shows the percent of the data set that
satisfies each
rule. There is also a table that shows the standard deviation for each rule
between the
groups of data (for example the different age bins). Rules that have a
standard
-187-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
deviation greater than 5% are highlighted. For each data group there is a Bar
Chart
showing the percent of the data set that satisfies each rule. One patient can
be
included in multiple bars within each chart; the bars do not add up to 100%.
For
example, if a patient satisfies both Rule I and Rule 2, that patient is
included in both
bars. There is also a bar displaying patients that did not satisfy any of the
rules. This
is useful for understanding what percent of the patients have no suggested
actions. As
illustrated in Figure 19, clicking on a bar creates a new window that shows
the data
records that make up that bar.
Application Design:
As can be seen in Figure 21D-20, there are three major steps involved in this
application:
= Get Rule Definitions and Calculate Rules for Each Patient (red)
= Grouping of Patients and Calculation of Standard Devation (purple)
= Creating tables and charts for reporting (green)
The determination of which rules are satisfied for each patient is done in the
same
way that the Single Patient rule determination is done. A Bar Chart is created
showing
how many patients in the full data set satisfy the criteria for each rule.
There is also a
table showing the standard deviation for each rule across the patient groups.
Each patient is placed in the appropriate bin and the number of patients
satisfying
each rule in each bin is calculated. Bar Charts are created showing the
percent of
patients within each group that satisfy each rule. Clicking on a bar in any of
the Bar
Charts creates a new HTML page showing the data that makes up the bar.
-188-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
9. Age Binned with Differences
Run Application From:
Pipeline Pilot Client
This protocol can also be deployed from Webport if desired. A user can then
run the
protocol without having to install the Pipeline Pilot client.
Protocols
= ProtocolsiWeb Services\ Wellstat_v3\Utilites\Age Binned with Differences
Input:
The protocol can be set up to point to the Full Data Source. Also the property
to bin
on, and the bin size can be entered via the protocol's top level parameters,
illustrated
in Figure 21D-22. These are accessed by clicking on the white space of the
protocol.
To have this report use a different property and bin size, the user can simply
change
the prop_list_bin parameter to the property of interest and the bin-size
parameter to
the desired bin size. Different bin sizes and/or properties can be selected
and the
protocol can be run multiple times, in order to illuminate how different
groupings can
affect the correlation values. In exemplary embodiments of the present
invention, a
"brute force" binning by every binnable property of interest can be
automatically
launched via multiple copies of this protocol operating in parallel and all
the
correlations thereby obtained out put to a cache or a list. Corresponding
correlations
can then be compared as to correlation values to isolate the best grouping
relative to
maximizing each such correlation.
-189-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
A user can also, set the upper and lower threshold for the standard deviation
to be
used to filter the results. These parameters can be changed to any desired
values. In
this example, the lower standard deviation threshold is set to 0.50 which will
limit the
output to only those attribute pairs that have this level of deviation. The
upper
deviation level is set at 2, which is the maximum deviation (-1 to 1).
Output:
As illustrated in Figure 21D-23, each chart displays the correlation values
for a pair of
attributes across all of the bins, in this example, age with a bin size of 5
years. One
XY Chart is created for each pair of attributes. Each chart displays the
correlation
value for a different attribute pair across the entire age spectrum (or
whichever
property the user specifies). This output allows the user to quickly identify
which
bins of the data have significantly lower or higher correlation values. Also,
patterns of
differences across groups can be identified by looking at the output graphs as
a whole.
Application Design:
As seen in Figure 21D-24, there are three major steps involved in this
application:
= Data prep and creation of correlation matrices
= Calculating the Standard Deviation of the correlation values for each
Attribute
Pairs
= Creation of XY Charts
The entire data set is read in and the data binned on the specified property,
in this
example age binned with a bin size of 5, giving 12 bins ranging from age 20-
75. A
correlation matrix is created for the data for each age bin.
-190-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The correlation value for each pair of attributes can then be compared over
the entire
age range and the standard deviation calculated. Also, the percent of the
population
contained within each bin is listed. This information can be used to identify
bins with
only a very small sample size.
Attribute pairs can be, for example, filtered out if the standard deviation of
the
correlation values across the age bins (value_StdDev) is not within the lower
and
upper stddev thresholds. Attribute pairs that are not filtered out, are
plotted in an XY
Chart, as shown above.
10. Automated Data Mining
Run Application From:
Pipeline Pilot Client
This protocol could also be deployed from Webport if desired. This would allow
users to be able to run the protocol without having to install the Pipeline
Pilot client.
Protocols
= ProtocolsiWeb Services\ Wellstat_v3\UtiIites\Automated Data Mining
o Protocols) Web Services\ Wellstat v3\Utilites\Automated Run -
Parallel
The Automated Run - Parallel protocol can be used to create a data cache of
the
correlation value, attribute pair and group information for all possible
groups:
The list of groups can be generated using a nested loop script, as described
below.
-191-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Automated Data Mining can then use this cache (described above) to create a
report
showing what grouping gives the highest correlation value for a particular
attribute
pair. In an exmeplary implementation, the Automated Run - Parallel protocol
was run
to create all the data caches that are then used by Automated Data Mining as
inputs.
This was done because running Automated Run - Parallel on a PC takes
approximately three hours with the CIP data base as an input. Thus Automated
Run -
Parallel is assumed to have been already been pre-run as a data prep step by
Automated Data Mining. Accordingly, pre processing the database and creating
the
cache allows for running Automated Data Mining in real time. With more
computational power, Automated Run - Parallel can simply be connected to
Automated Data Mining and both protocols can run from start to finish in an
automated fashion.
Input:
The protocol is set up to point to the Full Data Source. This parameter can
be, for
example, accessed by clicking on the white space of the protocol, as
illustrated in
Figure 21D-25.
Output:
Figure 21 D-26 illustrates the two outputs generated by this application. A
tabular
output shows all of the correlation values for a particular attribute pair
across all
groups, above the specified correlation threshold and standard deviation
threshold.
An absolute correlation value threshold can be set by the user. This report
allows
users to quickly identify how best to maximize the correlation value for a
particular
attribute pair.
-192-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The second output is created by starting with a particular group, for example
Measles_OD_DB_Int = 0 and looking at what additional grouping can be added to
increase the correlation values for a particular attribute pair. The graphs
show the
original correlation value as a green line and the improved correlation values
as red
dots, one dot for each new grouping. Each point on the graph has a mouse-over
tooltip
showing the grouping and correlation value. A table is also created displaying
the
data for each point on the graph.
Application Design:
As seen in Figure 21D-27, there are four major steps involved in this
application:
= Generation of all possible groups (blue)
= Creation of one correlation matrix per group
= Organization of the data by Attribute Pairs
= Creation of tables and XY charts
Because generating the correlation matrix data is more time consuming than the
other
data manipulation and generation of the reports, this application has been
broken into
two protocols. However, this is not a requirement. The Automated Run-Parallel
protocol performs the first two steps. Automated data Mining performs the last
two
steps. Automated Run - Parallel generates a correlation matrix for each
possible
grouping, from single property groups (such as, for example, sex or age) to as
many
groups as desired.
-193-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
After the Automated Run-Parallel protocol is used to create correlation
matrices, the
Automated Data Mining protocols can perfrom an analysis and generation of the
reports.
The Automated Run-Parallel protocol is designed to create one correlation
matrix for
each possible group. This data can be used in a variety of ways.
A simple nested loop, illustrated in Figure 21 D-28, can be used, for example,
to create
all of the possible data groups, including any binning necessary. This nested
looping
script adds all unique data groups to the prop_list property and the
associated bin size
information to the bin-Size property.
A file can be used to store the information about bin size for these groups.
The file is
composed of two fields, one for the name of the property and one for the bin
size of
that property. The grouping and creation of the correlation matrices can be
done in the
same way as for the other protocols listed above. Once the correlation matrix
is
created, the data contained in the table can be manipulated and displayed in a
variety
of ways.
In the example illustrated in Figure 21D-29, this was implemented for up to a
grouping of two properties, but any additional number of properties can be
added as
desired. For each possible group, the data can be used to create a correlation
matrix
and the data is added to a data cache which is used in the report creation.
Since there
are many possible ways to work with the correlation matrix data once it is
created, the
cache makes the correlation data accessible without having to rerun the part
of the
application that creates this information.
-194-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
As can be seen, the Automated Run - Parallel protocol has a subprotocol,
namely
Subprotocol 1, at the end of pipe 2 (pipe 2 snakes around for ease of
viewing). This
subprotocol is where a correlation matrix is created for a group. Subprotocol
1 detail is
illustrated in Figure 21D-30.
Detail of Filtered Heat Maps, a subprotocol in the third pipe of Subprotocol
1, is
illustrated in Figure 21 D-31.
Finally, subprotocol "for HTML", illustrated in Figure 32, (in the fourth pipe
of the
"Filtered Heat Maps" subprotocol of Subprotocol 1 of pipe 2 of Automated Run -
Parallel), is where the correlation matrix for each group is created. The data
is stored in a
data cache so it can be accessed by the Automated Data Mining protocol.
As noted, after Automated Run - Parallel has preprocessed the data, then
Automated
Data Mining begins processing. Figure 21D-33 illustrates its high level
processes.
The Data Prep subprotocol takes the data from the cache created by Automated
Run -
Parallel and groups and cleans up the data in preparation for creation of the
report. This
step could be moved to Automated Run - Parallel in order for all data
preparation to be
done in that protocol. Detail of the Data Prep subprotocol of Automated Data
Mining
is illustrated in Figure 21D-34.
-195-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The two subprotocols belonging to the Data Prep subprotocol of Automated Data
Mining are illustrated in Figure 21D-35.
In the exemplary implementation the data is displayed in two ways, in HTML
output and
in a PDF file. The protocols to do this, Create Graphs and Create Tables, are
illustrated
in Figure 21D-36.
-196-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
7. Data Files Used in Pipeline Protocols Described Above
As noted above, to run the abovedescribed protocols a number of data files are
drawn
from. The exemplary files used were a set of Excel spreadsheets placed in a
data folder,
as illustrated in Figure 21 D-37.
As can be seen from the screen shot of the Data\Wellstat folder, the files
used were
CIP_data.xls, Conditions.xls, Literature.xls, Rule_Definition.xls, Rules.xls,
and
Rules_Keywords.xls. These files, their contents, and their functions are next
described.
CIP_data.xls
This is the complete CIP database described above, the first 330 records of
which are
provided below in Appendix A.
Literature.xls
This is an internal database created to capture known information from
literature and
other sources regarding any conditions, diseases or symptoms that a particular
bioassay
result, or a combination of such results, may be a marker for. An exemplary
literature
database was constructed to catalog the indications of bioassay results
provided in the
CIP database
Copies of the Conditions.xls, Rules.xls, Rules_keywords.xls, and
Rule_Defininition.xls spreadsheets are provided in Exhibit A as well.
-197-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Conditions.xls
The Conditions.xls spreadsheet was used in applying defined rules to the
Single Patient
Vaccine Recommendations protocol. It can also be used, for example, in Patient
Population Rule Mining embodiments.
Rules.xls
The Rules.xls spreadsheet was used in applying defined rules to the Single
Patient
Vaccine Recommendations protocol. It can also be used, for example, in Patient
Population Rule Mining embodiments.
Rule Definition.xls
The Rule_Definition.xls spreadsheet was used in applying defined rules to the
Single
Patient Vaccine Recommendations protocol. It can also be used, for example, in
Patient
Population Rule Mining embodiments.
Rules_keywords.xls
The Rules_keywords.xls spreadsheet was used in applying defined rules to the
Single
Patient Vaccine Recommendations protocol. It can also be used, for example, in
Patient
Population Rule Mining embodiments.
-198-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
8. Complete Copy of Exemplary Pipeline Pilot Code (Provided on CD)
A complete copy of the Pipeline Pilot code is provided in Appendix D hereto,
which is
submitted on a CD for ease of viewing. Further information regarding the
contents of the
CD is provided in Appendix D below.
G. EXEMPLARY INTERNAL HYPOTHESIS DATABASE
Appendix B exemplifies the type of hypothesis database that can be constructed
using scientific articles and other literature regarding antibody markers to
assist an
ImmunoScore database user in building serological correlates for immunologic
and other
information stored in an exemplary database (such as, for example, the CIP
database). So
named as a "hypothesis database," the database can provide a user with all
available
information regarding two variables found to be correlated. The known
information may,
but often may not, explain the observed correlation. If it cannot, it can at
least marshall
whatever is known rearding the variables impicated in the correlation.
In exemplary embodiments of the present invention, a hypothesis database can
be
constructed using a software spreadsheet application, such as, for example,
Microsoft
Excel, Word Perfect Quattro Pro or Lotus 1-2-3 . Spreadsheet columns can be
set up
to record data obtained from the scientific articles as follows, for example:
Spreadsheet Column Information Recorded
A Assay (or Marker)
B Assay Family
C Assay Code
DL Assay Test Type
E Sample Size
-199-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
F Percent (%) Support
G P Value/Confidence Interval (CI)
H Interpretation
I Disease/Condition
J Source
K Notes
Upon reviewing a scientific article about an assay or marker, and making a
determination
that it contains data suitable for entry into the database (i.e., for example,
that the study is
not based on just one or only a few study subjects, that the study reports P
values and/or
confidence intervals, that the study provides numerical data supporting the
results
reported), each specific assay or marker discussed by the article can be
entered in a
separate row of Column A of the spreadsheet. The assay family to which the
assay or
marker belongs can be entered in Column B of the spreadsheet. Each specific
assay (or
marker) can be assigned an assay code, which can be entered in Column C of the
spreadsheet. The specific kind of test performed to identify the assay (or
marker), such
as, for example, enzyme-linked immunoassay (ELISA), particle agglutination,
sandwich
enzyme immunoassay, neutralization assay, solid phase enzyme immunoassay,
molecular
enzyme immunoassay, can be entered in Column D. The total study sample size
(including controls) can be entered in Column E of the spreadsheet. Numerical
data
supporting the study results reported, such as, for example, the number of
study subjects
out of the total subjects exhibiting the specific assay (or marker), the
percentage of study
subjects exhibiting the specific assay (or marker), the number of study
subjects having
serum concentration levels less than, equal to, or greater than a certain
amount, can be
entered in Column F of the spreadsheet. P values and/or 95% Confidence
Interval values
correlating with the numerical data supporting the study results can be
entered in Column
G, such as where only one study result is entered in Column F. Where more than
one
-200-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
study result is entered in Column F, p values and/or 95% Confidence Interval
values, can
be included in Column F together with the study results to which each such
value
pertains. The interpretation of the study results by the scientific article's
author(s), such
as, for example, a particular antibody level may indicate an active infection
with a
particular disease or acquired resistance to reinfection with a particular
disease or may be
associated with the presence of another disease, can be entered in Column H.
The
particular disease or condition in connection with which the assay (or marker)
has been
detected, such as, for example the detection of anti-diphtheria antibodies in
HIV-1
infected subjects, can be entered in Column I. The source of the data entered
in Columns
A through I and in Column K, such as, for example, the author(s) and title of
the
scientific article from which the data was obtained, as well as the name of
the journal,
volume number, issue number (if available), page numbers and year in which the
article
was published, can be entered in Column J. Other data reported in the specific
scientific
study identified in Column J, such as, for example, the geographical location
where the
study was conducted, the nationality of the study subjects, the gender
breakdown of the
study subjects, whether the study examined more than one assay (or marker),
and the age
of the study subjects, may be entered in Column K.
Appendix B, attached hereto, is a printout of the exemplary CIP hypothesis
database created using Microsoft Excel. Because of printing constraints,
Columns A-H
of each row of the database can be found on odd-numbered pages, while Columns
I-K of
each row of the database can be found on even-numbered pages.
-201-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The various automated data mining protocols described above (and whose
complete code is provided in Appendix D), can, for example, draw upon the
information
captured in the exemplary CIP hypothesis database, as next described.
In one embodiment, an exemplary system can retrieve data entered in the
exemplary CIP hypothesis database to perform a single patient analysis. For
example, if
serological assays indicate that an individual patient has both filarial
antibodies and
strongyloides antibodies, the exemplary system can retrieve the data from the
exemplary
CIP hypothesis database rows 2, 3, and 5. Row 2, for example, contains data
captured
from an article entitled "Predictive markers for development of
strongyloidiasis in patient
infect with both Strongyloidiasis stercoralis and HTLV-1," by M. Satoh et al.,
and
published in Clinical Experimental Immunology Vol. 133: 291096 (2003) (Column
J).
The captured data includes, for example, that particle agglutination (Column
D) was used
to test HTLV-1 (a retrovirus, in the same class of virus as the AIDS virus,
and is
associated with a rare form of blood dsycrasia known as Adult T-cell
Leukemia/lymphoma (ATLL) and a myelopathy, tropical spastic paresis) antibody
titer
(Column A) in 44 (Column E) patients infected with Strongyloides stercoralis
(a
nematode) in Okinawa, Japan, and that 31 patients (18 males, 13 females) were
HTLV-1-
positive and I 1 patients (7 males, 4 females) were HTLV- I -negative (Column
K).
Antibody titer in the direct fecal smear-positive group (8,192 median ranging
up to
28,672) was higher than in the direct fecal smear-negative group (4,096 median
ranging
up to 15,360) (P < 0.05) (Column F). There was a significant correlation (p =
+ 0.566, P
< 0.01) between the HTLV-1 proviral load and the antibody titer, and an
inverse
correlation between HTLV-1 proviral load and EBNA (Epstein-Barr Virus)
antibody titer
-202-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
(detected by anticomplement immunofluorescence) (p = -0.387, P < 0.05)
indicating that
increased HTLV-1 proviral load was especially related to lowering of immune
status
perhaps resulting in an increase of S. stercoralis load via immunity
impairment (Column
K). The authors concluded, inter alia, that the activity of HTLV-1 infection
influences
the results of direct fecal method of measuring HTLV-1 antibody titer in
patients infected
with both S. stercoralis and HTLV-1 (Column H). Row 3, for example, contains
data
captured from an article entitled "L3 antigen-specific antibody isotype
responses in
human strongyloidiasis: correlations with larval output," by N.S. Atkins, et
al., and
published in Parasite Immunology, Vol. 21: 517-26 (1999) (Column J). The
captured
data includes, for example, that immunoblotting (Column D) was used to test
for IgA
antibody (Column A) in 34 patients (Column E) consisting of two groups of
chronically
infected (for more than 30 years) ex-Far East Prisoners of War with and
without
detectable Strongyloides stercoralis larvae (Column K). IgA reactivity with
six
immunodominant S. stercoralis antigens was significantly elevated in
individuals with
undetectable larval output (Column F) (P < 0.05 for three antigens and P <
0.01 for one
antigen) (Column G), and IgE recognition of four S. stercoralis antigens was
significantly
higher among individuals with larval output (Column F). The authors concluded,
inter
alia, that the results were consistent with an IgA-mediated immune effector
mechanism
in modulating larval output (i.e., inhibiting worm fecundity and egg
viability) and IgE
playing a prominent role in acquired resistance to reinfection (Columns H and
K). The
authors postulated that parallel upregulation of IgE and IgG4 responses to
certain
antigenic components suggests IgG4 blockage of IgE-mediated allergic responses
and
may be central to establishment and persistence of asymptomatic chronic
strongyloidiasis
-203-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
(Column K). Row 5, for example, contains data captured from an article
entitled
"Detection of filaria-specific IgG4 antibodies and filarial DNA, for the
screening of
blood spots for Brugia timori," by P. Fischer, et al., and published in Annals
of Tropical
Medicine & Parasitology 99(1): 53-60 (2005) (Column J). The captured data
includes,
for example, that Brugia rapid (BR) (an antibody-detection dipstick test) and
PCR
(polymerase chain reaction)-based assays (Column D) were used to test for IgG4
antibody (Column A) in 66 individuals (Column E) from Alor island, East Nusa
Tenggara, Indonesia (which is an area highly endemic for Brugia timori)
(Column K).
Thirty-seven (37) of the 66 individuals (56.1%) of filter-paper blood spot
eluates were
positive using the BR test (32 strongly and 5 weakly), while the plasma
samples of 47 of
the 66 individuals (71.2%) were positive; 9 (23.4%) of the BR filter-paper
blood spot
eluates positives were PCR-positive (Column F). The authors concluded, inter
alia, that,
in general, the presence of microfilaremia is associated with relatively high
titers of anti-
filarial IgG4 (Column H).
In another embodiment, the analysis of population heat maps can be
supplemented by drawing upon the information captured regarding the assays (or
markers) implicated in a given correlation from the exemplary CIP hypothesis
database.
For example, if analysis of population heat maps generated by the Pipeline
Pilot software
reveals a correlation between hepatitis C virus (HCV) and schistosomiasis, as
is
described below, the exemplary system can retrieve the data from the exemplary
CIP
hypothesis database on schistosomiasis from row 4 and on hepatitis C from rows
44, 47-
48, 51, 61-124, 136, and 148-150. Row 4, for example, contains data captured
from an
article entitled "The antibody responses to adult-worm antigens of
Schistosomiasis
-204-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
haematobium, among infected and resistant individuals from an endemic
community in
southern Ghana," by Y. Osada, et al., and published in Annals of Tropical
Medicine &
Parasitology, Volume 97(8): 817-26 (2003) (Column J). The captured data
includes, for
example, that enzyme-linked immunoassay (ELISA) (Column D) was used to test
for
IgE, IgG, IgA and IgM antibodies (Column A) in 27 individuals (Column E)
infected
with Schistosoma haematobium (11 endemic normal subjects; 16 patently infected
subjects) in Okyerko, the Gomoa district of Ghana, where S. haematobium is
endemic
(Column K). Endemic normal subjects were generally older than patently
infected
individuals (P < 0.001); the male-female ratio was higher in patently infected
individuals
than in endemic normal individuals, although the difference was not
statistically
significant (P > 0.05); and for patently infected individuals and for patently
infected and
endemic normal individuals combined, S. haematobium egg output was negatively
correlated with the water-contact index (Column K). Endemic normals have
similar
levels of IgM antibody, higher levels of IgA (P < 0.05) and lower levels of
IgE (P < 0.01)
and IgG (P < 0.05) than patently infected individuals; for combined endemic
normals and
patently infected individuals, males had levels of IgM, IgA and IgE similar to
that of
females, but significantly higher levels of IgG (P < 0.01); when the patently
infected and
endemic normals were considered as a single group, S. haematobium egg outputs
positively correlated with levels IgE (P 0.01) and IgG (P < 0.001); in the
patently
infected group alone, only the correlation with IgG was statistically
significant (P <
0.01); the P values for the positive correlation of IgG and IgE in endemic
normals only,
patently infected only, and endemic normals and patently infected combined
were 0.01,
0.05 and 0.001, respectively; positive correlations between levels of specific
IgG and IgE
-205-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
were statistically significant only when data for endemic normals and patently
infected
individuals was pooled (P < 0.05); levels of specific IgA and IgE were
positively
correlated in the patently infected group (P < 0.01), but not in the combined
group; and
levels of specific IgA and IgG were positively correlated in the endemic
normals (P <
0.05), but not in the combined endemic normal and patently infected group
(Column F).
The authors concluded, inter alia, that the relatively high levels of IgG and
IgE may
directly reflect "active" current infection or that the high level of specific
IgG seen in the
patented infected group may reflect the presence of blocking antibodies; and
that IgE and
IgG antigens can be used as markers to reflect current infection intensity and
that anti-
worm antibodies do not act as protective antibodies in the natural course of
urinary
schistosomiasis (Column H).
A sampling of the kind of data regarding hepatitis C captured from the
articles in
rows 44, 47-48, 51, 61-124, 136, and 148-150, is exemplified by the data from,
for
example, rows 44 and 51. Row 44, for example, contains data captured from an
article
entitled "Viral markers and use of factor products among Finnish patients with
bleeding
disorders," by F. Ebeling, et al., and published in Haemophilia 7: 42-46
(2001) (Column
J). The captured data includes, for example, that ELISA (Column D) was used to
test for
hepatitis C antibody (Column A) in 193 patients (Column E) with bleeding
disorders
(hemophilia A or B, type 3 von Willebrand disease or factor XIII deficiency)
in Finland,
179 (93%) of whom were males (Column K). Fifty-one percent (51%) of the
patients
were anti-HCV positive (Column F), and the authors interpreted this positivity
as being
associated with blood transfusions (Column H). Row 51, for example, contains
data
captured from an article entitled "The clinical epidemiology and course of the
spectrum
-206-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
of renal diseases associated with HIV infection," by Lynda A. Szczech, et al.,
and
published in Kidney International, Volume 66: 1145-52 (2004). The captured
data
includes, for example, that hepatitis C antibody (Column A) was identified in
89 HIV-
infected patients who underwent renal biopsy during the course of clinical
care at six
major medical centers in the United States, 47 of whom had lesions other than
HIV-
associated nephropathy (HIVAN) and 42 of whom had HIVAN lesions (Column K).
Patients with lesions other than HIVAN were less likely to be black (37/47 vs.
42/42, P =
+0.02), less likely to have hypertension (24/26 vs. 31/24, P = 0.03), more
likely to have
greater creatinine clearance at the time of biopsy (60.6 vs. 39.0 mL/min, P =
0.008), and
have greater CD4 lymphocyte count at time of biopsy (287 vs. 187 cells/mL, P -
0.04);
all patients with HIVAN were black (Column K). Unadjusted survival curves
demonstrated better renal survival among patients with non-HIVAN lesions (P =
0.002)
(Column K). Patients with lesions other than HIVAN tended toward being more
likely to
be infected with hepatitis C (25/41 vs. 17/41, P = 0.08); and the presence of
hepatitis C
antibody was associated with a faster time to the institution of renal
replacement therapy
(HR 2.60, P = 0.01) (Column F). The authors concluded, inter alia, that
patients with
nephropathy other than HIV-associated nephropathy were more likely to have
hepatitis C
antibodies (Column H).
In another example, if analysis of population heat maps generated by the
Pipeline
Pilot software reveals a correlation between measles and hepatitis, as is
described below,
the exemplary system can retrieve the data from the exemplary CIP hypothesis
database
on measles from rows 9, 12, 27-31, and 34 and on hepatitis from rows 44-53,
and 55-157.
-207-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
A sampling of the kind of data regarding measles captured from the articles in
rows 9, 12, 27-31, and 34, is exemplified by the data from, for example, rows
27 and 34.
Row 27, for example, contains data captured from an article entitled "Measles
antibody
in vaccinated human immunodeficiency virus type 1-infected children," by
Stephen M.
Arpardi, et al., and published in Pediatrics, Volume 97(5): 653-57 (1996)
(Column J).
The captured data includes, for example, that ELISA (Column D) was used to
test for
measles antibody (Column A) in 81 perinatally HIV-infected children with prior
documented receipt of measles vaccine (i.e., monovalent measles or combination
measles-mumps-rubella), with a median age of 42 months at the time of study,
and a
median age of 14 months at first vaccination (Column K). Overall, 58 subjects
(72%)
had measles neutralization assay antibody titers > 1:5 (Column K). Children
studied
within 1 year of vaccination were significantly more likely to have detectable
measles
antibodies than those studied more than 1 year after vaccination (83% vs. 52%,
P < .01),
and the proportion of children with detectable measles antibody was greatest
for children
with no evidence of immunosuppression and lowest for children with severe
immunosuppression (Column F). The proportion of children with detectable
measles
antibody was significantly lower for children with CD8 greater than the 95`h
percentile
for their age and for children with lower CD4/CD8 ratios (Column F). The
authors
concluded, inter alia, that the prevalence of measles antibody in vaccinated
HIV-infected
children was considerably lower than in healthy children (only 72% of
previously
vaccinated children had measles antibody detected by neutralization assay, in
contrast to
95% among healthy children), that the proportion of children with detectable
measles
antibody declined over time, and that revaccination for seronegative HIV-
infected
-208-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
children was not likely to be effective once immunodeficiency was established
(Column
H). Row 34, for example, contains data captured from an article entitled
"Clinical
presentation of subacute sclerosing panencephalitis in Papua New Guinea," by
Charles S.
Mgone, et al., and published in Tropical Medicine & International Health,
Volume 8(3):
219-27 (2003) (Column J). The captured data includes, for example, that enzyme
immunoassay (EIA) (Column D) was used to test both serum and cerebrospinal
fluid for
measles-specific IgG antibody (Column A) in 95 children with a clinical
diagnosis of
subacute sclerosing panencephalitis (SSPE) from within Eastern Highlands
province of
Papua, New Guinea (Column K). Twenty-eight children had had measles, 28 had
not and
14 were uncertain; the verified mean age for contracting measles was 8.8 2.7
months,
the majority of children who had measles had contracted the infection in the
first year of
life, the mean age at which SSPE was manifested was 7.9 2.6 years, and the
time
between the measles illness and the onset of SSPE was 6.2 1.9 years (Column
F). The
authors concluded, inter alia, that high titers of measles antibodies are
found in the serum
(> 200,000) and/or cerebrospinal fluid (> 2000) of SSPE children (Column H),
and that,
although the pathogenesis of SSPE is not fully understood, the accumulated
evidence
suggests that it arises from persistence of altered measles virus in the brain
(Column K).
A sampling of the kind of data regarding hepatitis captured from the articles
in
rows 44-53, and 55-157, is exemplified by the data from, for example, row 133.
Row
133, for example, contains data captured from an article entitled "A
Seroprevalence
Survey of Hepatitis B Markers among Haitians in a Southwest Florida Farming
Community," by Michael D. Malsion, et al., and published in American Journal
of Public
Health, Volume 75(9): 1094-95 (1985). The captured data includes, for example,
that
-209-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
radioimmunoassays (Column D) were used to test for HBsAg (hepatitis B surface
antigen), anti-HBc (antibody to hepatitis B core antigen) and anti-HBs
(antibody to
hepatitis B surface antigen) (Column A) in 123 Haitian women attending a
prenatal clinic
in Immokalee, Florida, a small, migrant farm-worker community during a 12-
month
period (Column K). Twenty-eight out of 51 (55%) Haitian mothers had one or
more
HBV markers; 2/51 (4%) asymptomatic mothers were HBsAg positive and children
of
these women (aged 1 to 3 years) were negative for all HBV markers; 4/7 (57%)
of the
infants less than 6 months old and their mothers were antibody positive, but
none were
HBsAg positive; 3/54 (6%) of the children 1-4 years old were antibody positive
and none
were HBsAg-positive (Column F). The authors concluded, inter alia, that a
large
proportion of the Haitian women in Immokalee have been previously infected
with HBV,
and a small percentage are probably chronic HBsAg carriers; of the 7 HBsAg-
positive
women identified, only 1 was HBeAg positive; the infants of 2 HBsAg-positive
women
were negative for HBV markers, suggesting that the risk of perinatal
transmission of
hepatitis for infants born to HBeAg-negative women is low; and the small
portion of
children 1-4 years old with HBV markers suggests the risk for sib-to-sib
transmission in
this age group is also low (Column H).
If a correlation is found that is hitherto unknown (and thus interesting), a
user will want
to try to best explain the correlation as well as why it is or is not uniform
across various
segments of a given population. Providing such a user with both information
from
internal databases, as well as information obtained from external databases,
is thus very
useful. Such internal database could include, for example, the exemplary CIP
hypothesis
-210-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
database. External databases can include all known sources of immunological,
medical,
epidemiological, and related information, such as NIH, Medlin, PubMed, patent
databases, etc. An external database search can be launched, for example,
using a real
time external text analytics tool, such as is provided as a protocol in the
Pipeline Pilot
Software described above.
Thus, in exemplary embodiments of the present invention, after running an
automatic
data mining program on a given database, the results in the form of a set of
"interesting"
correlations can, for example, be generated. These correlations can then be
further
automatically processed, by running an internal as well as external text
search on them to
associate with each variable in the correlation known information that
hypothesizes a
basis for the correlation found. If no such hypothesis is available, which is
generally the
case for truly counter-intuitive and novel correlations, such information as
is then known
regarding each of the variables in the correlation can be marshalled via such
searching,
and can be output in a report of a user. This can assist such a user in
formulating a
hypothesis or in ruling one out.
H. EXPLANATION AND BASIS OF EXEMPLARY RULES CREATED FOR
PROCESSING CIP DATABASE
As described in the Automated Data Mining section above, a rules database was
created
for processing individual records or populations from the CIP database. This
was called
"Rules.xls" and is provided in Appendix A below. The following describes the
rationale
and basis for various ones of these rules.
-211-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
For example, Table 1 displays an exemplary set of the exemplary Canadian
Immigrant
Population assays and their interpretations.
Table 1: CIP Assays and Possible Interpretations
CIP Assay Possible Interpretation
Measles +/-/equivocal
Mumps +/-/equivocal
Rubella +/-/equivocal
Varicella +/-/equivocal
Tetanus OD --~ IU -> no interpretation
Diphtheria OD -* IU -~ no interpretation
C tome alovirus (CMV) +/-
Stron loides +/-/equivocal
Filaria +/-/equivocal
Schistosoma +/-
Hepatitis A +/-
Hepatitis B
HBc Ab +/-
HBs Ab + (not apparent)
HBs Ag reactive/non-reactive
HBe Ab reactive/non-reactive
HBe Ag reactive/non-reactive
Hepatitis C
HCV Ab +/-/"grayzone" = equivocal
HCVPCR +/-
HCV LIA +/-
IL-1 alpha reactive/non-reactive
IL-1 beta reactive/non-reactive
IL-2 reactive/non-reactive
IL-4 reactive/non-reactive
IL-5 reactive/non-reactive
IL-6 reactive/non-reactive
IL-8 reactive/non-reactive
IL-10 reactive/non-reactive
IL-12 70 reactive/non-reactive
IL-13 reactive/non-reactive
IL-15 reactive/non-reactive
IL-17 reactive/non-reactive
IL-23 reactive/non-reactive
-212-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
IFN- amma reactive/non-reactive
TNF-alpha reactive/non-reactive
TNF-beta reactive/non-reactive
Exemplary Rules for Measles-Mumps-Rubella (MMR) Testing
All patients are tested for measles-mumps-rubella antibodies. Positive results
require no
further action. Negative or equivocal results in any one of these assays would
indicate
need for immunization with MMR vaccine. Equivocal results in any one assay
should
require a booster immunization, while negative results would indicate a series
of two
immunizations.
As with all live virus vaccines, women known to be pregnant should not receive
the
MMR vaccine, and pregnancy should be avoided for four weeks following
vaccination
with MMR. However, women who are breast-feeding can be vaccinated. Children
and
other household contacts of pregnant women should be vaccinated according to
the
recommended schedule.
Severely immunocompromised persons should not be given MMR vaccine. This
includes
persons with conditions such as congenital immunodeficiency, AIDS, leukemia,
lymphoma, generalized malignancy, and those receiving treatment for cancer
with drugs,
radiation, or large doses of corticosteroids. Household contacts of
immunocompromised
people should be vaccinated according to the recommended schedule.
-213-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Although persons with AIDS or HIV infection with signs of serious
immunosuppression
should not be given MMR, persons with HIV infection without symptoms can and
should
be vaccinated against measles.
Exemplary Rules for Varicella Testing
All patients are tested for varicella antibody. Positive results require no
further action.
Children under the age of 13 with no history of chicken pox and negative
results should
receive two immunizations four to eight weeks apart. Children with equivocal
results and
no history of disease should receive one booster immunization and be retested
after one
year.
Herpes zoster (shingles) is a currently a risk for patients over 60 years of
age. Patients in
that age category, who do not have an immunodeficiency, and have a negative or
equivocal result, should receive one dose of zoster vaccine specifically
formulated for
adults. An immunodeficiency in these individuals would include a history of
primary or
acquired immunodeficiency states including leukemia; lymphomas of any type, or
other
malignant neoplasms affecting the bone marrow or lymphatic system; or AIDS or
other
clinical manifestations of infection with human immunodeficiency viruses. This
vaccine
is not indicated for women of child-bearing age and should not be administered
to
pregnant females.
Exemplary Rules for Tetanus Testing
All patients should be tested for tetanus antibody. Patients with less than
the minimum
protective level of 0.01 International Units (IU)/mL should be given a booster
dose of
-214-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
tetanus vaccine. Individuals over seven years of age receive the vaccine in
combination
with diphtheria vaccine (Td). Those children younger than seven years of age
can be
boosted with a vaccine containing a pertussis component (DTaP). People who had
a
serious allergic reaction to one dose of tetanus toxoid should not receive
another.
Persons with a moderate or severe acute illness should postpone receiving the
vaccine
until they are improved.
Exemplary Rules for Diphtheria Testing
All patients should be tested for diphtheria antibody. Patients with less than
the
minimum protective level of 0.01 IU/mL should be given a booster dose of
diphtheria
vaccine. Individuals over seven years of age receive the vaccine in
combination with
tetanus vaccine (Td). Those children younger than seven can be boosted with a
vaccine
containing the pertussis component as with the tetanus vaccine described
above. People
who have had a serious allergic reaction to one dose of DTaP, DT, Td, or Tdap
vaccine
should not receive another. Persons with a moderate or severe illness should
postpone
receiving the vaccine until their condition has improved.
Exemplary Rules for Cytomegalovirus (CMV) Testing
All patients should be tested for CMV antibody. A negative result would
require no
specific action.
As previously described, a positive result may be indicative of possible
immunosupression in these individuals as they age. Periodic diagnostic
monitoring of
-215-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
these patients with positive antibody levels should be triggered and
furthermore, increase
in frequency as they age.
As annual flu immunizations and periodic booster immunizations against
pneumococcal
infection have little effect in the elderly with high CMV titers, patients
with very high
levels of CMV antibody potentially may have different vaccine recommendations
than
the general population:
= Patients over 65 years of age with repeatedly high CMV levels should have
their
younger contacts (children and grandchildren) vaccinated annually against
influenza. In addition, these same contact individuals should have their
pneumococcal vaccinations up-to-date. Depending on recommendations by the
ACIP and AAP, these individuals may possibly be recommended to not receive
the annual influenza immunization.
= Patients over 65 years of age with repeatedly high CMV antibody levels
should
also be regularly screened for routine Thl/Th2/Treg/Th17 cytokine levels to
assess immune balance and any autoimmune diseases should be closely monitored
by regular ImmunoScore diagnostic screening.
= Patients 50-65 years of age with repeatedly high CMV levels should have
regular
influenza vaccinations and be checked every 2-5 years for antibody levels to
pneumococcal polysaccharides used in vaccines currently marketed. Similar to
the elderly group, patient contacts (children and grandchildren) should also
have
up-to-date influenza and pneumococcal immunizations.
-216-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= Patients 50-65 years of age with repeatedly high CMV antibody levels should
also
be regularly screened for cytokine levels as described above. Onset of
autoimmune disease and flares should be monitored closely by the health care
providers.
= Patients younger than 50 years of age with repeatedly high CMV antibody
levels
should be regularly immunized against influenza, and should be examined every
2-5 years for antibody levels to pneumococcal polysaccharides used in current
vaccines.
= Patients younger than 50 years of age with repeatedly high CMV antibody
levels
should be regularly screened for cytokine levels.
= Patients younger than 50 years of age with repeatedly high CMV antibody
levels
may be an ideal group to test with immunotherapeutics under development.
Exemplary Rules for Strongyloides Testing
All patients should be tested for Strongyloides stercoralis antibody. Negative
tests
require no follow-up action. A positive or equivocal result would indicate the
further
examination of stool samples. Microscopic examination of stool specimens is
insensitive; estimates for a single positive stool examination in cases of
uncomplicated
infection range from 0 to 66%. To overcome this lack of sensitivity,
investigators have
recommended examination of up to seven stool specimens; use of more sensitive
and
labor-intensive methods of stool examination; use of agar plate cultures; and
collection of
alternate specimens, such as duodenal aspirates.
-217-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Due to these difficulties in diagnosing the progress of strongyloides
infections, reported
efficacies of drugs used to treat strongyloides infection vary widely.
Chemotherapy is
advocated and considered an effective control measure for the reduction of
morbidity
resulting from intestinal nematode infection. The current drug of choice for
strongyloides is the benzimidazole compound, thiabendazole. This drug requires
a three
day regimen. Another drug being considered is ivermectin, which may be
preferable,
because it requires only one dose. Post-treatment follow-up testing
recommendations
would require stool sampling 30 days post-treatment.
Exemplary Rules for Filaria Testing
All patients should be tested for antibody to filarial worms, Wuchereria
bancrofti and
Brugia malayi. Negative tests require no follow-up action.
Filarial worms reside in the lymphatic system, and therefore likely have a
great impact on
the body's immune defense systems. Though infection is usually acquired early
in
childhood, filarial disease can take years to manifest. Many infected
individuals develop
no clinical symptoms underscoring the need for routine diagnostic testing in
endemic
areas.
Individuals that are positive for filaria antibody should be treated with a
combination of
albendazole with either diethylcarbamazine or ivermectin. This treatment has
been
shown to be over 99% effective in removing microfilariae from the blood for a
full year
after treatment. Seropositive individuals should be screened one year after
treatment.
-218-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Patients that have equivocal antibody levels should be re-tested after several
weeks to
determine if they are in the early stages of infection. If they then have a
positive result,
they should be treated as above. If negative, no further action is required.
Exemplary Rules for Schistosoma Testing
All patients should be tested for worms that cause schistosomiasis. Patients
that have
negative tests require no follow-up action.
Patients are infected by contact with water used in normal daily activities
such as
personal or domestic hygiene and swimming, or by professional activities such
as fishing,
rice cultivation and irrigation. Schistosomiasis is endemic in 74 tropical
developing
countries. Some 600 million people are at risk of becoming infected.
Population
movements and refugees in unstable regions contribute to the transmission of
schistosomiasis.
Patients who have a positive antibody result should be treated depending on
the
manifestation of illness. Praziquantel is used to treat all forms of
schistosomiasis.
Oxamniquine is used exclusively to treat intestinal schistosomiasis, and
metrifonate is
effective for the treatment of urinary schistosomiasis. No further
intervention is typically
needed following treatment for 2-5 years. According to the World Health
Organization
(WHO), treatment of schistosomiasis must be accompanied by health education to
preclude re-infection.
-219-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Exemplary Rules for Hepatitis A Virus (HAV) Testing
All patients should be tested for total antibody to HAV. HAV testing evaluates
total
antibody levels to HAV among patients. If a patient tests positive for anti-
HAV, a further
test for anti-HAV IgM is performed to determine presence of acute infection.
Currently
approved assays do not detect less than 100 mIU/mL of antibody, yet levels as
low as 10
to 20 mIU/mL are thought to confer protection. The CDC does not currently
recommend
revaccination of healthy individuals with undetectable antibody levels.
The most likely time for an HAV-infected person to spread HAV to others is
during the
two weeks before the infected person develops symptoms. Clearly, if a person
doesn't
even know that he or she is infected, it makes it difficult to protect others
from getting the
infection. The risk of spreading HAV becomes smaller over time and can still
be present
one week or longer after symptoms develop (e.g., yellowing of skin and eyes).
Infants are
more likely to be capable of spreading HAV for longer periods of time.
If an unvaccinated person thinks that he or she might have been exposed, he or
she
should call their health professional immediately to schedule an appointment
to
determine whether a real exposure has occurred and whether Ig should be
administered.
Ig is a concentrated dose of human antibodies that includes anti-HAV. In most
cases, this
preparation can protect an exposed person from developing HAV infection.
People at increased risk for exposure to HAV infection or those who are more
likely to
get seriously ill if infected with HAV should be vaccinated. According to CDC
recommendations, these individuals include
-220-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= All children at age 1 year (12-23 months)
= People aged 12 months or older who are traveling to or working in any area
of the
world except the United States, Canada, Western Europe, Japan, New Zealand,
and Australia
= Men who have sex with men
= Illegal drug users, both oral and injecting
= People who have blood clotting disorders
= People who work with HAV-infected primates or with HAV in a research
laboratory setting. No other groups have been shown to be at increased risk
for
HAV infection because of occupational exposure.
= People with chronic liver disease are not at increased risk of getting
infected, but
are at risk for developing serious complications if they get infected.
= Any person who wishes to be immune to hepatitis A
Hepatitis A vaccine is NOT routinely recommended for healthcare workers,
sewage
workers, or daycare providers. Children who are not vaccinated by age two
years should
be vaccinated as soon as feasible.
-221-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Exemplary Rules for Hepatitis B Virus (HBV) Testing
All patients are tested for anti-HBs (antibody to hepatitis B surface antigen)
and anti-HBc
(antibody to hepatitis B core antigen). Depending on the results of these two
tests,
additional testing will be performed as follows (Fig 1):
POSITIVE/POSITIVE If the patient has positive results for both tests, the
HbsAg
(hepatitis B surface antigen) is measured. This test is done to determine
whether patients
are chronic carriers of HBV infection. In cases in which the HBsAg test is
negative, no
further tests are performed, and the results are interpreted as indicating a
patient exposed
to hepatitis B virus that has cleared the virus. If the HBsAg test is
positive, the person is
identified as a chronic carrier of hepatitis B. Patients positive for HBsAg
are further
tested for the presence of antibody for hepatitis B core antigen IgM (IgM anti-
HBc),
hepatitis B e antigen (HBeAg), and antibody to hepatitis B e antigen (anti-
HBe) to
evaluate the level of viral replication.
POSITIVE/NEGATIVE Patients testing positive for anti-HBc only are then tested
for
HBsAg. Whether the HBsAg test is positive or negative, the plasma is further
tested for
IgM anti-HBc, HBeAg, and anti-HBe to evaluate the level of viral replication.
Any
positive result in this series of tests indicates that the patient is a
chronic HBV carrier.
NEGATIVE/POSITIVE Patients testing negative for anti-HBc but positive for anti-
HBs
may have been either vaccinated or exposed naturally to HBV, or this may
represent a
false-positive result (no exposure). These individuals will be retested the
following year.
-222-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
NEGATIVE/NEGATIVE Patients testing negative for both anti-HBs and anti-HBc
have
not been exposed to HBV and will be retested the next year.
It is estimated that about 1 out of 3 of the nearly I million Americans with
chronic HBV
infection acquired their infection as infants or young children. Those with
chronic HBV
infection are most likely to spread the infection to others. Infants and
children who
become chronically infected have an increased risk of dying prematurely from
liver
cancer or cirrhosis.
In contrast to other vaccine-preventable diseases of childhood, HBV infection
in infants
and young children usually produces no symptoms. Thus, the small number of
reported
cases of hepatitis B among children represents the tip of the iceberg of all
HBV infections
in children. For every child with symptoms of hepatitis B, there are at least
100 HBV-
infected children with no symptoms - hence the increased risk to spread the
infection to
others without knowing it.
Second, early childhood infection occurs. About 16,000 children under 10 years
of age
were infected with HBV every year in the United States before routine infant
hepatitis B
vaccination was recommended. Although these infections represented few of all
HBV
infections in the United States, it is estimated that 18 out of 100 people
with chronic
HBV infection in the United States acquired their infection during early
childhood.
Clearly, infections occur among unvaccinated infants born to mothers who are
not HBV-
infected. In addition, unvaccinated foreign-born children account for a high
proportion of
infections. More effort needs to be placed on vaccinating these unprotected
children.
-223-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Hepatitis B vaccine, usually a three-dose series, is recommended for all
children 0-18
years of age. It is recommended for infants beginning at birth in the
hospital. All older
children who did not get all the recommended doses of hepatitis B vaccine as
an infant
should complete their vaccine series as soon as possible. Most states require
hepatitis B
vaccine for school entry. Adolescents who are just starting their series will
need two or
three doses, depending on their age and the brand of vaccine used. Adults at
increased
risk of acquiring HBV infection should also be vaccinated. In addition, the
vaccine can be
given to any person who desires protection from hepatitis B.
Groups of adults at increased risk of HBV infection
= Healthcare workers and public safety workers with reasonably anticipated
risk for
exposure to blood or blood-contaminated body fluids
= Men who have sex with men
= Sexually active people who are not in long-term, mutually monogamous
relationships
= People seeking evaluation or treatment for a sexually transmitted disease
= Current or recent injection drug users
= Inmates of long-term correctional facilities
= People with end-stage kidney disease, including predialysis, hemodialysis,
peritoneal dialysis, and home dialysis patients
-224-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= Staff and residents of institutions or group homes for the developmentally
challenged
= Household members and sex partners of people with chronic HBV infection
= Susceptible (non-infected) people from United States populations known to
previously or currently have high rates of childhood HBV infection, including
Alaska Natives, Pacific Islanders, and immigrants or refugees from countries
with
intermediate or high rates of chronic HBV infection.
= International travelers to regions with high or intermediate rates of HBV
infection.
In addition, any adult who wishes to be protected from HBV infection should be
vaccinated without having to acknowledge a specific risk factor.
Exemplary Rules for Hepatitis C Virus (HCV) Testing
Testing protocol begins with screening patient sera for the presence of
antibody to HCV.
POSITIVE If the patient tests positive for anti-HCV with a ratio of optical
density of
sample signal to optical density of cutoff signal (S/C ratio) between 1.0 and
3.0, a
confirmatory test is done to rule out a false positive anti-HCV result.
Confirmatory tests
are usually both performed (either PCR and LIA as in the CIP study, or PCR and
RIBA).
If the PCR result is positive, then positive anti-HCV will be considered a
true positive.
If both the confirmatory tests are negative, the antibody result is considered
to be a false
positive. If the RIBA is indeterminate and the PCR is negative, then the
interpretation of
-225-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
the positive anti-HCV result is uncertain. It could be a false positive, or
the person could
be chronically infected with HCV or in the process of seroconversion.
All patients who are found to have no or questionable evidence of past
exposure to HCV
(Figure 3) will be retested the following year.
NEGATIVE If the anti-HCV test is negative, PCR testing for the presence of HCV
RNA
is conducted to rule out a false negative anti-HCV test. If the PCR test is
positive, the
patient is considered HCV infected and no further testing need be performed.
If the PCR
result is negative, the patient will be retested the following year.
Figure 21D-38 depicts an exemplary algorithm for Hepatitis A Virus (HAV)
testing, Fig.
21D-39 depicts an exemplary algorithm for Hepatitis B Virus Testing, and Fig.
21D-40
depicts an exemplary algorithm for Hepatitis C Virus (HCV) Testing, according
to an
exemplary embodiment of the present invention.
1. INTERPRETATION OF CERTAIN RESULTS OF AUTOMATED DATA
MINING
Heat Map Correlations
Parasitic Worms and Hepatitis C
The various automated data mining protocols described above (and whose
complete code
is provided in Appendix D), can, for example, create population heat maps from
the
exemplary CIP database. These population heat maps may show positive, negative
or no
correlation between and among various assays (or markers) or between and among
assays
and other variables within the CIP population. For example, the population
heat maps
-226-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
revealed a correlation between antibody reactivity to parasitic worms and
hepatitis C
viruses. To explain this an exemplary system could, for example, automatically
consult
an internal hypothesis database to search for possible explanations for the
correlation. As
can be seen from the CIP hypothesis database described above and provided in
Appendix
B, none of the parasitic worm references discuss hepatitis and none of the
hepatitis C
references discuss parasitic worms. Thus, in exemplary embodiments of the
present
invention, the exemplary system could, for example, then launch an internet
search and
access various internet databases, such as, for example, PubMed, MedLine,
Science
Direct, and NIH, as well as the Internet in general, to find any information
that might
offer an explanation regarding the observed correlation. In the present
example, a search
of scientific articles available in the PubMed database provided a basis for
this observed
worms-hepatitis C correlation as described below.
Hepatitis C virus (HCV) infection is the main cause of chronic liver disease
in Egypt and
is largely associated with schistosomiasis. Concomitant infection with HCV and
schistosomes can cause aggravation of liver damage. These two infectious
agents have
been shown to have similar adverse effects on the immune system, as manifested
by their
action on cytokine production by Thl and Th2 cells. Patients coinfected with
hepatitis C
virus and Schistosoma mansoni show high incidence of viral persistence and
accelerated
fibrosis. It is possible that enhancement of a Th2 response in co-infected
individuals
plays a role in persistence and severity of HCV infection in patients
concomitantly
infected with S. mansoni.
Patients infected with schistomsoma frequently show a high seroprevalence of
anti-
hepatitis C virus (anti-HCV) antibodies. The exact underlying mechanism by
which
-227-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
schistosomiasis enhances HCV seropositivity is unknown. In Egypt, evidence
suggests
that individuals over the age of 40 have been exposed more to the risk of HCV
infection
through inadequately sterilized needles used in mass anti-schistosomal
treatment
campaigns conducted from the 1960s through the 1980s. Some researchers have
postulated that patients designated low positive for HCV antibody perhaps were
falsely
positive due to the generation of autoantibodies in connection with
Schistosoma mansoni
infection.
A striking clinical feature of HCV infection is that more than 50% of patients
with acute
HCV develop chronic infection. It has been noted that activation of Th2
responses seems
to play a role in the development of chronicity in these patients. However,
the possibility
of helminth or nematode co-infection in these patients was not examined.
Another
research group investigating the possible diagnostic role of IL- 10
measurement,
postulated that elevated IL-10 correlated in HCV-positive schistosomal
patients with the
development of morbidity. Co-infected individuals appear to have increased Th2-
related
cytokines, and schistosomiasis may down regulate the normal stimulatory effect
that
HCV infection would have on Thl cytokines, leading to the chronicity of HCV
infection
and playing a role in unresponsiveness to interferon therapy in co-infected
patients. The
same group of researchers found that HCV infection correlated with an
alteration of
serum immunoglobulins in patients with chronic liver disease.
A murine model of schistosoma/hepatitis co-infection demonstrated that
suppression of
the antiviral type I interferon response by schistosome egg Ags in vivo
predisposed the
liver to enhanced viral replication with ensuing immunopathological
consequences. It is
-228-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
though that this model might be paralleled in human schistosome/hepatotropic
virus co-
infections, including hepatitis B and hepatitis C viruses.
During the time of egg deposition, schistosome-infected mice exhibit a down
regulation
of interleukin 2 and gamma interferon production toward parasitic antigens,
mitogens,
and foreign non-parasite protein antigens. A group of researchers found that
mice
infected with virus alone rapidly cleared the virus, while in animals co-
infected with virus
and S. mansoni, viral clearance was delayed by as much as 3 weeks in the liver
and by
several days in the spleen and lungs. These observations suggest that helminth
infection
may influence immune responses to concurrent viral infections.
A cohort study conducted in the Philippines found that males infected with
schistosomes
consistently produced higher levels of Th2 cytokines, and also had a higher
prevalence of
liver fibrosis.
A case-controlled study was recently undertaken to describe the prevalence of
Strongyloides stercoralis infection among patients with autoimmune liver
disease, such
as primary biliary cirrhosis, autoimmune hepatitis, and primary sclerosing
cholangitis.
The authors of that study hypothesized that immunomodulation by S. stercoralis
infection
may lower the incidence of autoimmune liver disease.
Strongyloides stercoralis infection has been shown to be related to increased
risk in
alcoholic cirrhosis. The same study found no increased risk for non-alcoholic
cirrhosis.
Measles and Hepatitis
In another example, the population heat maps revealed a correlation between
antibody
reactivity for measles and hepatitis viruses. The exemplary system first would
-229-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
automatically consult the internal CIP hypothesis database to search for
possible
explanations for the correlation. As can be seen from the CIP hypothesis
database
described in Appendix B, none of the measles references discuss hepatitis and
none of the
hepatitis references discuss measles. In exemplary embodiments of the present
invention,
the exemplary system would then launch an internet search and access various
internet
databases, such as, for example, PubMed, MedLine, Science Direct, and NIH, as
well as
the internet in general, to find any information that might offer an
explanation regarding
the observed correlation. In the present example, a search of scientific
articles available
in the PubMed database, provided a basis for this observed correlation as
described
below.
The hepatotropic viruses, measles and herpes viruses have been shown to act
presumably
as a trigger in patients with autoimmune hepatitis.
Adult syncytial giant cell hepatitis (GCH) is an uncommon and often fulminant
form of
hepatitis that may be caused by infection with a novel paramyxo-like virus. In
situ
hybridization studies showed that the disease agent was genetically related to
the measles
virus. One group of researchers concluded that paramyxoviruses should be
considered in
patients with severe sporadic hepatitis.
Epstein-Barr virus has a seroprevalence of more than 80% worldwide and is
known to be
associated with hepatitis. However, little is known about the underlying
pathogenesis
and immune system mechanisms and there are no standard diagnostic criteria for
diagnosing EBV-hepatitis available.
-230-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Viral infections of the mesenteric microvascular endothelium have been
hypothesized as
pathogenic factors in inflammatory bowel disease. The detection of anti-
measles virus
IgM in the majority of patients with Crohn's disease and in about one-half of
ulcerative
colitis patients as compared to a very low prevalence in patients with other
chronic
inflammatory disease is consistent with the hypothesis that the measles virus
has
pathogenic implications in inflammatory bowel diseases.
J. EXTENSION OF DATABASE AND AUTOMATIC DATA MINING
FUNCTIONALITY
In exemplary embodiments of the present invention, the database could be
augmented by
utilizing electronic medical records, such as Google's Personal Health Record,
for
example, to supply the non-assay information. In such embodiments, an
individuals
health records could be automatically downloaded to his database record each
time they
are updated.
In exemplary embodiments according to the present invention, a database and
analysis
system can be fully integrated with any computer system used to perform any of
the
applications described below in Section III, supplying, as it were, the back
office number
crunching conclusiosn to be oprated upon by any scoring, decision making, or
other
application system or software as may be useful.
Additionally, using the correlation matrix and data mining techniques
described above,
algorithms designed to operate on images, such as, for example, pattern
recognition
algorithms and other image processing algorithms, such as, for example, edge
detection
-231-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
and morphology, could be used to refine correlation regions in a "variable
space" to
automatically further process the data to find correlations of interest, as
well as to find the
precise segmentation in said variable space to maximize the value of the
correlation and
thus find the group of individuals that has the most in most common as to that
correlation. By repeating this process for every identified correlation and
comparing the
results, i.e., the different optimal segmentations of the database for each
correlation, the
most can be learned about what is driving the correlation.
Finally, derived variables, such as the rate of change of antibody levels
through time, the
ratio of various bioassay results, etc., can be added to the database and
correlations
identified with respect to these derived "second order" variables. It may be
that a
connection or phenomenon only manifests at the level of such seconday, or
tertiary or n-
ary variables, and only an automated process that methodically processes the
data over
and over with complex algorithms can bring to light all the information buried
therein.
In exemplary embodiments of the present invention, the time rate of change of
an assay
variable over time, as well as the second derivative with respect to time of
that assay
variable. Algorithms can easily be crafted that track any changes in the assay
values over
time, as well as the second derivative of such assays with respect to time.
This can be
especially useful in analyzing cytokine data, which tends to fluctuate with
inflammations,
colds and flu, but which often exhibits a baseline balance between Thl and Th2
categories, for example.
-232-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
K. EXEMPLARY ANALYSES PERFORMED ON CIP DATABASE
In exemplary embodiments of the present invention, a sample can be first
analyzed by
assaying all or a plurality of cytokines, and then, based on the individual's
"cytokine
signature" automatically spawning an additional panel or superpanel of assays
to
perform. Such an exemplary embodiment utilizes the cytokine's systemic
immunological
qualities to predict or locate potential areas for further study. Such a
predictive algorithm
can be based, for example, upon where an individual's cytokine signature lies
within the
Thl-Th2-Th17-Treg two dimensional space shown n Fig. 5D. For example, an
individual with a high ThI and high Th17 profile could have his serum
automatically
tested for automimmune markers, or for example, antibodies to bacterial
infections.
Another individual with a high Th2 and high Th 17 profile could have his serum
automatically tested for markers of allergy or atopic disease. Yet another
individual with
high Thi and Treg profile could automatically be tested for a chronic
mycobacterial
infection, while another individual with high Th2 and Treg could be tested
automatically
for parasitic infection.
Figure 21E-1 shows the interpretation of immunoassay results of a random
sampling of
the Canadian Immigrant Population (CIP) database by Pipeline Pilot,
categorizing the
results as either reactive (positive), non-reactive (negative), or equivocal
(marginal results
interpreted as neither positive nor negative). Significant percentages (18-20%
of total) of
the results of two of the immunoassays for parasites from the panel (filaria
and
strongyloides) were classified as equivocal, possibly indicating a nascent
infection in
these individuals with these parasites.
-233-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Figure 21 E-2 shows several signature Th2 cytokine assays compared with
results from
parasite antibody assays. For filaria (Figure 21 E-2a), the highest amount of
reactive
results for both IL-4 and IL-5 occur in the filaria equivocal patients. The
same is true for
IL-5 assay results for strongyloides equivocal patients (Figure 21E-2b). If we
are to
logically assume that equivocal results for these assays were to become
positive results in
the near future for these patients, the increased levels of IL-4 and IL-5 as
signature Th2
cytokines may be indicative of early stages of infection with both of these
parasites. It is
conceivable that as the ImmunoScore database grows, results such as this might
be seen
as indicative of early intervention against parasitic infection in areas where
such
infections are endemic. A patient with an equivocal filaria result might not
ordinarily be
treated, but elevated Th2 cytokine levels in individual patients might call
for treatment in
patients that are equivocal for filaria and have elevated levels of both IL-4
and IL-5, for
instance. In strongyloides patients, the initial flare of Th2 cytokines seems
counterbalanced by increased expression of inflammatory cytokines, IL-6 and
TNF-a as
shown in Figure 21 E-2c. An increased percentage of patient population
positive for these
cytokines is seen in samples that are not negative (e.g. reactive and
equivocal
strongyloides assay results).
Patients with serum antibody reactive to Hepatitis B core antigen are
considered to have
an active hepatitis B infection. This patient population had higher levels of
inflammatory
cytokine TNF-a than did patients non-reactive to core antigen (Figure 21 E-3).
Seropositivity to cytomegalovirus (CMV) has been linked to an aging immune
system,
and inability to deal with debilitating infection in the elderly (e.g.
influenza and
Streptococcus pneumoniae). Figure 21 E-4 shows the CIP database examined for
-234-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
seropositivity to CMV, drawing a distinction between very high seropositivity
(>250) and
lower serum antibody levels. It can be observed that females are more likely
than males
to have very high serum antibody levels to CMV, likely due to more
interactions with
very small children. Examination of the region of origin of the CIP draws
clear
distinctions in this population. Individuals from sub-Saharan Africa, Southern
Asia, and
North Africa are more likely than not to have very high levels of serum
antibody to
CMV, while individuals from the Latin America/Caribbean region, Eastern
Europe, or
Southeast Asia are more likely to have fewer serum antibodies to CMV.
Examination of
the age of the total population shows a clear increase in CMV seropositivity
correlated
with advancing age.
Figure 21E-5 examines the trend for increased serum cytokine levels plotted
against
CMV reactivity. The general trend for six of the eight cytokines examined is
for
increased serum levels of cytokines correlated with increased levels of CMV.
One
notable significant exception in this population is observed in the levels of
IFN-y, which
are inversely correlated with CMV seropositivity. It has been noted that
elderly
individuals are more prone to viral infection, and decreased levels of IFN--y
in these
individuals could be highly significant.
Serum cytokine levels were similarly examined vs. percent of CIP that had
serum
antibodies to filaria (Figure 21 E-6). These results indicated higher levels
of IL-4, IL- 10,
TNF-a, IL-17, and TNF-(3 to be associated with filarial infection, with no
notable
increases in the other cytokines. The increased level of IL-4 might be
expected based
upon reports of parasitic infections being correlated with overall increase in
Th2
cytokines. IL-10 is associated with Treg cells and could also be expected as a
regulatory
-235-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
factor in dampening the Th2 response. The increased levels of TNF-a, TNF-(3,
and IL-17
could be indicative of inflammation and could also be a cause for the
increased levels of
IL-10. It is interesting that the apparent patterns of IL-17 and IL-23 are
somewhat
incongruous, seeing that it is reported that IL-23 expression is necessary to
maintain a
TH17 response. Other cytokines, reported to be pro-inflammatory, such as IL-8,
IL-1,
and IL- 15 are not remarkable in their pattern of expression as related to
filarial infection.
Figure 21E-7 shows a decrease in IFN-y levels in individuals seropositive for
hepatitis A.
This is somewhat unexpected in that IFN-y is considered crucial for the combat
of viral
disease. Similar results were seen with individual positive for hepatitis B
core antigen
(data not shown).
An interesting cytokine profile is presented in Figure 21E-8 of individuals
possessing
serum antibodies to Strongyloides. These individuals show decreasing levels of
TNF-
a,TNF-0, IL-6,IL-17, IL-15, IL-8, IL-2, and IL-5, with a concomitant increase
in IFN-y,
IL-23,IL-10 and IL-4 levels. The IL-4 increase could be due to increased Th2
expression, but this leaves the decrease in IL-5 expression more difficult to
explain. The
IL- 10 increase with the concomitant decrease in many other cytokines could
possibly be
a demonstration of the suppressive effects of Treg cells expressing increased
amounts of
IL-10. Many of the pro-inflammatory cytokines are decreasing correspondent to
Strongyloides antibody positivity, but there is an increase in IFN-y and IL-
23. Clearly
the pattern of cytokine expression is complicated and worthy of further study.
Figure 21E-9 shows a multi-variate analysis of components of the CIP database
incorporating the IL-6:IL-2 cytokine ratio vs. serum levels of anti-CMV
antibody (top
panel) and anti-hepatitis B antibody (bottom panel) and color coded by age.
This
-236-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
particular representation is not necessarily informative, but rather
demonstrative of the
types of analyses that can be accomplished by ImmunoScore technology.
CMV Reactivity/Non-reactivity
Fig. 21E-11, Table 1, shows the percentages of the CIP that are positive for
at least one
cytokine in the four cytokine categories previously described - Th1, Th2,
Treg, and
Th17, and analyzes this positivity in relationship to positive serological
tests for disease-
specific antibodies. The mean values of the entire population are indicated in
green, and
significant deviations from the mean of the entire population greater than 1.0
are
indicated in yellow. Thus, Fig. 21 E- 11, Table 1, shows that patients that
are non-reactive
for CMV antibody have significantly lower levels of Th1, Th2, and Th17
cytokines than
do subjects that are seropositive for anti-CMV antibody. Also significantly,
these same
patients that are non-reactive to CMV have higher levels of Treg cytokines.
Taken
together, these results indicate that these patients are far less likely to
suffer from chronic
inflammatory conditions based upon their cytokine profiles. Patients that are
seropositive
or CMV antibody show the opposite - that is, they have elevated levels of TH
1, Th2, and
Thl7 cytokines and reduced levels of Treg cytokines. Due to the large number
of CMV
positive individuals (94% of CIP database), however, these results do not rise
to the level
of significance.
Hygiene Hypothesis Revisited
The hygiene hypothesis has described a state wherein parasitic infections in
third world
countries help prevent patients from atopic/allergic conditions like asthma by
tipping the
Thl/Th2 balance toward a more Th2-like state in those individuals. According
to the
hypothesis, we would expect to see a relative boost in the levels of Th2
cytokines in
-237-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
patients that are seropostive for infections with filaria, strongyloides and
schistosoma.
The results in Fig. 21E-11 indicate that while there are increases in levels
of Th2
cytokines for patients infected with strongyloides and filaria, those patients
with filarial
infections do not rise to the level of significance. Curiously, patients
infected with
schistosoma actually show a decrease in the level of Th2 cytokines that also
does not rise
to the level of significance. Also counter to the stated hygiene hypothesis,
the levels of
Thl cytokines are increased in patients seropostive for strongyloides
(significantly) and
filaria. Levels of Th17 cytokines are increased in patients seropositive for
filaria
(significantly) and strongyloides. Perhaps the most striking of the
observations is that all
three parasitic infections showed increases in the levels of Treg cytokines,
with patients
seropositive for filaria and strongyloides showing significantly higher
increases than the
rest of the CIP database. It is likely that the story behind the hygiene
hypothesis is more
complicated that a shift in Thl/Th2 cytokine balance and needs to consider the
contributions of Thl7 and Treg cytokine-producing cells.
Mining the CIP Database
The Canadian Immigrant Population Database is organized by large geographic
regions
of origin of the subjects. These regions are defined as follows:
= Region 1 = Sub-Saharan Africa
= Region 2 = South Asia
= Region 3 = North Africa
= Region 4 = Latin America/Caribbean
= Region 5 = Eastern Europe
-238-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= Region 6 = Southeast Asia
The Data Mining Tool was used to examine cytokine levels together with
antibody levels
to vaccine-preventable diseases, three parasitic infections (filaria,
schistosoma, and
strongyloides), viral hepatits infections (A, B, and C), and infection with
cytomegalovirus. Patients were grouped according to region of origin and
gender, and
place in bins based upon their ages, separated by 20 year increments. The age
categories
in this instance were from 10-30 years of age, 30 through 50, 50 through 70,
and 70
through 90. The tool generated 39 separate heat maps (figures 21E-10.1 through
21E-
10.21) based upon these classifications. The heat maps produced by the data
mining tool
display very interesting patterns of correlation based upon region of origin,
gender, and
age. Heat maps produced by population analyses with inadequate sample size
produce
very predictable, and uninteresting heat maps. These would include those heat
maps in
figures 21E-10.3 (top), 2113-10.4, 2113-10.8, 21E-10.12 (top), 21E-10.15, 21E-
10.18, and
21E-10.21.
Individuals in the CIP database originating from Sub-Saharan Africa are shown
in figures
21E-10.1 through 21E-10.4. Discounting those patients in Figure 21E-10.4 due
to small
sample size, the other individuals show very interesting patterns of
correlation between
cytokine expression and antibodies to the various antigens tested. In the
female
population under 30 years of age, cytokine patterns are largely inversely
correlated with
antibody reactivity to measles, hepatitis A, and hepatitis B core antigen.
Male subjects in
this same age category show some negative correlations with some cytokines and
varicella, and a separate group of cytokines and antibody to hepatitis B core
antigen. As
the population ages, these heat maps change dramatically (Figures 21 E-10.2
and 21 E-
-239-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
10.3). By the time the males reach middle age and older (50-70), there is a
striking
inverse correlation between all cytokine levels and parasitic exposure (e.g.
filaria,
schistosoma, and strongyloides shown in the bottom panel of Figure 21 E-
10.3). This
same group of individuals has a high degree of correlation among the parasitic
infections
themselves (red grouping at very bottom right of the bottom panel of Figure
21E-10.3).
In addition, there are interesting negative correlation patterns among the
cytokine
expression in this group itself - in particular, IL-8 and IL-5 expression are
inversely
correlated with a number of the other cytokines. Curiously, IL-8 (a pro-
inflammatory
cytokine) is inversely correlated with a number of other inflammatory
cytokines (IL-1.
IL-17 and IL-23), as well as anti-inflammatory cytokines IL-10, IL-4, and IL-
5.
Individuals in the CIP database originating from southern Asia display very
different heat
maps (Figures 21E-10.5 through 21E-10.7). The gender differences in heat map
profiles
as the population ages are striking. Isolating on the upper left quadrants (or
"cytokine
quadrant") in the six heat maps, one can see that the younger males (under age
50) have
better correlation values for the cytokine expression in general than do the
females. From
the total population analysis, it is not certain which of the genders would be
considered
more fit immunologically, but the differences are striking. Both of the older
groups
(Figure 21E-10.7) show many areas of negative correlation among the cytokine
quadrant,
particularly the females. At this point, it is unclear what the uncoupling of
cytokine
correlations mean for an aging population. Further study and analyses may
yield
important information regarding aging immune systems. It is possible that the
uncoupling of cytokine correlations is a sign of immune health in this
population - there
-240-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
is a wealth of information that could possibly be gleaned from ImmunoScore
population
analyses.
Regional geographic differences in population immune profiling can be seen
when
analyzing those individuals in the CIP database originating from North Africa
(Figures
21E-10.9 through 21E-10.11). The data in figure 21E-10.11 is likely not as
reliable due
to small sample size, but the younger individuals show striking cytokine
profiles. The
younger individuals (Figure 21E-10.9) are remarkable for their lack of overall
positive
cytokine correlation. Again, this may or may not be a sign of decreased immune
health,
but the profile changes in the older individuals in this population.
Particularly in the
males, there is much more positive correlation in the cytokine quadrant than
in the
younger group. The females from North Africa show very minor positive
correlations in
the cytokine quadrant when compared to other regions around the globe. This
becomes
obvious when examining the female populations from Latin America/Caribbean
(Figures
21E-10.12 through 21E-10.14). The females in this group have striking positive
correlations in the cytokine quadrant particularly in the 10-30 and 30-50 age
groups, but
the positive correlations are still prevalent in the 50-70 age group. In
addition to the
across the board positive correlation for cytokine values, there is a very
noticeable
negative correlation among cytokine values and hepatitis A infection in 30-50
year old
Latin American females (Figure 21E-10.13 - top). The younger Latin American
males
show some interesting negative correlations between cytokine values and
filaria
infection, and to a lesser extent, rubella, hepatitis A, measles, tetanus, and
varicella
(Figure 21 E-10.12 - bottom).
-241-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The demographic groups from Eastern Europe and Southeast Asia were smaller,
but
interesting nonetheless. In the younger groups from Eastern Europe (Figure 21
E-10.16),
there were interesting negative correlations between the cytokine assays and
the CMV
antibody assay. CMV infection is not as prevalent in Eastern Europe as some
other areas.
of the world, and it would be interesting to investigate this relationship
further,
particularly in the light of the role CMV infection plays in immune
senescence.
Hepatitis A and rubella also appear to have significant areas of negative
correlation with
cytokine expression profiles in these groups. Finally, in the group from
Southeast Asia,
the younger males (Figure 21E-10.19 - bottom) showed an interesting pattern of
reactivity between the cytokine assays and the hepatitis B_e antigen assays.
There were
areas of strong positive correlation (e.g. IL-1, IL-4, IL-5, IL-6, and IL-17)
and also areas
of strong negative correlation (IL-2, IL-8, and IL-23). There is an
interesting un-coupling
of the Th 17 cytokines in this instance, namely the pattern of IL-23 is vastly
different from
that of IL-17 and IL-6.
L. EXEMPLARY RESULTS USING DATA MINING PROTOCOLS ON CIP
DATABASE
Using the various automated analysis and data mining protocols described
above, various
analyses of the CIP database were performed. In a first set of analyses, a
series of
cytokine based searches were run, using a library of cytokine anlaysis
protocols that were
created for this purpose. These include the cytokine analyses described in the
previous
section, as well as the creation of predictive models, based on bayseian
modeling, that
-242-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
take as inputs a plurality of cytokine assay values and out put a prediction
for various
non-cytokine immunological virus, parasite or other exposures.
Figs. 21 F-1 through 21 F-6 depict the results of predictive models built
using cytokine
data according to an exemplary embodiment of the present invention;
Figs. 21G-1 through 21G-12 depict the results of running an exemplary patient
population rule mining protocol according to an exemplary embodiment of the
present
invention;
In exemplary embodiments of the present invention, a first assay panel
containing a
plurality of cytokine assays can be administered and the results processed.
Based on
automatic analyses of the cytokine data, a second tier or set of assays can
then be run on
the same individual. The cytokine assay results being used to inform the
contents of a
second assay panel. In this manner, it is not necessary to fund the assay of
entire
superpanels unless and until the expense is justified. This exemplary
embodiment can be
particularly useful in a health care management or insurance embodiment, as
described
below, especially where assay costs are igh or must be carefully controlled.
The cytokine
data can be analyzed in a variety of ways, and a "cytokine signature" can be
generated
and stored in the database. Such a cytokine signature can then be an input to
a series of
algorithms, the output of the totality of which is a set of secondary assays
to be run for
that individual (or population). The secondary assays and the cytokines can
then together
be processed, for example, in any of the ways described or disclosed above, or
in the
code provided herewith in Appendix D.
-243-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Also very useful in an insurance underwriting or health care management
application, is
the exemplary single patient vaccine recommendation protocol. Although
originally
named for its vaccine recommendation capabilities, it actually tallys all of
an individual's
assay results, calculates and presents where within the database they fall
percentage wise,
and displays therapeutic, diagnostic and other recommendations or observations
based
on the assay and other variable values in that individuals' record.
Such a protocol can be used, for example, to calculate an immunoscore, an
overall
immune status, with one or more sub-immunoscores, and an underwriter can use
such a
protocol from their desktop computer to immediately look up the electronic
file on any
patient or insured, and can approve health care procedures, intelligently
audit for
premium or rating purposes, and generally have command of the individual's
entire
health picture at a glance, all assisted by the system intelligence.
Figs. 21H-1 thorugh 21H-10 depict the results of running an exemplary
individual patient
vaccine recommendation protocol according to an exemplary embodiment of the
present
invention for 10 randomly selected individuals' within the database.
Exemplary Automated Data Mining Complete Output
Fig. 21I is an exemplary output from an exemplary automated data mining
protocol
according to an exemplary embodiment of the present invention, segmenting an
exemplary database by Region of origin, Sex and the cytokine assay IFN-gamma.
The
output includes
-244-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
-245-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
USES OF IMMUNOSCORE INFORMATION AND AUTOMATED
DATA MINING RESULTS IN VARIOUS COMMERCIAL,
RESEARCH AND GOVERNMENTAL CONTEXTS
In exemplary embodiments of the present invention, ImmunoScore information
(including, for example, results of assay panels, individual history and
records of health
care visits and treatments administered or undergone) processed in an
exemplary system
and stored in an exemplary database can be used in a variety of commercial,
research and
governmental applications. These uses can range from optimizing the health
care costs of
a medical insurance underwriter to facilitating immunogenicity studies for a
pharmaceutical manufacturer, or, for example, to tracking the incoming and
subsequent
immune status of immigrants. In what follows, descriptions of several
exemplary
business methods which leverage or exploit the use of ImmunoScore informatics
are
presented.
A. HEALTH INSURANCE UNDERWRITING AND MANAGEMENT
In exemplary embodiments of the present invention, systems and methods
according to
the present invention can be used, for example, to optimize the business of
health insurers
as well as healthcare providers, who are essentially self insurers. In
general, a health
insurance underwriter or a health insurance provider has a population of
individuals,
generally called insureds or plan members, whose medical care costs are
reimbursed or
paid for directly by the healthcare insurer or the healthcare plan. In such
contexts, it is
useful to monitor the health of the population of insureds or plan members,
especially
those who are older and in those years, generally, for example, starting at
age 60, when
individuals begin to encounter greater health and medical problems.
-246-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
In exemplary embodiments of the present invention, each plan member or
insured, or, for
example, each plan member or insured above a certain age, can be assayed, and
the
results can be used to determine whether any prophylactic therapy should be
administered
to these individuals. Sometimes the decision is as simple as identifying
vaccine
preventable diseases for which the individual does not have sufficient levels
of
antibodies. In that case, the prophylactic therapy would be the administration
of the
vaccine in question. More complicated decisions could include identification
of diseases
or of biochemical markers therefor, that an insured or plan member is
susceptible to that
do not have a direct and economical prophylactic therapy. In that case, there
can be, for
example, a more complex algorithm which decides what to do given (i) assay
results and
(ii) the relative costs of assuming the risk that the insured will contract
the disease versus
the costs of prophylactic therapies to prevent the disease or diseases
implicated. Such
algorithms could, for example, be implemented in a system such as is depicted
in Fig.
2A, where, for example, in addition to database 203 where the results of
assays
conducted on individuals are stored, there can also be a business rules
database 220
which can also supply inputs to a central processor 204 which implements such
analysis
and algorithms. The inputs to such algorithms can then be, for example, not
just assay
results, medical history and demographic information, but also a set of
business rules
allowing a decision to be made or facilitated, taking into account the
relative costs and
benefits of administering prophylactic therapies. Such benefits to be
considered can, for
example, be those inuring to the individual as well as those inuring to the
members of the
health care plan as a whole, or, those which seek to maximize profits or
efficiencies. In
exemplary embodiments of the present invention such a healthcare insurance
-247-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
optimization method could be implemented as is illustrated in process flow
diagrams
Figs. 22 and 23.
As can be envisioned from the CIP database, it appears that the level of anti-
Rubella
antibody is uniformly lower in those individuals from SE Asia. Rubella is a
generally a
mild, self-resolving infection except in pregnant females, in which instance
there are
undue complications to the newborn, known as Chronic Rubella Syndrome (CRS).
In an
immigrant population such as the one documented in the CIP database, if women
of
child-bearing age from SE Asia were demonstrated to be susceptible to Rubella
infection,
health care authorities, as well as those underwriting insurance policies
would benefit
from such information. Not only are those women more at risk during pregnancy,
but
this particular immigrant population would be more likely to infect native
Canadians of
child-bearing age (assuming that their own antibody levels had waned). The
general
health of the population, therefore, would be well-served making sure that
these
individuals were appropriately vaccinated to avoid Rubella infection and
possible
complications to child-bearing women. These data reveal that Canadian
authorities (and
by extension, those in the United States) could, for example, be well served
and fiscally
responsible in the long run by testing and immunizing the immigrant population
against
Rubella and other vaccine-preventable diseases.
Fig. 22 depicts an exemplary process flow for a health care management
application.
With reference thereto, at 2201 an insured's immune status can be examined,
for example
by conducting one or more assays or panels of assays such as, for example,
those that are
described above. At 2202, for example, the results of those assays can be used
to identify
diseases that the insured is susceptible to, and moreover, the risk of
contraction of each
-248-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
disease for that individual can be calculated. At 2203, prophylactic therapies
that could
prevent each identified disease can be identified, and at 2204, for each
identified disease
a decision can be made by calculating the expected costs of treatment (such
as, for
example, by taking the known costs of treatment multiplied by the probability
of
contraction) and the costs of associated prophylactic therapies. Finally, at
2205,
prophylactic therapies that cost less than the expected costs of treatment can
be required
for the insured as a condition of maintaining his or her insurance coverage or
membership
in the health plan. For those prophylactic therapies whose costs are greater
than expected
treatment costs but neveretheless desired by the insured, the cost
differential can be born
by the insured rather than the health insurer.
Fig. 23 depicts a particular subset of the process flow illustrated in Fig. 22
where the
prophylactic therapies are simple and the ailments identified are vaccine
preventable
diseases. Beginning at 2301, an insured or plan member's immune status is
examined by
conducting one or more assays or panels of assays such as those described
above. At
2302, vaccine preventable diseases to which the insured is susceptible are
identified
based on an analysis of the results of the immune status from 2301. At 2303,
the insured
can be, for example, required to obtain vaccines for the identified vaccine
preventable
diseases. At 2304, follow-up examinations of the insured's immune status post-
vaccination can be made, again by conducting one or more assays or panels of
assays,
and these results can also be stored in the database. At 2305, the follow-up
examination
results can be used to evaluate the efficacy of any administered vaccines to
provide the
necessary immunity to the identified diseases for this individual. When
extended to an
entire population, such as, for example, the insureds of a health insurance
company or the
-249-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
members of a health plan, this can, for example, provide a means of evaluating
the
efficacy of vaccines in an aging population. This can also be very useful in
the context of
measuring and dealing with immunosenescense, as described below.
Next described are a number of process flow charts which illustrate exemplary
process
flow according to various embodiments of the present invention applied to
healthcare
management applications. Fig. 24 is an alternative process flow to that
depicted in Fig.
22, and is concerned with adjusting an insurance premium or an HMO
participation fee
for an individual based upon identification of potential diseases that an
individual is
susceptible to using ImmunoScore diagnostics.
The context of Fig. 24 could arise, for example, where an insurance company or
HMO
requiring an annual ImmunoScore diagnostic panel as a condition of maintaining
insurance coverage or participation under a healthcare plan. Such annual
requirement
would be akin to the annual information questionnaires that automobile
insurance
companies require of all of their insureds wherein an insured must state if he
has had any
serious health problems, if he has been involved in any accidents, or if other
out of the
ordinary events have occurred. With reference to Fig. 24 at 2401, the
individual's
immune status can be examined and at 2402, based upon the results of such
examination,
all diseases to which the individual is susceptible can be identified. 2405 is
a decision
tree which is applied to each disease identified at 2402. Thus, at 2405, for
each disease a
decision is made as to whether a prophylactic therapy is available. If there
is no such
available therapy, the flow terminates at 2410 where the insured's premium is
adjusted
upward, to account for the additional risk the insurance company is taking in
continuing
to cover this individual. If, at 2405 there is a prophylactic therapy
available then the flow
-250-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
moves to 2406 where it is determined whether to administer or approve the
prophylactic
therapy. Based upon this decision, the premium can also be adjusted.
Fig. 24A is a more detailed version of the analyses described in connection
with Figs. 22
and 24. With reference to Fig. 24A, at 24A01 the immune status of an
individual can be
examined, and at 24A02 the initial total cost can be set to zero. 24A02
through 24A35
are then applied in a loop which cycles over all of the diseases for which an
individual is
tested in the examination at 24A01. Such identified diseases can be, for
example, those
indicated by analyzing the results of assays conducted and other data
associated with the
individual or various populations to which he/she belongs, as described above.
For each
potential disease, at 24A05 it can be determined whether the individual is
susceptible or
not based upon the assay results. If the individual is not susceptible,
process flow can
terminate as to that disease at 24A20 and no incrementation of cost occurs. If
the
individual is susceptible, the flow moves to 24A 10 where it is determined
whether a
prophylactic therapy exists. If a prophylactic therapy does not exist, at
24A30 the total
cost is incremented by the cost of treatment. If such a therapy does exist, at
24A05 it can
be determined whether the treatment cost from the disease is greater than the
cost of the
prophylactic therapy. If the treatment cost is greater than the cost of such
therapy, then at
24A35 the prophylactic therapy can be offered to be reimbursed up to the
treatment cost
and the total cost can be incremented by the treatment cost. If the cost of
prophylatic
therapy is greater than treatmet cost, then at 24A25 the individual is
required to take the
prophylactic therapy and the total cost can be incremented by the prophylactic
therapy's
cost. After looping through all of the potentially relevant diseases, at 24A50
the premium
can be adjusted based upon the total cost. The computation of total cost and
prophylactic
-251-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
therapy cost at both the disease specific level and the over-all levels can be
given by the
following rules:
Disease specific:
Computation of TC: P(CDIIS and not PT) * C(TjCD and not PT and IS)
Computation of PT Cost: P(CDIIS and PT) * C(TICD and PT and IS)
Overall disease-related healthcare costs:
TC = P(CDilnot PT and IS) * C (TiICDi and PT and IS) + C(PT) (in all diseases)
PT = I P(CDilnot PT and IS) * C (TiICDi and not PT and IS)
The various exemplary implementations of healthcare management described above
have
considered each disease individually. Fig. 25 addresses a more complicated
situation
where all of the potential diseases are identified and all prophylactic
therapies available
for all of the identified diseases are also identified in all possible
combinations of
diseases and prophylactic therapies are analyzed using a cost benefit
approach. Thus,
with reference to Fig. 25, at 2501 a panel of assays can be conducted. At
2502, based
upon the results of such assays all diseases the individuals are susceptible
to are
identified. At 2505 all prophylactic therapies which are available for each of
the
identified diseases can also be identified, and at 2510 a cost benefit
analysis of all
possible combinations of prophylactic therapies and diseases can be, for
example,
undertaken using business rules. Implementation of this functionality
represents a much
more complex level of analysis as it is necessary to first define all possible
combinations
of diseases and prophylactic therapies. For example, if the individual is
susceptible to
five diseases and a prophylactic therapy exists for each of them but these
prophylactic
therapies vary widely in cost, it can be, for example, useful to a healthcare
manager or a
healthcare insurance underwriter to know whether it may be more economical to
only
-252-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
administer some of the identified prophylactic therapies and run the risk of
the individual
contracting the diseases for which prophylactic therapies are not
administered. For each
of the possible combinations a cost in terms of cost of administering the
prophylactic
therapy and expected cost of treatment without the therapy is assessed and at
2515 one or
more therapies can be approved and/or the insured's premium or the
individual's
insurance premium adjusted.
It is understood that in the description of the various possible algorithms
which can be
used in an ImmunoScore analysis for healthcare management that the term
individual,
insured, and healthcare plan participant are functionally equivalent. While
some
algorithms are expressed in terms of health insurance context, the same
analysis
represented by them can easily be applied to HMO management or management of
other
healthcare plans. As will be described below, the same techniques can be
applied where
the entire population is covered under a healthcare plan, such as, for
example, in a
socialized medicine jurisdiction. Alternatively, the same techniques can be
applied
where a large population of some mutual affinity is covered by a single
healthcare plan
such as, for example, United States Veterans whose healthcare is provided by
the U.S.
Veterans Administration. Thus, it is understood that any particular algorithm
or method
described in one context also applies to any other contexts.
Fig. 25A is identical to Fig. 25 except that it offers an additional option.
At 25A20, if, in
fact, the minimum cost, which is simply the total cost of the least costly
permutation at
25A10, is, for example, too great for underwriting limits or healthcare
management
criteria at 25A20, the participant can, for example, be canceled from the
plan.
-253-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Fig. 26 depicts an exemplary process flow for use in healthcare management
applications.
Fig. 26 is not concerned with dollar costs but rather cost in terms of quality
of life. Such
an analysis would be useful where dollar cost is less important than quality
of life, such
as, for example, in exemplary embodiments where a supplemental insurance
company
insures a minimum quality of life and undertakes to provide for whatever
healthcare costs
are necessary to maintain that quality of life. Additionally, a socialized
medicine
jurisdiction, for example, could have a minimum quality of life which it seeks
to provide
to each citizen as a basic human right which that jurisdiction sees all of its
citizens as
having. With reference to Fig. 26, at 2601, an immune status of an individual
can be
examined and the quality of life can be set to zero. For the purposes of Fig.
26, a higher
quality of life score translates to a higher quality of life. At 2602 all
diseases to which
the individual is susceptible are identified and a decrease in QOL score can,
for example,
be assigned to each disease. The scoring data (i.e., a map of identified
health scenarios to
some QOL metric) can, for example, be stored in a business rules database such
as is
depicted in Fig. 2A. Such a decrease in quality of life score can be, for
example, a
measure of unexpected pain and suffering, a measure of how many sick days are
generally associated with it, or, for example, whether the sick days are at
home, taken at
the hospital, or taken while still at work, and finally whether surgery is
involved. At
2605, all prophylactic therapies which are available for all of the identified
diseases at
2602 can also be identified. At 2610 for each identified disease and each
possible
combination of identified diseases (assuming that the individual could
contract more than
one disease, either simultaneously or in succession) the probability of
contracting the
disease can be computed and from that probability an associated expected
decrease in
-254-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
quality of life can be, for example, computed. As provided in Fig. 26, an
exemplary
formula which can be used in this context:
E(QOLDEC) = Prob(Disease) * AQOL;
QOL = QOL - E(QOLDEC)
At 2615 an increase in quality of life can be assessed for each identified
disease or
combination of identified diseases for which either prophylactic therapies or
therapeutic
therapies exist. Thus, in exemplary embodiments of the present invention, the
quality of
life score can be incremented by looping through each disease and adding the
expected
increase in quality of life associated with either (i) providing a
prophylactic therapy or (ii)
a therapeutic measure to mitigate the loss and quality of life due to
contracting the
disease. For example, not every disease for which there is a prophylactic
therapy can be
totally obviated. Some diseases to which individuals are susceptible can be
mitigated but
not prevented by prophylactic therapies. For example, when people feel the
onset of a
cold they often take echinacea. Echinacea tends to lower the amount of time
one is
symptomatic but rarely totally prevents contracting the cold. Alternatively,
if a
prophylactic therapy completely obviates the individual from contracting the
disease then
the E(QOL;,,c) should exactly equal the E(QOLdec). If the prophylactic therapy
happens,
for example, to bestow other benefits besides preventing the disease, then the
expected
increase in the QOL associated with undergoing the prophylactic therapy would
exceed
the E(QOLdec). Similar computations would apply to various possibilities. At
the end of
process flow in Fig. 26 a net quality of life figure can thus be computed.
Fig. 26A is a more detailed process flow for the example illustrated in Fig.
26 with the
exception that in Fig 26 an improved QOL is indicated by a more positive score
and in
-255-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Fig26A an improved QOL is indicated by a more negative score. At 26A01 immune
status can be examined and at 26A02 the quality of life can be set to zero. At
26A10 the
probability of contracting a disease given the immune status obtained in at
26A01 can,
for example, be computed. At 26A20 the probability of contracting the disease
given the
immune status can be multiplied by a "badness" score. At 26A30 this product
can be
added to the quality of life score. 26A10 through 26A35 can then be repeated
for each
disease for which susceptibility could be examined, given the assays
administered at
26A01. In this exemplary process flow a better quality of life is associated
with a lower
number which is the opposite convention of that adopted in the process flow of
Fig. 26.
It is for this reason that a "badness" score is assigned to each disease and
an expected
"badness" is added to the quality of life at 26A30. Additionally, at 26A15,
all possible
prophylactic therapies for the identified disease (it is noted that 26A15 and
26A35 are
within the for-each-disease loop as well) can be generated and mitigation
scores can be
assigned for each physical therapy or combination thereof. At 26A35, the
mitigation
score can be, for example, subtracted from the quality of life score and once
flow is
looped from 26A10 through 26A35 for each disease, at 26A40 a total quality of
life score
can, for example, be output. Using this total quality of life score, at 26A50
the best set of
prophylactic therapies in terms of higher quality of life can be offered to
the individual
with the stated quality of life improvement.
It is noted that in the schema of Figs. 26A a badness score is associated with
each
contracted identified disease. An exemplary badness scoring system is
presented in the
upper right of Fig. 26A and comprises, for example, +l for a home sick day,
+10 for a
hospital sick day, +'/2 for a work sick day, and +100 for a surgery.
Accordingly, the
-256-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
quality of life score would dramatically decrease if the individual was found
to
susceptible to a number of diseases each of which required surgery if
contracted.
Fig. 27 is a final healthcare management exemplary process flow chart. Fig. 27
addresses
a newly discovered HPV vaccine that is 100% effective in preventing cervical
cancer in
women. The question is who should receive the vaccine and when should they be
tested.
From the point of view of society as a whole, perhaps everybody who has not
contracted
HPV should be vaccinated to prevent them from ever contracting it and thus
prevent the
females amongst them, and females in contact with the males amongst them, from
ever
contracting cervical cancer. Of course, this has a greater cost than simply
vaccinating
women prior to their exposure to HPV. Therefore, the decision as to who
receives the
HPV vaccine will often depend upon who is managing the healthcare of the
population in
question. This will be described in connection with the final decision at
2715.
With reference to Fig. 27, beginning at 2701, an assay panel containing an HPV
assay
can, for example, be conducted relative to one or more individuals. At 2705 it
can be
determined whether that individual is seronegative or seropositive to the HPV
virus. If
seronegative, the individual has not yet contracted HPV and flow moves to
2710, where
the decision as to the individual's gender is made. If the individual is a
male, is not
seronegative, and is seropositive to HPV, then flow can terminate at 2706 and
any
therapeutic treatments that are available can be administered. Continuing at
2710, if the
individual is a female flow terminates at 2711 and the HPV vaccine is always
administered. Whether the healthcare manager is an insurance company, an HMO,
a
socialized medicine jurisdiction or a large scale healthcare management entity
such as the
Veterans Administration, any female whose healthcare is being managed should
be
-257-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
vaccinated to prevent any healthcare expenditure in treatment for cervical
cancer.
However, what about males? The only utility derived from vaccinating males is
that
females in sexual contact with them will not contract HPV. If those females
are managed
by a different healthcare entity there is little utility in protecting "our"
men. If those
females are protected in the same healthcare management entity, then there is
utility in
protecting them. Alternatively, even if the females are not provided
healthcare or
healthcare insurance under a given plan, a government regulating that plan may
see a
social benefit in wiping out cervical cancer, or at least those cervical
cancers caused by
HPV, which are the vast majority of such cancers. Accordingly, given all of
these
concerns, at 2715, the HPV vaccine can be administered if the utility value of
the
prophylactic effect is greater than the cost of treatment, which is simply the
cost of the
vaccine. The utility value will, as noted above, be a complicated function of
a number of
factors, the most prominent of which being who is responsible (financially,
politically or
morally) for the healthcare of the females that this male may come in contact
with.
B. HEALTH CARE/HEALTH INSURANCE CREDIT EXCHANGE
The applications that have been described thus far relating to heathcare
management all
assume that in the cost benefit analysis, additional costs can be passed to an
insured, or,
for example, if too high, the insured or member of a health plan (such as an
HMO) can be
canceled. While this may maximize profits for the health plan or the health
insurance
company in the short run, it can result in dissatisfied insureds and
eventually loss of a
certain percentage of the insured base of individuals. Loss of customers is
never a good
thing, even if under certain analyses they are unprofitable customers. One way
of solving
this problem is, instead of passing costs through to consumers, i.e., to
insureds or health
-258-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
plan members, to set up a means by which they can procure credits in years
when they
are predominately healthy and use those credits when not costs but - debits -
are assessed
against them as per the exemplary analyses described above in connection with
Figs. 22
through 27. Thus, in exemplary embodiments of the present invention, a health
care
provider, a health care insurance company, or other financial intermediary in
conjunction
with the health insurance provider or health care provider, such as an HMO,
can set up a
health insurance credit exchange. Such an exchange can operate in a fashion
similar to
those government programs which have rules against excessive energy use or
excessive
pollution derived from an entity's activities. An entity which is a polluter,
or an
"excessive" user of energy or a natural resourse such as water, for example,
can purchase
credits from other individuals or entities who have a low energy use, low
water use or are
low polluters. In this fashion, those individuals or entities who exceed a
certain threshold
of some desirable metric, such as, for example, low energy use, low water use,
or other
"green" factors, can purchase, negotiate, trade or otherwise procure credits
from those
who are below such a threshold so as to avoid fines or negative consequences
from
violating the environmental or natural resource use standards.
Thus, in the health care context there are always some individuals who are
sick more than
others. Individuals do not know whether they will be in the underwriting bin
of more
sickly than average or less sickly than average. Insurance companies try to
spread the
risk of the more sickly amongst a larger population which obviously includes
those who
are less sickly, and charge an essentially average health insurance premium to
everyone.
However, as underwriting becomes more granular, using exemplary embodiments of
the
present invention, it can be predicted, even decades in advance whether a
particular
-259-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
individual is more or less likely to contract a disease, such as for example,
autoimmune
diseases as described above. For example, as described above in Section I,
certain
autoimmune diseases have markers which are harbingers 7-10 years in advance of
their
eventual symtomology. Thus, using exemplary embodiments of the present
invention,
health care plans, health care administrators and health care insurers will be
able to divide
the population into many more bins of insureds and associate with each of them
a more
accurate health insurance premium cost. This can cause those in the more risky
bins to
have a much greater insurance cost. One way of ameliorating this is to
encourage people
to join health care plans early in their lives when they are healthy and
before even the
onset of eventual disease emerges, such as via a marker or predictor in an
Immunoscore
assay result marker context. In so doing, people who are healthy can receive
credits
which they can bank within the system or buy, sell or trade. If regular
Immunoscore
audits of individuals reveal that someone is moving from a less risky bin into
a more
risky bin, and a cost would be added to their health insurance premium (i.e.,
a debit),
instead of paying an extra premium they can procure a credit through a health
care credit
exchange either from their own account which they banked in earlier years or
from other
healthy peoples' accounts which are presently available for exchange.
Thus, in exemplary embodiments of the present invention, an insurance company
could,
by setting up and maintaining such a healthcare credit exchange, retain more
customers
as well as encourage customers to join its ranks of insureds early on in their
lives so as to
be able to bank for the future and/or sell credits for being healthy. By
acting as
intermediary, an exemplary system can make a market for such health care
credits, and
not have to wait for a particular debit holder to find a particular credit
holder willing to
-260-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
exchange. Acting in some ways as a securities market maker, an Immunoscore
based
third party can buy credits and sell debits.
Thus, in exemplary embodiments of the present invention, insureds can thus be
induced
to pay higher premiums when they are younger and more healthy which would
therefore
give them extra protection against being assessed debits later on should they
become sick.
This results in a net flow of capital to the health insurers, or the HMOs,
because they can
charge higher premiums than the "true" or correct "premium" with the full
consent of the
insured in exchange for allowing and facilitating participation in the health
care credits
exchange. On the other hand, they can also retain more customers because
people who
are subject to debits as a result of more granular analyses of their overall
health via
Immunoscore diagnostics can simply use credits they have accumulated earlier
in their
lives or procure credits from other insureds which would ultimately be cheaper
for them
than having to find substandard coverage. Additionally, the insurance company
is not
faced with canceling bad insureds and then having to spend client development
money to
procure new "good" insureds, rather, it can more or less retain its insured
base as well as
generate additional profits from the maintenance of the healthcare credit
exchange.
Further, if a healthcare management entity sets up a health care credit
exchange it can, in
exemplary embodiments of the present invention, require immunoscore
diagnostics, such
as set forth in Section 1 above, at various significant life points in each
insured's lifetime.
This can have the effect of positive feedback in the amount of data that an
immunoscore
database has available and thus, an improvement and greater accuracy and
predictive
value that the algorithms of the Immunoscore analysis can provide to the
insurer. Over
-261-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
the course of time an insurer will tend to make more money and have more
accurate
predictive models than its competitors who do not use such an Immunoscore
system.
Finally, ImmunoScore databases lend themselves to storing health care credit
and debit
information as part of an individual's record, making it nearly seamless to
create
algorithms to track such credits/debits and manage the exchange. After all,
ImmunoScore is the tool being used to generate the very granularity that
assigns the
credits/debits and makes the entire business possible.
C. VETERANS HEALTH CARE MANAGEMENT (VARIANT OF HEALTH
CARE)
A special instance of health care management relates to veterans care. In the
United
States, the Veterans Health Administration (VHA) provides a broad spectrum of
medical,
surgical, and rehabilitative care to its customers. Individuals that qualify
for veterans
healthcare services include, for example, returning Active Duty, National
Guard and
Reserve service members of Operation Enduring Freedom (OEF) and Operation
Iraqi
Freedom (OIF). The vision statement of the VHA states that it needs to be a
comprehensive, integrated healthcare system that provides excellence in health
care
value, excellence in service as defined by its customers, and excellence in
education and
research, and needs to be an organization characterized by exceptional
accountability and
by being an employer of choice.
In exemplary embodiments of the present invention, veterans, with their
special
requirements based on service, can be well served by ImmunoScore diagnostics
and data
management. As previously described in Section I, soldiers have very specific
vaccination requirements based on their deployment and area of expertise.
ImmunoScore
-262-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
diagnostic panels can be tailored to the needs and context of the individual
soldier based
upon his or her previous exposure to immunization and also to different
infectious agents
depending on the relevant theater of deployment. In addition to immune
response to
infectious agents, veterans are likely candidates for measurement of immune
system
perturbations induced by, for example, Post Traumatic Stress Disorder (PTSD),
exposure
to unique chemical agents (e.g., Agent Orange), Gulf War Syndrome, and
recovery from
injuries sustained in service.
As described above in connection with the CIP database, linear regression
analysis of a
patient database could yield valuable information pertinent to appropriate
treatment of
veterans after their years of service. Those analyses displayed possible
correlations
between, for example, measles and mumps immunity and immunity to varicella
infection.
Any possible associations between service locale and adverse agents could be
documented and analyzed by an exemplary ImmunoScore data mining process in
similar
fashion.
The VA Research and Development program (The Office of Research and
Development)
aspires to lead the Veterans Health Administration in providing unequaled
health care
value to veterans. The ImmunoScore technology can help contain healthcare
costs for
veterans by monitoring and analyzing immunologic information.
D. SOCIALIZED MEDICINE MANAGEMENT
A socialized medicine jurisdiction is essentially a health care provider or
insurer for an
entire population. Thus, the health care management applications of
ImmunoScore
described above can also be implemented in a socialized medicine jurisdiction.
Countries
-263-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
with socialized medicine, such as the UK, New Zealand, and particularly
Canada, present
opportunities to stress preventive medicine for the good of the populace
(i.e., by
maximizing QOL for a given health care budget) and the advantages of lower
cost
healthcare as represented by ImmunoScore managed healthcare. These governments
could be provided with healthcare management services via an implementation of
the
ImmunoScore system.
The CIP database discussed in Section II above, has revealed the utility of an
exemplary
ImmunoScore database for a country with an immigrant population. There has
been
much concern regarding outbreaks of mumps in the United States and Europe.
This
disease has clearly been shown to spread from contact with travelers (CDC,
2006). The
CIP database indicates a degree of relatedness between patients that have
antibodies to
both Rubella and Mumps. If this type of analysis were to be extended to
geographic
regions and associated with specific genders, a government that supported
socialized
medicine could, for example, be very much in favor of assuring that an
immigrant
population was properly immunized, for the protection of that immigrant
population, as
well as the native population.
E. SUPPLEMENTAL INSURANCE (AFLAC MODEL)
AFLAC is the leading provider of supplemental insurance, which provides help
with
expenses not covered by an individual's major medical plan. The company is the
number
one provider of guaranteed-renewable insurance in the United States and Japan.
Its
products provide protection to more than 40 million people and go beyond the
traditional
insurance by directly paying claimants with cash benefits.
-264-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
With the cost of health care rising, the challenge for most employers is to
satisfy the
specialized needs of each employee without having to fund expensive new plans.
AFLAC provides products including, for example, the following: Accident
Disability;
Short Term; Disability; Cancer Benefit; Hospital Indemnity.
ImmunoScore diagnostic testing and database storage can provide information
for use in
just such supplemental insurance programs. ImmunoScore can, for example,
provide an
individual with immune status testing that could be monitored over time and
offer the
peace of mind that would come from knowing that that patient had a "healthy"
immune
system. In addition, an insurer would be better able to underwrite premiums
for
supplemental health insurance with a sounder understanding of the patient's
health status.
Additionally, in exemplary embodiments of the present invention, a
"immunological
insurance plan" could be offered. Such a plan could provide all immunological
monitoring and therapeutics to each insured for a fixed annual premium and
guarantee a
certain defined quality of life to each insured. Such a plan could utilize one
or more of
the health care management processes described above.
To be able to effectively underwrite such supplemental insurance, supplemental
insurance firms need to be aware of relatedness between immune parameters as
revealed
by database analyses. For instance, the CIP database revealed tendencies for
Hepatitis A
antibody to be present in individuals from certain geographic regions.
Supplemental
insurance coverage could benefit from insuring that travelers to these regions
were
assured of their own immune system's ability to combat Hepatitis A infections
in regions
where the disease is endemic. Or, for example, the CIP database revealed a
possible
-265-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
suspension of Tetanus immunity amongst individuals reactive to CMV. In
exemplary
embodiments of the present invention, a health insurer (whether supplemental
or primary)
would take special care to take such a factor into account.
F. IMMUNOSCORE AND THE WELLNESS INDUSTRY
In 1994, the U.S. Congress laid the groundwork for the Wellness Industry by
passing the
Dietary Supplement Health and Education Act (DSHEA). This Act set new
standards for
the manufacturing, testing and marketing of nutritional products. Products
that meet
strict government standards earn the title of nutraceuticals. Blurring the
line between
conventional foods and drugs, nutraceuticals are defined as foods or parts of
food that
confer health or medicinal value, including the prevention and treatment of
disease.
The Food Policy Institute (http://www.foodpolicyinstitute.org) has defined
drivers of
nutraceutical industry growth. The nutraceutical market was once viewed as
largely a
counter-culture "back to nature" phenomenon, but is now buoyed by a number of
solid
fundamentals.
Changing consumer demographics. Americans are living longer and emphasizing
the
importance of quality of life in their later years. As the baby boomers
approach ages
where personal health becomes more paramount, the demand for mechanisms for
conveying health will grow.
Increasing ethnic diversification. The mainstream U.S. nutraceuticals industry
is a
relatively new phenomenon. However, the use of foods, herbals, and other
natural
products to convey health and medicinal values has a long history of
acceptance by many
of the world's cultures.
-266-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Paradigm shift in personal health. Americans are taking more responsibility
for their
personal health, embracing the concept of health maintenance and wellness.
Thus, the
paradigm is shifting away from disease treatment and towards disease
prevention.
Dissatisfaction with Western healthcare. Americans are becoming more reticent
about
accepting the side effects of synthetic drugs and remedies. Similarly, rising
healthcare
costs are encouraging Americans to explore alternatives to traditional
orthodox medicine.
Increasing acceptance of alternative healthcare practices. There is a growing
acceptance
among Americans of alternative or complementary therapies and wellness
modalities.
Recent years have witnessed increased use, for example, of chiropractic care,
vitamin
therapy, aromatherapy, meditation and relaxation techniques, and acupuncture.
Increased understanding and awareness of diet-disease relationships. Many of
the
leading causes of premature death in the U.S. are diet-related. Examples
include heart
disease, diabetes, and many types of cancer. The USDA estimates that diet-
related
disease and death costs the U.S. in excess of $250 billion each year.
The Food Policy Institute has identified challenges facing the nutraceutical
industry.
Few farmers are producing herbals and other botanical inputs (due to limited
market
knowledge, technical requirements and other obstacles.
Limited access to finance and capital constrains industry development and
expansion.
Ambiguous regulatory framework for ensuring product standardization and
efficacy.
Regulatory restrictions on marketing products via health claims impede retail
efforts.
-267-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Raw material supply issues (consistency of quality and availability) for
botanical
manufacturers.
Limited endorsement by traditional healthcare practitioners.
Consumers can not differentiate between high and low quality products and are
not
sufficiently educated to make informed decisions about proper product use.
ImmunoScore diagnoses and database could provide the answers to these
challenges.
Individuals and populations could be studied with respect to the efficacy of a
nutraceutical diet. ImmunoScore would either pave the way for more growth in
curtain
nutraceuticals, or perhaps point out the sale of "snake oil." Individual
products, or
product lines could be endorsed as valid by ImmunoScore measurements.
The Wellness Industry is expected to grow. The Wellness Industry includes the
concept
of "wellness insurance" to lower health care costs to individuals. This may
provide yet
another opportunity to leverage ImmunoScore testing and data storage into the
insurance
industry.
In addition, workplace wellness as a concept has been used extensively in
recent years by
management in business and industry, health professionals, fitness experts,
and others.
Well-designed and administered programs deliver positive outcomes for
employers as
well as employees. Because healthy employees cost less than employees
suffering from
illness, ImmunoScore can be a part of employee insurance offered by employers
wanting
the best and most affordable health care for their employees.
-268-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Analyses of the CIP database have shown the development of positive and
negative
relationships for one variable with respect to another variable (for example,
Rubella
antibody and Hepatitis A antibody levels as is illustrated in Fig. 20D, and
for example,
Mumps antibody vis-a-vis Hep A, Measles and Rubella, as shown in Fig. 20E).
This type
of analysis could be extended to other variables regarding "wellness." For
example,
fitness measurements could be incorporated (body mass index, cardiac function,
etc.)
into an overall immune fitness relationship.
Virtual PhysicalsTM - Incorporate ImmunoScore Diagnostic and Database
The Virtual PhysicalTM is a comprehensive diagnostic screening procedure that
uses state-
of-the-art technology to take a global look at a patient's body and identify a
variety of
conditions at early stages where intervention can be most helpful. A Virtual
PhysicalTM
may also be viewed as an integral component of a holistic, behavioral medicine
program,
where the body, and one's diet, exercise, and lifestyle habits are viewed as a
whole,
determining where problems may exist and where changes might be required.
The Virtual Physical'sTM early detection capability can uncover asymptomatic
and often
life-threatening diseases generally not detectable by physical exam or
standard screening
tests. This allows the management of disease in early stages, where medical
therapy and
treatment options are typically less costly, less invasive and more effective.
Virtual Physical'sTM comprehensive scan of an individual's body is
significantly more
detailed than an X-ray. It covers: (a) the heart and arteries, identifying
near microscopic
amounts of plaque; (b) the lungs at the air cell level showing the earliest
stages of smoke
damage, emphysema, or lung cancer; (c) the spine, evaluating for osteoporosis,
disc
-269-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
disease and other back problems; (d) internal organs for detection of tumors,
stones and
cysts of all sizes; (e) aneurysms in the abdominal and chest cavities; (f)
thyroid and
parathyroid disease; (g) joint disease; and (h) uterine, ovarian, and prostate
disease.
In the interest of determining a patient's "totality of health," ImmunoScore
screening
could accompany a Virtual PhysicalTM to add an immune health component to the
virtual
screening. It is possible that insurance will cover a Virtual PhysicalTM in
the future, and
ImmunoScore testing and data storage could be incorporated into the patient's
records
that could be transferred to the patient's primary care physician or
specialist.
G. WOMEN OF CHILDBEARING AGE/SCREENING OF PREGNANT
WOMEN
A superpanel for women of childbearing age was described above in Section I.
In light thereof, ImmunoScore diagnostic tests and database storage
availability in the
offices of obstetricians would greatly enable appropriate immunization of
pregnant
women as well as find correlates of prenatal interest. In addition to
screening pregnant
women for their immune status regarding vaccine preventable diseases,
ImmunoScore
diagnoses and data management could also be of value in determining the immune
status
of pregnant women regarding, for example, group B streptococcal infection,
cytomegalovirus (CMV) infection, and other infectious diseases that may
adversely affect
the newborn, yet are treatable prenatally. Early onset GBS infection has been
the leading
cause of death attributable to infection in newborn infants for over three
decades, with
over 6,000 cases a year in the United States (Vallejo, et al. 1994).
Antibiotics have been
used to good effect to prevent newborn GBS infection. There is also promising
preliminary data on an effective intervention to prevent CMV infection in
newborns in
-270-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
pregnant women that has been published recently (Nigro, et al. 2005). All
these
treatments can be more advantageously administered using ImmunoScore
technology.
Fig. 28. depicts an exemplary process flow for managing the immune status of
women of
child-bearing age. Beginning at 2801 the immune status of a women of child-
bearing age
is examined. At 2810 the vaccine preventable diseases to which the woman is
susceptible are identified as well as the woman CMV infection status and
pregnancy
status. At 2820 these three variables are used to generate healthcare
recommendations, as
follows. If the woman has not been infected with CMV and is not pregnant, she
is
advised to obtain immunizations for the identified vaccine preventable
diseases. If she is
an insured under a healthcare insurance plan, or her healthcare is provided by
an HMV or
socialized medicine entity she can be, for example, required to obtain these
immunizations to save future treatment costs as well as to serve the utility
of having a
healthy population. If she has not been infected with CMV but is pregnant, she
can be
informed of extra precautions regarding CMV status and pregnancy. Moreover, no
immunization with attenuated vaccines is recommended or should be performed.
However, other immunizations should be recommended based upon current CDC
guidelines. If the woman is seropositive to CMV and is not pregnant, she can
be advised
or required, as the case may be, to obtain immunizations for the identified
vaccine
preventable disease. Finally, if she seropositive for CMV and pregnant, no
extra
precautions should be taken regarding the CMV status unless there is an active
primary
infection. Moreover, no attenuated vaccine should be recommended or
administered.
However, other immunizations can be recommended or required based upon current
CDC
guidelines. At 2830 a follow-up examination of the women's immune status post-
-271-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
vaccination can be conducted, and, if she is not pregnant, the information can
simply be
stored in a system database. If she is pregnant, a post-natal follow-up can be
recommended or required, as the case may be, comprising MMR vaccination of the
mother and monitoring of CMV status of the child. Finally, at 2840, based upon
the post-
vaccination follow-up at 2830 the efficacy of the administered vaccines can be
evaluated
as to whether they provide the necessary immunity to the vaccine preventable
diseases
identified at 2810.
The CIP clearly points out the need for antibody measurements in women of
child-
bearing years. The obvious antibody to be examined is that for Rubella, to
which the
women of SE Asia were shown to have levels below average. Other important
antibodies
in women of child bearing years are, of course, those to group B Streptococcal
organisms
and others that affect fetal development or those associated with neonatal
illnesses. From
an insurance and public health perspective, these are extremely important
issues.
H. VACCINE-O-MATIVACCINE DISTRIBUTION NETWORK
In exemplary embodiments of the present invention, ImmunoScore technologies
can be
used to facilitate the easy dispensing of vaccines to the public as well as
giving the public
access to their immunologic information. Therefore, in exemplary embodiments
of the
present invention a business analogous to the "Fotomat" photograph finishing
stores,
once located in malls and strip malls across America, can be created. For
purposes of the
present description, this exemplary embodiment of the present invention can be
called
"Vaccine-o-Mat." Vaccine-o-Mats can be located in small buildings in corners
of malls
and strip malls, as concessions in large chain stores such as Target or Wal-
Mart, or they
can be located almost anywhere in appropriate markets and one day be as
ubiquitous as
-272-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Starbucks Coffee centers. At a Vaccine-o-Mat a member of the public can have
his or
her immune status checked and can receive any vaccines that he or she may be
deficient
in. If an individual steps on a rusty nail and doesn't remember the last time
he or she had
a tetanus booster he or she can simply drive to the nearest Vaccine-o-Mat,
have a panel of
assays containing tetanus and any related compliments as conducted and
determine then
and there whether he or she needs a vaccine. What makes the Vaccine-o-Mat
business
possible is instruments which can process large numbers of assays in a
relatively short
period of time, as noted above. One such instrument is the cobas e 411
analyzer (Roche
Diagonstics).
Fig. 29 depicts an exemplary process flow for use at a Vaccine-o-Mat. At 2901,
the
customer's immune status is examined for vaccine preventable diseases and
related
immunologic information. It is further contemplated that a particular customer
may want
to have his or her bodily fluids assayed for a wide variety of immunologic
tests and not
have them restricted to vaccine preventable diseases. Therefore 2901 need not
to be
strictly directed towards vaccine preventable diseases. At 2910, within 90
minutes the
assay results can be processed to generate recommendations for appropriate
vaccines.
This functionality depends upon, as noted above, instruments which can process
a large
number of assays in a relatively short amount of time. This concept allows for
partnering
with large chain stores or malls where customers could make their first stop
at the
Vaccine-o-Mat to have their blood tested. They could then continue shopping
and then
return at the end of their shopping excursion to receive any necessary
vaccines and report
regarding their immune status. At 2920 appropriate vaccines can be
administered to the
customer on site, and at 2930 the customer can be provided with a printout of
the assay
-273-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
results the updated vaccination record and his or her database record from the
ImmunoScore database along with instructions on how to access that information
in the
future. Finally, at 2940 all of the additionally required customer information
resulting
from that particular visit is stored in the database for future reference.
One of the benefits of the ImmunoScore technology is the ability to link
diagnostic
testing of the immune system with rapid delivery of medication at the point of
care
(ideally, during the course of an office visit). Thus, in exemplary
embodiments of the
present invention a vaccine distribution network can be set up, for example,
to link
vaccine manufacturers to physicians' offices - or other authorized vaccine
dispensing
personnel equipped with diagnostic facilities. Vaccine distribution can also,
for example,
become part of the ImmunoScore database tracking specific manufacturers' lots
numbers
to points of sale. This can be important in getting timely information
incorporated into
the Vaccine Adverse Event Reporting System (VAERS).
Fig. 29A depicts exemplary envisioned interactions between various parties
according to
an exemplary embodiment of the present invention directed towards vaccine
distribution.
Information gathered to an exemplary ImmunoScore database can, for example, be
shared with the various agencies responsible for dictating vaccination
decisions.
Unsuspected or unknown relationships regarding immune health or function can
be, for
example, "fished" or "mined" from a system database using appropriate queries
and
analysis. In addition, in exemplary embodiments of the present invention,
suspected
adverse events from vaccination could be addressed and acknowledged or
dismissed,
based upon information gleaned from the system database.
-274-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
With reference to Fig. 29A, various entities and institutions which can, for
example, be
involved in vaccine distribution or vaccine distribution network are depicted.
They
include any vaccine manufacturers 29A05 who through vaccine sales provide
vaccines to
physicians or healthcare providers 29A10. The physicians or healthcare
providers 29A10
also receive diagnostic testing kits and research services, such as, for
example,
ImmunoScore vaccine diagnostic panels 29A01. The government 29A15 has a
variety of
roles in a vaccine distribution network, including subsidizing or providing
economic
incentives to create or build a supply of vaccines by a transfer of funds to,
or via tax
incentives to, vaccine manufacturers 29A05. The government can further
subsidize or
fund HMOs 29A25 and in this context the Veteran's Administration, described
above can
be considered one of them. Additionally, the government 29A15 can mandate
vaccine
benefits to certain segments of the population and those can be provided by
HMO 29A25
or equivalents. Finally, the government 29A 15 can itself access personalized
immune
status data as to individuals or populations or sub-populations 29A12 for a
variety of
research or health management purposes. The CDC and ACIP 29A50 can receive
input
from Physicians/Healthcare Providers 29A10 as well as from a vaccine status
database
29A30. Vaccine status database 29A30 can be generated from an Immunization
Registry
29A40 set up by the CDC, ACIP or other similar institutions or bodies to
maintain
immunization records for the population so as to better know who should be
vaccinated.
Figs. 29B and 29C, described below illustrate improving connectivity between
entities
and organizations who could access and utilize ImmunoScore information in this
context,
allowing the benefits of ImmunoScore to be ubiquitously available.
-275-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
1. CONSUMER ACCESSIBILITY TO IMMUNOLOGIC INFORMATION
Americans are playing a risky game of sexual roulette, according to a new poll
finding
that only 39 percent of respondents always ask a new lover if they are
infected with HIV.
The poll, taken by Zogby for MSNBC.com also found that 73 percent of
respondents
were involved in a monogamous relationship, and that 66 percent of those
surveyed had
had unprotected sex while under the influence of alcohol. While 39 percent of
respondents said they always asked whether a new partner is infected with HIV
or other
sexually transmitted diseases, 31 percent said they never discuss the touchy
issue with a
new partner. Moreover, the survey found that 15 percent of Americans had paid
for sex,
35 percent of respondents said they had been with between one and five sexual
partners,
and 19 percent said they had had more than 25 partners.
In exemplary embodiments of the present invention this "risky business" can be
ameliorated. Accordingly, at the Vaccine-o-Mat described above, individuals
can have
their immune status tested by conducting, for example, an STD assay panel, as
described
in Section I above, which can then be shown to potential sexual partners to
fully disclose
the immunologic risks that may be involved in any proposed liason. For
example, a
couple can stop at a Vaccine-o-Mat near a romantic restaurant of their choice.
They can
have the assays conducted and go off to dine. If things are going well, by the
time their
coffee has arrived they can obtain each other's immune status and be off -
either alone or
together - depending upon the ImmunoScore results.
Alternatively, for example, someone worried by past promiscuities can
routinely procure
his or her immune status at the local Vaccine-o-Mat in 90 minutes, and put any
worries to
rest, or at least know what they are facing.
-276-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
J. IMMUNOSCORE CONNECTIVITY VIA INTERAPPLICATION
TRANSLATOR/DATA INTEGRATOR
In many exemplary embodiments according to the present invention, the power of
an
ImmunoScore diagnosis and database lies in the interaction of the database
with many
different organizations, as shown in Fig. 29B. Use of a web services
interconnector to
provide this connectivity is illustrated in Fig. 29C, next described. The CDC,
the
government (or governments, for that matter), health maintenance
organizations, vaccine
manufacturers, and physicians would all be able to interact with the database
and each
other to make the best possible decisions regarding the health and welfare of
the
citizenry.
With reference to Figs. 29B and 29C, a number of entities and organizations
who could
access and utilize ImmunScore information are shown. Fig. 29B shows a
complicated
information exchange structure wherein each entity involved has to set up a
separate
communications line or pathway to each of the other entities in the network.
This can
easily be remedied, as shown in Fig. 29C, by utilization of an
Interapplication
Connectivity Provider 29C50 which can interconnect the various individual and
sometimes proprietary computer systems, computer networks, databases, and
applications
of each of the individual entities participating in the vaccine
distribution/creation network
so that they can talk to each other. This technology is often referred to as
interapplication
connectivity or interapplication translation. One example of such a
interapplication
connectivity provider is the IBM, in particular the IBM Web Services Centers
Of
Excellence. Additionally, Enterprise Computing service companies, such as, for
example, EDS also provide products which link different and disparate
computing
platforms so that they can exchange data and information in an efficient
manner.
-277-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The CIP database has only scratched the surface of what can be captured and
shared by a
large ImmunoScore database, but important information can be gleaned from this
database, such as it is, of use to government sources, patients, physicians,
and insurers.
Demographic information regarding crowding and sanitary facilities has been
shown to
correlate to degrees of protection to vaccine-preventable diseases in the
populations
examined. If the database were to also include information regarding the
movement of
patients (for instance), much useful information could be shared among these
concerned
groups.
K. IMMUNOLOGIC INFORMATICS BASED LIFE INSURANCE
UNDERWRITING
In the exemplary embodiments of the present invention ImmunoScore data can be
used to
optimize the underwriting of life insurance. Additionally, assuming that
regulatory
restrictions are not preclusive, ImmunoScore data can be used by companies
which
provide both life and health insurance to the same clientele. The use of
ImmunoScore
technology for these purposes is depicted in the exemplary process flow chart
of Fig. 30.
With reference to Fig. 30, at 3001 an individual's immune status can be
examined and
any diseases to which he or she is susceptible identified. At 3015, by
accessing Business
Rules Database 3010, the probability of death of the individual given the
immune status
identified at 3001 can be computed. At 3016 the cost of insuring that
individual, based
on the probability of death of years to death calculated in at 3015 can be
computed and
premiums can be set at 3020. It is noted that the term "death" appearing in
Fig. 30 is
shorthand for "years remaining until death."
-278-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Additionally, at 3002 all combinations of possible prophylactic therapies can
be
generated given the immune status obtained at 3001. From these combinations,
at 3005,
the probability of time (generally in years) to death given the immune status
and the
various combinations of prophylactic therapies can be computed. Such
computation, at
3005, exchanges data with Business Rule Database 3010. For convenience, two
Business
Rules Databases 3010 are depicted in Fig. 30; in exemplary embodiments of the
present
invention there could be one or many Business Rules Databases each devoted to
a
specific informational domain. In the depicted exemplary embodiment of Fig. 30
they
could most likely be combined inasmuch as they are providing information which
allows
a system to compute the probable time to death given an immune status.
However, the
Business Rules Database on the right side of the figure may require more
complex
information in order to also factor in the available set of possible
preventive therapies for
each identified disease.
At 3016, the outputs of 3015 and 3005 are input to allow the exemplary system
to
compute the cost of insuring the given individual. At 3021 the system can
select the two
or three best sets of prophylactic therapies from the information generated at
3002, and at
3025 it can offer these prophylactic therapies to the client with a proviso
that the life
insurance premium set at 3020 in absence of factoring in prophylactic
therapies could be
lower by (x) if the client chooses to undertake the prophylactic therapies.
Alternatively,
at 3030 it may be in an insurance company's interest to pay for the
prophylactic
therapies, i.e., offering them to the insured for free, if the cost of the
prophylactic
therapies is less than the present value of the expected savings to the life
insurance
-279-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
companies by the insured having the prophylactic therapies performed. This can
be
expressed, for example, as:
PT cost < PV{(death benefit)*[(Prob (deaths no IS, no PT) - Prob (deaths IS,
PT)]}
Thus, if at 3030 such an offer is made, any premium adjustment at 3020 can be
diminished or completely reduced. The function of 3030 is to increase the
profits to the
life insurance company by not only identifying the premium which it would
charge the
insured but also, based on the immune status data obtained during the
underwriting
process (or during an annual audit process), to identify prophylactic
treatments that could
be offered to increase the time to death for the same individual thus allowing
the
insurance company to continue to earn the return on the cumulative premiums
prior to
having to pay the death benefit to the survivors.
It is also noted that at 3021 where the 2-3 best sets of prophylactic
therapies are found the
term best is really a function of how much the probable time to death is
increased.
Finally, the availability of probable time to death given a certain immune
status and
certain prophylactic therapy can be computed using the following equation as
noted in
Fig. 30:
Prob (deathlIS and PT) _
P(CDIPT and IS) * P(DICD and IS) +
P(not CDIPT and IS) * P(DInot CD and PT IS)
When offering prophylactic therapies to an insured, unique opportunities arise
for
insurance companies providing both life and health benefits. A healthier
insured lives
-280-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
longer and uses less health care, resulting in twofold savings for an insurer.
Because
such a life insurance company also approves health care expenditures, there is
no red tape
or customer effort spent on securing approval for any offered or recommended
prophylactic therapies. Thus, in such contexts, the real world optimizations
can actually
converge on the theoretical optimizations calculated by an ImmunoScore
analysis as
depicted in Fig. 30. This can, in exemplary embodiments, increase QOL for
insureds and
profits for the insurers, as well, hopefully.
Patient commonalities, as revealed by analyses of the CIP database, could be
visualized.
For example, if a population immigrating from Eastern Europe were shown, in
general, to
have lower protection against a specific disease or diseases, that information
could, for
example, be of interest to health/life insurance companies.
Described below is a second exemplary embodiment of the present invention
wherein
ImmunoScore data can be used to optimize the underwriting of life insurance.
The
second embodiment can be used in combination with or separate from the method
described above.
The underlying concept of this method involves ascertaining an individual's
likelihood of
suriving an unkown, unanticipated, or otherwise unaccounted for disease. For
example,
the disease may be one that is not considered at step 3001. For example, new
diseases
emerge (e.g., HIV in the 1980's) or old diseases become resistant to therapies
(e.g.,
antibiotic resistant forms). Thus, ImmunoScore is used to assess the ability
of an
individual's immume system to react favorably to one of these challenges.
-281-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
For example, the magnitude and direction of the combined response of Th 1,
Th2, Th 17,
and Treg can be used to assess an individual's ability to resist new diseases.
In some
embodiments, the probability of death given an immune status (Prob(deathlIS) )
is
inversely related to the magnitude of the combined response. The larger the
magnitude
of the combined response (as trended over time or at one time point), the more
out-of-
balance the immune system is, increasing the chances the individual will
contract
diseases, thereby increasing the appropriate life insurance premium.
This use of ImmunoScore technology may be easier to implement than the other
life
insurance model, because less information is required regarding diseases,
prophylactic
therapies, and how they affect the individual's remaining lifetime.
Nevertheless, this
second method can still improve the stratification of individuals to set
and/or adjust
premium levels. The two methods can be used in combination. The first method
can be
used on all known diseases or on a subset (perhaps only 1 - 10 diseases) to
improve the
results obtained from only using the second method.
L. DIAGNOSING AND MANAGING IMMUNOSENESCENCE IN THE
ELDERLY
Human aging is associated with progressive decline in immune functions and
increased
frequency of infections. Morbidity and mortality due to infectious disease is
greater in
the elderly than in the young, at least partly because of age-associated
decreased immune
competence, which renders individuals more susceptible to pathogens (Pawelec,
et al.
2005). A decline in immune function is a hallmark of aging that affects the
ability to
resist influenza and respond to vaccination. An accumulation of dysfunctional
T cells
-282-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
may be detrimental under conditions of chronic antigenic stress (chronic
infection,
cancer, autoimmunity). The most important changes occur in T-cell immunity,
and are
manifested particularly as altered clonal expansion of cells of limited
antigen specificity
(Fulop, et al. 2005). This is most marked in the CD8+ T cell subset, which
displays a
decrease in both responsiveness and normal function. Normally, CD8+ T cells
appear to
be strongly associated with cytolytic activity, either by direct killing of
antigen-bearing
target cells by granule-mediated exocytosis or Fas-mediated cytotoxic
mechanisms. In
addition, it is suggested that antigen-activated CD8+ T lymphocytes can
eliminate or
control viral infection by secretion of antiviral cytokines, such as gamma
interferon (IFN-
y) and tumor necrosis factor alpha (TNF-a). IFN-y production by CD8+ T cells
can have
both local and systemic consequences, whereas cytotoxins such as perforin are
cytolytic
for the cells that come in direct contact with the cytolytic T lymphocytes
(CTL).
The output of the T cell pool is governed by output from the thymus and not by
replication (Aspinall and Andrew, 2000). As thymic T cell production
diminishes with
age, a decline in contribution made by thymic emigrants to the naive T cell
pool occurs
(Mackall, et al. 1995). Diminution in the size of the naive T cell pool is a
common
finding with aging, and is a consequence of reduced thymic output (Kurashima,
et al.
1995). Thymic atrophy is thought to result from a failure of the thymic
microenvironment to support thymopoiesis in old age and recent evidence
suggests that a
decline in interleukin-7 (IL-7) expression may limit thymocyte development by
restricting combinations of survival, proliferation and rearrangement of the
beta chain of
the T cell receptor (Andrew and Aspinall, 2002). Therapeutic intervention with
IL-7 and
derivatives has been shown to reverse thymic atrophy in old animals and also
lead to
-283-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
improved immune function compared with age and sex matched control animals
(Aspinall, 2005).
The CD8+ T cell repertoire becomes less diverse in old age due to reduced
thymic output
and the accumulation of clonally expanded memory CD8+ T cells as a consequence
of
prolonged antigenic stimulation. Clonally expanded T cells are usually CD8+
and show
an increased incidence with age, so far it seems that clonal expansion is not
due to
malignancy but may follow antigen stimulation. It has been suggested that
repeated or
persistent infections with viruses such as influenza, cytomegalovirus (CMV),
and
Epstein-Barr virus (EBV) may drive responses that result in large T cell
clones.
Longitudinal studies suggest that a set of immune parameters including high
percentages
of peripheral CD8+ CD28- CD57+ T cells, low CD4+ and B cell counts, and poor T
cell
proliferative responses to mitogens is associated with decreased remaining
longevity in
the free-living very elderly (>85 years). CMV seropositivity is closely
associated with
increases in the size of the CD57+ CD8+ T cell pool, which is thought to
represent a
highly differentiated population of late memory cells. Furthermore, CMV
seropositivity
is associated with increases in CD8+ count in old age and has been documented
to have
negative influences on immune parameters in the very elderly. A group
concluded that
the "obsession" of a large fraction of the entire CD8+ T cell subset with one
single viral
epitope may contribute to the increased incidence of infectious disease in the
elderly by
shrinking the T cell repertoire for responses to other antigens. Like CMV, EBV
manages
to persist for the lifetime of the infected host. During chronic asymptomatic
infection in
healthy individuals, EBV resides in memory T cells. Expansion of peripheral
CD8+
CD28- T cells in response to chronic EBV infection has been linked to
rheumatoid
-284-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
arthritis. The clinical consequences of these changes are as yet not well
defined, except
for their extremely important negative impact on defense against infections.
Considering
the public health consequences of decreased immune competence in old age,
strategies
for immune response modulation are desirable to decrease the health burden for
the
elderly and improve their quality of life.
Features of successful aging have been associated with well-preserved immune
function
while poor survival is predicted by high CTL counts, low numbers of B cells
and poor
responses by T cells to polyclonal stimulation. The phenomenon of replicative
sensescence has been associated with these changes and relates to a finite
number of
doublings (25-30 cycles) after which cell cycle arrest occurs. In CTLs, this
growth arrest
is associated with increased production of several pro-inflammatory cytokines,
resistance
to apoptosis and loss of the co-stimulatory molecule, CD28, required for
optimal
stimulation of CTLs. In older adults, greater than 50% of CTLs fail to express
CD28 and
these cells are resistant to apoptosis.
The loss of CD28 expression due to replicative senescence has been associated
with a
number of the adverse effects of aging on immune function. Although the
frequency of
influenza virus-specific CTLs does not appreciably change with age, the
decline in CTL
activity against influenza may be due to a loss of antigen-specific
proliferation and/or
diminished lytic activity. Normal loss of CD28 expression during CTL
activation and the
potential for these cells to undergo activation-induced cell death, may be
confused with
the loss of CD28 with replicative senescence and resistance of CTLs to
apoptosis.
Furthermore, the role of cytokines (such as, for example, IL-2, IL-7, and IL-
15) in
preventing activation-induced cell death and age-related changes in the
production of
-285-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
these cytokines create a complex array of interactions that may confound the
interpretation of in vitro experiments. Understanding the complexity will
provide an
opportunity to optimize the CTL response to vaccination by manipulating CTLs
that
retain their replicative capacity in response to appropriate antigenic
stimuli.
Currently, influenza vaccination of elderly individuals is recommended
worldwide. A
recent study looked retrospectively at influenza vaccine efficacy in
individuals aged 65
years or older. They found that in homes for elderly individuals, that
vaccines were not
significantly effective against influenza, influenza-like illness, or
pneumonia. More
encouragingly, vaccine performance was improved for admissions to the hospital
for
influenza or pneumonia, respiratory diseases, and cardiac disease. This group
concluded
that the usefulness of influenza and pneumococcal vaccines was modest. On the
same
day the Jefferson report was published online, the American Medical Directors
Association released a special announcement regarding the Jefferson study and
influenza
vaccine recommendations for the elderly
(http://www.amda.com/newsroom/092205 vaccines.htm). While not disagreeing with
the tenets of the study, they continued to recommend for vaccination of the
elderly
because influenza vaccination is effective at preventing severe illness,
secondary
complications, and deaths. They also reiterated that the CDC recommends
influenza
vaccination for people age 65 years and over and for all persons in long-term
care
facilities (http://www.amda.com/newsrooni/092205 vaccines.htm). Both groups
concluded that better influenza vaccines that offer more protection in older
persons are
desirable and a high priority of influenza researchers.
-286-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The threat of pandemic influenza has increased with the direct transmission of
highly
pathogenic avian H5N1 viruses to humans. Continued reliance in killed virus or
subunit
vaccines will leave adults at significantly higher risk of illness, disability
and death in the
event of an influenza pandemic. Research that increases our understanding of
how
immunosenescence affects the cell-mediated response to influenza and vaccine
responsiveness is critical to the development of effective pandemic influenza
vaccines for
older people. In the absence of influenza vaccines that target these defects,
an influenza
pandemic will have a significant impact on older people and quickly overwhelm
the
health care system.
On August 8, 2005 the CDC has stated that the effectiveness of inactivated
influenza
vaccine depends primarily on the age and the immunocompetence of the vaccine
recipient and the degree of similarity between the viruses in the vaccine and
those in
circulation. When the vaccine and circulating viruses are antigenically
similar, influenza
vaccine prevents influenza illness among approximately 70-90% of healthy
adults aged <
65 years. Children aged > 6 months can develop protective levels of anti-
influenza
antibody against specific influenza virus strains after vaccination, although
the antibody
response among children at high risk for influenza-related complications might
be lower
than among healthy children. In addition, no efficacy was demonstrated among
children
who had received only one dose of influenza vaccine, illustrating the
importance of
administering two doses of vaccine to previously unvaccinated children aged <9
years.
Older persons and persons with certain chronic diseases might develop lower
post-
vaccination antibody titers than healthy young adults and thus remain
susceptible to
influenza infection and influenza-related upper respiratory tract illness
-287-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
(http://www.cdc. og v/flu/professionals/vaccination/efficacy.htm). While
current vaccines
are cost-saving, new influenza vaccines will likely be needed to avoid the
crisis
anticipated in health care related to the general aging of the population.
Another component to the aging immune system is the relationship between
innate
immunity and inflammation. During evolution the human was set to live 40 or 50
years;
today, however, the immune system must remain active for a much longer time.
This
very long activity leads to a chronic inflammation that slowly but inexorably
damages
one or several organs. This is a typical phenomenon linked to aging and it is
considered
the major risk factor for age-related chronic diseases. Alzheimer's disease,
atherosclerosis, diabetes, sarcopenia, and cancer to name several, all have an
important
inflammatory component, though disease progression seems also dependent on the
genetic background of individuals. Inflammatory genotypes are an important and
necessary part of the normal host response to pathogens in early life, but the
overproduction of inflammatory molecules might also cause immune-related
inflammatory diseases and eventual death later.
Most age-related diseases have complex etiology and pathogenic mechanisms. The
clinical diagnosis and therapy of these diseases requires a multidisciplinary
approach
with progressively increased costs. A body of experimental and clinical
evidence suggest
that the immune system is implicated, with a variable degree of importance, in
almost all
age-related or associated diseases. Both the innate and the clonotypic immune
systems
are usually involved in the pathogenesis of these chronic diseases (Caruso, et
al. 2004;
Pawelec, et al. 2002). Several functional markers of the immune system may be
used
either as markers of successful aging or conversely as markers of unsuccessful
aging. A
-288-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
combination of high CD8+ and low CD4+ and poor T cell proliferation has been
associated with higher mortality in very old subjects. Old men carrying an
anti-
inflammatory IL-10 high-producer genotype or a pro-inflammatory IL-2 low-
producer
genotype show the lowest values of CD8+ cells. This study, however, did not do
a
functional assessment of T cells.
In a mouse model looking at T cell subset patterns, researchers found that a
composite
combination of subset values was a significant predictor of longevity among
genetically
heterogeneous mice, with a strength of association higher in older mice than
among the
young. Developing useful biomarkers of aging has proven to be remarkable
difficult, in
part because many age-sensitive variables tested as candidate biomarkers are
sensitive to
genetic and nongenetic influences other than aging. Any individual assay, for
example a
test of a specific T cell subset in a single blood sample, is likely to have a
good deal of
uncertainty, but the combination of results from related tests may increase
the signal-to-
noise ratio and thus provide stronger predictive power than any single assay
by itself. In
humans, ImmunoScore testing would help build the models of T cell subset
patterns.
Possible courses of therapy would then be ideally tailored to meet the needs
of the
individual and not a "best guess, one size fits all" course of treatment.
Clearly, the population aged > 65 years would be better served by ImmunoScore
diagnostics rather than the current state of affairs. A blanket recommendation
for an
influenza or pneumococcal vaccination for the entire elderly population may
not be in the
best interest of an individual being immunized. ImmunoScore diagnostic tests
could, for
example, first reveal levels of protective antibody to vaccine-preventable
diseases. Of
particular interest would be antibody levels against influenza, pneumococcal
infection,
-289-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
tetanus, diphtheria, pertussis, hepatitis, varicella, CMV, and EBV. Just as
important as
determination of antibody levels in elderly patient sera, ImmunoScore
diagnostic tests
could reveal the status of cellular components of the immune system. The
proportion of
naive/committed T and B cells would be crucial for further recommendations by
the
attending medical staff. As therapeutic interventions are developed for
dealing with
immunosenescence, the ImmunoScore diagnostic information regarding individuals
and
compiled database information will shed valuable light onto the effects of
treatments on
the immune system. As the population ages, strategies for immune response
modulation
are desirable to decrease the health burden for the elderly and improve their
quality of
life.
A preliminary immune risk phenotype (IRP) has been developed from longitudinal
studies of the elderly. Immune system measurements consisted of determinations
of T-
cell subsets, plasma IL-6, IL-2 responsiveness to conconavalin A, and CMV and
EBV
serology. Regression analyses indicated that the IRP and cognitive impairment
together
predicted 58% of observed deaths. This type of analysis would be a valuable
adjunct to
assessing insurance premiums.
The following table captures exemplary desirable analytes to monitor in the
population as
individuals age. A database storing the results of such assays could ensure
that a given
individual's analyte levels could be tracked over time rather than merely
captured as a
snapshot.
Alteration Anal to
CD45RO cells
-290-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
CD95 cells
CD28 expression
CD152 expression
killer cell lectin-like receptor G 1
apoptosis of CD8 cells
apoptosis of CD4 cells
IFN-y production
IL-2 production
telomere lengths
telomerase induction
DNA damage
DNA repair
stress resistance and heat-shock protein expression
Table 1: Alterations in the T-cell compartment with age
Thus, in exemplary embodiments of the present invention an Immunosenescence
superpanel can be defined, comprising the following panels:
Meningococcal Diagnostic Panel;
Persistent Immunity Induced by Childhood Vaccines; and
Immunosenescence Diagnostic Panel
The first two panels are defined in Sections IA1 and IA3 of the Immunologic
Information
Patent, and an Immunosenesence panel can, for example, be defined as follows.
Human aging is associated with progressive decline in immune functions and
increased
frequency of infections. A decline in immune function is a hallmark of aging
that affects
the ability to resist influenza and respond to vaccination. The most important
changes
occur in T cell immunity. An accumulation of dysfunctional T cells may be
detrimental
under conditions of chronic antigenic stress (chronic infection, cancer,
autoimmunity).
-291-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Exemplary Alterations in T-cell compartment to monitor:
Typical Analyte
Alteration
Increased CD45RO cells
Increased CD95 cells
Decreased CD 28 expression
Increased CD 152 expression
Increased Killer cell lectin-like receptor GI
Decreased A o tosis of CD8+ cells
Increased A o tosis of CD4 cells
Decreased IFN-y production
Decreased IL-2 production
Decreased Telomere lengths
Decreased Telomerase induction
Increased DNA damage
Decreased DNA repair
Decreased Stress resistance and heat-shock protein
expression
Other analytes of particular interest in an immunosenescence assay panel can,
for
example, include:
= Antibody to CMV
= Antibody to EBV
= Antibody to influenza
= Antibody to pneumococcal disease
= Antibody to pertussis
= Antibody to tetanus
= Antibody to diphtheria
= Plasma levels of IL-6
= Thl/Th2 components as described below:
Cytokines Receptors 04okines Receptors
INF-y CCR5 IL-4 CCR3
TNF-u CXCR3 IL-5 CCR4
IL-2 CCR1 IL-6 CCR8
IL-12 IL-10 CRTh2
IL-13
-292-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Fig. 31 depicts an exemplary process flow for managing immunosenescent
individuals,
either in a health care provider or a health care insurer context.
In exemplary embodiments of the present invention immunosenescense in an
individual
can be managed using the process exemplary flow depicted in Fig. 31. With
reference
thereto, at 3101, an elderly individual's immune status can be examined. This
can be
accomplished by conducting one or more assay panels as described above in
Section I.
At 3110, the vaccine preventable diseases that the elderly individual is
susceptible to can
be identified at the same time the individual's CMV infection status together
with other
relevant markers of an immune system competence can also be determined. At
3120
vaccine and/or other healthcare recommendations can be made based upon the
immune
status examined at 3101. Additionally, a separate T cell compartment can be
assessed.
At 3130, the individual can be immunized for vaccine preventable disease based
upon his
or her immune system's ability to response to vaccination. Using the
ImmunoScore data,
the individual can be classified as either (1) immunocompetent (2) immuno-
deficient or
(3) somewhere inbetween immunocompetent or immuno-deficient. At 3130 an immuno-
competent individual can be vaccinated as recommended by current ACIP
recommendations. An immuno-deficient individual would need to be managed using
different measures than routine vaccination. Such measures could include, for
example,
adoptive transfer of a compartment of T or B cells or extraordinary hygiene
measures.
The individuals who fall somewhere between immunocompetence and immuno=
deficiency need some kind of hybrid health management between standard
vaccination
and immunoadjuvant therapies such as adoptive transfer of T or B cells and
extraordinary
hygiene measures. At 3140, the elderly individual's immune status can be
followed up
-293-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
post vaccination or post treatment and these results stored in the system
database. At
3150, this information can be used to evaluate the efficacy of the vaccination
or other
therapies as to their abilities to provide the necessary immunity to the
identified diseases.
M. FROZEN STORAGE OF NAIVE IMMUNE CELLS (IRP
CONSIDERATIONS)
As previously described, the immune risk phenotype (IRP) is an emerging
concept -
predicting mortality based on CMV seropositivity (Pawelec, et al. 2005).
Pawelec, et al.
maintained that the manner in which CMV and the host immune system interact is
critical
in determining the IRP and is hence predictive of mortality. The consequences
of IRP is
early expression of immunosenescence. Immunosenescence leads to: a) decreased
T- and
B-cell responses to foreign antigen; b) increased responses to self antigens;
c) increased
morbidity and mortality to infectious disease; and d) decreased response to
vaccine
antigens.
Greater elucidation of the IRP and its consequences is to be expected in the
future.
Genetic screening at a very early age could be predictive of immune health at
a much
more advanced age. The ImmunoScore diagnostic screen could be performed from a
heel
stick done at birth, and a child's baseline immune status could almost
instantaneously be
generated. Pre-natal screening tests could also be developed in the future as
an
immunodiagnostic tool.
Concerned parents may wish to store their child's cord blood as a source of
hematopoietic progenitor cells that could be stored (at a cost to the parents
or the
insurers) for that child for treatment of developing IRP symptoms much later
in life.
Umbilical cord blood (UCB) is currently used as a source of these
hematopoietic
-294-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
progenitor cells as an alternative to the bone marrow or peripheral blood for
treatment of
several onco-hematological diseases (Adami, et al. 2005).
On April 18, 2005 the Institute of Medicine (IOM) issued a report recommending
that a
new cord blood coordinating center - similar to the existing National Marrow
Donor
Program - be set up to ensure a standardized and interconnected national
system to cost-
effectively store and distribute these cells.
ImmunoScore diagnostics shows the need for storing cord blood.
Another application for ImmunoScore diagnostics is to link storage and
analysis of naive
cells of the immune system (innate or adaptive), as next described.
T cells currently used for adaptive immunotherapy trials are selected for
their capacity to
produce high levels of IFN-y and for their ability to efficiently and
specifically lyse
relevant target cells (Dudley and Rosenburg, 2003; Yee, et al. 2002). However,
it was
found that CD8+ T cells that acquire complete effector properties and exhibit
increased
anti-tumor activity in vitro are less effective at triggering tumor
regressions and cures in
vivo (Gattinoni, et al. 2005). While the progressive acquisition of terminal
effector
properties is characterized by pronounced in vitro tumor killing, in vivo T
cell activation,
proliferation, and survival are progressively impaired. These findings suggest
that the
current methodology for selecting T cells for transfer is inadequate
(Gattinoni, et al.
2005). It is clear that new solutions are needed to generate more effective
anti-tumor T
cells for the development of experimental human adoptive transfer-based
therapies.
-295-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The indication is that storage of naive T and B cells is important for
individuals who will
become immunocompromised later in life, whether those cells come from that
individual
or from another source. Naive cells would also not necessarily be isolated
from cord
blood, but could also be isolated from bone marrow or peripheral blood. In
addition,
screening methods can be used to characterize those immune cells regarding
cell surface
characteristics and cytokine expression. Here too, ImmunoScore can be used to
a distinct
advantage.
N. VACCINE USE OUTCOME/DESIGN
Currently, what the public considers vaccines are designed as a prophylactic
means to
avoid illness caused by infectious disease. In practice, agents used to
promote an
immune response as a therapeutic course of action for cancer or immunotherapy
have
also been termed "vaccines." It is the intent of the ImmunoScore design to be
able to
monitor changes in an individual's immune system in relation to a prophylactic
or
therapeutic vaccine and enable the individual patient and his physician to
make the best
possible decisions regarding the patient's immune system health regarding
prophylactic
vaccination, therapeutic vaccination, or other therapeutic treatment in
attempt to "shift"
the immune system of that patient. In addition, the ImmunoScore database will
compile
important population data regarding demographics and population genetics.
0. RESEARCH SERVICES
In exemplary embodiments of the present invention ImmunoScore technologies can
be
used to provide research services, such as, for example, for clinical trials
in the following
areas:
1. vaccines;
-296-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
2. transplants;
3. adaptive immunotherapy;
4. population modeling; and
5. government applications.
P. IMMIGRATION CONSULTING
Testing the immigrant population for vaccine-preventable diseases is another
embodiment of the invention. Governments are very interested in the
immunization
status of individuals and families immigrating into their countries. The
invention can
rapidly provide the results of assays to governmental authorities for all
required
immunizations. There would be no need to rely on paperwork - a diagnostic
examination
would yield more suitable data regarding immune status. The current
vaccination
requirements for immigration into the United States are for measles, mumps,
rubella,
polio, tetanus, diphtheria, pertussis, influenza, hepatitis B and any other
vaccinations
recommended by the Advisory Committee for Immunization Practices (ACIP).
Current
recommendations of the ACIP also include varicella, Haemophilus influenzae
type B,
and pneumococcal vaccines. The current law requires all individuals applying
for status
as a lawful permanent resident (either by applying for an immigrant visa
abroad or for
adjustment of status in the United States) to establish that they have been
vaccinated.
Nonimmigrant (temporary) visa applicants are not required to comply with the
vaccination requirements as a condition of visa issuance, but must comply if
they apply
for adjustment of status at a later date (Immigration and Naturalization
Services, 2001).
-297-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
One or more exemplary ImmunoScore diagnostic panels could, for example, be
provided
to INS or other immigration authorities as a means to determine the immune
status of
immigrants. In practice, ImmunoScore diagnostic testing can be more cost-
effective than
a paper record trail and more likely to be reliable as an accurate assessment
of immune
status of individuals relocating to the United States.
Additionally, the Institute of Medicine (IOM) has concluded that the United
States
quarantine system is in need of a strategic overhaul. The IOM reports that the
United
States once had 55 federal quarantine stations, but the perception that
microbial threats
had been controlled led to dismantling of most of the system in the 1970s.
However,
nearly 40 new infectious diseases have been identified since 1973, and
bioterrorism has
become a serious concern. The 25 stations that will make up the expanded
quarantine
station system now receive more than 75 million international travelers a
year, according
to IOM reports. The stations screen travelers, refugees, immigrants, animals,
and cargo
for disease agents shortly before and during their arrival. However, the
quarantine system
relies on a much broader network that includes local public health agencies,
hospitals,
customs agents, agricultural inspectors, and others, the IOM said.
The IOM recommended the following:
The quarantine stations, the CDC, and the DGMQ (called the quarantine core)
should
lead the effort by developing a national strategic plan with uniform
principles and
outcomes. The quarantine core should shift its main focus from inspecting
people and
cargo at ports to leading the activities of the overall quarantine system. The
strategic plan
-298-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
should help participating government agencies and other groups in the system
to
prioritize activities and focus resources on the greatest risks.
The quarantine core should work with partners in the quarantine network to
define or
redefine each group's roles, authority, and communication channels.
The quarantine system needs enhanced skills, more people, more training, more
space,
and better use of technology to fulfill its evolving role. An example of
technology cited in
the news release was targeted use of passenger locator cards that could be
used on flights
to and from countries with disease outbreaks. The cards would log passenger
seat
numbers and contact information in a scannable format. This could simplify
tracking of
passengers potentially exposed to disease, such as those who new to the United
States
from Sierra Leone in 2004 with a man who later died of Lassa fever.
The quarantine core must review its methodology periodically to ensure
stations are in
the best places and appropriately staffed.
The quarantine core must have plans, capacity, resources, and "clear and
sufficient legal
authority" to respond quickly to surges in activity at one or more ports.
The core must define and fund a research agenda to measure the effectiveness
of its
procedures. The committee found that many routines at quarantine stations are
based on
experience and tradition and lack a scientific basis.
The core must use scientifically sound methods to measure the effectiveness
and quality
of its operations, including assessing performance of critical functions
throughout the
system. It must also address any shortfalls that come to light.
-299-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
(http://www.nap.edu/books/030909951 X/html).
ImmunoScore technology could be useful at such immigration port of entry
screening
points. There is a need for global health that can not be understated. The
cost of failure
could be extremely high. There are people moving around the globe and among
the
states with clear health needs, and they are currently moving without the
ability of
government authorities to track them.
Additionally, ImmunoScore technologies can be used to discover links between
immunological phenomena. For example, from the results of Greenway (discussed
above
in Section I regarding the immigrant panel) a possible link between TB
infection and
HepB prevalence can be investigated by analyzing sera from an immigrant
population for
both active TB and HepB seropositivity. It is possible that more than one co-
infection
may be found in this manner. For example, in the following study, A high
prevalence of
hepatitis B virus infection among tuberculosis patients with and without HIV
in Rio de
Janeiro, Brazil, Blal CA et al Eur J Clin Microbiol Infect Dis. 2005
Jan;24(1):41-3, such
a correlation was in fact found.
The Blal study sought to investigate the prevalence and exposure factors
associated with
hepatitis B infection in tuberculosis patients with and without HIV type 1 co-
infection.
The presence of hepatitis B virus serological markers was investigated in a
retrospective
study. The seroprevalence of hepatitis B virus in patients with tuberculosis
only was
14.6%, and in tuberculosis patients co-infected with HIV it increased to
35.8%. In
patients with HIV and tuberculosis co-infection, homosexuality constituted the
principal
exposure factor, while in tuberculosis patients without HIV, a gradual
increase in
-300-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
hepatitis B virus seroprevalence was noted along with increasing age. These
results
demonstrate that hepatitis B infection is highly prevalent in tuberculosis
patients in Brazil
and suggest that a vaccination program for the general population should be
considered in
order to prevent further hepatitis B infections.
Q. DISASTER SURVIVORS: IMMUNIZATIONS, RECOVERY, PROGNOSIS
AND TREATMENT
In exemplary embodiments of the present invention, rapid response services to
disaster
survivors can be provided. Fig. 32 depicts an exemplary process flow for such
an
application.
At 3201 a disaster survivors' immune status can be examined using one or more
ImmunoScore assay panels as described above in Section I. At 3210 the vaccine
preventable diseases to which the survivor is susceptible can be identified
and
simultaneously the cellular component of his or her immune system can be
assessed to
get an immediate post disaster baseline. At 3220 vaccination and healthcare
recommendations can be generated based upon antibody levels to the identified
to the
assay vaccine preventable diseases. At 3230 immunization can be carried out
and at
3240 follow-up examination of the survivor's immune status can be administered
and the
results stored in the system database. Further screening of T cell components
of the
immune system is recommended for all survivors regardless of their
psychological state
at the time in order to develop data regarding post-traumatic stress disorder.
Finally, at
3250 the efficacy of the vaccine and/or therapies can be evaluated as to their
ability to
provide necessary immunity to the identified diseases.
-301-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
There are many different possible responses of an individual to an event
perceived as
potentially life-threatening. It is difficult to predict long-term responses
to trauma based
on the acute response to a traumatic event. If physiological risk factors are
important in
understanding how psychopathology develops, then ImmunoScore measurements can
provide invaluable research information and possibly identify treatments yet
to be
defined. This could pave the way to personalized medicine. Fig. 33 illustrates
possible
responses to trauma.
With reference thereto, at 3301 a Disaster Trauma occurs. There are two
pathways
leading from 3301, namely, Normal Response Factors 3305 and Pathological
Response
Factors 3303. A Normal Response Factors 3305 pathway from Disaster Trauma 3301
leads to Recovery at 3310. However, Pathological Response Factors 3303 lead an
individual from Disaster Trauma 3301 to Post-Traumatic Stress Disorder 3320.
It is the
job of healthcare personnel to put the individual on a Pathway to Recovery
3310. In
exemplary embodiments of the present invention ImmunoScore technologies can be
used
to determine possible therapies 3315, as well as to track immunological
correlates of
PTSD to verify diagnosis and evaluate therapeutic efficacies.
In the immediate aftermath of a traumatic event, most people experience a
combination
of the following symptoms: (a) difficulty sleeping, (b) difficulty
concentrating, (c)
irritability, (d) nightmares, (e) recurrent thoughts of the trauma, and (f)
distress at the
reminder of the traumatic event. The question in the determination of a
pathological
response is when does the continuation of these "normal" responses become
pathological,
and have serious effect on the health of the individual's immune system?
-302-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
There are different possible outcomes of trauma exposure. There is an
increased risk of:
(a) Post-Traumatic Stress Disorder (PTSD), (b) major depression, (c) panic
disorder, (d)
generalized anxiety disorder, (e) substance abuse, and (f) other somatic
symptoms or
expressions of physical illness including hypertension, asthma, and chronic
pain
syndromes. The differential outcomes may rely on different physiological
parameters.
Pre-existing cognitive factors may or may not be the cause, result, or
correlate of pre-
existing biological alterations, either or both setting the stage for an
extreme response to
the trauma. Clarifying the precise nature and biological correlates of
symptoms that
appear in the immediate aftermath of a trauma will assist in developing models
for
potential prophylactic interventions and early treatments. In this regard the
ImmunoScore diagnostic panel could initially be used in a research application
to track
immune system markers and relate them to specific conditions. As a system
database
evolves, ImmunoScore panels can, for example, be used as a guide to
therapeutic
treatment.
Individuals currently at the greatest risk for developing PTSD following
trauma are those
individuals with (a) a family history of psychopathology, (b) a history of
childhood
abuse, (c) prior trauma exposure, and (d) the cognitive factors of lower IQ,
female
gender, and poor social support. There is increased concordance for PTSD in
monozygotic twins compared with dizygotic twins lending support to the genetic
pre-
disposition argument of PTSD.
-303-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
R. MONITOR ADOPTIVE IMMUNOTHERAPY/TRANSPLANTS
After adaptive transfer, several events must occur for T cells to cause the
regression of
established tumors. T cells must be activated in vivo through antigen-specific
vaccination. They must then vigorously expand to levels capable of causing the
destruction of significant tumor burdens. Finally, anti-tumor T cells must
survive long
enough to complete the eradication of all tumor cells (Overwijk, et al. 2003).
It has been
found in an animal model that the progressive differentiation of T cells to a
terminal
differentiated effector stage results in a series of phenotypic and functional
changes that
make them less "fit" to perform these functions (Gattinoni, et al. 2005).
In patients under consideration for adaptive immunotherapy and/or
transplantation,
history and analyses of exposure to CMV, EBV, West Nile Virus, and viral
hepatitis in
both the donor and recipient are crucial. ImmunoScore diagnoses of both the
donor and
recipient would examine the immune history of both individuals.
S. ELECTIVE SURGERY
Many patients opt for elective surgery - plastic surgery, facial plastic
surgery,
dermatology, cosmetic dentistry, vision, urology, and infertility among
others. Whenever
undergoing surgery, there is a risk of nosocomial infection. Common organisms
that
cause nosocomial infections are Apergillus, Candida, Staphylococcus aureus,
Staphylococcus epidermidis, Pseudomonas aeruginosa, and Bordetella pertusis.
Prior to
elective surgery, it would benefit the patient and the attending surgeon to
know the level
of antibody protection to these infectious agents. An ImmunoScore panel could
be
tailored to meet these diagnostic needs and immunizations could be provided to
those
agents with available vaccines. If the patient's immune status is sufficiently
poor, a
-304-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
recommendation not to have the surgery may be given. In these cases, the
expected value
of costs of complications rising from infections outweighes the expected value
of the
benefits from the surgery. In addition, following surgery patients could be
screened for c
reactive protein (CRP), tumor necrosis factor-alpha (TNF-a), IL-6, and soluble
IL-2
receptor (sIL-2R) as possible early indicators of inflammation leading to
sepsis. It is
important to screen for a panel of analytes indicating sepsis, as one analyte
is often not
enough to get a proper diagnosis.
T. SERVICES TO CHARITABLE FOUNDATIONS PROMOTING
IMMUNOLOGICAL WELL BEING
Currently, the lack of accurate, affordable, and accessible diagnostic tests
significantly
impedes global health efforts. The Global Alliance for Vaccines and
Immunizations
(GAVI) was created in 1999 to protect health and save children's lives
throughout the
widespread use of modern vaccines. GAVI is a partnership of governments,
international
organizations, major philanthropists, research institutions, and the private
sector that
work together to: (a) improve access to sustainable immunization services, (b)
expand the
safe use of all needed cost-effective vaccines, (c) accelerate research and
development
efforts for new vaccines needed in developing countries, (d) make immunization
coverage a key indicator of development, (e) promote sustainability by
adequate
financing, and (f) reinforce global and national immunization goals including
eradicating
polio, eliminating maternal and neonatal tetanus, reducing measles, and
increasing access
to vitamin A.
Underlying all health care tools - including therapeutic products, vaccines,
and other
preventative tools - are "platform" technologies that define and facilitate
their use. For
-305-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
example, immunochromatography is a technology platform that has enabled the
development of affordable, easy-to-use dipstick format diagnostic tools. The
ImmunoScore diagnostic panel, a platform technology, can be used to great
advantage by
GAVI to improve global health efforts
GAVI issues requests for proposal (RFP5) to support research efforts to create
diagnostic
technology platforms and tolls that enable improved prevention, treatment, and
surveillance in developing country settings. The foundation issues RFPs to
support the
systemic evaluation of sets of genes, proteins, and cellular pathways to
determine their
potential role in contributing to the development of new vaccines,
diagnostics, and drugs
for GAVI's priority diseases and conditions. One area of concern is population
genetics
and how to design drugs and vaccines to discourage the emergence of resistance
and to
discover how genetics affects the efficacy of drugs and other interventions.
The
ImmunoScore database would be an ideal tool for GAVI to use to evaluate
genetic
parameters and immune response to vaccines and drugs under consideration. A
.second
area is applied immunology. Here systematic approaches, such as that provided
by the
ImmunoScore technology, are needed to measure the human immune response to
guide
vaccine design and define biological signs that identify early or latent
infection.
U. DISCOVERY OF UNWANTED IMMUNOGENICITY OF
THERAPEUTICS
There is potential of the human immune system to identify biological
therapeutic
products as foreign and mount an immune response. There are three main areas
of
concern with the production of antibodies against biological therapeutics in
humans:
-306-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Safety - assurance of safety involves the assessment of whether antibodies
induced could
have adverse clinical implications.
Efficacy - can be affected by the presence of antibodies binding to the
product and
reducing its potency.
Measurement of phamacokinetic/pharmacodynamic parameters - the presence of
antibodies can alter these clinical parameters and also interfere in the
assays used in their
assessment (Koren, et al. 2002).
The immunogenicity of therapeutic proteins can be influenced by many factors,
including
the genetic background of the patient, the type of disease, the type of
protein (human or
nonhuman), the presence of conjugates or fragments, the route of
administration, dose
frequency, and duration of treatment (Schellekens, 2002). Manufacturing,
handling, and
storage can introduce contaminants, or alter the three dimensional structure
of the protein
via oxidation or aggregate formation. Various means have been suggested by
which
therapeutic proteins might be modified to reduce their immunogenicity,
including
PEGylation, site-specific mutagenesis, exon shuffling, and humanization of
monoclonal
antibodies (Schellekens, 2002). In the future, it may be possible to predict
the
immunogenicity of new therapeutic proteins more accurately, using specifically
designed
animal models, including nonhuman primates and transgenic mice.
ImmunoScore diagnoses and database storage could be instrumental in the
development
of analytical techniques to monitor both the drugs and the patient population.
An
individual's tendency to mount an immune challenge to a protein therapeutic
could be
revealed prior to initiation of the treatment based upon the patient's
ImmunoScore
-307-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
profile. In addition, once therapeutic treatment began, ImmunoScore diagnoses
and
database management could track a patient's immune response to the drug. The
drug
manufacturers would be able to use the ImmunoScore technology to conduct
clinical
trials and also to select an appropriate population in which to test the drug.
Based upon
ImmunoScore population data, the drug could be designated for use based upon
the
genotype of the individual being treated.
Fig. 34 depicts an exemplary process flow for an ImmunoScore immunogenicity
study in
exemplary embodiments of the present invention. The exemplary study is
directed to
immunogenicity of therapeutic proteins.
With reference thereto, at 3401 a prospective patient's immune status can be
examined to
obtain a baseline ImmunoScore. At 3410 patients for whom treatment would not
be
advisable (based upon immune system profiling) can be identified, and
therapeutic
treatment for a patient group for which therapy is advisable can be initiated.
At 3420
patients' further treatment and health care recommendations can be made, based
on
careful periodic monitoring of antibody levels to therapeutic proteins. In
addition,
cellular components of the immune system would warrant careful monitoring -
particularly in regard to the antigenic components of the therapeutic
compound. At 3430
patient data can be compiled for drugs in clinical trial. Population data can
also be
compiled to assist in drug design. At 3440, follow-up examinations of
patients' immune
status post-treatment can be implemented and the results stored in a system
database.
Further screening of antibody levels and T-cell components of the immune
system can be
implemented for all patients. Finally, at 3450 the efficacy of therapies to
provide
-308-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
necessary treatment to patients can be evaluated, and the extent of
undesirable
immunogenicity can be determined.
V. TWO-SIDED MARKET APPLICATIONS
A two-sided market is a market wherein there are two sets (at least) of
customers that, in
effect, need each other. Each type of customer values the market more if the
other type
of customer also buys the service. Businesses service such markets by acting
as
"matchmakers."
Examples of two-sided markets:
= computer operating systems
o software developers write applications
o computer users run applications
= video game console manufacturers serve
o video game players
o video game designers
= payment card companies
o consumers
o merchants
-309-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
These businesses all produce platforms that make matches between two or more
distinct
groups of consumers.
The description of two-sided markets likely came about from payment card
companies
and legal ramifications of what may have been perceived as monopolistic
business
practices, but was actually the demonstration of two-sided marketing
practices. A key
aspect to the business model for most of these industries involves the optimal
pricing
structure: the division of revenues between the two sides of the market that
gets both
sides on board. The need for pricing structure as well as pricing level
distinguishes
industries based on a two-sided market from the industries ordinarily studied
by
economists. In two-sided markets, the product may not exist at all if the
business does
not get the pricing structure right. Currently, there is no appreciable market
for adult
vaccines, other than those for influenza and pneumococcus. ImmunoScore
diagnostics
can, for example, likely reveal lapses in protection for vaccine preventable
diseases, such
as pertussis, tetanus, diphtheria, mumps, measles, and others. Diagnostic
testing can thus
reveal a large marketing potential for vaccine manufacturers.
Both the ImmunoScore diagnostic application and the ImmunoScore database
management modules can be considered as two-sided marketing opportunities, in
that
none of the participants (patients, insurers, researchers, primary care
physicians, vaccine
manufacturers, or government entities) may necessarily be willing to enter
into a
beneficial marketing alliance without direction provided by the ImmunoScore
platforms,
as illustrated in Fig. 35. ImmunoScore can act as a "matchmaker" for these
different
groups of consumers. An ImmunoScore diagnostic platform can, for example,
serve to
link patients, physicians, and vaccine manufacturers and help to illustrate
the need for
-310-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
continuing vaccine coverage in adults and children at risk. As an ImmunoScore
database
module grows from an ImmunoScore diagnostic module, insurers, research groups
(both
academic and commercial), and groups responsible for vaccine decision-making
(ACIP
and AAP) and tracking (CDC and VAERS) can be able to take advantage of the
data
generated from assessing the immune status of patients.
Network effects. A network effect arises when the value that one user receives
from a
product increases with the number of other users of that product. It goes
without saying
that the value to governmental decision-making, insuring, and research
interests can
expand enormously with the increase in size of an ImmunoScore patient
database. Health
insurers can also be involved at ImmunoScore diagnostic platform level.
Insurers that
would be interested in providing insurance based upon an individual patient's
ImmunoScore would benefit most from the ImmunoScore database platform. Most
network effects arise because a product tends to be two-sided. ImmunoScore,
having
more than two interactive sides, would demonstrate large network effects that
should
benefit society as a whole, with better health for the population at large and
decreased
costs for the insurers. Information garnered from an ImmunoScore Database can
enable
the performance of vaccine researchers and the vaccine decision-makers in the
government tremendously.
Survey of Two-Sided Markets
Diverse industries:
= credit cards
-311-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= computer operating systems
= video games
= corporate bond trading
= residential real estate
Firms in these industries have adopted similar business models and pricing
strategies for
solving the problem they have in common - getting and keeping two sides of a
market on
board. The intermediary helps customers complete a transaction by providing a
platform.
The transaction occurs when both sides get together. Currently, there is a
real need to get
adult patients and vaccine manufacturers together for the betterment of public
health.
ImmunoScore diagnostics will be an effective facilitator of this interaction,
with the
medical insurance companies being a third beneficiary. The intermediaries
succeed in
the businesses by figuring out how a pricing structure internalizes the
externalities
between the two sides. In the case of ImmunoScore Diagnostics, the health
insurers
should be willing to pay for the diagnostic testing as well as the cost of
vaccination, as
those costs would be less than those to treat debilitating diseases otherwise
preventable
by judicious use of vaccination.
A market is two-sided if at any point in time there are:
= two distinct groups of customers - with ImmunoScore diagnostic and
ImmunoScore database platforms, there would be patients, vaccine
manufacturers, health insurers, vaccine researchers, and vaccine decision-
-312-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
making organizations that would benefit from the two-sided ImmunoScore
platforms.
= the value obtained by one kind of customer increases with the number of
the other kind of customers - as the number of patients are added to the
database, the database would increase in potential utility to researchers,
vaccine decision-makers, insurers, and vaccine manufacturers. The more
the database grows, the better it would be for the patient population as
physicians would better be able to determine individual patient's immune
status based on knowledge accumulated over the entire patient population.
= the intermediary is necessary for internalizing the externalities created by
one group for the other group - there is currently no real push for adults or
older children to have diagnostic screening related to vaccine-preventable
diseases. As ImmunoScore data accumulate, there should be an added
impetus for adult and adolescent vaccination coverage.
Researchers have examined the pricing and production strategy of a firm in a
two-sided
market. Consider the case in which both sides of the market are buying a
"transaction"
and in which the seller incurs a marginal cost for consummating that
transaction. The
prices charged to the buyers and sellers are two prices. The buyer's demand
depends
only on the price faced by the buyer and the seller's demand depends only on
the price
faced by the seller. The demands can be thought of, roughly speaking, as the
number of
buyers and sellers using the system. The transactions that a seller engages
in, and its
benefits from those transactions, increase proportionally with the number of
buyers on
-313-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
the system. The same holds for an individual buyer. Total demand equals the
product of
the two demands. Thus, if there were 500 sellers and 100 buyers, there would
be 50,000
transactions. The assumption of a multiplicative demand between the two sides
actually
understates the importance of the indirect network effects. It ignores the
fact that the
value each side obtains from the other side increases with the number of
customers on the
other side. In the cases of ImmunoScore diagnostics and an ImmunoScore
database, the
benefit to all sides of the market could increase dramatically (presumably
something
more than a multiplicative effect) as the number of consumers grows. Feeding
information to the database can only assist patients, physicians, vaccine
decision-making
bodies, vaccine manufacturers, and health insurers.
None of the conditions for determining the price level or the price structure
in two-sided
markets corresponds to marginal revenue equaling marginal costs on either side
of the
market. Such conditions have no meaning in two-sided markets because there is
no way
to allocate the increases in revenues from changes in prices to one side or
the other.
Changes in prices result in more "transactions" from which each side jointly
benefits.
Business Models in Two-Sided Markets. Both sides need to be brought on board.
For instance, there would be no demand by households for payment cards is they
could not be used anywhere, and no demand by retailers if no one had them to
use.
Investment and pricing strategies are key to getting both sides on board. Even
with
both sides on board, businesses have to carefully balance their two demands.
They
have to consider how changes on one side of the market will impact the other
side of
the market. The need to balance the needs of the various consumers will be of
utmost importance to the careful development of ImmunoScore diagnostics and
-314-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
ImmunoScore database management as two-sided markets. Currently, patients,
physicians, and vaccine manufacturers seem painfully unaware of the need for
diagnostic testing and boosting for vaccine-preventable diseases.
Getting both sides on board. One way to get both sides on board is to obtain a
critical
mass of users on one side of the market by giving them the service for free or
even
paying them to take it. Another way to solve the chicken-and-egg problem is to
invest in
one side of the market to lower the costs to consumers on that side of
participating in the
market. Providing low prices or transfers to one side of the market helps the
platform
solve the chicken-and-egg problem by encouraging the benefited group's
participation -
which in turn, due to network effects, encourages the non-benefited group's
participation.
Another effect of providing benefits to one side is that this assistance can
discourage use
of competing two-sided firms. In the case of the ImmunoScore diagnostic and
database
platforms, initially the medical insurance industry would likely bear the
burden of any
associated costs, but the benefit to this industry in increased wellness of
their clientele
should offset any up-front costs. In addition, to those patients who are seen
to have an
unfavorable ImmunoScore the supplemental insurance industry should be
available and
able to come in and insure those individuals with special needs.
Pricing strategies and balancing interests. Firms in mature two-sided markets
still have to
devise and maintain an optimal pricing structure. In most observed two-sided
markets,
companies seem to settle on pricing structures that are heavily skewed towards
one side
of the market. Certain customers on one side of the market may be extremely
valuable to
customers on the other side of the market - "marquee buyers." In the case of
the
ImmunoScore diagnostic and database platforms, the "marquee buyers" could be
seen as
-315-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
large HMOs that would in truth benefit from having a healthier patient
population. The
existence of marquee buyers tends to reduce the price to all buyers and
increase it to its
sellers. Acceptance of ImmunoScore platforms by large insurance organizations
and
government agencies would enable "bringing on board" smaller insurance
agencies. A
similar phenomenon occurs when certain customers are extremely loyal to the
two-sided
firm - perhaps because of long-term contracts or sunk-cost investments.
Multihoming. Most two-sided markets in the real world appear to have several
competing two-sided firms and at least one side appears to multihome.
Multihoming
affects both the price level and the pricing structure. Not surprisingly the
price level
tends to be lower with multihoming. The possibility of multihoming may
encourage
firms to lower their prices on the side of the market in which multihoming
could occur.
By lowering their prices, firms discourage customers on that side from
affiliating with
other two-sided firms. The firm can then charge more to customers on the other
side, for
whom fewer substitutes are available.
Two-Sided Markets and Social Welfare. A relatively small number of firms tend
to
compete in two-sided markets. That is because these markets have network
effects and
usually incur substantial fixed costs for getting one or both sides on board.
Larger firms
have advantages over smaller firms because larger size delivers more value - a
bigger
network - to consumers. Firms in concentrated two-sided markets may have
opportunities to earn supra-competitive profits - i.e., profits that exceed
those necessary
to attract capital to the industry after accounting for risk. Several factors
affect the extent
to which this can happen over time.
-316-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
7. The extent to which firms are competing to become established in a two-
sided market. This results in investment to court customers, to provide
them with subsidies in the form of equipment, and to offer them low or
negative prices. Vaccine manufacturers and physicians offices might
initially need to be coaxed into the ImmunoScore diagnostic and database
markets, but should see the benefits as the structure grows.
8. The extent to which there are first mover advantages in getting either side
of the market on board Then, the competition to make these investments
should reduce the opportunities to earn significant supra-competitive
returns. Savvy Health Maintenance Organizations could be the first to
realize the benefits to their coverage that ImmunoScore diagnostics and
database platforms could provide, and as such may be eager to get into this
opportunity at the ground level. The governmental organizations could
also be made to see the benefits of diagnosing and cataloging lapses in
vaccine-preventable disease conditions.
The consequences of having relatively few competitors in two-sided markets,
and the
existence of network effects, raise familiar issues concerning the efficacy of
competitive
markets and the possible roles for government intervention. The pricing and
investment
strategies that firms in two-sided markets use to get both sides on board and
balance the
interests of both sides raise novel questions. These pricing and other
business strategies
are needed to solve a fundamental economic problem arising from the
interdependency of
demand on both sides of the market. In some cases, the product could not even
exist
without efforts to subsidize one side of the market or the other. In the case
of the
-317-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
ImmunoScore platforms, the patients would likely need to be subsidized by the
participation of health insurers.
Researchers have compared the pricing structure adopted by firms in two-sided
markets
to the pricing structure that would maximize social welfare. Interestingly,
they find that a
monopoly firm, a firm with competition, and a benevolent social planner would
adopt
similar pricing structures. The precise relative prices would differ somewhat.
They
found that the pricing structure adopted by the market is not biased towards
one side of
the market or the other side of the market compared to the pricing structure
that would be
adopted by the benevolent social planner. ImmunoScore diagnostic and database
platforms may be thought of as a benevolent social plan. The welfare of the
patients is
paramount, and there would be additional benefits presented to vaccine
manufacturers,
research groups, and government organizations.
Two-sided markets are an increasingly important part of the global economy.
Firms that
provide platforms for multiple consumer groups are a critical part of many
interrelated
segments of the computer industry. In most industrialized countries a large
fraction of
payments takes place through firms and associations that provide platforms for
merchants
and customers to exchange money. The increased importance of the Internet for
household-to-household, business-to-household, and business-to-business
transactions
and the emergence of e-pay systems on the Internet will increase the fraction
of payments
going through commercial payment platforms. ImmunoScore diagnostic and
database
platforms would help bring health care into the 21St century. There is a
tremendous need
for portability in health care record-keeping, and the ImmunoScore database
platform
-318-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
would be instrumental in the transfer of health care records from primary care
physician
to specialist.
Two-sided firms have to come up with the right price structure and the right
investment
strategy for balancing the demands of the customer groups they must get and
keep on
their platforms.
In many industries, platforms court two (or more) sides that use the platform
to interact
with each other. The platform may charge interaction-independent fixed fees.
For
example, American Express charges yearly fees to cardholders. In the case of
video
games, platforms charge game developers fees for development kits on top of
royalties
per copy sold, and they charge gainers for the console. In the case of the
ImmunoScore
database platform, it might be appropriate to charge academic and commercial
research
groups for use of the information captured by the database modules.
Managers devote considerable time and resources to figure out which side
should bear
the pricing burden, and commonly end up making little money on one side (or
even using
this side as a loss-leader) and recouping their costs on the other side.
Marketing
managers for the ImmunoScore platforms will need to carefully balance many
consumers' needs and the applications of fees.
Pricing Principles for Two-Sided Platforms. Departures from standard business
strategies that result from the platform's internalization of the other side's
welfare (the
linkage between the two sides from the platform's viewpoint). This linkage is
most
apparent when the platform makes no or loses money on one side. A factor that
is
conducive to a high price on one side, to the extent that it raises the
platform's margin on
-319-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
that side, tends also to call for a low price on the other side as attracting
members on that
other side becomes more profitable. In the case of the ImmunoScore platforms,
it is
imperative to bring patients on board, but their participation might be
encouraged by the
dual factors of their curiosity as to their personal ImmunoScore and also the
participation
of their insurer in the platform.
Platforms must perform a balancing act with respect to their price structure
as well as
other policy dimensions; quite generally, they encourage positive
externalities and
discourage negative ones and to do so usually constrain one side to the
benefit of the
other. While asymmetric information and the concomitant rent extraction
concerns keep
the platform's price structure neutral, it is nonetheless a source of sub-
optimal trade
among end-users. The platform has an incentive to cap or alter through a
subsidy the
price charged to buyers so as to boost buyer's surplus and their willingness
to join the
platform. Then the platform behaves pretty much like a public utility
commission that
addresses a market power problem by setting a price cap or by subsidizing some
services
through a fund levied from other services.
The rationale for constraining the price charged by the seller to the buyer
would vanish if
the industry were organized according to the vertical view: were the platform
not to deal
directly with buyers, the platform would want to provide sellers with the
maximal profit
in their relationship with buyers and therefore would grant them maximal
commercial
freedom. It is only because the platform can extract surplus on the buyer side
that it is
willing to displease the seller side by constraining it.
-320-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
End-users often care not only about the price (that they pay to the platform
and to the
other side), but also about the quality of the interaction. In health care,
the quality of the
physician-patient interaction assumes particular importance. An ImmunoScore
Diagnostic platform will help nurture the doctor-patient relationship and
focus on the
patient's "wellness" rather than strictly on "treatment."
While price regulation is complex or inefficient, the platform may still make
itself
attractive to one side of the market by encouraging competition on the other
side.
Competition on the other side brings prices closer to marginal cost, and the
volume of
interactions closer to the efficient volume; it also protects against the hold-
up of one's
specific investments. An ImmunoScore diagnostic platform could encourage
competition
among vaccine manufacturers on behalf of the patient population. The
manufacturers
should still realize greater sales, but their prices should remain competitive
for the
insurers and patient population. Accordingly, a two-sided platform benefits
from
allowing competition on a given side as it can at least partly recoup benefits
on the other
side.
Dynamics. To create a two-sided market, a "chicken or egg" problem has to be
solved: to
convince some buyers to adopt a certain intermediation platform, it is
necessary to first
convince some sellers; but, to convince the sellers, there must be some buyers
on the
market. In most models, this problem is avoided by assuming the simultaneous
arrival of
agents on the two market sides, in a rational expectations equilibrium.
However, there
are circumstances in which one market side has to intervene before the other
one. The
most cited case is the one of videogame consoles which, to get customers, must
appear on
the market already equipped with a complete range of games and complementary
-321-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
applications. There appears to be a growing need for the determination of a
patient's
immune status. There is a current outbreak of mumps disease in the Midwest in
individuals that have received two MMR immunizations. The incidence of
pertussis
continues to increase. Travel has now been related to the spread of Severe
Acute
Respiratory Syndrome (SARS), influenza, measles, tuberculosis, and mumps. The
time
is appropriate for the introduction of the measurement of the immune status of
individual
patients, and the tracking of information regarding each individual's immune
status.
ImmunoScore diagnostic and database platforms can tip the balance from a
display of
need to a mode of action going forward.
W. DRUG HYPERSENSITIVITY
Incorporating Drug Hypersensitivity into a Two-Sided Business Model
Adverse drug reactions are common. Identifying true drug allergy, however, can
be
challenging. Drug hypersensitivity is a clinical diagnosis based upon
available data.
Drug hypersensitivity is defined as an immune-mediated response to a drug
agent in a
sensitized patient. Identifiable risk factors for drug hypersensitivity
reactions include
age, female gender, concurrent illnesses, and previous hypersensitivity to
related drugs.
Monitoring drug hypersensitivity in patients and incorporating those data into
an
ImmunoScore database platform is another example of a two-sided market
opportunity.
As with the other examples of two-sided markets, the medical insurance
organizations
would likely initially cover most of the expenditures to bring the other
market
components into the market that would be beneficial to all participants. Other
"sides" of
the market would involve patients, physicians, researchers for both the
pharmaceutical
industry and allergy specialists.
-322-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
The Gel and Coombs classification system describes the predominant immune
mechanisms that lead to clinical symptoms of drug hypersensitivity (Table 1).
This
classification system includes IgE-mediated Type I reactions, cytotoxic Type
II reactions,
Type III reactions involving the formation of immune complexes, and the
delayed, cell-
mediated Type IV reactions. However, some drug hypersensitivity reactions are
difficult
to classify because of a lack of evidence supporting a predominant immunologic
mechanism. These include certain cutaneous drug reactions and specific drug
hypersensitivity syndromes.
Diagnostic testing for these reactions remains somewhat problematic. The
current types
of tests and therapeutic considerations are for each of the four types of
hypersensitivity
reactions described in Table 2 below. Confirmation of suspected Type I
hypersensitivity
reactions requires the detection of antigen-specific IgE. Currently, skin
testing is a useful
diagnostic procedure for reactivity to penicillin. With other drug agents, a
negative skin
test does not effectively rule out the presence of specific IgE. Further IgE
tests for other
agents await development. The sensitivity of ECL technology as embodied in an
exemplary ImmunoScore diagnostic platform can be a very effective tool to
enable
researchers to better study IgE populations specific for drug component
antigens.
Currently, the diagnosis of drug hypersensitivity is usually based upon
clinical judgment
because definitive, confirmatory drug-specific testing is often difficult.
Once the diagnosis has been established, appropriate documentation should be
included
in the medical record specifying the causative drug and the nature of the
adverse effect.
Immune-mediated drug hypersensitivity reactions typically pose a predictable,
more
serious health risk with re-exposure to the drug. In this application of the
technology, an
-323-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
exemplary ImmunoScore database platform can, for example, capture all
pertinent
information related to any adverse drug reaction. This would not only be of
benefit to the
patient, but also as data was accrued, pharmaceutical companies would benefit
from
statistical information gathered from mining the database. Real drug
hypersensitivity
would also be separated from reactions that may not be hypersensitivity.
Instead of
relying on patient recall and a faulty data collection system, an exemplary
ImmunoScore
database can only include documented case histories. Patient medications can,
for
example, be tracked via an ImmunoScore database and real hypersensitivity can
be
officially documented.
The most important drug-related risk factors for drug hypersensitivity concern
the
chemical properties and molecular weight of a drug. Larger drugs with greater
structural
complexity are more likely to be immunogenic. Heterologous antisera,
streptokinase and
insulin are examples of complex antigens capable of eliciting hypersensitivity
reactions.
Another factor affecting the frequency of hypersensitivity drug reactions is
the route of
drug administration; topical, intramuscular, and intravenous administrations
are more
likely to cause hypersensitivity reactions. These effects are caused by the
efficiency of
antigen presentation in the skin, the adjuvant effects of repository drug
preparations, and
the high concentrations of circulating drug antigen rapidly achieved with
intravenous
therapy. Oral medications are less likely to result in drug hypersensitivity.
Most medications, because of their small molecular size, are unable to elicit
an immune
response independently. Drugs must first covalently bind to larger carrier
molecules such
as tissue or serum proteins to act as complete multivalent antigens. This
process is called
haptenation, and the drugs act as haptens. The elicited immune response may be
-324-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
humoral, with the production of specific antibodies, cellular, with the
generation of
specific T lymphocytes, or both. Frequently, the identity of the metabolites
is unknown,
making it impossible to develop accurate diagnostic tests for drug allergy
(Solensky,
2006).
A thorough history is an essential component of the evaluation of patients
with suspected
drug allergies. The history helps guide the clinician in the choice of
diagnostic tests and
the decision whether it is safe to reintroduce the medication. Typically,
years or decades
have passed since reactions occurred, and, as a result, these records are
usually
unavailable at the time of consultation.
Patients labeled penicillin-allergic are more likely to be treated with more
expensive and
broad-spectrum antibiotics, a practice that leads to the development and
spread of
multiple drug-resistant bacteria and higher direct and indirect health care
costs. Among
patients. with a reported history of penicillin allergy, 80-90% have no
evidence of IgE
antibodies to penicillin on skin testing and thus avoid penicillin
unnecessarily. The
discrepancy between claimed and real penicillin allergies probably results
from several
factors. The reaction may have been predictable or due to the underlying
illness and
hence may have been mislabeled as allergic from the onset. Another contributor
to the
discrepancy is the tendency of patients with type 1 penicillin allergy to lose
penicillin-
specific IgE antibodies over time. Insight into the immunochemistry of
penicillin has
allowed for the development of validated skin-test reagents to detect
penicillin-specific
IgE.
-325-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Together with penicillin, cephalosporins are the antibiotics most widely used
for treating
common infections, and like penicillin, can cause immediate reactions.
Manifested
clinically by urticaria, angio-edema, rhinitis, brochospasm, and anaphylactic
shock, such
reactions are generally IgE-mediated and are among the most dangerous.
Although the
incidence of severe immediate reactions to cephalosporins does not seem to be
much
different from that to penicillin, studies of cephalosporins as allergens are
not nearly as
numerous or thorough as those on penicillin, and very few have been dedicated
to the still
little known determinants responsible for allergic reactions.
Unpredictable adverse reactions to aspirin and NSAIDS fall into several major
categories.
Respiratory reactions occur in patients with underlying asthma, non-allergic
rhinitis, and
nasal polyposis. The preferred term for this disorder is aspirin-exacerbated
respiratory
disease (AERD). The reactions typically involve the entire respiratory tract,
with
symptoms of rhinitis, conjunctivitis, and bronchospasm. Patients who have AERD
exhibit cross-reactivity with all non-steroidal anti-inflammatory drugs
(NSAIDS), but
they can tolerate cyclo-oxygenase 2 enzyme (COX-2) selective inhibitors. No in
vitro
tests to detect aspirin sensitivity exist, and oral challenge remains the gold
standard
diagnostic test for AERD.
True hypersensitivity reactions to local anesthetics are uncommon and usually
consist of
delayed contact dermatitis; anaphylaxis from local anesthetics occurs rarely
if ever. Most
adverse reactions are vasovagal, psychogenic, toxic, or predictable side
effects of
epinephrine that is often used in combination with local anesthetics. Large-
scale studies
have found that, following full evaluation, virtually all patients with a
history of allergy
to local anesthetics are able to tolerate these drugs. Unfortunately, patients
who
-326-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
experience any adverse reaction to local anesthetics are frequently labeled
allergic and
told to avoid all "-caines" in the future. Because evaluation of these
patients invariably
finds them able to receive a local anesthetic, such evaluation prevents them
from being
subjected to the increased risk of general anesthesia or, alternatively, to
pain from the
absence of anesthesia. Evaluation of patients with a supposed allergy to local
anesthetics
is also important because it serves to alleviate dentists' or physicians'
legal (malpractice-
related) concerns regarding use of a drug to which the patient is listed as
being allergic.
Allergic drug reactions compose a small percentage of adverse drug reactions,
yet they
are commonly encountered in clinical practice, and physicians are taught to
routinely
question patients about these reactions during history taking. Medical history
taking is
critical in the evaluation of antibiotic allergy and in distinguishing
allergic reactions from
other adverse reactions. This information is important, since over-diagnosis
of allergic
reactions can lead to unnecessary use of more costly antimicrobial agents and
may
promote the development of resistant microorganisms. Whenever possible,
patients who
are being evaluated for possible antibiotic allergy should be encouraged to
provide all
medical records related to previous adverse drug reactions.
Treatment. For drugs that are presumed to be mediated by IgE, drug
desensitization my
be performed if the implicated agent is required for treatment.
Desensitization involves
the administration of increasing amounts of the antibiotic slowly over a
period of hours
until a therapeutic dose is reached. The mechanism by which clinical tolerance
is
achieved is unclear, but it is thought to involve antigen-specific mast-cell
desensitization.
Since maintenance of a desensitized state requires the continuous presence of
the drug,
desensitization must be repeated if the antibiotic is required again later.
-327-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
For reactions that are not considered to be mediated by IgE, management
depends on the
clinical manifestations of the previous reaction. For macropapular eruptions,
the
specialist may consider a graded drug challenge, which is equivalent to
provocation
testing. Initial starting doses are generally higher than those used for
desensitization, and
the interval between doses varies, ranging from hours to days or weeks. The
decision
whether to discontinue an antibiotic if a reaction occurs depends on the
nature of the
reaction; bullous lesions or those involving mucous membranes warrant
withdrawal of
the drug, whereas it may be reasonable to treat through milder reactions, such
as
maculopapular eruptions, with the use of antihistamines, corticosteroids, or
both as
needed.
Cephalosporin in patients with penicillin allergy. Penicillins and
cephalosporins share a
P-lactam ring structure, making cross-reactivity a concern. Whereas most
patients who
have a history of penicillin allergy will tolerate cephalosporins,
indiscriminate
administration cannot be recommended, especially for patients who have had
life-
threatening reactions. For patients with a history of penicillin allergy who
require a
cephalosporin, treatment depends on whether the previous reaction was mediated
by IgE.
If testing is positive and a cephalosporin is considered necessary, then
desensitization
should be performed with the use of the particular cephalosporin chosen for
treatment.
Areas of Uncertainty. (Gruchalla and Pirmohamed, 2006) The mechanisms
underlying
antibiotic allergy have not been clearly elucidated. This understanding is
needed to
facilitate the development of better diagnostic tools and drugs then are less
immunogenic.
Better understanding is needed of factors mediating individual susceptibility
to allergic
reactions to antibiotics. Some patients have reported adverse reactions to
many
-328-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
chemically unrelated antibiotics. The existence of the so-called multiple drug
allergy
syndrome is controversial, and accepted diagnostic tests are needed to
document drug
allergy in these patients.
Recommendations. Patients who report a history of antibiotic allergy require a
careful
assessment of the nature of the reaction to determine if the likelihood that
it was
immunologically mediated. For patients whose history suggests and IgE-mediated
reaction to penicillin, skin testing is indicated. If the test results are
negative, the 13-
lactam agent may be administered. If the test results are positive or testing
cannot be
done, the drug should be avoided or a desensitization procedure should be
performed.
ImmunoScore and Drug Hypersensitivity. Exemplary ImmunoScore diagnostic and
ImmunoScore database platforms can be seen as examples of two-sided markets in
both
the diagnoses of drug hypersensitivity as well as in the retention of an
individual patient's
drug hypersensitivity testing and records for future health care medication
decisions as
shown in Fig. 36. In such cases, it would be predicted that the health
maintenance
organizations and pharmaceutical manufacturers (seen at the base of the
diagram,
propping up the platform structure) would belong to the side(s) of the market
most eager
to subsidize the other partners. Patients, physicians, and allergy specialists
are natural
partners to exemplary ImmunoScore diagnostic and database platforms regarding
drug
hypersensitivity. Because the diagnoses of drug hypersensitivity reactions are
in their
infancy from a scientific standpoint, research groups developing diagnostic
assays are
also natural customers for the two ImmunoScore platforms.
-329-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Initial patient histories should include a recording of all prescription and
non-prescription
drugs taken within the last month, including dates of administration and
dosage. This is a
real example of the proposed utility of the ImmunoScore database platform,
wherein
patient medications could be tracked and also easily transferable from primary
care
physicians to specialists.
For the HMOs and other insurers, drug hypersensitivity diagnoses and
cataloging by
ImmunoScore are a natural marriage. There are dual concerns in health care
regarding
the expense of exotic antibiotics and the development of antibiotic-resistant
strains of
organisms. Real patient information regarding drug hypersensitivity (as
opposed to
patient recall and limited health records) would certainly be welcomed by the
medical
and insurance professions.
-330-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Table 1 - Gell and Coombs Classification of Drug Hypersensitivity Reactions
(Riedl and Casillas, 2003)
Immune reaction Mechanism Clinical manifestations Timing of
reactions
Type I (IgE Drug-IgE complex Urticaria, angioedema, Minutes to hours
mediated) binding to mast cells bronchospasm, after drug
with release of pruritus, vomiting, exposure
histamine, diarrhea, anaphylaxis
inflammatory
mediators
Type II Specific IgG or IgM Hemolytic anemia, Variable
(cytotoxic) antibodies directed at neutropenia,
drug-hapten coated thrombocytopenia
cells
Type III (immune Tissue deposition of Serum sickness, fever, 1 to 3 weeks
complex) drug-antibody rash, arthralgias, after drug
complexes with lymphoadenopathy, exposure
complement urticaria,
activation and glomerulonephritis,
inflammation vasculitis
Type IV (delayed, MHC presentation of Allergic contact 2 to 7 days after
cell-mediated) drug molecules to T dermatitis, cutaneous drug
cells with cytokine maculopapular drug exposure
and inflammatory rash
mediator release
-331-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Table 2
Diagnostic Testing and Therapy for Drug Hypersensitivity (Solensky, 2005)
Immune reaction Laboratory tests Therapeutic considerations
Type I (IgE-mediated) Skin testing Discontinue drug
Radioallergosorbent test Consider epinephrine,
(RAST) antihistamines, systemic
Serum trypase corticosteroids,
bronchodilators
Inpatient monitoring, if
severe
Type II (cytotoxic) Direct or indirect Coombs' Discontinue drug
test Consider systemic
corticosteroids
Transfusion in severe cases
Type III (immune Erythrocyte sedimentation Discontinue drug
complex) rate (ESR) Consider NSAIDS,
Complement studies antihistamines, or systemic
Antinuclear antibody, coricosteroids; or
antihistone antibody plasmapheresis, if severe
Tissue biopsy for
immunofluorescence
studies
Type IV (delayed, cell- Patch testing Discontinue drug
mediated) Lymphocyte proliferation Consider topical
assay corticosteroids,
antihistamines, or systemic
corticosteroids, if severe
X. HEALTH CARE TRANSPARENCY AND COMPETITION
Currently, health care in the United States consumes $2 trillion per year. Out-
of-pocket
costs for those who have insurance have nearly tripled in the last six years,
and 46 million
Americans are uninsured. Unpaid and unpayable health care bills account for
the
majority of all personal bankruptcies in the country. Eight criteria for
improving health
care can be articulated as:
1. Consistent high quality
2. Lower cost - follows from high quality. Higher quality is often naturally
less
expensive. Providers improve quality by honing their organizational
-332-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
processes to become more efficient and effective - to avoid error and to do
things right the first time.
3. Available to all - for ethical, political, systemic, and business reasons,
health
care must be available to everyone.
4. Single model - every provider in the system must compete to offer the best
product at the best price.
5. Shaped by market forces - the consumer market has the sustained systemic
power to bring consumers more for less.
6. Practical - the solution must arise from present realities.
7. Progressive - dramatic change can not occur all at once.
8. Self-reinforcing - as any part of the health care system moves toward a new
reality, that movement must allow and encourage other parts to move
forward as well.
Competition thus far has failed to work the same wonders in health care that
it has in so
many other industries. In Redefining Health Care: Creating Value-Based
Competition on
Results, Michael Porter and Elizabeth Teisberg argue that this is because
competition has
taken place at the wrong level and over the wrong goals. Further exacerbating
the
problem is the complete absence of feedback loops. Very little in health care
has a real
price or a real measurable result. Competition in health care has consisted of
health
plans' and providers' attempts to push cost and risk of themselves and onto
each other or
onto employers - and now, onto the consumers. Consumers are not looking to
embrace
an institution, but are looking for a solution for a particular problem. One
can envision a
world in which health care is organized mainly around products tailored to
particular
medical conditions. Such products can be delivered by medically integrated
practice
units made up of teams that work together on the same medical condition over
long
periods of time. In this particular vision, transparency drives quality.
Health plans could
steer patients toward the providers who offer the best results for the least
money.
-333-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Referring physicians could refuse to recommend any specialist or package with
quality
scores in the lower quintiles, for fear of being sued for malpractice
themselves. When
health care providers compete at the level of the medical condition, on real
prices and real
results, feedback loops can become extremely compelling. Offering the highest
possible
quality at the lowest possible price will no longer be voluntary, and health
plans will also
be forced to compete on the basis of real results and genuine customer service
at the
lowest price, rather than at their current modus operandi - which can include
denying
coverage and shifting cost and risk to employers, consumers, and providers.
New structures for public reporting of medical results are popping up on
federal, state
and regional levels. In many of these initiatives, process measures are
starting to give
way to results measures. In a number of regions, new tiered payment models use
co-
payments and other means to encourage patients to use the providers with the
lowest cost
and highest quality scores. Such models also reward more efficient systems,
those that
beat their risk-adjusted cost targets, with higher reimbursements, and punish
those less
efficient providers with lower reimbursements.
The pieces - transparency, integrated products, and true measurement - are
coming into
play in the health care marketplace. Once it becomes common for health care
providers
to post actual prices and actual results in standardized ways that produce
comparable
data, it is hard to see how consumers, insurance companies, and referring
physicians
would ever choose low quality at high prices.
In exemplary embodiments of the present invention, ImmunoScore diagnostics and
database management can, for example,
-334-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= keep score not only of patient's immune data, but effectiveness of
treatments/vaccine
= tie records of physician recommendations relating immune status to fiscal
responsibility and patient well-being
= provide data for insurers
= provide data for providers
= provide data for consumers
Major decisions about health care in the U.S. have traditionally been made by
employers,
who determine for their employees which benefits and forms of coverage are
needed,
what types of providers are included in the network, and which organizations
administer
the benefits. But this paternalistic approach effectively allowed the consumer
to be a
passive participant in his or her own health care. The consumer to this point
has had no
economic incentive to seek the best care at the fairest price, or to give up
unhealthy
habits. It has been written (Knott, et al. 2007) that new health care formats
and
competitors are gaining traction, with MinuteClinics and RediClinics - low
cost walk-in
health care centers for common ailments at one end of the spectrum, and highly
personalized "concierge care" at the other. In addition, companies that are
not traditional
health care players are leveraging their capabilities to create entirely new
offerings that
enable and encourage the move toward health care consumerism. Fidelity, for
example,
is developing products and tools that exploit the emerging health-wealth
intersection,
such as a calculator that helps predict out-of-pocket health care costs.
Standardized data
on cost, service, and outcomes has the power to establish a new basis of
competition.
Payers are also pushing for new payment mechanisms, such as pay-for-
performance, that
base reimbursement on outcomes or adherence to broadly accepted clinical
guidelines,
known as evidence based medicine.
-335-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
To make competition and innovation among payers and suppliers possible, an
exemplary
system could include the following:
= consumers who live healthy lives and plan for their future health care
needs
= a fundamentally restructured supply side that provides consumers all the
information they need to make wise choices and is quick to respond to
changing consumer demands
= new kinds of intermediaries to help align the supply and demand sides and
help consumers navigate the complex system
Much of what is needed on the demand side is in place today or likely to
emerge in the
near term. Consumer-directed health plan (CDHP) enrollees offer an early
glimpse of
subtle changes in a retail market. CDHP enrollees are more likely to be aware
of price
and quality differences in products and services and more likely to have seen
information
and shop around; they are more likely to ask for prices up front, more likely
to negotiate
prices, and more willing to trade convenience for lower prices. They are also
more likely
to plan ahead when making health care decisions and to invest dollars now to
prevent
problems later.
The overall design of the ImmunoScore technology is one in which preventative
medicine takes the forefront in treatment options. Vaccine status is the most
obvious
application, and patients lacking protective antibody levels can be
vaccinated. Other
levels of immune preparedness would also be similarly assessed and
preventative
measures could be undertaken prior to clinical manifestations of autoimmune
disease,
cancer, or immunosenescence. Similarly, evaluations of physicians and health
plans
could be readily facilitated using the ImmunoScore database. .ImmunoScore can
be used
to discover and define fundamental relationships, such as, for example, (i)
optimal
-336-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
ThI/Th2 balances, or (ii) lack of any members for immunosenescense or
autoimmune
decease, that can serve as indicators of overall imune system harmony. Such
relationships can, for example, be quantified as one or more "ImmunoScores."
Patients
with healthy ImmunoScores would point to their primary care physicians and
their
insurers as providers of admirable health care practices. Prevention being
much more
cost-effective than treatment would provide the best of all worlds to the
patients,
physicians and insurers. Physicians whose patients had consistently lower
ImmunoScores would raise a cautionary flag and those doctors and practices
could be
scrutinized for provision of first rate health care (or something less). If
the records were
transparent, patients as consumers would use their dollars to pay for the best
possible
health care rather than pay for poor care at high cost. For example, in 2005,
Aetna began
testing tools that allow consumers to compare physicians on actual cost, so
that they can
gauge their out-of-pocket expenses. WellPoint has embarked on a pilot program
at the
suggestion of General Motors to provide complete comparative cost data for
hospitals on
"episodes of care." A number of employers are also finding success with
wellness
programs. Typical wellness programs feature free or low-cost health screenings
and
other sorts of preventive care.
It has been approved proposed that additional investments in health
information
technology and greater connectivity among providers will be needed to ease
sharing of
patient information and enable consumers to better manage their own health.
ImmunoScore database management can, for example, serve as an excellent tool
to
address connectivity issues that patients, physicians, and insurers would
have. Thus,
ImmunoScore technology proposes to be a new intermediary in health care
connectivity.
-337-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Patients would have more control over their own health care decisions -
spending as well
as courses of treatment.
Public health and data collection. In public health, the current underlying
assumption is
that good data will lead to better decisions, which will result in enhanced
population
health. In practice, no necessary linear sequence exists from good data to
better health
(AbouZahr, et al. 2007). Various types of data are obtained at different
levels of the
health system, to be used by several actors for many reasons. Providers
generate and use
information in the context of patients' care; managers need data to enhance
efficiency
and effectiveness; planners rely on statistics for operational decisions; and
policymakers
use information for prioritization and resource allocation (AbouZahr, et al.
2007). There
are different data sources currently used to formulate public policy - each
with
advantages and disadvantages:
= Routinely reported service data. Routine and administrative reports are
generated as a by-product of patient-provider interactions and health
facility functioning. Health facilities are a primary source of data for
notifiable diseases and are at the heart of a country's surveillance and
response programs, although facility case reporting needs to be
complemented by active case seeking strategies to generate a complete
picture of epidemic risk. No matter how many data elements are routinely
reported, information is inevitably biased by patterns of service use and
non-use, and the extent or direction of bias is impossible to ascertain
without recourse to other sources of data. Services delivered (number of
-338-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
immunizations, antenatal visits, outpatients seen, etc) do not necessarily
equate to population need.
= Population based data. Mistrust of service-based statistics has fuelled
interest in household surveys that can generate unbiased data for
populations as a whole rather than just the sections that use available
health services. These household surveys have several disadvantages.
They need large investments in human and financial resources and
therefore are usually funded externally, resulting in bias towards the
interests of donors or well-sponsored programs. They are also time-
consuming and are undertaken only occasionally, and generate results
spanning a period, rather than the immediate past. Samples are rarely of
sufficient size to deliver nationally valid results. Growth in surveys to
generate health related data has been fuelled by their ability to deliver
statistics on child mortality, population coverage, and certain risk factors.
In the past few years, scope for measurement of health status with
household surveys has greatly expanded owing to cheap and reliable
diagnostic tests that can be used in the research setting to generate
population-based estimates of disease prevalence. But surveys are not as
effective for measurement of adult mortality, which is a relatively rare
event compared with child mortality.
ImmunoScore would relieve much of the concern regarding public health and data
collection. As stated above, there has been concern regarding bias in the
routinely
reported service data. As ImmunoScore grows in size and popular usage,
concerns about
-339-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
bias can be alleviated. Services delivered to any individual patient would be
based solely
upon the needs of that particular patient, and tailored to that individual
patient's
immunologic needs, with no regard for social stratum. Political justifications
for mis-
representation of data would be eliminated by the automatic and mechanical
nature of the
data acquisition.
The tremendous requirement for human and financial resources for collecting
population
based data would also be alleviated by ImmunoScore technology. Data can be
collected
at hundreds of remote locations and transferred back to an exemplary
ImmunoScore
central database. There would be no need for third party human resources -
data
collection would occur at the hospital, clinic, or physician's office and
stored for future
use.
ImmunoScore Tracking of Medical Services (ImmunoScoreKeeping)
As has been described, heath care processes are very complex, involving both
clinical
and administrative tasks, large volumes of data, and a large number of
patients and
personnel. Health care processes are also very dynamic. As new processes are
initiated,
changes in health care treatments, drugs, and protocols can invalidate current
methodologies, requiring reparative actions. ImmunoScoreKeeping can, for
example,
capture all of such complicated dynamic components and provide accurate
performance
measurements ("ImmunoScoreCards") not only for individual patients, but also
as a
means of tracking relative efficiencies of other complex components of the
health care
system.
-340-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
For example, upon a visit to a provider using an ImmunoScore system, the
patient's data
can be captured by an "ImmunoScoreKeeper." (an exemplary POC assay reader
connected to a system database, as described above). Not only demographic and
test
data, but also treatments/drugs prescribed, physician's ID number, and insurer
can be
stored. Any additional testing or measurements (blood chemistry, X-rays,
physical
therapy, etc.) can be entered into the remote ImmunoScore data collection
system at, for
example, the physician's office. A critical requirement for efficient
management of
health care is the management of the quality of service. Appropriate control
of quality of
service leads to the creation of quality care services; these, in turn, can
fulfill patient
satisfaction.
Traditionally, health care services have been managed using limited forms of
workflow.
Some examples of these are clinical and administrative protocols. However,
these
protocols have remained limited in their usefulness in part because developers
have rarely
incorporated both clinical and administrative activities into one
comprehensive care
protocol. This lack of integration hinders the delivery of care, as the
effectiveness of
protocols is often dependent on many administrative tasks being properly
executed at the
correct time.
Thus, in exemplary embodiments of the present invention, ImmunoScoreKeeping
can
enable medical practices to provide better quality care at reduced costs.
ImmunoScore
can, for example, maintain keep ImmunoScoreCards on:
= individual patients
= physicians
-341-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
= groups of physicians/managed care organizations
= insurers
In addition, as vaccines, drugs, and therapies prescribed can all be monitored
and tracked,
an ImmunoScoreKeeper can also monitor the efficacy of the vaccines, drugs, and
treatments prescribed. As these data are compiled, they can be shared and
submitted to
appropriate oversight organizations (FDA, CDC, ACIP, Physician's
organizations, drug
and vaccine manufacturers, etc.) to better enable these groups to make clear
decisions
and/or recommendations. Such organizations would be consumers of ImmunoScore
data.
Thus, an ImmunoScoreKeeper can allow insurers to rate physicians and enable
their
customers (the patients) to make better informed decisions regarding their
choice of
physician. An ImmunoScoreKeeper can track effectiveness of treatments to
patient
outcomes. Prescription drug and vaccine efficacies can be monitored not only
in
population-based samplings, but longitudinally in individual patients with
repeated
ImmunoScore diagnostic testing protocols. Physicians can thus monitor the
efficiency of
the practice that they are associated with, and thereby make the best career
choices to
advance their careers in the most efficient practices. Hospitals could be
measured for
effectiveness in patient care against other hospitals and groups of
physicians. Types of
hospital settings could be evaluated prospectively. Causes of nosocomial
infections
might be tracked, for example, to certain types of hospital environments.
Insurers can be
measured against common metrics and be forced to compete for business via
accurate
ImmunoScoreKeeping.
-342-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
In exemplary embodiments of the present invention, an ImmunoScoreKeeper can,
for
example, provide a means of integrated monitoring of individual patient
treatment and
also administration of both physicians and insurers practices. The
ImmunoScoreKeeper
can, for example, generate a numerical value for each component of the health
care
system upon which real competition among providers and insurers could be
generated.
This competition would thereby provide substantial increases in health care
quality and
decreases in health care costs.
Y. USER ACCESS VIA DATA NETWORKS AND ON-LINE ADVERTISING
In exemplary embodiments of the present invention, users can, for example,
access an
ImmunoScore Database via computer data networks. Such networks can include,
for
example, VPN's or the Internet. In exemplary embodiments of the present
invention, an
ImmunoScore database can be accessed, updated and queried via one or more web
page
portals. With a substantial ImmunoScore subscriber base of clinicians, health
care
management professionals, individuals, health insurance managers and
executives, and
pharmnaceutical company researchers and management personnel, a given
ImmunoScore
embodiment can serve as an indispensable portal for anyone involved in the
health care,
health insurance, life sciences and related industries. This creates an
opportunity for
targeted on-line advertising.
Online advertising is growing quickly. Recent research predicts that global
spending on
Internet ads will overtake radio in 2008. It has also been predicted that the
rate of
spending on Internet ads will grow six times faster than that for traditional
media between
2006 and 2009, a trend already taking shape in the Middle East and Europe.
Even
companies that do not engage in e-commerce, such as, for example, Unilever,
nonetheless
-343-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
want to create better user experiences for people online. They want to improve
their
brand presence.
Many other companies have started to use Web 2.0 technologies, such as, for
example,
blogging and video clips, to increase brand awareness. Last year (2006) in
China, Pepsi
invited people to write screenplays for company spokesmen and a famous pop
singer.
And when it launched its Qashqai car in the UK last month, Nissan offered a
game
website where people could try to shoot the car; it broadcast video clips of
the car, which
could be linked to blogs and social networking sites; and it ran banners over
some Yahoo
sites.
This summer, two companies, Joost and BabelGum, will start to broadcast entire
TV
programs free over the Internet. The content owners in effect have their own
channels
and advertising will pay the way. In effect, "people will watch Friends on a
website. We
may thus see the death of the TV station and the birth of the network station.
Given this state of affairs, in exemplary embodiments of the present invention
an
exemplary system according to the present invention can be used as an Internet
portal for
everyone associated with health care in the broadest system (thus encompassing
any
consumers or providers of any of the business models described above). In the
same
sense that individuals utilize online search engines, such as, for example,
Google or
answer.com, to research a topic, anyone remotely connected to health care,
can, for
example, access an exemplary ImmunoScore webportal. Whether the individual is
simply an individual who has records in an ImmunoScore database, whether the
individual is an executive of a health insurance company or an HMO, whether
the
-344-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
individual in question is a physician or hospital administrator, or whether
the individual is
a researcher or someone who works in sales or the technical side of
pharmeaceutical
developments, an exemplary ImmunoScore system and an associated ImmunoScore
web
interface can become a ubiquitous tool used each and every day by millions of
people.
Targeted online advertising can then be used to deliver business to business,
or business
to individual in the case of individual to consumers, advertising to a market
which is
already attuned to the benefits of technology which is applied to healthcare
and
understands the value of preventive medicine, individualized medicine, and a
granular
approach to health care provision and analysis and follow-up as to efficiency
and efficacy
of health care.
For example, individuals whose ImmunoScore record, after the appropriate
analysis, has
disclosed a potential likelihood of having an autoimmune disease can be
provided
literature, products, news of new drugs, experimental clinical trials, etc.,
for their review
and potential participation and/or purchase. Similarly, health insurance
companies
offering a healthcare credit exchange program, as described above, can also
advertise
such programs and enhancements to such programs to a target audience of
sophisticated
health care consumers. An interested third party operator of an exemplary
system
according to the present invention could even, for example, offer to create
and operate a
health care exchange program to all health care insurerance companies using
ImmunoScore! All of these examples list just a few of the possibilities. Thus,
the more
data that an exemplary ImmunoScore database obtains and learns how to best
process to
extract all of the information latent therein, the greater impact Immunoscore
technology
can have and the greater draw an ImmunoScore based web portal can provide.
-345-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Z. PROPHYLATIC THERAPIES DURING SURGERY
Before a surgical procedure, the patient's ImmunoScore can be used to aid in
the
execution of the surgery. For example, if the patient's ImmunoScore indicates
that the
patient's immune system is in good balance, fewer or less powerful
prophylactic
antibiotics may be prescribed and/or administered. Benefits include reducing
cotss, and
maintaining the patient's natural gut flora (which for example, may avoid
Clostridium
difficile infections). Conversely, if the patient's immune system is weak,
additional or
more powerful antiboitcs may be help prevent a surgically-induced infection.
As another
example, if the patient's ImmunoScore indicates substantial allergies,
additional testing
may be useful to determine if the patient is allergic to certain surgical
materials (e.g.,
latex) so that alternates can be used.
AA. CONTRAINDICATIONS FOR BIOLOGICAL ACTIVE THERAPUTICS
Biologically active therapeutics may present new issues for patients. These
therapeutics
include priobiotcs. These therapeutics include viruses that have been
engineered to
selectively attack only some cells in the body (e.g., cancerous cells). When
administering
these biologically active therapeutics, there is some chance of unexpected
consequences.
By examining an individual's ImmunoScore, unexpected consequencs may be
avoided by
contradindicating the biologically active therapeutic.
The present invention has been described in connection with exemplary
embodiments
and implementations, as examples only. It will be understood by those having
ordinary
skill in the pertinent art that modifications to any of the embodiments may be
easily made
without materially departing from the scope and spirit of the present
invention which is
defined by the appended claims. Such modifications can include, for example,
using
-346-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
other appropriate assays or tests, other rules or analyses of the results
thereof, as may be
known in the art to assess the immune status of individuals or populations.
Additionally,
such modifications can include, for example, using various assay devices and
techniques
as may be known, using various available methods of storing and analyzing data
(including various "data mining" techniques) as may be available, and defining
various
alternative demographic groups and various sets of ImmunoScore test panels to
be
administered thereto.
-347-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
APPENDIX A
Exemplary CIP Database
1. List of CIP Database Variables/Fields and Additonal Notes
2. First 330 Records From Exemplary CIP Database
3. Copies of spreadsheets used along with CIP database as
inputs to data mining protocols:
Conditions.xls,
Rules.xls,
Rules_keywords.xls, and
Rule Defininition.xls.
-348-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
1. List of CIP Database Variables/Fields
-349-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
()
L
O
I
co
0
0)
r-
0
U
N
E <
c0 W
0
c I
0 0 O n O
N 'V'
d c
E
N M c Y C O
} N J M
II a) > N N c
o d c"
_
a) (n
Z c f ) .9 R a) O} () N c C c c 0
C) 0 z -n U) 0 3 3 c 0 Y
o
z n n m
z
c
a) N II c) p } Y c c D c
M }; ~= Z 0 11 D
c c II c
c: cli
C O N U a)
Q c c0 1 Z 3 'N Q d!diItI i
c0 11 N C N 11 C 7 V ~ I 11
Z ~-- p Q cu
C '~ U =~ II }
cu 0 0 .o U "" c0 t0 N N Z E Z O O O O Z c0 0
E O C 0 U c0 U c 5 0- 11 ,8 II Z Z X Z C)
O C N r co U 70 c o D N C a C) 11
c 2 C) O O N J J - CL
11 a) c0 U) 0 a) E a) E C' (n c0 C XO U)
C C U 0> >` a) -0 >+ C O U) = O Q M O O
CU aT M. C u) -a Y C 'c7 a) a) J 'C c .~ O 0- a) N N N C C co O E a (0 U a) w
7 Q Y Y
m m a) o 0 0 ~ 2 a W > W Y (D cu E 75 U ~ a)
M c c c
E n >, c o 0 0 'o E o co o cco o o g o o V o o p 0 0
o o n a`> c c .o E aci T o >. o o a) o o o i>l o o o 0 o 0 o o m Ca
X a) a) C 3 7 O) .0 E N > >' > >' a) a) a) a) () a) 'U U
(0 a) c0 O) 0 0 0 a) CO 7 ' c0 0 c0 0 a) 2 a T 0) N 0) =D N 0) Mn o . N 0) 0-
0 (n O Q a U U W O Z U 2 2 0 Q = 2 Q 2 Q 2 Q 2 Q = Q O O
O U C C L t M
N 0 a3 N co ) OD N M co cam- .0 (N m -0 N CO . ~I r1 U) .0 (C) .0 4)
> r v LO U) U N- .L. O N N U) V- U) c0 U) M U) LO U7 u7 U) co N
> N > > N [t ( LO N N N U') N N U) N U)
> > > > > > > o > > >> > > > > > >
E
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
C
0
C
C
(0
N
a)
C U
11 -0
I.f) 3
0
CO
C
C 0
L
U w
a) Co
v n-
Z`
~.ca
C v a)
0 0 M
c
M 0
n
N 75 '-'
N
O M
N C
C 0 E > > }
Z ` a)
11 n o CL '0
Y - 3 C.= N a)
11 U CO Co U L
N N D Z C) a) 0 `- U)) d
to to II C Q II L - II 'C7 3
2) cn
cu 0 C N _ 3 a) p
1 cl)
C:) E L M r O C C .O .Q F- I OO 0 'C 1
E =3
Q p O >- U C U .L - C O N -
O N M LO f- M C3) C Z Z > 0 c~u p 3 0 = o U
N In CO C? -1 r r r r N Z LL Ll 11 II W O d '0 W 0 =C O N C
i i i i i 1 Z Z O O O C W C FL O a)
J - .> U
-J J J J J J J J LL
Cl) .N p C 0 0 Z= Z Q O D W
0 0 0 0 0 0 0 0 0 0 0 0 0 0
N U) II O Q Z U) z 11 0 .0 C C C C C C C C C C co C C C= p.
=3 0 0 > O N 'p 0 j E ~ o
OO w 00000O000O00 0= 0a 0Q 0 a- C O cH0 0 T
c~ v c~ v v v v v v c~ v c~ 0 0 0 0 0 0 0 0 C Y O` a~i 0 (1) (1) o
w.. '++ : + 'rte i.+ O N O N a) >' j L O Q Q a) a) =~ U
Q Q Q Q Q Q Q Q Q Q Q Q Q Q CT .- C7) 0) 0 0 3 O >, 3 T >> C a)
00000000000000=<X<M<oJZ~ 61-01- << Q. W
U
o E c :u c t _ _
cu U)
N LI? (9 00 ~ m r-
CL r- r- ~o m
N E .9 m 'a 000 a CO (Y) M CU cu U LO Ica Q' C7
J _ J J J _ J J J cu (a to U') U') co V) M qr I-
LL Z N N to N Ul > M Q CO M M M> C'7 M W
J lJ ~ ~ > ' U J
W
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
a)
> a) >
m
O O 0
to D_ 0_ Il
0 N II N
11 N
Cl 11 O 1111
O A
A N N U U
O N 0 O 0
N
> 0 = j C
Q O a) a) aa)) E
a) II II II E
II A N N II
N C) O O LO
> O U V II v n
V
a3 'O 0
C C Cu Q)
N O
C ~
0 N C) O
p O n o
N
A 0 > C14
11
11 A v7 dj A a) II > v~
V > V N> ) - Cm C)
C O
C N N O O C a)
a) Cu C O C C
C O) II O II > N O O C 0 0
II O O > > >
Cu > Cu CU
V V_ V o U C a) >' C N N O O
C:) > C) C C C C >
O O C) N C) O> O O
> C C) C O C > +:' to N p w O O O 0) }. N
r C' L C C L 'C O) C C) 7 O N p p 0 > D
V 0) 0 0 0 C` O CO C Cr U) C p 0 o U O 0 O U D_ O
}O Cl) m N N 0 O 7 U 0 U> C 0.> (D (1) CD
> > CL ...., Q)
0 O ui O U O N N N vi (0 C C c 'U L > D OC M O cVa N O N C v0,
o coN ODpv o ca (1) O m a)t Q ao a) vo mQ OQ
C D V N
o c ._ y p D CU M H s a p ZZ 4--
w >-
c 0 a C r- < < < (n
a) U) C w (n ()
C
p O C 0 C in C to C O C 0 0 0 0 w, . o a a [n C O C N 0) > 0 E
N 0) N 7 0) C O - w., C` `~ Jo N C m O N m O O m p U a
a) 0)
a v p p N p p
o =E o f (n 4) a N N= C= C= =
=3 CD p 3 >.
a)
a) :. N p M ID D c0 p 'O O N O 't 4! O O O 0 N
O a v Q (D CL .N N cn fn w
C aa) C m d
O O (p D N C m C m m 0 O O O O
as OOHOwO D-'=' CQ C a) C a) C.~? mw-mmm Z -m UU
fn fn fn fn N V) N fn O ~ cv a) m ca p O 0 O 0. -Cu Cl) m Cl) w m V) m Y co V)
V) q)
~. N N N to cn M M - - O () () a) a) a) z ;_
(0 f0 fn N !n to N N U U U a) a) 0` >
` Q- m C- N m Q- m CL m Q- N a m ur
D_ d (0 f0 (0 m m (d .2 .2 'C D :. N ~+ 0) a) D. a) D_ D_ Q) D_ N D_ 0) D_ N
D_ D_
m Cu :3 CL CL
!n
N C n `L
z CO . `- _ > `- _ C wp C t z Z Z Q Z a Z; <
CL o 0 -`aC7 o~'~v > Zaa (D a)>Za~.
m> t>3 n. O N C N Q a m a Co a U> >
i a a W CU a) N~~ E D> N> ao) 2> m 0 CO M CD m a): = >
E E E E > W opt U2=====___ __
CZQ 0
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
N .O
gyp. U to to 0 p Ln
O by as to O > O 0
c~ 1104
4-4
co o o :5
0 40 j O s cep tl.
_0 0 > 4=r O N CO C~
0 c/)
73
N cad U ~~ Q U
m V)
Cd 4"
CIS U
O 4-r Cd
V) C4-4 M qi
ti U Cd 0 Q. E U U O
C 'C
fr y C). c .i: -0 c d d En X x R.
y 4] .. a) O ~ Cd .d .. bA `) 4-4 X
V) V)
V) :Z
Q O 0 Cd 0.) 'o 0 .~ G
V i'y ~, O p cd y yCWd ai ' v O
~. > > > U . O s En 0
by x+ O O '~ U U > En
-c,3 > CC
, V) ca
U cq3 cd Cd 4; U U sU. GQ
Ow c d 'n
u C41
Cd
>, N
(4=4 0 O U In
p O U 4 U m d O U C
cd >' - U N p N bA 0 rA YO 4 M u
y U, t cd CU =O U m Cd
~" + v_~i O 4] p= bhp U x ~" G O h 0
Ln 7:1
xi p U Cd O p d) 3 vUi C d E
=C~ "O O m O O YO -U-. = CU O 0 0 p v>'i
E 03
ow "0
o o q .n U cd x -~ cd o
' cu Q. bA () a) 4-4 U p x x
r. cd ::3
0 '0 o cn V/ LL U d t r. bA o
> .5 r, U
Y >. ¾ d N C to d
CJ m ::s Lr
o o O U
c c c x_ O U y ~' 4~ y W y O
a c .2 o o Ca ~, U 0 X O U s. x p>
o b M CYO M O M YO 4-4
'b N U
RS (O c0 N N 4(4-4 5 , m O
U U U O. Q Q h +, (4-4 En c's
O 'd
d a Q 0 L C U Q~~ cC N Cd ~C].
- ow C,3 C-) C
O O O c c c rn > `~ p cd O p y> X Ca?
0 rn cc
.u N is cd vii =U O
c c Cn c m cd
O cu 1E U U G> 'w rte'" '5 C's 0 4 U ~ w o
O Cl) cd Cn sU+ V) M 0. cd
) ¾ o ca M
I c c_ o v V) V)
o v v t7 ~~ G~ a) O.
U_ LL L` Z x o CA
x. r~r a, x d U a x .0 b
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
j a~ O
>, 4=~ 0 a~>i ~GQ -O -0 U
O ~, o cd a j ~, cd cd C
P.4 np
cd ' a csd
c~ U O O M c
N
C~s
C'S V) cd
to P.
V) O O .-. cd U LZ a)
4 4-4 ' U y s~ C
404 $" >1 4- 13
O
o > o .!Z c;3 > _
4-4 Q. cu V
cd V) '6
4-4 CIS 1.0 0
'3 > 4- 0 V) 75
0 >
V) En $-4
N 4-+ ~' cn o > N w ' .b cn U O
O bA N o cl. cd O o o > JR. 3 ~, a~ x Q N o a~ 0. cz
on O U
V) U
OU 111) -0 0
(U z -0 O
Al o a~ o 0 3 0 o U 3 a.
Q= . M
cn C's .- 4-4 >=.o o V ~ `er
4~ 4U4
Q) u
as 0 b oo 0 U 4-4 0
O
t4-4 >i 4-
`~ 0 -Z: 3 Q 0' x o a~ d
c;j
Q c, CC ~-+ N O N N !~, N cd N
.fl ~' O. o ;?
N CA o N v, cd ^ "C7 bA O M O"
C, co Q) 0 0
C,3 U
0 4- 0 0 o
m 'd 4- 40.
cd (D -0 b
0 O O O p 4y y o +~-' 0 3 4 y ti a.
O
O 'd > CA
N 'C N s-
to 0
cad b~A ~ rcd o ^ a~ a~ -0 0 0 .0 vim, = au
cd+ 6 .~ o - cC 0
: d . r. 4Gy " r vii
co 0.
U8 >
r_i CIS Q -.=
o 0 0 y N x 3 == x OD '.- vi k
v) o Q. ti. x v~ N o >> vi Q `~ b
cd U i 0 0 Q 0 '.~ to
03 M
U (D GQ O y O> a ca N O cn L G c
co 0 4., cCd 'c~ > y Q 'O
ca 0 . y, cd w ~+=0-d () c~ `n O H O O r .--~ i.. -~ V
O cn G~ rn .. 4 , cd ^C3
Cfj O Q" O ~. .ter c~ cN p
ell U
A cC 'b y
O N vii o U U y c ` ~ V V A I
r. d cd Cl) 0 b b as a~
o 0 ai w id A O a~ Q=
0 >Qx > x ao v~ x ~ 0> U o~ 0 H
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
0 0 'b
0
>, c O
cd 0 0 O O _ 0 0 O' U (IS
J~ :z
cd ~+ 0 O
p O
0 0 0
Q bq
k U 0 u 00 0) Q i> c 0 U b4
0 V) > cn cd 'b ~ cC~~ = ~ L7 p ~ 7-.
y bA 0 C 0 0
Ed
Q) +r
co, Z p vi
1 b0A
cd O En 4-4
- y 'C >, p U> 0 U 40 0 t1+ `~ 0 0
^O _ 0 0 0 >,
4 s O" O 0 p 0 cd > 0 'D 0
+5 r.
0 0 O
U Sy +r +' U CS +^0, 0 U
w cad o o o S .
0 0 0~ 00 4 o co
a..U o o ~:~tnCZ a) -o
o - a.
o ¾ +,'
- Co ? v; U o 0 0
0
ncs
0 03 to Q)
0
bn y o s0 +' 0 cd 0- F-' N
s,' 0 4-.. 0 cod 0 0 C
0 }, 0 0 cd n 4- ... bb to N v7 cqj V) C44 Q 0
4~ =~ p' 0 p O 4 -' m
0 cd
0 .. > cd w 0 0 4 V O O
64 40. 0
z
04 +- 0
En - 4. cj
cd O En o a? to ~' 3 3 0 -e:- o C5 u
4) 0 C4;; En u W)
x +~ a
V) V) o 120
0 (u r.
Q. 0
-0 0 0 404
o i o b
> cd n .d U 0 +0+ 'O
Q a cC s. 0 4~ 0 y ~. O 7a U,
0 0
0 0
> ~. 0 O> 0 0
[~ s0, R 0
> .., U > O 0 r 0 0 tom, Q U V
0 0 v, N 0 4 a 0 0 .y 0
CA C4 0 En
o I.. '
O
ice! V2 0 =.. p r0+ rn 0 '~ y
p yV sue. 8 `" F-~ C o-0
0~ Q Q U,
= i. i. v,
Fr G~ 0 O p
oo v, II-li
0 cd 4- (1) cd r- (m
Q cC cC v1 4 v "o v0i r. z b W '~ 55 E c6 UJ~ 4 rn -C '"U a - - -
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
ai
Z
N ¾' Q, E O
cd 0 X
4~ N - cd
~ h U
4 O
Cl)
,
U Ll, +~ 0 O 0 0
H U U U 0 q
=- M N N N H O cd
44
O O S. N L ~." C O -
H H O E o
H a
0-4
O b~bAA O O ~+-~ O - ; ~' U H O.
cad . . c M
CC cd C cd bb O r- 7~
bA to U 0 bi
cn rn 7)
cn N t3 _ _ cd G) H cd co O +' cd z
----=----, =--En =~-i ~ fi
a a a a 4 O 404
4 00
O z3 re)
M vl r- L3 C2 ~
a 0-4 0-4
a a X ,_a a a a ~. ~,
~. ~, t-. ~, I. ~- N r r r-+ -., ~+ ~-- 40 401
40, 40 400 40 c0 0
40,
O O
~ d d 4 4-r 404 O O
I Gam) a)
.0 N U U 4) N .~ .O
,--
.b
O O O
O 0 0 O ~n O
4.1 N
r., 0 . 0 0 O N 0
cn V) V) V,
Q N ~~+ U U
U 4) U U N ~" N G." C C v~ C
f3 Q. P. CZ. f1. c U Q. G) cad cad S _~ N
04
d d d d d dH d; d d d d -d
c c
iivi 0
Cl) V1 r~/1 Cd O y 'jS Cd C~ C~ Cd y O
h o
C~ C~ cC C~ CC to , ~ CC CA v) vl) h Cl CC
M 1f) [~ Z3 =~ OZ Cd V M Cl C9 'N C~ C1
.-~ .--i ~--~ U N N ~p O 00 W W O
a a a a o a V a a 0 a a a a z z
0-4 V) 0-0
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
r
Q
't3
O
U
bA
i-.
4)
c
bA ~
rhc'
co?
U ~
~ H
to
}., N
b En
0
C
Q. U
to
Q H
O
rte. ~
Gz~ ~
(~ LY
H 0)
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
0 N N 0 N N y OD a)
txO N a) N OD Q1 y p O
N y- L N Y 4- 0
N N C i `~- 'j N C v-= '3 L +' L Vhf O O Q
m> +O+ 'a 0 a' u> Y 'C o - m C F- 0(U
O
m (C C a) N a 'a on
N 7 4- cu C U' C 0 00 C vii C o p '- = 3 'j > O a) 0 O
C E Q N N '_' c 0 ) N '+.., != 'a O w a 0 0
U 'Y m e u
O N 0 +- Y p u N o Y ~+ a) u Fv '~ u H m O c U >
.N Y C= f6 'U N :+ C. m U N t~if Q) L a) > -a
a 0 v 'a v- 0 u v "O m c m v c y > 'a Q U c
E E 0 c a, c aui 0 a) c v u c 3 0 ,n o
-0 0 E o o v 0 (Es o o 0 > M > U C 3 c
N m
Co X 4 1 V N L Q ' X Z: u a) L O. m O N w O -a in m a) C2.
1 4J m a1 i+ o f N m a1 : 0 E =` L O 3 -a i m m C_ L (i,
'tn U O C U C. i Ou O C U C. i Ou == a) 'O C a m a) a L
t+'if co Q) H W , C U41
' m i) N a) `~ C L Q 3 .u =y a+ >
0 a) X c 0 o E v v X o v E v C. 0 u a) c a) 0
O > Q) C- E > c a) m Q' E a) Y > Y U C m U N N - cu
o v u, v - E n m N v E .- 0 a ,Nn 3 0 :6 C
aci v>
c 3 L > c 3 a S C N L ro
C. a) Y ,~ Y ^( +i a) Y Y 0 +m+ 0 w Y a)
C Y C - > U d
C C U i=+ a) C
+~ C u N u L L O C u =N U L 0 O C v U ((0 T O (u a,j C
(Y E C .N O Y cu a) v OL O y p a) w ON' C C v (U a) U +L-~ m U d
0 m y 'a > +L=' L >- aai a> +L- L> m E +' Y >- 0 > (ca W
N x L C. o u N 3 m C. 0 V m CU 0-c41 O a) E Yn
+O O a) 1> a) a O a) L> v o E O u C c u o E CL 'A
C C
N (>a w C d N ((0 Om0 C G) V O O E >>i C U 7 O 1 O C-
a te a N
(YO p +p~ L ,~ - .5 a) 00 41 L O` C . a OD C CO N m 0 $ o >~ c C
W +~ O c m LA O tYi( c u L Y L m C +~ 41 m
N Y N O aJ Y N O aJ d4 a1 N a) (O c a) N E C L U N
C C (n 3 N L O > E N N p => E u 3 m > p E aL.. ~ C N
() L N m a) N CU i o m (~ aJ
N V N C ` -O N N H i N -
> a) L U y =C ' a~ C.. 41 N O E Y> ~ o c o a) 00 > O o c o O s C o a 0> >p
C Q m o
u CL
a) N 'C C Q O a-- N V) = C Q. V 1(=- a) N Y m E > N C. N O +-' o
w W OD a) -Fu
0 4-1
M w > c f6 u H E 0 Q U C c an
(U 72 a) y E a c .u
X m > c ac) E
a) x o :~ c> X -r c m 0 o '5 c_ X m t co 3 . 3 L '~ a E C = ND 3 c
00 "0 u u 0 +_+ a) 0p u u 0' a) m 3 al a) o t! >_ >, p O 41 , _a V) i C N C OD
a 1 C ~ , OD (1) C N C OL -0 c 'a . O a) C a)
3 O N O N C O N (C 0 C O C 3 C m N > Y L
~^ m O
0 C) 0 a~N- 0 O 00 (gyp O E E 0 =V' Y
a) 4 a -w :3 (U v a m N E o ~ 0 0 ~ - > o p c (u o L- to w o =a CU E Y a,
C L 'u ,C O OA u ,[ 0 _ 00 m C m LL C5 >
w 4-1
'~ Y U N -C Y L N Q) U y( -a (9 Y i o U d 0 E O (u q- N =L z E
0 m m C ro a c ^ c -a c C. ra a c .'_^ c -a c m m` >- c u +, a v>
H N N L N > L O N N L 'N >. w O > > a) v +m-~ o a C c V
E C L 0 L Y a) cl. Z O t +-' T tea s m r- aL=+ > C r- L O m (^ '~
41 CU -0
m i 0 Y a) ~+ L U Y 0 Y a= i U Y 0 a) 0 0 Y 3 E
X a) m N E 3 v a) 0 c n a E v d o c > c N `) v- v Y> 75 O u c c a) o E> u -C
Q a) E> u C F C v a 3(0 0 a o r- o E c u a u c
Q s y ' EF= o E o o N EF- 40' E E a`r '> ' E t _
a E E Y v a) a) > c a) p Y
m N U ~ C 0 a) O a1 O a1 a) Q a) O > N u N m c c 0 Y U () N
4T a) +' > Y N L - V '+O > Y N L E u 0 0 cu > p 'u l7 C c N C vi
E c L v m O c a) U .E a) m 0 c a) L) a N 0 m O
u v a) a) +'
_ O a) O
C t C ' i L a) 00 ~' (0 L Y "O O m t' L C
ON E a) = f i Y >, `~ y to Y j, ` - a) OD 3 a) O = j +L=+ *' > }' U Y L C N
+r
E C i _ a. N 0
v y a m L 0 L O CU -0 L c 3 O m +-J c u a =c m
LA Ln o O L X o m 'a a) (a N X\ a a) m vi > c c +, +-+ 3 = O Y C.
v OD N a " N >
t.0 3
-0 w 0 aY) Q0-0 3 3 0 c u 0
U N N w
u t.0 -0 N s U CO
O N 0 4 U 4-1 (U
~ a C 0 - Q Q C C. - a) L N 41 v .C CU
.`_ v ' C O m w 3 0 0 m d 3 to N c 0 > 0 a Q m v
ON - Q E {U// 0 > N > v +L-. vui C) > 0 > v +L==~ C 'a U U Y C E c
L o E c c o E *' c c o 0
m - > a a~ v 01 0 a (Eo c C
co 0 a n o c v o n t c v-> (-tea o 5 E aai . c 0 E
Q) z CU Ln a) Y CU Y IA -0
m 0 4- u O_
(0 r- M CO O vi OD a) c 4- w aa)) Q N N OD W (U a v 3 >' C > L U O_ -0 T '^ U
_O
~ N cu
+, O u +' v C r9 a) a) c +~ v E m a) : O +' y O c: 'A 0 3 0 >
-0 M -.4 tw C n M E v v =~ m m v v v Y ) v r a > Q) a 3' o a)
E c -a :a E -0 -0 N cr -O cu 0 p E -0 a) vmi O 0= p 0 ~mif 0 0' w C p ~ m-0
c m u Q
v o 0
Y x
O O a) "C >' U a) "a >- w }, m a) m c -o 3 =o~ (+= L c
+' ((0 (0 E 3
N U i O vO- 3 N C N O ~ N c to m OD L N 0 +L=+ N H L.., t u j. u (0 3 a)
L m r O m O O y _ c +B E L
3 N? > C in L ++ d N C L+ F a - a GJ N pa N> N N C Q (a - O` 0 m
N L +L-~ vvi "a C U V41 ) 7 y o 'a C U OD C +L00 m v 3 O E L .>
` 'a '3 c w=' 3 m 41 O 'On c Y O (^ m W O 4 j 0 j L c E c .0 ` L2 U O 4-1 C +%
O - ~ +.. O u =E of a) N O O U in a)
L- 0 N Q 0 c Q p L_ u Q. O m
O C "~" C C `' U C " >. m u c m N p yi
-a a) m v -0 o 3 0 ?r a) 0 m 0 m(U CO c: N
CO a) m a) (a CO Q Q a Y o +' -a c c c Q a) O o c c v o v a) d 3> E .c L n 0
+>-, C E E Yn c E N E Y a) p E E Y N p m c_ m > Y C3 c
0) 4-1 -0 . LL
v o m +~ E o o u -a cLa > E o o 'v a > v v u 'v o ~; a Y o C E v 00
N õ,
a) p m p a) a) p m u > > m
p N > O
3 . U X C +,,, O N t C X N t C L y m L aJ -- 0 X Z m
U ) a a m 1- W w 3 to N 3 E 3 E Y 0 l 7 u >( u a o 0 L-6 w L n
0
-'
a)
Q' .- N M M CD r- 00 0)
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
H w a) U)
w U) 0 a)
C" ca w E 2 v > N m o 0 c L c 'o 0 0 C p
L= cu 4- U 2 ~Q - m 0)
Q~.So=_ 0 0 mm c a)
LL o a) =o = cu =~ 0) gy 2 a) N ~
07 N c >, ova c r E
a)H ooo0 04 Q f~ ~ ~umi~NOf6 0 0 '~~c oN.cvO> a)
c a) Q Cam nvi Npw'NEcca
my 3~ m a) c .n 0o0<2.__Qm 0.r30)C omm
CU a) a m C . M m c CO E w O E N ` N O 0 U p N M U N C N Qj) aa)i N vi 3 a
C O to vi =' a O > N m 0 .o _ vOi C .S t"n L ui N Q O N z Q C N N C M N N C
1 -0
B 00 y cu E 0`- -p -0 Q 0 N 0 L m CU U m o p c p .N Q p j N N' O
W .- L L _ (n E U o m O L 3 4- cl
(D r- 0 m O '^ na Cl)
uc E E m Ez wN ou).C o) N avi Q o 3 0 aNi >=> c a) 3 c a) O E -CO2
0 oQ"">,-z o
:L' o r- U c on c > v m c_ a-'õ a. NUg" 0 a) m'm_Uw 0Y 0 QC) CLCl avi E a)
> 0 3 E c 3 E o 02 a ) (D 16 ~0N0 0L- og o aa) ~~ E E- p
a0) E c c y N Q fa CO 0 E c 0 N C E u N 0 O C Cl) OV N ~
c
N E N E.
~[ to l0 0 of D m r O 0 N m Z 7 m E 0) N N V O O N m O a) Q c .N _ c
a) >, C m O Y\ .~+ .`. C C C = Q DI 3 N N 0 C 0' N
amv ~, X0-6 pccNc)o E_poEcovEv~ ~c3m
75 L u .fl 0 E O C m -0 0 M N Y 0 0 M N V L C' a N .~ C c .N v .N v T 3 N 0-
:z c 0 >" O C Y 2 O E 3 ^ U) - O C 3 0 O U) -0 Vj 20 N M C -0 0 m U 0 =w (D
+1 M c ca 0 au) Q j a) 0 o o V 0~ O O m' N C a) O E a) CO 7) >, L O L rn p
0 m (n
n a) ~- E E -- C V) FL N Y c 0 0 =,r L 0 E E c -J 0 cn cn v V` .- c 0 L Q-
cu (U
LA W m u v E O O 0 C `p O C .! v Q L C 0 N N w 'd -0 (D a) m m C fn j,
v N 3 0 Q C U) E N N '(U 0 +4 O Q -Ile Q O 0 N O Q io t 0 0
0 R
0 4- a p > =C 3 E E N o M rn: p cn m m a) - Q.. Q Q v E
co c o) ` o u E r' M ~L-0 = c =c c 3 cZ a`> a>iv 0 O m co= E> a)
o ==0 E 0 a O DLO. N E w p w u N O 0 ? C N j .-= j t E 2 0
-r 0 N o E- a m o w on u j-- 0 w u) o E E (D r_
E m r . _
N 0 m> am) EvLt~C c.-
N 0 CL - U) E in
` O. N N to +r C a) - f6 O p c
e-1 a) L .E C a 0 a; u t: m o
`~ .O E of L 0 +-' m O E m m N 0 3 m O V N 0 r-. 3 N' o a) O o *7 C C L 0 L '
0 Q C - - p y 0 N o a) 0 C Q. v CL U) j C =C 0 0 E N L) =L
E n m cL- o m c 30` ~'N o m~c o f m cLH- oa T a) o E'N aci ai.~
a =~ v> Y E o p } o N N o >,~ o N N o ~>,f cu c0) cc E E5 E E 'x w m a) (a
(D
C m m . 0 a) 0 O a C C E
0 -he (v In cu "- J 3 Q u 6 0 0 0 0 3 O j >, 0 -0 U)
M (D a W L%
C O O N 7 -p 'p o U z q E o in m c 0 0 >, a> c 3 m N
0 p C }, m E 0 N O C U O 0 U) " a) m Q Q. 'O m c 0._ N N
v c m ~- o Q~Ea) ^ 0 .... co Em LDY 0v 0
. m E E U V 0 L, s> 3 v -0 3 Q E E N O c m 3-2 > Q.
'~^ E E E ) c o v o vmu cv0cty icnww 0)aa))c0aa)))c o a) > E .4 y + n on i1
0` O 0 p m M.2 = O y 4- m a) v C C L L V
0
cr-
E on > o 0 v> tifl L L L c (D ' a) > a) N 0 3 0- n
0
c p W0 f p Y m cc co oN a) 0a 3 ca)'o`aa) (D ccorEEX
M
E - N a C :3 1- C L m- U) o 0 0 j 0 3 0- L- 0) -- N 0 p m in C E N == 0 a) a)
M a)
0 ^ E m o N C .p m 0 ca O E 0 T m in 0) O N ' n~
a O N E 0 Q m 0 m 3 E 0 0 0 .c O a) m CO C .2 C o m N C m 0 0 3
C 'H y, > .- N CO U .- N 3 'O 0 E O C O M C m .0 C O E> O
0 y C Li O 0 0 O A L C 0 0 >, > O Q .C X C > 0 0 0) tf L E Q:. 0
CL Q c E v o W o a) v r E o 0. E QY m ai~ U C< cm)~ o
E- > 0 m = M M U) o c m~=E E o 0~in c v~= -) c~ na)'-~ >,c
cv a) 0) -0 0 m a) a 0 E=c 0 N C
E m au n c N c v u , w a) co c 0 Q)
N a) 0 v a > co i V-) M :0-- cn E O C U L .c m c O 0 w 3 N m c v p C (n
a) 0
-0 L.1 0 0
o N d y 0 - C N O E N 0 2 w C 0 0 O N N N O N= m Y CO CO N p w m
cn(D Q) Ew-oEU~ a)
ar on =a c E a, a- m E E m CD v) m 0 2rn~N o c -
E
C C 3 0 0 ~+ L E C u 4-
O (D m U) 2 m 0 E 0 N v 7 L v N' N 0 0
Q C c E Q 3 C U O E _O U O 3 C N O N o ON.. ' (n Q x c CL c ami 3 E E m a o m
E rnao 0 e Em c~ c o to Co y o ~ N a)
a) " a) cn a)-CL m c co ~ m c vo Loa ms w CL) a) 01
Y N E o E o m c tm 0 -.9 0 C6 >,
O CD C v :. v i m 'n o Q o V o v Q_ .... C 3 C O
tea, 0 m E m =oL000 E a) N Do
0 0 a0o v00 o a W- O aC) O' 0) Z? O N c = C - j O ) - N C E 0 i0 a) N 'C E
0 o a
w~~r~~CUNCV- ~0~. uCLM 0 cu E
vac 4'~ CL-0 E-0 M Q v>r
V, o Ou >m -0p
4- -0 >0 0 0 E 0 U) C N 'C N Q m p = c0 2 -0 tj U) U U)) v O Z N - t
c Cl) o 0) 0 a) o E Q> a, Y o 0> 0 0 E >,o o m vi m o 0 c .~ m E 3
M C-0 m041 Ea)2ovELwom0>- m0-'cEv cnymXc ?MN
OD 3 +~' a) x Z O 0
-5> 0 Q Q 0) N 'O (0 C ai ) E W n 0) D 'N m E 0 o O 0
E 'C v c o o v E~ (D 3: in *0 U) 00 >o Ns -0 -0 0 ou o vN D)o m c
p O - Q 3 w N . - 0 L 0 0 0 O_
00 4(u, aN' !- 3 C 0 O Ou c O V E C o O` C M N
cu :3 N V N U Uj N Q N 0 X N
0
c 0 E >' o ! > E 0 cc CO ~c N LA > Q 'c E ~ E m > a) N > E~ o can ~~ 1-0
c E +- m E E o
a ,~ _ 3 v 2 n ~ E u E 7 o >0 O E
E E o p v12z Q Q NN ~~~ o m ~s
E v a o m in E 3 N
M Q. v a) 0 m 0 > m .. a m2 m C W N >,.2 o a) >
x v 0 0 0 0 m E p x 0 C L O c 0 M 0 C V m p C > 0 D 0 L m
w Q m a C C 0 m E E 2 w 2 m.. 3._Utv._'v 3 v z t~._~ my m._ QE N EL
O N
r r
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
C c o a~i o C a
O) Y u) =O a) N w a) a) cn 'p O y `1 a
W O C C N C ma) U U Z c J N cn o~ O c-1 ` +, 1 C
o f - m m
m m~~ o o cn o 0 V, cu
O CO L)) o aid U) 0
U C cYn 0 U w ~' C U O u L 0 a) o a) > O 'X p r- C cu
v; M a C tw Co- n rn m 7 a) U w a) >' Q Q U E Y o E E p al L , N a)
a) 3 C U E .o Y ai 0.9 O> O C m a) C C ++ >+ Q.~ p E m U C LL N N fC -0 Q V)
W N C =C =V o a) U L a) =C N .a m E m a L C m E p E O o -be =C v ()
Y O Q y C N 7 U ~` ~'- a) j cn O) E =a L =- 0 C O a) 0
>. ~ w U m a 7 01 U Q (0 +L-' m m vi L J a) -6 0 a o co a) C o E r0 m L O N
O Q 0 C m a) - m a, U Y
c m N 3 n E 'C LM) E CU 0 0" C C > ~ .r v L) N c a) > vi O ra
O L a) C U> Q Q O~ - C mO - N' M U) O C ~ C p.- C t c L a) v
in=~ E m ~o p U m c y ) o p a>i ~.-M c Q~rna~ ami E E r v ~e > a >-
O o 3 v c E c.~ L U o- o C- o U +. o o M
c a N ' N p -0 O vi c O c -_ y N N m p).cn - a) m O C U 'c mu 0 >, E
C :-C -le _. m- N 't +==0 0 a) w .-+ o U L7 J U C t^ =N
LL w L O O o C C N N cn
EY - U o Q v NFL ()-C CC C U >..Q > E -0 Z u o f a-C ~r6, C
a) LL C m m d m m O) ~ L M C a) vai LL N N N U)
C O U- C_ p C m
N p a) m LL O rn 3 a) Y C 3 j 'y a j >. O p m U Y M vii C O I, O L a
C > O) U Z 0) a o O a v 3 O M Q Q W L: U y 4.. m LL
0) uj :3 -0 r-I U) u Q
0) 0 Q M C U O c c >, p O N O J E 0 - v 0 H a E Z
E m0^ m c U co :3 (o CD cn C m a) U a ro (Q6 a p U Cn L rn cu d m O C>
=` O o (C? o
E :3 O LL co (n . 3 co O 'a p U p aj (a N` - V) H e m
m a) z Qy o o ~c m c aD L aQ m w C) 3 E m c~
U En oF- o' cam. m E v m cocn`t a?' X-C ~C~ Q rnL U c E v o. 3
U _ m 7 C a; ,~ v) 0 3 C W N 3 a= co =a m rn C U m L u N rv u` U
c vo, m> E Ei ' E o> N E co t o m co E c`a a r C--C (D cn U-)
o cC ~.X ~~
(D -M o o co) a) a,c C c -5 Esv n c o ,? 3 ~? i Q
~p 4) U ~. a L M Y C o o m L H C m C 3 ~- U) a) O a) -a N
'- zQ vain 0 a E m To o T v m m m~ NC E(n a) -' ro J 41
C M C U Y
E E co- o () v m m a) ~~ C o m' o C fl-o ai - U i "-e
`~30mmcC maHUU~pca ~~o~ U)-0Q v m co m p o Qu U) Z. L v c +~
co ma omt (D C c Q0 >'E E g oWn E c 6
"
as 0
C o C
70 m= - ~L N o m o' y QEc a) Q cw 0- 0 i v v
cc 7 C U m> U N U C) N 3 C ,~ C m (n U N CU N p> L vim- F- Y O
%P a) ~ U U m N U) U O O V Q) V p 7 E CL o -a E m 0 p
a) LL w o o
O Z .~ L (n E U C m m N C 7 N >, co vU, E J a) L a V E ~. O
> U 7 E- L v) 'p m 'a -a m C co L U . ` C C CL -0) co m m L Q O L V) C 3 v W u
m M
U a) C X C a) L :. a) a) m .C o o W > `C 'a - L C +.,
con c o
)p- a) co aci CoEa)'=U >Jviaciaa)Occ 3 doN C)'tn v
C U o o o
LL a7 U C v) p 4+ L a ,n O t-~
a)a)mSt > mmy - E CO oE o m N :r 4-, c
0) m
3 0 o m rn0 ~~ rn.` m r E ' v m C v
r`n c a) z=, 0-5 a U)
a)Y NF a) o CL cE mcc a)UcJ=` s C Qonce -0 0
N."
Q0 (o
tt- U)) a`)'o umi c 0 E a) N '- m 0.o E o E ocw cn E E
L 0F' U `m C ai C U) ~La 0)2.0 Uw on- aNi 0 o C'C c g- v =C v
L) U) W O O L J O m N ~' L LL z r-
v) .^. Q C rL=. cn C a) L Y C N O
m ~ C c 3 '- vii
E O m
aL,~3a .rn~a) o'-mc o S.-a-EL 4-' a a a) 2Y-a C-
m F .~ c m w c 'E a w E $ t3 > a) m o a
` a1
M O > C 3 0 .p X -O C C C CU 'O O m m ,= L > 41
>' O '> C Y p 0 'O CO a) N Q- a) a) O O o m .> m a) c Q V) CU LL .0 C O E v 0
N C
ca -:p O U v) -.0 O N C E? ~ N Q O U .E N a) U O Z a) m f O D E N a1
~ , C "N U !n y r Q C L C 0 C)
E C c m "C O a) L Y -72-U)
C iJ C tw
m >' a) L 0 O O a) C a) a J m a) 4) Y C N a) N m r 4- - 40
`-- C O>
M co
~ E'~ c ca vi c Q E aC U o vai y = vC n O 0 aa) a+,) `- a) v t
3~ a n a) () a)cNX CM v m.oc >.=o- m cp o a) 3 a0o m ro 0
CL N
v T N 4-1
CLL. m Cc~ rC ~ N ~-a a) ES E 2 E a 0 o o10'- m
a) a
3 cn w- 3 CL 'C C V N co N C >, a) t U) O - a) tA :3
(n (n D J p> O c Q 0 v 0 v m
N O O^ t 0 'c Q m a) r- 3 7 LL (D 3 .C to m 0) N a) Z' v) O L C j C= 41 a) C Q-
C C C O a) >+ Z In a) C C C> C m W N Q E v) N w Q a) v; 0 C E a1 V)
-'~ - c m ai.c o~ cL ~ N ~v E=~ w ~ ~ N ~ m 0
u v
~Moa)cQ Cap~>'Wa)yp_a mE a, v c u v a,
,.~ ~-"= coax mm3 3 c `a)am m C o
cu C %u o ~ NL U QY~+~`~
p cmi d H U c s c p v N Q vai N C N 0 ) 3 0 c iv J a) =E (-~i a V "~ 0 co O'er
) N p C -Q- 0 a)'ot O C_ cca o O C E> c W N j C o V Q U Q u
C C U C a) -0 p N o C 'O .U O C j . U o J C 6-S Su C DNj ' N r6 4- 0 m 7 O N
(d 0- N v) m a) C C U C 7 o E 0 >, m U a) C .0 (n a a to m r- m Q O C a= 0 0 0
L a) U C ;a U> y > -0' m =' ~, w (fl Q 7 M Q) a) 0 a o- a,LLocca 'r~np n~yaci~
v>uirn Q a~c Dmo`
0 CL CU O ~ C cn o E E U+~c~o -; cm czo~~Eo(imEa)c ). a) O '~c~~am0MLD2XCc 0)
M p a) U 0 a,Z Ln
o E a a ~ E'= U m- o o cv C v; Qc C
. V)
a a, =c a E
75 = v ~=X m C'C< a) W C o'C m a m-aa a~ a -;~-1 m ' u c rEo E m a)
O C .- U U C m (p a Z L > m .~ 0 O O C L X O 2 Q m 0 X y i O X Q C E O C
.E m~ QQF- ~ - m w am a Q~ w ra E w ro
M to
r r r
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
N
c a co CD
L` o a) is a)
a3o S `~ (D c o a) ca a) - a)
N a9 =- L.0 W 070 U N
+' - O c p 0 7 f6 O 0 N O
a) N C C N > a) E2 U 0) O
L rL C d a) 0 0 .C O N p O C
i_ 0.0 0 N C O
C 0 0
3 N o m I~ a 0 a) a) Hh
N N N a) -0 a) Co a -0 w CD
E CL L J U 0 > O p m.~w
iJ -pL
m cm C a) zn
mC f0- N O
7
a) N C L O L O O O N p O LL 0)
a) 0 w c "~ a) c- 0 Z O
a.. N L O TL O- E C p
C U 3 w C "..' U O :3 C L O E p
a 0C'as2 N Im a) 0 4)c E
U) E
E
a)
Z 0 O i a) ) a) O >, O w
C E 0 U =U E o- L ~ vi o O 'p
0 :3
C N C
0 to V .U L a) (7 a) O` O N j O
v Cu va'i~co3ccH 3> LL_C
a) z
o
Ev~P p F-Z=S
M 7 E E U N O - O `
~ QL O O ' Q U
E O.-L d N 0 O () U)
-0 0 ..- N- O c U E 0
L 0 a) E C
(I) C O >, L O - m
- L M (A .C O N O -0 0 0 cm LA M O N C `> a) O tO 0. 0 C O E
co N Y O O c L) N E
0) (D
E > 2 Y N N L -0 > +==a) cL0 Y N
7 :) O a) p (9'0 N O N C
0)-D .
cu 0) 0) >, H-0 . 0 L C N
C -_ U -6 C_ (0 -O C N C a) O m
(1) a E c m a) , m (U > 41 C 3 3 .~ 0 0 a3 1~ a) p v) p .> O 0-0 (L)
C a) a) 'o N 7 L O Id w E
U U C >, O>O (n O a) 7 a) O 'p
+ 2 C O -O ..+ O N ; p C c N c N L C
U) tan N 3 Z c c a U 1' L- N LL ca
v S E ) u s ( 1 ) co3E mom Cu Z'N
~c ,~ tom N a r- > L L H a)
a) iri Q 0 () :3 c 0 J .- CM a) O
U) Z' E
c) Uc~ )ca) BOO U C om0
c0 0 0 N t: 1 co -0 C O C O
0 N fa O (n 0 L J fB L a) i N N` C
M (1) ` N
0 CU p L O f0 N O
O) O) 0
CL N c c - L p .0 O c C-J 0) .N N m p V c O H
c 3 3 a) m E O~ C U c a) 0
O O N- a) O N 3 C2 ' a' ~C <0 N O (L) N
t3. - 0 =C w N N C - Q E O L .U U O p j
0 L L >. i 7 ..., ..., q- O)
E~ m ~= rno ca.0 >, c c E E
3: a O O w. vE n NvUv;3 - o
O L Q C C 0 a) N C E =N J H 'U C N N N 0 d p
CL m CO) CO) 2 U C; C 6 E 0 C O 3 E N t O m
CU o 2 m u ,~ a) w o p c E E
in D. a 0 (A n. U) E'E uvi 'm c o -Q) C X x ` = lL > O-0 2 O Q O U p E N
E N () a) O O a) Q'C N E W m
O
~a i tea) Ev~>. 2N ~cInm (1)
3ami acai uNi`nU>aNioZ.L _ I M 0L 0o
E d LL. 3 U c p 0 a) 2 0 N a
c 1) a>+ m C C >, 0 0 0 N N N C 0 N CD M c Y
M (n N E O V ,C N C
CC) 0 C
E Q =N. O =CL) U C a) N E c > N, M O
E N U O c vi 0 O) N- 0-0 M.9 co 0 o -J () = N
E E? U m a)s >,yv c NU 8o v E
E= a) o' a L O N O Cu in ' N N U Z m `"
E N a) o f0 N N Y - CL :3 o N O J d > ca c
-' cc ) y 0 E E ?0oNT> ,0Nmrn~>a(?c`0io
,~,~, ri o > =0 C v ) D N N c0 a) cd . NO. c)- co C C N
o O o m E
E c a c a o m y ui >a? -
n. NMI- M 0 C2 CO MIT fa E O r- J L E N O
of ~+ L.L N rrtn LLL N N E ~+ a) CL 'Z~
a) Q E U z O Z ' i M. E> 'c rn N L acj m h1UU
W. 0 0 0 0 O O O 0. 0 O 0 0 O O O O O O.W W. o O (d -0 N W (0 i~ CC O)
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
9-
~_ > ,c ca 0
a) O U O =N a) W
C U > O C 4) O .N c '3 L m
U m U O j m V Y N N N E O
U p N 'O O .G _ a) E
m a 0 N C N O CL N O Q
m
0) C N C U p a N M C - a t 0 C or a)
N _ C C 3 E O F C N- E
0) ma) m a y V Q p C O 0 u - o N L
_ c
u U N L U > U w N C .C > N> U -0 E
U 3 Ow ~ w N 0 31 CL o c -O C' C N O m Q m m
o E
N (a N C C aJ C N N < D C O p 6>, N 0' N E
- C> o a)cE t
N '~ > D O p O w N Q Q~ c c N N J N
E c~ ~Yu 0 5m ac- cp c umi 3 N N~ ~c Oct
O C aa)i co O 0 O U) -0 CL Y p cLn E a) N lf a) N
C O C m> 1 0) m a) a) _a N C N C m
L r J c
U w E m ~- 3 '~ C U N m+-'mu m a)m p N a)
a) N
coo 3 H ~.N aci m a a) arnco f
E
E N a=+ a) O L N =C E o.
u CD U CU ~'~'c0c`)CCa o~'om >,>,a)~ -t6
pE
4 1 :3 -u
' p C p C.C ~L U 3 m E 4-:
N m
0
Op p
0 > (D or) a- CU C a)
(D a 4)
a E 2 N a O C O m a)
N ~ C p a)
U cn =- -
a) N> C E N
U N N a) N N '6 -a N U) O O N
O a
0 ca U m 0)- U 0 U m 0 N p0 > w
(D -C NC aOm' ON_Um COC m V 0C' LNC NN
a) OO rn a) mL
c o 0 E u) rnot o~ a) o f u-) m
c o E vi m n> 0) 0. E a~ E a") m c a) Y
E 7:
cm a`) Ec w-Nc Nm 0 0 0 N o
wmU E 0 2 -6 3 0 Om C
Ea)C mi o c c ) w v Nw_ aa)i
L N w a N CL - m m Z O - O -a 0)=~ C
CO O U C C a U w C O LL O U a) C m m =0 (D >cCU UU) ca Q 0 3.5 OC CM > m ( -
a') ~ ca A6 :5 c_
y m CC m 3E :3
a) N Z' ' LLL E oC a ) E
ID o 0 0 CUJL C -co o N E
. N a)
C o 3 a ) - m LC O E.0 wC" C'
ca - m C N N' N C O >a. a) m
co ~, cm V E
ca c 3 m (6.C 3 0 3 3 .. m~ rn E
a) In ~ o EH ~ a ) E c c c)a-) E c
N" E L C E m L
N O Y m L m
L N a) U
O m` m N p p N m .~ +.
~ w C 0-0 E m au c u 0 )c NN o f N o
0)co -.5 m Ev a -- 2 m vim N O
`O. E .C m > m .C N a) - =O (D N a 4) m a cai
O C N< 3 0 ca :3 (D y C N U n N N O 0 '` V m C C 'L" ~
a) (D N U) O U N _ v3) m 'p) C O m E U m a) 0 0 CD
N U C u 0 0 co a) U O N N N U .C E 0 -0
fn O N C a) a)C N C .Q N C t C m a cc (D '0 '0 N E Q a) p L m
C Q p N O N O 0 C '-' .- p C O =C E U O N
mC->:. O O 0-0 Q) a).. aO C-0 mw 2
)
a) O a) L a li N co m U m N M C 0 - 0
O O Q O p) m m U 0 . 2 N C 0 N U a, a) p L d C a)
m M L E 2 O
E )a 'i E a E6 (6 a))~
m C O(D m y y a) U E O a) L E y C U 0 0 0 N 0 w
o
0 w 5, co m m N 0 : 3 M< I- o Q e n C N c 3
a) En
L) a) 3a) or cu EELCUU >
N
C O N E 0
m U U a) O N O a) U - t C N C L) L O LU N a T+ N
0 o m L E~ v ~ m rnp N 0-0)a) a)a o o m
CL t: `
0 M -C
M Ca
a) C 0 -a) CM C a) C C m a) O -p j ai C_ U 0 3 w C .N
c
E U c o
c 2 2 N a O M U c~ v v U= p O
Ea 'aYcoa)NC Ern'y~CEca2mc2m
Ca .0 C r X O U N o N W , X! N "a x in E aa)) ._ Cl)
N s L
X
w m? W aU C N a-0Y W m UL W `a ,C m aY
O - N
N 04 N N
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Poy, Rose J.
From: Haleva, Aaron
Sent: Monday, November 10, 2008 2:26 PM
To, Poy, Rose J.
Attachments: Variable _key.xls; Conditions.xls; Literature.xls;
Rule_Definition.xls; Rules.xls;
Rules_keywords.xls; bin_sizes.xls
NO
Variable_key.xls Conditions.xls (14 Literature.xls (126 Rule_Definition.xls
(37 KB) KB) KB) (22 KB)
I need three copies of each of these files printed out
R R r1,
Rules.xls (60 KB) Rules_keywords.xls bin_sizes.xis (26
(29 KB) KB)
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
VZV Varicella Int
MPGI Measles OD DB Int
v4 Sex
V5 DOB
v6 Age
v14a Interpreter
v14b Interpreter-Language
v15a Country_of_Origin
v15b Region_of_Origin
v17a Date Canada
v18 Citizenship'
v21 Medical Problems
v22 Hospitalized
v23 Medications
v24a Pregnant
v251a Measles history
v252a Mumps_history
v252b Mumps_Age
v253a Rubella history
v254a Chickenpox_history
v254b Chickenpox Age
v255a Tetanus history
v256a Diptheria_history
v257a Hepatitis history
v257b Hepatitis - Age
v258a Pertussis_history
v259a Polio history
v26 Vaccination- record
v33a Education
v33b Years-Education
v33c School_type
v34a Indoor toilet
v34a1 Indoor-toilet-type
v34c outdoor toilet
v34c1 Outdoor-toilet type," !
v35 Water_supply_type
. v37a People_household
v37b Rooms household
:crowding Persons-Rooms
UNIVERSITY UNIVERSITY
ELECTRICITY ELECTRICITY
MEINT Measles _Int
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
HABIPN HepAInt
v20 h Refugee_camp
v24b-h months _pregnant
v251b h Measles_age
v255b h Tetanus_Age
v256b h Diptheria_Age
v258b h Pertussis_Age
v259b h Polio_Age
v2 -h Data recruited
v253b h Rubella_Age
v15aO Country_of_Origin_name
New-Rube Rubella Ab
meaod 1 Measles OD DB
meati 1 Measles titre DB
meodr 1 Measles ratio Z
mezod 1 Measles OD Z
mumod_1 Mumps_OD_DB
vzvod 1 Varicella OD DB
vzvti Varicella titre DB
OpDenTetanus Tetanus OD
IUTetanus Tetanus Ab
OpDenDiph Diptheria_OD
IUDiph Diptheria_Ab
HAVAB HepA_OD
MUMTI Mumps titre DB
months ca Months-CA
new rubin2 Rubella Int
CMVIg CMV_OD
CMVIgINT CMV Int
HBcAb HepB_cAb_CID
HBcAbINT HepB_cAb_Int
HBSAb HepB_SAb_CID
HBSAg HepB_SAg_OD
HepBeAb HepB_eAb_ratio
HBeAbINT HepB_eAb Int
HepBeAg HepB_eAg_ratio
HBeAgINT HepB_eAg_lnt
HCV HepC_Ab_OD
HCVINT HepC_Int
HCVPCR HepC_PCR
HCVLIA `HepC_LIA
CID_Strongy Strongyloides_Ab_CID
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
OD Filaria FilariaAb OD
OD Schisto Schistosoma Ab OD
final Schisto Schistosoma int
fins! Strongy Strongyloides_Int
final Filaria Filaria ant
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Poy, Rose J.
From: Haleva, Aaron
Sent: Monday, November 10, 2008 2:26 PM
To: Poy, Rose J.
Attachments: Variable_key.xls; Conditions.xls; Literature.xls;
Rule_Definition.xls; Rules.xis;
Rules_keywords.xls; bin_sizes.xls
Variable_key.xls Conditions.xls (14 Literature.xls (126 Rule_Definition.xls
(37 KB) KB) KB) (22 KB)
I need three copies of each of these files printed out
IN
Rules.xls (60 KB) Rules _keywords.xis bin_sizes.xls (26
(29 KB) KB)
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Condition_ID Attribute Value operator
1 Schistosoma_Int Reactive eq
2 Filaria_Int Reactive eq
3 Filaria_Int Reactive eq
4 Strongyloides_Int Reactive eq
Strongyloides_Int Reactive eq
6 Tetanus Int Reactive eq
7 Diptheria_Int Reactive eq
8 Varicella_Int Reactive eq
9 TNF_alpha_int Reactive eq
I FN_gamma_int Nonreactive eq
11 IL 13 int Nonreactive eq
12 IL 10 int Nonreactive eq
13IL_6_int Reactive eq
14 IL 12 i_nt Reactive eq
IL 17 i_nt Reactive eq
16IL_2_int Reactive eq
17 IL_1_alpha_int Reactive eq
18 IL 1 _beta _int Reactive eq
19 IFN_gamma_int Reactive eq
IL 10 int Reactive eq
21 IL_23_int Reactive eq
22IL_5_int Reactive eq
23 IL_4_int Nonreactive eq
24IL_8_int Reactive eq
TNF_alpha_int Nonreactive eq
26 IL_6_int Nonreactive eq
27 IL_23_int Nonreactive eq
28IL_4_int Reactive eq
29 I L_13_i nt Reactive eq
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Poy, Rose J.
From: Haleva, Aaron
Sent: Monday, November 10, 2008 2:26 PM
To: Poy, Rose J.
Attachments: Variable_key.xls; Conditions.xls; Literature.xls;
Rule_Definition.xls; Rules.xls;
Rules_keywords.xls; bin_sizes.xls
Variable _key.xls Conditions.xls (14 Literature.xls (126 Rule_Definition.xls
(37 KB) KB) KB) (22 KB)
I need three copies of each of these files printed out
qqpp 3 ~p
43 'Hy ~C3 ~`
Rules,xls (60 KB) Rules_keywords.xls bin_sizes.xis (26
(29 KB) KB)
1
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
RuIe_ID Keyword
1 Schistosoma
1 Schistosomiasis
2 Filaria
3 Filaria
4 Strongyloides
Strongyloides
6 Tetanus
7 Diptheria
8 Varicella
8 Chicken Pox
9 Hepatitis C
Pregnancy
11 Autoimmune
12 Rheumatoid Arthritis
13 Systemic Lupus Erythematosus
14 Cancer
Multiple Sclerosis
16 Inflammatory Bowel Diseaase
16 Ulcerative Colitis
16 Crohn's Disease
17 Diabetes
18 Breast
18 Cancer
19 Cancer
Cancer
21 Atherosclerosis
22 Allergies
23 Asthma
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
Poy, Rose J.
From: Haleva, Aaron
Sent: Monday, November 10, 2008 2:26 PM
To: Poy, Rose J.
Attachments: Variable_key.xls; Conditions.xls; Literature.xls;
Rule_Definition.xls; Rules.xls;
Rules_keywords.xls; bin_sizes.xls
t Li{I' i
Nz'
Variable_key.xls Conditions.xls (14 Literature.xls (126 Rule_Definition,xls
(37 KB) KB) KB) (22 KB)
I need three copies of each of these files printed out
Rules.xls (60 KB) Rules_keywords.xls bin_sizes.xls (26
(29 KB) KB)
1
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
prop_list bin-size
Age 40
Mumps_Age 40
Chickenpox-Age 40
Hepatitis-Age 40
Months-pregnant 40
Measles-age 40
Tetanus-Age 40
Diptheria_Age 40
Pertussis_Age 40
Polio-Age 40
Data recruited 1000
Rubella - Age 40
Rubella Ab 200
Measles_OD_DB 1
Measles_titre_DB 10000
Measles_ratio_Z 5
Measles_OD_Z 5
Mumps_OD_DB 1
Varicella_OD_DB 2
Varicella_titre_DB 2000
Tetanus OD .2
Tetanus Ab 30
Diptheria_OD 2
Diptheria_Ab 3
HepA_OD 2
Mumps_titre_DB 2000
Months-CA 40
CMV_OD 100
HepB_cAb_OD 2
HepB_SAb_OD 400
HepB_SAg_OD 100
HepB_eAb_ratio 1
HepB_eAg_ratio 1
HepC_Ab_OD 50
Strongyloides_Ab_OD 1
Filaria_Ab_OD 1
Schistosoma_Ab_OD 1
years - education 10
education 4
rooms_houshold 5
people-household 10
persons-room 1
IL_2 1
IL_4 1
IL _5 1
IL_6 1
IL 10 .1
IL_12p70 1
IFN_gamma 1
TNF_alpha 1
IL 13 1
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
IL_17 1
IL 23 1
TN F_beta 1
IL_8 1
IL_1_alpha 1
IL_1_beta 1
IL_15 1
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
APPENDIX A
2. First 330 Records From Exemplary CIP Database
-358-
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
(0 co O N U) O V O N 0 U) CO O O ) N (0 O O O O 0) (0 O N O
O 0) 00 U) N O CO M N- CO N 00 IT M (0 co
CO V' - O '- co Nr co M I- ''- CO N 'IT
U) 0 0 co I,- co N O M CO ~ I- M co O CO O co
O U) O (0 U) M U) 0) co 00 N M 'q O U) O M (O U)
r U) r- U) CO 0) N- CA I'- ti N N- =-- '- N O
I- U) O CO M '- LO co co N- co O M U) ' M
J CO CO 0) N'- U) - CO O N ~- N N M 0) O N
CO co '- to M I,- (0 CO N N qv- 0) N U) M U) co U)
O'- CO O co O N C) O' CY) ~-- ~- O '-- 0) CO O
M O O N O O ,- O'- O 0 0'- O O O
O co -- 0 0 0 0 00 N U) O N O O) M O O O 0) N O O O
N co U) U) (0 N r- ~- 00 0) co U) U) qq qT M n
~t N 0) 0) 00 1- N CO CO N N ti 0) 0) O ti U) co
O M co M N 0) 1 0) N CO N- N O N 0)
06 O O co M O) 00 IT I- co O 0) v M co N 'IT N 00
I- N- N N 0) r- 0) 00 N ' U) U) O) U) N I- N O
l 1q, N- (0 CO 11 N - co N CO co U) O U) v I-
M U) 00 00 U) 0) r- - CO CO M co 0) f` (O - 0) 00
M co 00 00 N O U) N- co M q CO co co U) '- N- N-
(O U) 14) U) O ~-- M N M co M "T 00 M i-- LO N N-
U) CV) CO ;t U) I "T O =- N qT C )l
(.0 r- I,- 1,- O CO O CO 0 0 O) O M n N O O (0 O O O O (0 O O O
'-CO U) 0) ~- =- 00 0) O co U) N [I-
I- co I%T N It 0) N- CO 00 N U) N N '7 00
0) N U) 00 co I- U) N- O 00 M U) O O) N co
O (0 O r- r- O N- M 0 CO N ~' co 0) f- M
I- N co (0 U) O) O N - CO '- co (N O N
o0 N O N N U) M N N- N LO CO N U) co
(0 ~-- O c- N 'IT CO '- Ct CO M co co CO CO
0) ti O CO qq- (0 N d' O co 00 s- U) N
N l qT M U) -4 O- `- M N- CO N-' (fl 4
O N CO r CO N M co r- co N'- r- ~
O O N O O O 0) .'- O O O CO O M N O O O Cr) 0 0 0 CO O N- O O O O
CO ti co N COQ) d' N U)
co - '- U) U) U) 0) 0
00)) N 0) N N
0) O 0)
O co CO co C3 LO LO
0) 00 O O
f~ f- (0 co co U) co M tf)
CO O O (0 '- O
0 0 0 0 0 0 0 0 0) M O O O O CV) co 0 0 0 0 0 0 O O O N
r r 0
I- U) N- ti -t -t It
O C*) O U) (0 - 1- N N nl
~- O '- CO O) N U) co M
0) M 0) N r- CO CO Nr I-
0) M 0) 0 O) N N co 0)
N 0) N CO '- N 0) co 00
N 0) N O M O M U) CO
Iq U) O LO co I,-
CO N M O i-- O
O O O O O M O 0 0 0 0 0 O O O O O O O O O O O 0 0 0 O O O
N
co
N O
N
J C)
O
N
CO CO O O (M O O O 0 N I- ti O O O O N- 0) M O O (0 O O U) O O O
00 O I- M N qT M 00 CO CO '- CO
`7 CO N N- O M N co Iq U) '- CO
co O co O 00 (0 CO CO CO co CO O U)
I (0 O .- '- N co
0 0) ' CO 0) 'IT N co
CO r ti C) CO CO 'q co N '- CA
~- r U) co M (O I,- U) I- M O) N- ti
N- 0) ti U) O 0) U) Iq U) U) 0) O M
co O U) '- co (O O co O CO O) M U)
O O O O O O O O 01 1 O O
i- - U) CO 0 0 0 0 0 0 0 r- ti CO O O O O O M O 0 0 0 0 0 0 0 0
(N CN CO N U-)
- ,t N CO 'q 000 O U)
M M 0) I- co N CO U)
'T N- O 00)) N- N O co
r^ LO OMO qt O co
co M
J N- U) U) O U) 0) OO CO
N M- O M O O O
'O
Z
1v
L
0
U)
O '- N CO 'I O 0) O N M U) CO f- CO NO O r- N
N M U) (0 a0 0 ~- ~- - CO t LO U) CO 10 N N (V N N N N N M cM
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
G) m (D (D a) a) a) (L) a) a) (1) a) a) (L) (D (D Q) () a) a) a) () a) a) a)
a) a) a) a) a)
> > > > > > > > > > > > ! > > > > > > > > > > > > > > > >
N 0) N v N N CA N co N N to (0 CA cn N N N N 0) 0) 0) (n (0 0) U) N (0 N N (o
0)
0 0 0 0 0 0 a) 0 0 0 0 0 0 a) 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CL a. a M a. a z M a _ a a. a. a. z a. M m a. a. a. M M a a. n. a. a. a a_ a.
a_ a
0000000000000 o)0 0000000000000 000
r- O
co CF)
IT N
00 N- 0)
cli
L 000 r
O
H M M N-
O O O
O IT O M 0 0 0 0 0 CD 0 C 0 0 0 0 0'- 0 0 0 0 0 0 0 0 O O O
t0 CO O O N
C N '- C)
co M (f)
1 LL (0 O (D
z I- O
17 N CO
O O O O
0 0 0 0 0 0 0 0 0 0 O O O O O O O O O O O O 0 0 0
O O N 0)
M LO
I E f
N N
r 00
O
:Z iLL
N
0) 0) 0) M ~--- 0) ti O ti O N 0) CO N CO N IT c M O M N 0 0 N- M (~ M a)
O s-'- (D IT O O r- O O O N- O r- M 0 0 co (D N M CD 0) (0 0)
O N N O O) N- N IT M Nt 00 N N- CO N r- N r` 0) 0) co M 0) N 0) (a
CO 0) O CO N O 0) ti O ti r- 0) r- '- rI- M M Cr) N [- M N (0 N
CO O M (D '- 0) 0 0 ti (t) O O N O N N O O'- CV) 'tT :t
O O N N 0) O 0 O'- (D O O CO O 'IT 0) - N 00 O co N O N-
CD O N - (1) CO ti c- O O 0) (f) ' Iq 0) I- 'IT V- C) r- Iq O
J O (D co CO N (0 CO O O N 0) O ",T N V) O O) NT M co ao 'cr (0 co
O (0 N N- O 0) CO N- co N- O IT - O Cl) N 1- (0 w- c.,) N N- 0)
O O C-) - M - 0) 0) co to to U) N M - N O (D (0 O
I- N M ' O O '- N - M O = - O N N- N
N- O It (0 0) co I-T 0 0 0 0 r- O co co co 0 0 0 0 0 0 0 0 0 I- (0 N O
CO 'cr (0 Iq N M co 0) co co co m 14, O (0
ti O r- 00 O co m CO 00 CO =- ti CO O
N 0 I- CM C') co M O O
M (O O - r- d' (0 - - N '- IT M N N
(D 0) CO CO O R M lq* (D (D ti 00 N M
(D CV) IT N - N CV) 0) O O N N N M CO 00
J O O - - O '- 0) 0) 00 IT ICT '- s- N 0)
r It 00 O 0) IT (D 'IT 0) rt `7 i- (D O O
O OD O N O (D O O ti N N N M f-
C) O 0) 0 IT O 0 V* 0 0 0 11:11 1 0
O co O (D (D 0) M N (D N Iq r` CO OO I- M N N N N 0 O 0) O O CO N O co
M O M O'- M (0 N M CD N- (D O Iq co O O O O CO
(O M CO d CO O Co O ti CO (D O N- N (f) 0) M CO N O I- '
i'` (D [t to (D M '- CO 00 M 0) 00 IT P- I- O r N r` N- co co (0 M 0) O
O O N- M M (0 ~-- N (D O O M ~- ~- M M O 0) M N 0) ti N- (f) '- C'M
N O d' O M C) 'IT N M I- CO N- O 00 N N O M N' M (D to t7 O O
o) cM CO C N' O O M O CV) 'IT ' N 0) IT () 0) N N N C) O
J 00 N- N CO M 0) I~ N O 0) O (0 1- O (0 U) IT O (0 r- (0 I~ Iq - N-
CO O (D q V CO U) M O N- M N (0 O (N =- N- CO I- =- Iq (0 O M N O
O I- M M O O 0) '- N O CD O M M O to O o) O V- V O O N
. . . . . . . . . . . . . . . .
=- M O M N M N M N M M O N N - N M
0 0'- 0 0 0 IT tT O 0) O N 0) CO N 0 0 0 0 V* 0 0 0) O O O d '.- O
O IT 0) 'IT N r- N N O O M r~ 't (0
lq- O Q '.-- M N CD N O 00 N N (N
0 to O 0) 0) - c0 ICT M (D O 0) v 0) (0
M M CD N 0) O 0) U=) '7 N- M O O ' CV)
o '.-- O O co IT (D O O (0 (D O
r r 00 00 co M CO co m CO CO co co
J N M (D N- O co 0) co It ti N 00 ti co
O lq- N M O CO N co M 0) 07 1` 00 M 0)
00 M O 0) N M O M N co O 0)
. . . . .
O O O O c3* C) 0 0 O O O O O N
M N N I- O O O I- O 0 M d' 0 I M O M (D M M O O N ~-- N- ~-- C0 N (D
M O co CD co O O (D O O (D 00 O M v v CM O [f 00
M V) I- '.- M O) O 9T O 0) CD O NT O O f` O 0) 0) M 0)
0 ti N 0) co (0 0) O N co '- CO N 0) N M IT 0) (0 co O N O
O (D O N co 0) O N O 0 ' 0 O CO ' 0) '- N 0) N r` O O O
N 0) M O 0) co O O O co 0) co O 0 -.-' O O co I- M d'
O 1- 0 N- O O O ICT M M v (D O O 0) O N- O O co I-
N IV M O V, (D CO '.- CO (0 (D M O O ' N v N N IT 00
N O (O O e- 00 co v C0 CO CO N R (D N O CO (I- O
N M CO N M O N M CD M N co O v 0 O N N ti O M L
N N '- '- O N O =- O i- '- 0 N 0 0 '- 0'
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
0 0 0 r O 0 o O r O o 0 CD 0 0 0 0 0 0 O C) O r o 0 0 o r o 0
M M M M M M M M M M M M M M cM V' M O M M M M M M M M M C~) C)
CO
>
lf')
U d' IT Lo M LO CD r N M CD (D N (D CD O LO U) '.T Iq .q 0 In LO U) U) to 't
U) U)
LO
>
O CO OO (O O CO 00 (O f~ r O (D CO r 00 CD CD aD CO O CA O CD O (O CO O CD O
CC) CO CO
co
U)
>.
M
a)
0)
(0
a) a)
E E
> ( N
C C
=. w
~_ CU
> >
O O O O O O O r O 0 0 r 0 0 0 0 0 0 O O O O O O O O O O O O O O
O) O M D) I- U') (D r O r I- P- O O M N CO N UC) M (D O) CO O) N N CO CD O M
U)
M N N N N N M N N M M M M M M M M M M M N M M N M M N N ~t M M
r r O r r O O r r r r r r r O r r r 0 0
N N N V O N (U N N CU Q) N> O N> O N N CV ID N> N N N
U) N U) 7 fn in N (j) a) U) U) 7 D) (1) 0) U) U) CT N N U) O U) 7 CO c" N U)
U) 7 U)
1 O O O CT O O O O O O O Q O O 4) O O N O 0 0 CT O O CT O (1) O 0 1 0 ZT O
a a a w a a a a a a a w a z a a z a a a w a a w a z a If a w a
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
v LO O u) u) N
U) CO CO U)
~- r N
N N 0 0 0 0 O O N O N 0 0 0 N O '- ~-- N O
N
aN
N O O N O N O O O N O ~- O N O N O O N N O O N O O N N O
(Y)
NI
co co O O) It) N- - d
N
N
N N O- N O C) 0- C) 0 N O N O N O O O- N -'- O N C) O'- N'- w
cu
t0 d
N
N N O N N O O O O O i O N O N O r- O N =- O N N O N O
CU
; U)
0 0 0 0 0 0 0 0 0 0 c), CD 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
CN'
C) 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0'- 0 0 0 0'- 0 0
O O O O O O O O O O 0 0 C) 0
N
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
00 N 00 00 O r f- U) i~ M O O O M N (O I~ M t0 `7 U) O
r N r r N N r r r r r r r r r r - r s- N
cr)
to to M O M ~t M M C tL) to to O l1) V U)
to Lo U,) Lo
M
CM
O O O 0 0 0 0 0 0 O O O O O O O O O O O O O 0 0 0 0 0 0 0 0 0 0
O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O
.Q)
UIU )
u)
N N 0 0 0 0 0 0 0 0 O O N N O 0 0 0 O O O O O O O O O O O O O O
0)
(0
co
U
U) co
N
N
O N O O O O O 0 0 O O N O O O 'O O O O O O O O O c), 0 O O O O
;(0
N
La
O N O O O O O O O O O O N O O O O O O O O O O O O O O O O O N O
(U.
O N O O O O O O O O O O O O O O O O O O O O O O O O O O O O N O
(U
U)
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
I- V' U) U) U) N- 00 CD (0 r U') U') II' I'- Cf) r r M N U) Cfl N- I- N I- M M
M 1- (0 r
CD - U) N- N- (D M N (0 N M CO (D (0 M M M CD O 14- (D N- O (D O r (D CD N co
O M r r O co (D r (D M M M (D
(D (t) V) O (D M C0 M co CD CD M M M CD
(D CO (D M (D CD (D M M M CD
CD 00 O
(D N U-) (D co M (D (D (D M M M (0
(D (D (D (D M (D (0 CO M M M CD
(D LO (D M (D M (D (D CD M M M CD
(0 (D CD (D M co (0 (D M M M (D
U (D to CA co M r m (D (D r M M co w
O C) O O O r r O r r r -1 1 O O
CD I - r CO (D (O =- 00 N U) CD CO 'tt M (f) (D O (D CO Cn (D M CO V (D CD LO
'IT
r r r
Bch
'IT
(0 (D M co 114- ~t Cn m N I,- N co lq- 14- O Cn r (n co I- If IT (n IT co
(0
M
r r r r r r r r r r -r r r r r r r r r r r r r r r r r ~- r r r r
U)
O
N N (V r w
a)
(0
U
114*
co
O O 0 CD 0 0 0 O 0 CD 0(D 0 C) 0 0 O 0 CD 0 0 0 0 0 0 O 0 0
co
r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r
r?
r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r
N N N r N N N N N N N N r N N N N N r r N N N N N N N N
co
co
i>
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
r r r r r r r 10 )0) rn rn m 0 )0) C)rn rn rn rn rn rn rn rn rn rn rn rn rn rn
rn rn rn rn rn rn rn rn rn rn
1 .c~
r r r r [~ r r r r r r r r r r r r r r r r r r r r r r r r r r r
0) 0) 0) 0) 0) 0) rn 0) (D 0)
0) 0) (D 0) 0) a) rn rn rn a) 0) rn rn rn rn a) rn rn rn a) 0) 0)
'
r r r r r r r r r r r r r r r r r 0) M r r M U) M r r M M r r
a) a) O) O) O) a) O) a) 0) O) O) rn m O) a) a) O) O) 0) 0) 0) 6) a)
U-)
r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r
mmm
-Q
CN
>
r
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 (D C) (D a)
O)
C3
a
OO
~. > > > > >
m' a) a) U U U a) a) U Q) Q) a) a) a) a) a) N U a) a) a) N U U a) a) N a) U a)
a) a) N
> > Ca ca co > > Ct > > > > > > > > co > > > > N N > > > > cv .> > > >
_! U U ~ ~ ~ U U~ U U U U U U U U ~. U U U U~~ U U U U~ U U U U
cB Ct C C C Ct m c m m Ct m m ca Ct ca c Ca ca Ct w c C cu cu cu ca c co m co
Ct
t U a) 0 0 0 (l) (1) O 4) (1) a) (1) a) a) O N O a) () a) a) O O m a) (L) () O
(2) a) O a)
w z z Z w w z w w w w w w Of oc z w d' W Z Z W x W -.Z W.M.Of ly
3 L.
~ O co CU
> U (1) (1) ^ W W > (1) a) (1)
> > > > > > > > > > > > > > > > O > > 0 > > > > > > > > >
O O O () O O O O 0 1 0 1 0 O O 0 1 0 , 0 1 0 C O O a) cri O O O O O O a) O O O
c ZQ.aaaCL IL ELI aaaaaCL wa- azwaCL aIL aIL Zaaa
r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r
V
U
w
w
r r r r r r r r O r r O O O O r r O r O O r r r r r- r O r O r r
(/)
>
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
rE
s0.
.N
E
O O O O O O O O O O O O O O 0 0 0 0 0 0 0 0 O O O O O O O O O O
O I- O U) O O O O O O O M O O O O O N O O U) r(D 0 0 0 0 M U) 0 0 0
00 00 N r 0 0 0 N O M O) U) 00 U) (D N N c') C) N ~- N O) O) N. C) CO 'IT V'
O) O
r r M CO O I- M r 'q N CO (D N ~- r N co co N N (0 M N U)
r r r
O
M M CO O) 1- N U) O) N N N O) (0 U) 00 O) N. N 'd= N 0) 00 N U) O) 't7 (D 'q V-
00
1. U) N' I' ti O) (0 M (0 U) O U7 ' M N 'T (0 r c a N (0 (0 N O r M '- (D M
U7 M Nt O (O N 'It N 1- d. O N 1- Cl) r (0 It r U7 (O O M Cl) 1- (0 _ N O 00 1-
O
O O O O r ~- r r O O r O O r - 0 0 0 O 0 0 r r 0 0 O O O O
-v
0
(6
E
co
N co D ---
1. O V M In N r 'U) N M N N. (D N 0) O N (D r r 0') (D
O) `- O O O U) 00 N 0 0 U7 0 6 4 4 0 lf) I~ u7 .CO U) CO M r r N 1- v U) O M M
q (0 U7 r'- (0 O (0
p r r r N ~- o) N r r
a)
Z
O
O c (a c c ns T w M T T T co ca ca ,. c o c c c N =Q EL c ca c p E =c c c =E c
c c c c
a`) 5 a (- c rn vi E x 2 rn E E E E o L E
2) 2) a) 'O L 7 7 O p a) (a 7 0 0 0 Ol =)I (1) O O O
Cr r r r r r N r r r r r r r r r r
O) 0) O) 0) ) O) O) O) O) r O) 0) O) O) (3) O) O) O) O) O) O) O) O) O) O) O)
M
LO
r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r r
O (nO O ) OO O ) O ) OO)O)O OO)O)O D) O O) O O) O O O) O O) d) O O) O O O)
L
'L
O)
U7
r r r r r r r r -r r r r r r r r r r r r r r r r r r r r r r r r
O O O O O) O O O O) O O O O O O O O) O O O O O O O M M M O O O) O) O)
.D
00
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
F-:
0 O O O O O 0 O O O 0
O
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 CD 0 0 0 0 0 0 0 0 0 0 0 0 0
O O O O O O O O O O O O CY) O c') O O M 0 0 0 0 0 0 0 0 c) 0 0 0 0 0
M CO q- NT
't 0) M O M U) 0) CC). C14 U) N 0) N N (D CO N - O 1I- 0) - N M O ~-- O
CD N M 00 r- N -- i,- N V V tt 1- V- N- (v). LO. (Y) N CO V itll 1- 1- N 11
r Q:
= O r- 0 0 M r- i- M M M O (M O (D M N O W O N 0
M ~- O N O O N 00 O M N ~ M O O O ~ 0 0 0 ~ M O 00 N "T M
N e- (D 0) (D - 0 1~ O =- T71-71 (D O O r (D N- -7 7 0 7 N- 7 7 r r
0 0 0 0 '- 0 0 0 0 0 0 0 0 '- 0 0 0 0'-' 0 0 0 0 0 0 0 0
O CO N M CO O 0) N rl- (D 00 M 0) 1- CO N 0) N 0) (D M N N N LO M '-;T CO
0) O ~ 00 CO N ~ (0 (D 0) 0) - N- OD O N O 0) 0) CO M N U) Iq '- (D N IT 0) N
O- M O 0) 'IT (D N N (0 O M , 1,_ , N N r- ci 0 0 - 0 O- M N- N CD, L 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 O O 0 0 0 0 0 0 0 (D
0
0) (D =- 0) O to CD 'IT O CO O 0) CO 1- CO M (D M r O M O N ~- N N
0 0 0 0 0 0 1- N M ti O N M O O CO O N 00 CD 1- ~ 1- 0 0 0 0 M
N N `7 O (D CO M qq CO Q O N M 0) N M M N - O N M O N qq m N `q m M
0 0 0 0- 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
C
0
0)
00 N- co M . 'q M CO O (D N- 0) 0) N- co M 'CO CO N Lo 'q- (D 0) CO 1~ d' 0)
CO N- O N M 00 - O (D 0) ' IT O M M- 0) O (D O (D g O M ti O V- q 0) - Iq 00
U) N CO O 0 0) (D M w .-- O 0) O q N- O 0) O e- U) N O (0 M (D ' 0 - 0)
(D O - ~- N- M O O O O O O O O O O- O N ~- O M O O tl
0)
U) M O O 0) ti It O 0) (D (0 CO N (D CD r- M 1- 'gr (D CO (D N O 0) N- CO - O
O (D M M M s-- M N- CO (D M 0- CO O IT N IcT 0) O CV) N (D M V N- N N N- M
M 0) N C) N O '- N- 1` 0) N 0'- U) 0) co O cV N 1- 'IT - ti N 0) N O
I ca
.. N O N ~- ~- O O 0 0 0 0 0 ~- - 0 0 0 0 ~- ' 0
C
a)
0
0
O CO IT M N N 0) 0) CO CO 00 00 =- 11- CO r- O ti P- 'IT N 00 co co r ti
N N O (M O 'a 00 'T (D M M cM CO 'q Iq N co tt O O M O (D 1- O N M N M O CO
O O CO N CO M O 1- M r- 'cr 0 0 0 0 0 1- O N CD M CO N CO O O
O 0 0 0 0 0 0 0 0 0 ~- O ~- O O O ~-
O N 0) 1- N U) O N N (0 U) O (0 M O 0) 0) 00 CM v 0) 1- CO O 0) M O (0 M
1'- M N M N CO 0) r- q 0) C:) 'q C0 (D (D (0 O O O 1- (D Iq (D O (0 M N N (D
0 0 0) O 0 0 0 0 (D M r- O (0 O N M O O 0) O 7 N p 7 'IT O N M m .-
. . . . . . . . . . . . .
O O O O O 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
E
E
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
0) M
_ (0
O
O
CO' N
N
O 0000 (0 Co
co
O O O
(!)
m'
2 ~. N
00 0 0 0 0 0 0 0 0 0 - 0 O O O M O O C) O 0 0- O O O
0 0 0 0 0 0 0 0- 0 0 0 0 0 0.- 0 0 0 0 0) O O O O O O
Q) 0 0 a) a) Q) Q) Q) Q) Q) Q) N 0) Q) Q) a) Q) Q) Q) Q) Q) 0) Q) Q)
U O D U U U U U U U U N U () U U U U U U N U N U U U U U U 0) 0 a)
z, C (o r- C C C C C C C C C C C C C C C C c (u c (6 C C C C C C
0 0) 0 0 0 0 0 0 0 0 0 0)0 0 0 0 0 0 0 0 0) 0 0 0 0 0 0 0 0
Z t Z Z Z Z Z Z Z Z Z Q Z Z Z Z Z Z Z Z Q Z I Z Z Z Z Z Z O
Q)
0)
-0 Co
Q a
2 w N- O M N u) '- O CO (0 0) IT CO O N ti V' Oo O) (0 ti Co (0 N 0 O N (0
LO CO ~ 00 O 00 M O N- M N- O M O CO CO CO (0 0) (0 t- O ~ O O O NT 0)
O O O O t` 00 O O 00 O to 00 CO 0) ti 0) O O) O OO O 0) O 0) O) II- O
.
O - - - N '- '- N - O ' i- '- N '- O - O '- N =- N
I-
Z
CY) Q) Q) Q) Q) (1) N CD (1) N Q) Q) () Q) 0 a) a) a) a) Q) a) a) Q) (1) N (D
Q) Q) Q) Q)
5;. > > > > > > > > > > > 2t 2t > > > > > > > > > > > > 2t > >
U U U U m U U U U U U U U U U U U U U U U U U U U U U
C0 m CQ (u m (u CO m (v m m CQ m (u (u m m m (o m (v m CQ N to (o (a (a m
Q) 0 Q) Q) 0) N Q) 0 Q) 0 0 Q) Q) Q) Q) 0 ) 0) W 0 0 0 0 ) 0 0 0 0 0 ) 0 Q) 0
0 (D a) () Q)
O O O O O O O O O O O O O O O O O
O O O O 0 0 0 0 0 0 0 0 0 O O O O O O O O O O O O O O
V O O co O Nt. O LO I- I- . 0 0 0 0 . 0 0 0 O N- O 0 co LO 0 0 0 0 0
to I- N M O to 0) LO - N O O to O '- O U) O Lo - O O r to U. O co U) O
N lql M I' N M N'- N N N N N 0 N N N N N N N 0) N N N N N N
A_ N~ M A N A N N A A A A '- A A A A N A A'- A A A N A 'IT
0 0 0 0 0 0 0 0 0 e- 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
N
a)
C
co N IT N M I- O 0) N O [t ti ao O M oo to Qo O O O) ao ao
0) 0) 0) M O M O) to 0) Cf O N O (0 N =-- ' 00 N N O Co co ti t-- O M co (0
U CY) to r r r- c- M r a- N M N O
co
L
C
0
E
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
U.) t7 00 00 M O tl) In (0 CA (f) r r lC) In In M r N ti tC) t( ) r
(0 N- O N (A N r O M r N CD (C) M r N- O ti M U) N- N- I M CD r
(0 O M O r 0 0 (C) r r r 0) O O r r co O 00 O O O - r U N N 0
O O O O O O O N N O 0 0 co O r r O 0 . 0 0 6 0 0 0 O O
0 O O O O O O O O O
a
:0
U) N- r 0) M U) U) (f) M N (0 CO U) C() CO V) U) 00 U) 0) CO to (0 CD co V u)
00 M M O 00 M N- V) M U) 00 N M O N- (0 N N 00 CO M 0) M CO 0)
00 O O O 00 CO (D r O 0 0 N (0 O cM m O cM O CD I-T O N N M O
C O O O O r 0 O O 0 r 0 0 O 0 0 O r
2 0 O 0 0 0 0 0 O O O 0 0 0 0 O O 0 0 0
0
0
0
2
a)
F a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a) a)
a) a) a) a) ((00
Z > > > > > > > > > > > > > > > > > > > > > > > > > > > > > a
U N N N N N N (0 (0 (0 N N N N N c0 (0 m (0 m m (a f0 (d m Co m m ca (0
_' O O O O a) a) a) 0 2 0 2 0 0 2 0 0 2 0 0 0 2 0 0 0 2 0 O O O
C C C C C C c C C C c C C C C C C C C c C C C C C C C C
0000000000000 1 0 0 1 0 1 00000 1 0 1 000 1 00 0 0
Z Z Z Z Z Z Z Z Z Z Z Z Z Z Z Z Z Z Z Z Z Z Z Z Z Z Z Z Z
0
0) r 1 I-t N N O U) CD 0) O N O CD U) r r 0) 00 I- M N 0) M M M Il-
N r M r N N N N M "I" r N N N N "I" N r t} "I" N N N M C") N
O O O O O O O c). CJ O O O O O O O O O O O O O O O O O O O O
a)
Z >
CY)
;Q U
a) (0
m
O
Z
;a)
2 0
M
O
z
-0 N
m 0
CO
a)
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
N
a)
C)
L
1.~
C71 (71 C a) a) a)
C
'O
I
a W d oil
O w +
O
U
Co
co I I.M. .1+1. $is 111111. .1 11111.
00 e- I- o) 0) CO N 0) I- 00 U) N I- I- N =- C') LO I- (O N 't co
V- 14,
O N O O CD - V UC) - ~- O O N N OD N- N (M 14, Cr) CO I- O t[) O O
D O N O N O O d O O O O 0 0 0 0 0 0 0 0 Cl 0 0 O- O
0 0 0 0 0 0 0 0 0 0 0 0 O 0 0 0 0 0 0 0 0 0 0 0 0 0
L
{ U'
co
iO
O
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
00 0) 0) o () o v(0 C) o v N o 0 o aD o (0 0 0 CO O) U) (0 (0 ao
r 00 (Y) N (0 O 0) O M M O '-- (D CO =- M
O to C') cp M (0 00 N- CO (0 O N O M O N
0) N (0 N N 00 (D M (0 O Iq 0) co (0 N 00 O N N
C) N- M N 0) 0) N 00 IT O M (0 O CO '- C) N N M lq- N-
N 00 0) N- N- M M N. U) 0) U) (0 IT (0 (0 0) ti N 0) 'IT
U) 0) O U) U) (D V Iq (D (0 00 N N- M cf N
J N 'It N. - CO N- U) N- N. (0 l() r IT M M CD CO
I- (D (0 0) - N. N- 00 0) O N M U) U) N (D N. M
N- M N N O O U) M O N CO (D O 00 N NT O O U)
O [i O O - 0 0'- O O ' O O O O '- O '- O
0) 0) O M O CD O CD N- (O N O 00 O M O O CO O r- M O O co M U) O co co
IT co 00 N N CO 'IT N- CO 0 I- co CD 0) I-T I- M
0) fl- I- O - O Iq O M ti M co M co M '- 0) U) co 0)
U) N N 0) ~- .- O 0) M N co co I,- co O (D O U) C
M U) co CO 0) M to co 0) U1 M "T CO '- (0
co U) I- U) (O N 0) 0) I-T M M I- N- O co U) U) O N U)
J lql (D M N (D CO U) CD N N O I~ M t- co CD - M
CD (D N I '- N LO O '- 0) N M CD I- II- U)
CO co CD 0) (0 CO O O O co U) N U) O 0) M O U) M N.
O Nr IT M 0) O N O 'IT 0) CO O d '- clj OR O (? U)
U) t CO 4 '- O '- to M r O) O co N CO co V) (0 0) v O O O O O N M M v O) M O
U) ao N U) U) O O) O N O N M
00 14' N. co 0) IT 0) 0) .- '- 0) 0) 1 O CO N I- co
r r N- co co Cl) CD N 0) IT 0) 00 M 1- C0 O (0
CO U) co O I` co 00 co I- O N- M 0) 0) CO U) 0) U) U) N
ti M (D C0 (D CO '- 0) U) M M N co 0) co N N 14- U) I-T
rt N O If I- ti CD N lq- (D d' O . 0) (D N 0) co U)
O U7 M CO I. (0 O N '[r' O O'- I- N- O CD d O O IT
J U) ti M LO O) U) M O N '7 C0 00 00 0) N. 00 N N
O M 00 q N- O CO O) O (D O) O N co N N 0)
N M N M `- O CO M U) 0) N 0' 0''- O N CO
U) N N IM CO CO 1 (0 r M CD to V - O N
(") N (0 co 0 0' O I-T O co I-T N 0 0 (D 0 co O U) O O O O O 0) 00 O O
Fl- OD CD ~- ~- O) 0) (D O 0) I- 0)
0) 0) N O O s- C) C) CD N O O N
N'- O 0) 0) N N N co co N N. 00
L U) C0 M '- (D CO U) M M co lq- 0) N M
M CO N. C0 CD 00 N N lq- O 0) O (0
O (O x- N U) U) N- '- '- (D 0) co co U)
U) U) N. I- ' - M U7 U) 00 CD co O co
co N 0) V I- ti O CO 00 M 0) co co (0
N 0) (D 00 I. ti I-, m M OR C) M
M ~- M O CO O O
M
0) O O) O O O O M 0 0 0 0 0 0 0 O ti O O O O O O O O t~
a)
N N' '- N O U) N. co
`T N (D N U) N. 0)
CO '- I,- O O c- N 0) tt N
N =- .- '- N 00 M N- M 0)
0) CO to 0) IT M 'q- O U) M
I- V O 0) N- N M - N U) U7
M M 0) N co N M N 0)
0 ( CY) O 04 N N 0 IT 0)) N
0 CD
N
d N-' M O O '- N O
O O O O 0) O O O O O O O O O O O O O O O O O O O O O O O O O
I.
CO
N CO
00
co
M
N.
O
O 0) O O O N O O O O 0 0 0 0 0 0 0 0 O O 0 O O
co I- M ti 0) 0) U) M (D (N
CU N N. N. O O N- co N- '- N I-
+-' (O 00 N. 0) 'G' I- M O I- 00 IT
a) O M N v 0) co 0) (D (D M co
O N M CO - N '- N
7 N 0) 0) U) '- (D U) 0) 0) N N- N (0 M V M O
0) 0) '- 00 N- U)
J 00 O O) - O , O 0) N N- T7
O O O O e- O '- O O O
N- (0 O O O O O 0 0 C 00) O O O O O O O M O I-T O O 0 0 0 N. O 00 N O
(a 0) co N U) U) r r 00 Nr
L N N CD CO 0 00 0 M CO 00 (0
00 N. M 00 Nr 0) U) N. N
r- C.0 CY) m CD r- cy) a) 14-
00 - N OD G) 0) P- 14- U)
N O r- CM O N oo 0) CD 00
J 0) U) N
T7
N O M O O O O
0
.Z
v
0
a)
M '~ U7 CO r CO 0) O '- "Vr- M U) (O N. 00 0) O '- N M `ar U) CO I) 00 0) O r-
N M V
I ,;r U) U) U) U) O 10 U) U) U) (D (0 (D (0 O
.4T- 14, M M M M M M q
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
a> a> a> a) a> a) a> a> a> > a> a> a> > a> a> a> a> a> a> > a> (> a) a> a) >
a) a> a> a> a>
> > > > > > > > > .. > > > =.. > > > > > > .. > > > > > .. > > > > >
N N N N N N N N N O) N U) U) 0) N N N N N v) O) N N N N N O) N N N N N
x` 0 0 0 0 0 0 0 0 0 a> O O O a> 0 0 0 0 0 0 a> O O O O O a> 0 0 0 0 0
a M 0 a a a a a z m a a z a a m a. a a z a a a a M z a. a a M a.
0 0 0 O O O O 0) 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 O F Z-0
O CD N N
CO CD (D
N N 0) CD
LL M _ 000 to
00 (D 00 CY) N N
r r O
O O CD 0 0 0 0 0 CO 0 0 0 0 0 0 0 0 0 I,- O M O O O O O O O O O
(C> O LO r- co co
'IT CC) 'IT M
- (0 Co lq- (D
(0 co 000 co N-
- I LO LO 00 'IT
LLI I-T M 0)qqT
0) N- 0) ((0
Z CA CO I- N O
CO O N
CD N- O O O O O O(0 O 0 0 0 0 0 0 0 co 0 0 0 CO 0 O CCD M O O O O
E U) Ct j CCD ti 000 ONO I-
E O 00 Co co ~ M
00 '- M CD N M r-
tC) 00 'IT I` (D O 00
Z
U- C6 't N O )) M to
N co M N O co C) 0) M I- N N- 0) (D 00 O 0) C) O 0 It O C) I- V CO rtD co CO -
O C) co co M CD M M co N I- - - M CO O O 0) - M Cl)
co (3) O) N- N It CO N M CD N O 00 (D N- 0 O) I- CC) IT ti to c- O 0)
I- I- N- M M CO (D M (D '- I- 00 '- I- co N co 00 N- M N - ti (Cl
m O M 00 M O M M N OO '- t('> 00 CO O) N M O CO O O Cn '- CO N CO CC) M N It -
Iq CC) N I- M CO O U) N- (O LO 'q 'IT N- to '- O
00 't 00 (D - It LO co CA Co M Cn (D M C) M N- N OD r N 00 N- '- N-
J O M N ql O) LO - N M 00 M U) N N to qt 00 - CD M OD N- CD CO 00
O (D O) O) 00 It O 1 P M U) I- CO co to CD tri C r N- N O O) M N
N'- M N O M O) c) -q c) CD 00 0) M Co d' CA N M (O I` CO
N I- r v U) N Co - v O) N- "M v ' - MM I N -1(01
M I, N I- O C) O M M CD 0 0 0 0 0 0 0 O) 0 0 0 O 0 0 0 0 O O O
co N- O 0 0 0 N LO N C:) IT C)) 0 co 0 O
N- tC) N '- N- CD O I -
M N Co O O) O CD
r CD M C) U) O) Co Co CD N-
O CD CO I- I- C) N t0
J CC) 0) O) V- O) O CD 'T to
CD '- O 00 CO CD N N lq- U)
N U) N- O M r C) C) N CO
0 ~-- N =- co CD '- O O
I- LO M O O Ict ti M I~ (D 0 ) CO M O O I IT N Iq d CO co n N CO CO I -
N I- O O O) h O IT N'- CD O 00 1 (O I- M CC) N to N I- I- I- U)
co N I- Co fl- I - co co M CD CO U) M 'I C) O O I- M co M C co M co
to v 'IT m O M I- N CD N CO O (n '- O) CO M 'q- N'- O 00 N- N
tf) T O O) CO '- CD OD N CA U) LO C ) CO co co N O M CD M qql N I- N- CD d'
r- (D C) U) O I- I- i- 0 CO N CC) O I- I- M O O) O) - - U) r CA CD CO r-
M lq- CO O M N '- I- O) Iq ' to (O to `7 N- CO N N- U) V CO N
J CO I I` O U) (O IT Iq I- co N- I- CD CO 0- ICT O) CA 00 I-T O CD CA co
0 CO N to M N CO V O) CO O U) r U) '- I- - CO lqr Iq CO CO - q-
(0 (O N O N U) CO M Iq m C) O C) I- co N O) co M M N M N M
'V O CD N t O O N '- CD O 14 O U') M tD O N O M U) O O M
to U) to co 0 0 0 O co 0 to O O M O O O O M O C) 0 0 0 0 0'- O CD N
N =- CO N Co CO I- O O N N N M
N co co C) lqr to CO N- CA O N CO CO
O O 0) N W) N (D N- - - O IC) '-T
(Yj' U) V- O U) I - O) CA N I- tt'> I- (D I~
00 O) M O 0 co N co O) O) N I-
0 O O (D r- LO CA V) M CD N O
J co 'IT O (D O) IT N U) CO CO CD co N
N W) N O O co O CC) CD
CD co co I- N N- CO
m O O N '- ' 1'-1 I O
to '- CO CO O 0 M O O CO N '7 O IT N 0 0 0- 0 =- I- O O CD O co (D CD (D
C) N I- O) I- ti CA N N I,- M U) CO N- O
CO C) I- O O U) O U) N O co N CO N C) O CD
0 C) CO CO N IT O M Iq O co N C) O C) I- to '- CO
N O) N- O N co N- Cn CO O) CO s- i- M O) CO Il-
N CD co to co O O) N d N- CD ~ N M IT 1- CD I- U)
co co O 00 00 - N- N N co M LO U) O) 'IT CO r-
C) N co I- N- N N M CO C) co (D N Ih N to co
O co M M N 00 M co CD O (D O -- O I,- U') IT M
to CO O GO CO M '- I- co N I- N- OO N N (D
O M N '7 C'') N- N O O ~-- 041 1111, O O 1i-1 I O O
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
0 C) 0 0 0 o r o 0 0 0 0 0 0 0 0 0 0 0 o r o 0 0 0 0 0 0 0 0
O M CO C) M (V) CO CO M r r' M CO) O C=r) M co CV) co CV) CV) C1) m m Cr=) CV)
O co co M
co
aIt LO CO to LO LO LO to lO LO N N N N LO LO 0 CO CO M CO Cfl M M M 'LO CO "T
Ln
O 00 C70 W O 00 M O O O O M M N~ C10 00 O CO O M O M tO C1D to r C1O C10 N M M
fan.
LO
a
>
0)
c c
ci, MMai
W W
O O O O O O O 0 O 0 0 0 0 r 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 r O O
>
V- rl- O) CO r- O O) CO 0) I- 0) CO r CO 14- O M Irt O O N CO (0 0) ~-- O CO I-
m CO M
m co N co r) CV) M r N N C'O N C+) N M CO CO CV) CV) CV) C'M CV) CV) M IT N N
N N M co
~=- r r O r O O r O O r r r r r O r r r r r O r r r O r r O r r 0
>
~p ~p a~pp
N> N N N N 0 0 N N 0) g U CU 0 N N 0 O N N N N N 0 N N N N
CU >
U) a) V) V) U) V) N V) 'in- Fn 7 Cl) 3 7 Cl) O U) U) U) V) V) N N U) Cl) 'Fn
U) N V) V) 0)
O w 0 0 0 0 0 0 0 0 0" 0{7 0' 0 0 0 0 0 0 0 0 0 0 0 0" 0 0 0 0 0 Q)
Q- z a a a a a a a a w a w w a a a a a a a a a a a w o-
a a a a z
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
O) 00 Ul) co M (0 cf) N M M 'IT
O O O N O O O O O N O 0 0 0 ~- O N N N O O
CU:
N N N O N O O N N N O N N O O O N O O N N N N N N N N O
LO
ti U) co (D r) co
C'~)
N.
to:
Co
O O O O O O O N N O O N ~- O O N O O O O O N N O
cm
w
N
N O O -' ~- - O O N O O O CV ~-- ~-- O O- N- O N-- O O N N O
3'Cu
U')
0 0 0 0 0 0 0 0 0 0 0 O 0 0 0 0 0 O O O O O O 0 0 0 0 0 0 0
Cd
f~.
C : ) O O O O O 0 0 0 0 ~- ~- 0 0 0 0 0
CJ
CA 02710574 2010-06-23
WO 2009/061514 PCT/US2008/012658
M O N co N (D N. I- r 'ct N r M O O 0) N- U) N Cn ti N U) N N ti N O N- O CD N
r N r ~-- r ~- r ~-- r r r r r r r r r r r ~- r r N r r r r r
F;M
cr)
U) U) U) U) U) M v7 M co to U) U) U) M U) U) "t U) U) CO U) M M U) CO U) M
Co
co
M.
0 0 0 0 0 0 0 0 0 O O O O O O O O O O O O O O O O 0 0 0 0 0 0 a
(D
O O O O O O O O O O N O O O O O O O O O O O O N O O O O O N O O
CO
0)
to
N
O O O O O O O O O O N O- O N O O O O O O O O N O O O 0 N N O O
0)
(6
LO
M co
Q
i N
O O O O O O r O O r N O 0 0 0 0 0 0 0 0 0 0 0 N O 0 0 0 0 N 0 0
Co
rl-
0 0 0 0 0 0 0 O O O N O O O N O O O O O O O O N O O O N N O 0 0
CO
O O O O O O O O O O N O O O O O O O Q O O O O N O 0 0 0 0 O O O
(U
U)
UN)
-- I I I i I I I I I I I I I I I I -LL
DEMANDE OU BREVET VOLUMINEUX
LA PRRSENTE PARTIE DE CETTE DEMANDE OU CE BREVET COMPREND
PLUS D'UN TOME.
CECI EST LE TOME 1 DE 2
CONTENANT LES PAGES 1 A 391
NOTE : Pour les tomes additionels, veuillez contacter le Bureau canadien des
brevets
JUMBO APPLICATIONS/PATENTS
THIS SECTION OF THE APPLICATION/PATENT CONTAINS MORE THAN ONE
VOLUME
THIS IS VOLUME 1 OF 2
CONTAINING PAGES 1 TO 391
NOTE: For additional volumes, please contact the Canadian Patent Office
NOM DU FICHIER / FILE NAME:
NOTE POUR LE TOME / VOLUME NOTE:
