Note: Descriptions are shown in the official language in which they were submitted.
CA 022~8082 1998-12-10
W O 97/47271 PCTrUS97tlO238
CELLULAR VACCINES AND IMMUNOTHERAPEUTICS
AND METHODS FOR THEIR PREPARATION
~.T,~TED APPLICATION
The present application claims the priority benefits of U.S.
provisional application 60/019,639, filed June 12, 1996.
- FI~T~n OF TH~ INVENTION
The present invention provides a method for enhancing the
immunogenicity of weakly immunogenic or non-immunogenic cells in
order to provide the immune system with an immunogenic signal
capable of stimulating T cell activation leading to an effective
immune response. The method of the invention generates cellular
vaccines which are useful for the prevention and treatment of
diseases which develop and/or persist by escaping the immune
response triggered by T cell activation. Such diseases include,
for example, all cancers, natural and induced immune-deficiency
states, and diseases caused by infections with a variety of
pathogens.
R~CKGROUND OF T~ INVENTION
U.S. patent 5,484,596 by Hanna et al. describes using tumor
tissue as a vaccine. U.S. Patent 4,844,893 by Honsik et al.
describes arming IL-2-activated leukocytes with Mabs directed to
antigens preferentially expressed on tumor cells for killing the
target cells. Both patents are incorporated by reference herein.
Anti-tumor immune responses are mediated primarily by T
CA 022~8082 1998-12-lo
W O 97/47271 PCT~US97/10238
lymphocytes. Down regulation of both the major
histocompatibility complex ~MHC) and the molecules that
costimulate the immune response is associated with defective T
cells activation signaling by tumor cells (Luboldt et al., Cancer
3~_ 56:826-830, 1996; L. Chen et al., 1992, Gell 1: 1093; P.S.
Linsley, J. A. Ledbetter, 1993, Ann. Rev. Immunol 11: 191; G.J.
Freeman et al., 1993, Science 262: 909; C.H. June et al., 1994,
Immune. Today 15: 321; J. T. Gerge et al., 1993, Cancer Res. 53:
2374; Ostrand-Rosenberg, 1993, S. Curr. Opin. Immune. 6: 772; B.
E. Elliot et al., 1989, Adv. Cancer Res. 53: 181).
T cell receptor (TCR) recognition of MHC-bound antigen is
not a sufficient signal for T cell activation. Costimualtory
molecules, such as B7-1 and B7-2, are cell surface proteins of
antigen presenting cells (APCs), and other cells targeted by the
immune response, that provide critical signals for T cell
activation (for review, see L. Chen et al., 1995, Immunol. Rev.
145: 123i T. Tykocinski et al., 1996, Am. J. Path. 148: 1). B7
signaling via the T cell surface molecule CD28 appears to be the
major costimulatory pathway for T cell activation. However,
recent studies show that costimulation is a more complex event
which involves both cytokines and adhesion molecules (G. Yang et
al., 1995, J. Immune. 154: 2794; M. Kubin et al., 1994, J. Exp.
Med. 180: 211; Y. Li et al., 1996, J. Exp. Med. 183: 639).
Many approaches have been used to enhance the immunogenicity
of tumor cells (see, for example, the references cited in this
section). The major approaches presently under investigation
involve gene transfer. In this regard, most of the methods
CA 022~8082 1998-12-lo
W O 97/47271 PCTrUS97/10238
employed to date have required ex vivo or in vivo transfection
with genes such as MHC or B7, or modification of tumor cells with
antigen presenting cells (APCs) (Y. J. Guo et al., 1994, Science
263: 518; M. Tykocinski, 1996, A. J. Path. 148: 1; J. Young and
K. Inaba, 1996, J. Exp. Med. 183: 7; L. Zitvogel et al., 1996, J.
Exp. Med. 183: 87; C. M. Celluzzi et al., 1996, J. Exp. Med. 183:
283). These approaches are time consuming and problematic
- because of the poor transfectability of primary tumor cells and
because of the requirement for large numbers of APCs.
In vitro treatment of tumor cells with cytokines increases
the expression of MHC and adhesion molecules (R. Mattsson et al.,
1992, Biol-Reprod 46: 1176; R. J. Ulevitch et al., 1991, Am. J.
Pathol. 139: 287; F. Willems et al., 1994, Eur. J. Immune. 24:
1007; I. Saito et al., 1993, J. Clin. Lab. Anal. l: 180; R. A.
Panettieri et al., 1994, J. Immune. 154: 1358; M. Ikeda et al.,
1994, J. Invest. Dermatol. 103: 791). Transfection of tumor cells
with MHC, B7-1 and B7-2 genes converts low immunogenic tumor cell
lines to immunogenic cell lines (S. E. Townsend and J. P.
Allison, Science 259: 368; J. P. Allison et al., 1995, Curr.
Opin. Immune. 7: 682; G. Yang et al., 1995, J. Immune. 154: 2794;
M. Kubin et al., 1994, J. Exp. Med. 180: 211). Non-immunogenic
tumor cells are not responsive to transfection with the B7 gene
alone but can become responsive by co-expression of CD48
molecules at the cell surface (Y. Li et al., 1996, J. Exp. Med.
183: 639).
The costimulatory molecule B7 can under some circumstances
deliver a negative signal through its binding to CTLA-4, a second
CA 022~8082 1998-12-10
WO97/47271 PCT~S97/10238
receptor for B7 on T cells. Cross-linking CTLA-4 molecules in
vitro has been shown to inhibit T cell proliferation.
Furthermore, mice deficient in CTLA-4 develop severe T cell
proliferative disorders (K. Kawai et al., 1993, Science 261: 609;
J. P. Allison, M. K. Krummel, 1995, Science 270: 932; J.M. Green
et al., 1994, Immunity 1: 501). A recent report showed that the
introduction of anti-CTLA-4 monoclonal antibody (MAb), which
- blocks CTLA-4 mediated signaling, resulted in enhanced T cell-
dependent rejection of tumors in certain mouse models (D.R. Leach
et al., 1996, Science 271: 1734). These data provide evidence
that CTLA-4 may be counter-regulatory to the CD28 costimulatory
signal. Thus, transfected tumor cells expressing B7 molecules
may fail to elicit effective immunity due to CTLA-4 mediated
negative signaling.
In addition to T cell activation using B7 gene transfection,
bispecific monoclonal antibodies (Bi-MAbs) in combination with
pre-stimulated lymphocytes have been used to induce T cell
activation under certain circumstances. For example, one study
reports that costimulatory signals can be delivered by a
combination of Bi-MAbs to CD28:CD30 (CD30 is a Hodgkin's tumor-
associated antigen) and CD3:CD30 in combination with peripheral
blood lymphocytes (PBLs) pre-stimulated with the CD3:CD30 Bi-MAb
in the presence of CD30+ Hodgkin's tumor-derived cells; however,
the combination of CD28:CD30 and CD3:CD30 Bi-MAbs alone did not
induce significant in vitro cytotoxicity of resting human PBLs
against a Hodgkin's tumor-derived cell line, and stimulation with
the CD28:CD30 Bi-MAb alone was not effective (C. Renner et al.,
CA 022~8082 1998-12- 10
WO 97/47271 rCT/US97/10238
1994, Science 264: 833). Similarly, regression of ~odgkin's
derived tumor xenografts was observed only when both the
CD28:CD30 and CD3:CD30 Bi-MAbs were used in combination with PBLs
prestimulated in vitro with CD30+ cells and CD3:CD30 Bi-MAb; no
significant effect was observed in xenografts treated with either
of the Bi-MAbs alone, or a combination of the two Bi-MAbs without
prestimulated human PBLs (Renner et al., supra).
SUMMARY OF THE INV~NTION
The present invention features immunogenic tumor cells and
other immunogenic autologous cells, convenient methods of making
such immunogenic cells, methods of using such immunogenic cells
to activate or enhance immune response against diseased cells
with minimum effect on normal or healthy cells, and methods of
avoiding the negative T cell signaling pathway.
The present invention provides a method for enhancing the
immunogenicity of weakly-immunogenic or non-immunogenic cells,
resulting in a cellular vaccine that can stimulate T cell
activation, which in turn leads to an effective immune response
against diseased cells. The cellular vaccines of the present
- invention can be used as vaccines to prevent diseases and as
immunotherapeutics to treat diseases.
In summary, the method of the invention involves the steps
of (1) treating weakly- or non-immunogenic autologous cells
2~ (target cells) in order to amplify primary and costimulatory T
cell activation signals in the cells, and (2) attaching to the
treated cells a substance capable of binding to one or more
CA 022~8082 1998-12-10
W O 97/47271 PCTrUS97/10238
antigens on the treated cells and to one or more T cell
activation costimulatory molecules on the surface of T cells
(such as CD28), thereby providing the treated cells with the
capacity to physically link to T cells and to activate the
costimulatory signal. Such substances include, but are not
limited to, bispecific monoclonal antibodies (Bi-MAbs) targeted
to antigen on the treated cells and to CD28 and/or other
costimulatory molecules on T cells. The first step may be
skipped when the autologous cell is attached with (1) a bridge
molecule with two or more binding sites for T cell activation
costimulatory molecules on the surface of T cells, or (2) two or
more bridge molecules each with one or more binding sites for T
cell activation costimulatory molecules on the surface of T
cells.
Once the primary and/or costimulatory T cell activation
signals in the target diseased cells have been amplified by
cytokines or other means and the bridge molecules have been
attached to the target diseased cells, the cytokines and the
bridge molecules not attached to the target diseased cells may be
removed from the immunogenic composition before the target
diseased cells are administered to a patient. This additional
step minimizes adverse effects associated with administering
cytokines to a patient. It also minimizes the risk associated
with allowing bridge molecules not attached to a target diseased
cell into a patient, an event which may cause unwanted immune
response against normal or healthy cells.
The first step of the method up-regulates antigen processing
CA 022~8082 lsss-l2-lo
WO97/47271 PCT~S97/10238
capacity within the treated cells and amplifies the expression of
cell surface molecules involved in T cell activation. The second
step provides the treated cells with a means to physically bridge
to T cells via CD28 and/or other costimulatory molecules, thereby
providing optimal conditions for stimulating T cell activation.
Thus, in a first aspect, this invention features an
immunogenic composition for administration to a patient mammal
(including a human) having target diseased cells. The
immunogenic composition contains an autologous target diseased
cell which differs from the diseased cells in the patient in that
it processes and presents antigens characteristic of the diseased
cells more effectively. For example, the autologous target
diseased cell expresses one or more primary (e.g., MHC) and/or
costimulatory (e.g., B7-l and B7-2) T cell activation molecules
at a higher level (e.g., 50~ higher, preferably 2 folds higher,
more preferably lO folds higher). As described below, there are
different ways of enhancing the expression level of the primary
and/or costimulatory T cell activation molecules.
In addition, the autologous target diseased cell has
attached thereto one or more bridge molecules. Each bridge
molecule has one or more binding sites for one or more
costimulatory molecules on the surface of effector cells, which
include, but are not limited to, T cells, NK cells, macrophages,
LAK cells, B cells, and other white blood cells. Preferably,
though not required, the bridge molecules have one or more
binding sites for one or more antigens on the surface of the
target diseased cell and are attached to the target diseased
CA 022~8082 l998-l2-lO
W O 97/47271 rCT~USg7/10238
cells at the cell surface antigens. In another preferred
embodiment, substantially all (e.g., ~80~, preferably ~90~, more
preferably >95~) the bridge molecules in the immunogenic
composition are attached to the autologous target diseased cells
so that the composition is substantially free of bridge molecules
not attached to a target diseased cell. In a further preferred
embodiment, the immunogenic composition contains a
pharmaceutically effective amount of the target diseased cells
with bridge molecules attached thereto.
By "immunogenic" is meant the ability to activate the
response of the whole or part of the immune system of a m~m~ 1,
especially the response of T cells.
By "autologous" is meant that the target diseased cell is
from the patient mammal, or from another patient having a common
major histocompatibility phenotype. An autologous target cell
may be obtained from the patient mammal or another source sharing
the same MHC with methods known to those skilled in the art.
By "target diseased cell" is meant a cell causing,
propagating, aggravating or contributing to a disease in a
patient mammal. Target diseased cells include, but are not
limited to, tumor cells (including unmodified tumor cells, tumor
cells modified with different approaches, and primary culture).
The sources of tumor cells include, but are not limited to, liver
cancer, hepatocellular carcinoma, lung cancer, gastric cancer,
colorectal carcinoma, renal carcinoma, head and neck cancers,
sarcoma, lymphoma, leukemia, brain tumors, osterosarcoma, blade
carcinoma, myloma, melanoma, breast cancer, prostate cancer,
CA 022~8082 1998-12-10
WO97/47271 PCT~S97/10238
ovarian cancer, and pancreas carcinoma.
Target diseased cells may also be cells infected with prions
(which cause Mad Cow diseases among others), viruses, bacteria,
fungi, protozoa or other parasites (e.g. worms).
Viruses include those described or referred to in Fields
V;rology Seco~ t;on, l990, Raven Press, New York,
incorporated by reference herein. Examples include, but are not
- limited to, herpes virus, rhinoviruses, hepatitis virus (type A,
B, C and D), HIV, EBV, HPV, and HLV.
Bacteria include those described or referred to in Bergey's
~nual of Deter~;n~tive Bacteriology N;nth Editlon, 1994,
Williams and Wilkins, incorporated by reference herein. Examples
include, but are not limited to, gram positive and negative
bacteria, streptococci, pseudomonas and enterococci,
Mycobacterium tuberculosis, AeromonAs hydrophilia, Aeromonas
caviae, Aeromonas sobria, Streptococcus uberis, Enterococcus
faecium, Enterococcus faecalis, Bacillus sphaericus, Pseudomonas
fluorescens, Pseudomonas putida, Serratia liquefaciens,
Lactococcus lactis, Xanthomonas maltophilia, Staphylococcus
simulans, Staphylococcus hominis, Streptococcus constellatus,
Streptococcus anginosus, Escherichia coli, Staphylococcus aureus,
Mycobacterium fortuitum, and Klebsiella pneumonia.
Primary T cell activation molecules include MHC class I, MHC
class II and other molecules associated with antigen processing
and/or presentation. Costimulatory T cell activation molecules
include ICAM-l, ICAM-2, ICAM-3, LFA-l, LFA-2, VLA-l, VCAM-l, 4-1-
BB, B7-l, B7-2, and other cell adhesion proteins and other cell
CA 022~8082 1998-12-10
W O 97/47271 PCTrUS97/10238
surface proteins which can activate T cell costimulatory pathways
through T cell surface proteins.
By "bridge molecule" is meant a molecule or substance which
can bring two or more cells together by attaching to the cells
with its binding sites. Preferably, a bridge molecule can bring
an autologous target diseased cell together with an effector cell
and deliver a signal to the effector cell to activate or enhance
the effector cell's immune response against the target. A bridge
molecule has one or more binding sites for stimulatory and/or
costimulatory molecules on the effector cells. These binding
sites can be designed to activate a positive regulator of T cell
activation (e.g., CD28, 4-lBB~ but avoid stimulating a negative
regulator of T cell activation (e.g., CTLA-4). The binding sites
can also be designed to blockade a negative regulator of T cell
activation (see Leach et al., Science 271:1734-1736, 1996). A
bridge molecule may also have one or more binding sites for
antigens on the surface of the target diseased cell. Bridge
molecules include, but are not limited to, bispecific monoclonal
antibodies, fusion proteins, organic polymers, and hybrids of
chemical and biochemical materials. The antibodies described or
disclosed in U.S. patents 5,601,819, 5,637,481, 5,635,602,
5,635,600, 5,591,828, 5,292,668 and 5,582,996 are incorporated by
reference herein.
The antigen on the target cell serving as an anchor for the
bridge molecule need not be unique to the target cell when the
bridge molecule is attached to the target cell in vitro. Any
molecule on the target cell surface can be used to anchor the
CA 022~8082 1998-12-lo
W O 97147271 PCT~US97/10238
bridge molecule, including, but not limited to, proteins,
glycoproteins, lipids, glycolipids, phospholipids, lipid
aggregates, steroids, and carbohydrate groups such as
disaccharides, oligosaccharides and polysaccharides (see
"Molecular Biology of The Cell," pp47-58, pp276-337, Second
Edition, published by Garland Publishing, Inc. NY & London).
Examples include transferrin receptor, Low Density Lipoprotein
~LDL) receptor, gp55, gp95, gpll5, gp210, CD44, ICAM-l, ICAM-2,
collagen and fibronectin receptor, transferrin receptors, Fc
receptor, and cytokine receptors.
Costimulatory molecules on the surface of effector cells may
- be antigens, fatty acids, lipids, steroids and sugars that can
stimulate or costimulate these effector cells' functions to
destroy the target cells. Costimulatory molecules include, but
are not limited to, CDla, CDlb, CDlc, CD2, CD2R, CD3, CD4, CD5,
CD6, CD7, CD8, CD9, CD10, CDlla, CDllb, CDllc, CDw12, CD13, CDl4,
CDl5, CD15s, CDl6a, CDl6b, CDw17, CD18, CDl9, CD20, CD21, CD22,
CD23, CD24, CD25, CD26, CD27, CD28, CD29, CD30, CD31, CD32, CD33,
CD34, CD35, CD36, CD37, CD38, CD39, CD40, CD41, CD42a, CD42b,
CD42c, CD42d, CD43, CD44, CD44R, CD45, CD45RA, CD45RB, CD45RO,
CD46, CD47, CD48, CD49a, CD49b, CD49c, CD49d, CD49e, CD49f, CD50,
CD51, CD51/61 complex, CD52, CD53, CD54, CD55, CD56, CD57, CD58,
CD59, CDw60, CD61, CD62E, CD62L, CD62P, CD63, CD64, CDw65, CD66a,
CD66b, CD66c, CD66d, CD66e, CD67, CD68, CD69, CD70, CD71, CD72,
CD73, CD74, CDw75, CDw76, CD77, CDw78, CD79a, CD80, CD81, CD82,
CD83, CDw84, CD85, CD86, CD87, CD88, CD89, CDw90, CD91, CDw92,
CD93, CD94, CD95, CD96, CD97, CD98, CD99, CD99R, CD100, CDwlOl,
.. ..
CA 022~8082 1998-12-10
WO97/47271 PCT~S97/10238
CD102, CD103, CD104, CD105, CD106, CD107a, CD107b, CDw108,
CDwlO9, CDllO-CD114, CD115, CDw116, CD117, CD118*, CDll9, CD120a,
CD120b, CDwl21a, CDw121b, CD122, CD123*, CDw124, CD125*, CD126,
CDw127, CDw128, CD129, CDw130, LFA-l, LFA-2, LFA-3, VLA-l,VCAM-l,
VCAM-2, 4-lBB, cytokine and chemokin receptors. In a preferred
embodiment, the bridge molecule has a binding site for CD28 or 4-
lBB on the surface of T cells.
By "pharmaceutically effective" is meant the ability to
cure, reduce or prevent one or more clinical symptoms caused by
or associated with the diseased cells in the patient mammal,
including, but not limited to, uncontrolled cell proliferation
bacteria infection, and virus infection.
The immunogenic composition may be isolated, enriched or
purified for administration to a patient.
By "isolated" in reference to the immunogenic composition is
meant that the autologous target diseased cell is isolated from a
natural source. Use of the term "isolated" indicates that one or
more naturally occurring materials have been removed from the
normal environment. Thus, the target diseased cell may be placed
in a different cellular environment or in a solution free of
other cells. The term does not imply that the target diseased
cell is the only cell present, but does indicate that it is the
predominate cell present (at least 20 - 50~ more than any other
cells) and is essentially free (about 90 ~ pure at least) of
other tissues naturally associated with it in the body of the
patient. In a preferred embodiment, the composition is
substantially free of effector cells such as T cells. In another
CA 022~8082 1998-12-lo
W O 97/47271 PCT~US97/10238
preferred embodiment, the composition is substantially free of
bridge molecules not attached to a target diseased cell. In a
third preferred embodiment, the composition is substantially free
of cytokines outside of the target diseased cell.
By '~enriched'~ in reference to the immunogenic composition is
meant that the autologous target diseased cell constitutes a
significantly higher fraction (2 - 5 fold) of the total cells in
the composition than in the diseased tissue in the patient's
body. This could be caused by a person by preferential reduction
in the amount of other cells present, or by a preferential
increase in the amount of the specific target diseased cells, or
by a combination of the two. However, it should be noted that
enriched does not imply that there are no other cells present,
just that the relative amount of the cell of interest has been
significantly increased in a useful manner. The term
"significantly" here is used to indicate that the level of
increase is useful to the person making such an increase, and
generally means an increase relative to other cells of about at
least 2 fold, more preferably at least 5 to 10 fold or even more.
By "purified" in reference to the immunogenic composition
does not require absolute purity (such as a homogeneous
preparation); instead, it represents an indication that the
target diseased cell is relatively purer than in the natural
environment. The target diseased cells could be obtained
directly from the patient or from cell culture, with or without
modifications. Purification of at least one order of magnitude,
preferably two or three orders, and more preferably four or five
CA 022~8082 1998-12-lo
W O 97/47271 PCT~US97/10238
14
orders of magnitude is expressly contemplated. In a preferred
embodiment, the composition is substantially free of effector
cells such as T cells.
The immunogenic composition may contain a pharmaceutically
suitable carrier or excipient. Techniques for formulation and
administration may be found in Remington's Pharmaceutic~l
Sciences, 18th ed., Mack Publishing Co., Easton, PA (1990). The
immunogenic composition may be administered to a patient
systemicallyj e.g., by intravenous infusion or subcutaneous
injection. A composition of the invention may be administered as
a unit dose to a patient m~mm~ l, each unit containing a
predetermined ~uantity (e.g., about lx105 to about lx101~,
preferably about lx106 to about lxlO9, and more preferably about
lx107 to about lx108) of armed and/or activated autologous target
diseased cells calculated to produce the desired therapeutic
effect in association with the physiologically tolerable aqueous
medium as diluent.
The expression of primary and costimulatory T cell
activation molecules may be enhanced by various means, for
example, in vi tro, ex vivo or in vivo treatment of target cells
with cytokines or other factors capable of inducing the desired
amplification; and in vitro and in vivo transfer to the target
cells of MHC genes, adhesion molecule genes, cytokine genes,
and/or their respective transcription activators or enhancers.
Cytokines include those described or referred to in The Cytok;ne
Handbook, Thomson, A., (ed.), 1994, Academic Press, San Diego,
incorporated by reference herein. Examples include, but are not
CA 022~8082 l998-l2-lO
W O 97/47271 PCTAUS97/10238
limited to, IL-l, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9,
IL-10, IL-ll, IL-12, IL-13, IL-14, IL-15, IL-16, IL-17,
inteferons (e.g., IFN ~, ~, and y), tumor necrosis factors (e.g.,
TNF ~, and ~) and other chemokines and lymphokines. In a
preferred embodiment, IFN y and TNF ~ are used either alone or in
combination to enhance the expression of primary and
costimulatory T cell activation molecules in autologous target
diseased cells.
When a target diseased cell coated with bridge molecules is
administered into a patient, it will bind to costimulatory
molecules on the surface of the effector cells. The more densely
the target diseased cell is coated with bridge molecules, the
more effector cells it will be able to bind. In addition, the
more binding sites a bridge molecule has for the costimulatory
molecules, the more effector cells it will be able to bind.
In that regard, Applicant has found that a cellular vaccine
may be prepared without the need of cytokine treatment (to
increase the levels of primary and costimulatory T cell
activation molecules) when a plurality of bridge molecules are
attached to a target cell with binding sites for two or more
different costimulatory molecules on the surface of T cells.
Individual bridge molecules may be attached to different anchor
molecules on the surface of the target diseased cell. An
individual bridge molecule may also have two or more binding
sites for two or more different costimulatory molecules on the
surface of T cells.
Thus, in a second aspect, this invention features an
CA 022~8082 1998-12-10
WO97/47271 PCT~S97/10238
16
immunogenic composition containing an autologous target diseased
cell having attached thereto (a) a bridge molecule which has two
or more binding sites for two or more different effector cells,
(b) a bridge molecule which has two or more binding sites for two
or more different costimulatory molecules on the surface of
effector cells, (c) two or more bridge molecules each containing
a binding site for a different effector cell, (d) two or more
- bridge molecules each containing a binding site for a different
costimulatory molecule on the surface of effector cells, (d) two
or more bridge molecules each attached to a different antigen on
the target cells, or (e) a combination of two or more of the
above.
A pharmaceutically effective amount of an immunogenic
composition of this invention may be complemented by a
pharmaceutically acceptable carrier before administration to a
patient mammal.
Alternatively, a patient may be administered with a
pharmaceutical composition containing (l) a pharmaceutically
effective amount of a cytokine capable of increasing the level of
one or more primary and costimulatory T cell activation molecules
in tumor cells, (2) a pharmaceutically effective amount of a
bridge molecule containing a binding site for an antigen on the
surface of the tumor cells and a binding site for a costimulatory
molecules on the surface of T cells, and (3) a pharmaceutically
acceptable carrier.
In treating a patient, the autologous target cell may be
treated with cytokines or other means of increasing primary and
CA 022~8082 1998-12-lO
W O 97/47271 PCTrUS97/10238
costimulatory T cell activation molecules in vitro before the
target cell is administered to the patient. Alternatively, the
cytokines may be administered to the patient to increase primary
and costimulatory T cell activation molecules in vivo.
In a third aspect, this invention features a method of
generating cytotoxic leukocytes against diseased cells in a
patient mammal by contacting a population of effector cells
(e.g., white blood cells) in vitro with immunogenic compositions
described above for a time period sufficient to react with the
immunogenic compositions and collecting the treated effector cell
population. The cytotoxic leukocytes so generated can then be
administered to a patient to treat or prevent diseases.
The method of the invention is useful for the prevention and
treatment of diseases which develop and/or persist by escaping
immune responses triggered by T cell activation. Such diseases
include, for example, all cancers, natural and induced immune
deficiency states, and diseases caused by infections with a
variety of pathogens. The method of the invention is illustrated
herein by demonstrating its application to three different types
of human cancers. Cancer cells are by nature generally weakly
immunogenic, fail to trigger an effective T cell response, and
survive and grow as a result. As demonstrated herein, cancers
can be prevented, and established cancers may be cured, by
stimulating an effective T cell response using autologous tumor
cell vaccines of the invention.
Other features and advantages of the invention will be
apparent from the following detailed description of the
CA 022~8082 1998-12-10
WO97/47271 PCT~S97/10238
18
invention, and from the claims.
BRIF.~ D~CRIPTION OF T~ FIGU~
FIG. 1. Expression of MHC class I, ICAM-l, ICAM-2 and
VCAM-l antigens on cytokine treated hepa 1-6 cells. The results
are representative data from four comparable experiments.
FIG. 2. Stimulation and proliferation of syngeneic splenic
T cells in vitro induced by cytokine treated hepa 1-6 cells and
anti-CD28 MAb.
FIG. 3. Cytotoxicity of CTLs generated by in vitro priming
of naive splenic T cells with cytokine treated hepa 1-6 cells in
- combination with anti-CD28 Bi-MAbs or control MAb.
FIG. 4. Induction of protective immunity with cytokine
treated hepa 1-6 cells armed with anti-CD28 Bi-MAb.
FIG. 5. Tumor rejection following treatment with cytokine
treated hepa 1-6 cells armed with CD28:gp55 Bi-MAb.
FIG. 6. Cure of established hepatomas in vivo.
FIG. 7. Therapeutic effectiveness of y-irradiated cytokine
treated hepa 1-6 cells armed with CD28:gp55 Bi-MAb.
D~T~TT~n DF~CRIPTION OF T~F INVENTION
The present invention provides methods for immunizing
individuals against disease and for treating individuals with
established diseases using cellular vaccines created with a two-
step process described herein. The methods of the invention maybe applicable to any disorder involving a low- or non-immunogenic
response pathology, wherein effective treatment or prophylaxis
CA 022~8082 lsss-l2-lo
WO97/47271 PCT~S97/10238
19
requires an immune boost through activation of T cells. Such
disorders include, but are not limited to, all forms of cancer,
immune deficiency disorders (both natural and induced), and
infectious diseases caused by viral or other pathogenic agents.
The methods of the invention comprise modifying weakly- or
non-immunogenic autologous cells of the disorder (target cells)
by (1) treating the target cells in order to amplify primary and
costimulatory T cell activation signals therein, and ~2)
attaching to the target cells a substance capable of binding to
one or more antigens on the target cells and to one or more T
cell activation costimulatory molecules on the surface of T cells
~i.e., CD28), thereby providing the target cells with the
capacity to physically link to T cells and activate the
costimulatory signal. Such substances include, but are not
limited to, bispecific monoclonal antibodies (Bi-MAbs) targeted
to antigen on the treated cells and to CD28 and/or other
costimulatory molecules on T cells.
The first step of the method amplifies the expression of
cell surface molecules involved in T cell activation, such as MHC
and adhesion molecules, and up-regulates antigen processing
- capacity within the target cells by enhancing enzyme activity
involved in intracellular antigen processing. For the first step
of the method, any means which can amplify primary and
costimulatory T cell activation signals in the target cells
(i.e., the expression of MHC and adhesion molecules), may be
used. Such amplified expression may be achieved by, for example,
in vitro and in vivo treatment of target cells with cytokines or
CA 022~8082 1998-12-10
WO97/47271 PCT~S97/10238
other factors capable of inducing the desired amplification; and
in vitro and in vivo transfer of MHC genes, adhesion molecule
genes, cytokine genes, and/or MHC, adhesion molecule, and
cytokine gene transcription activators or enhancers to the target
cells.
In one embodiment of the method, amplification of primary
and costimulatory T cell activation signals in the target cells
is achieved using cytokine treatment. Target cells may be
treated with cytokines ex vivo or in vitro as described in the
examples herein. Alternatively, cytokines may be administered to
the target cells in vivo by, for example, intralesional
injection, intralymph injection, subcutaneous injection, etc., in
suitable pharmaceutical carriers or controlled release
preparations. Any cytokine or combination of cytokines which
results in the amplified expression of MHC and adhesion molecules
may be used to treat cells in the first step of the method. In
preferred embodiments, described more fully by way of the
examples herein, a combination of interferon (IFN-y) and tumor
necrosis factor-~ (TNF-~) is used in the first step. Preferably,
cells may be treated with concentrations of between about lO -
lO0 U IFN-y in combination with concentrations of between about
lO-lO0 U TNF-~, more preferably with lO0 U IFN-y, and 50 U TNF-~,
as described in Section 6.l., infra. However, the conditions and
specific cytokines most optimal for the amplification of
activation signals on the particular cells to be treated may vary
and may be determined essentially as described in Section 6.l.,
infra.
CA 022~8082 1998-12-lo
W O 97/47271 PCT~US97110238
The second step of the method of the invention provides the
treated cells with the capacity to physically bridge to T cell
surfaces via CD28 and/or other T cell costimulatory molecules,
thereby providing optimal conditions for stimulating T cell
activation. For the second step, any substance capable of
binding to one or more antigens on the treated cells and to one
or more T cell activation costimulatory molecules on the surface
of T cells may be used. Such bispecific or multispecific
bridging substances may comprise, for example, Bi-MAbs, proteins
and other macromolecules, and polymer materials, which contain a
functionality capable of binding to the targeted T cell
costimulatory molecule and activating, or inducing the activation
of, the costimulatory signal. In one embodiment, described by
way of the examples in Section 6, infra, Bi-MAbs are used as the
bridging substance.
One functionality of the bispecific or multispecific
bridging substance may be directed to a target cell-specific
antigen or any antigen expressed on the target cells. Optimally,
where the target cells are to be armed with bridging substance in
vivo, the target cell antigen to which the bridging substance is
directed should be unique.
However, the target cell antigen need not be unique to the
treated cells, since the attachment of the bridging substance may
be practiced in vitro. Accordingly, bridging substances attached
to the target cells in vi tro will not cross-react with the same
antigen on cells in the individual to be immunized with the
modified cells.
CA 022~8082 1998-12-10
W O 97/47271 rCTrUS97/10238
After such bridging ~ubstances are incubated with cells
treated according to the first step of the method, free bridging
substance may be washed away and bound bridging substance may be
cross-linked to the cell surface with polyethylene glycol (PEG)
or another cross-linking agent.
Another functionality of the bispecific or multispecific
bridging substance is specifically directed to a T cell
activation costimulator such as CD28. Thus, when the modified
cells are used to immunize an individual, the CD28 (or other
costimulatory molecule) binding sites of the attached bridging
substance are free, and will bind to CD28 (or other costimulatory
molecule) on T cell surfaces, ensuring that the modified cells
will become physically linked to T cells. This bridging
substance-mediated physical link also brings other molecules on
the surfaces of the modified cells, some of which have been
amplified by the cytokine treatment step, into contact with other
molecules on the surfaces of T cells, providing further
costimulation which thereby further facilitates T cell
activation.
The second step of the method may be practiced in vitro or
in vivo, depending upon whether target cells were treated
according to the first step of the method in vivo or in vitro,
the circumstances of the disease or lesion to be treated, and the
clinical objectives of the treatment. Where the first step of
the method is conducted in vivo, the treated cells may be armed
with the bridging substance in vivo as well. In this case, the
clinician may use a variety of known methods for administering
CA 022~8082 1998-12-lO
W O 97/47271 PCTrUS97/10238
the bridging substance to a patient. The best route of
administering the bridging substance to patients who have had
disease- or lesion- specific cells (target cells) treated in vivo
according to the first step of the method will depend on clinical
and/or other aspects of the disease or lesion to be treated as
well as on the site of the treated cells. For example, where
target cells located in lymph node have been treated in vivo,
direct administration of the bridging substance to the lymph node
i8 preferred. Similarly, for example, where tumor cells have
been treated in vivo by intratumor injection of cytokines or gene
transfer vectors, the bridging substance preferably should be
- administered directly into the tumor or to the local environment
of the tumor.
Where the first step of the method is conducted in vitro,
the treated cells may be armed with the bridging substance in
vivo or in vitro. Where the bridging substance is administered
in vivo, the same route used to administer the treated cells or a
similar route should be used, taking into account the same
factors discussed above regarding in vivo arming of in vivo
treated target cells.
When in vitro treated cells are armed in vitro, the treated
and armed cells (cellular vaccine) may be used in vivo for
treatment and prevention of disease, or in vitro for generation
of lesion- or disease- specific cytotoxic T lymphocytes (CTLs).
Arming treated cells in vitro provides the advantage of being
able to use a bridging directed to any antigen on the target
cell.
CA 022~8082 1998-12-lo
W O 97/47271 PCT~US97/10238
24
When used for treatment or immunization of a patient, in
vitro treated and armed cells may be administered to the patient
using a variety of methods known to those skilled in the art. In
a particular embodiment, described more fully in the examples
which follow, in vitro treated and armed cells are administered
subcutaneously. In another embodiment, the treated and armed
cells are administered by direct intralesion injection, an
- administration route that may provide advantages over
subcutaneous administration in certain circumstances (for
example, where the lesion to be treated is not well vascularized,
is inaccessible for biopsy, or cannot be disrupted without
creating further risk to the patient). In a further embodiment,
the treated and armed cells are administered by injection into
the lymph nodes. This method of administration requires fewer
treated and armed cells than may typically be required using
other routes of administration. Such intralymph administration
may be preferred in situations where only limited autologous
tissue can be obtained from a lesion using thin needle biopsy
techniques (i.e., inaccessible/inoperable cancers). Moreover,
intralymph administration is likely to enhance the interaction
between the cellular vaccine and T cells given the large number
of T cells within lymph nodes. Applicant's initial experimental
data indicates that a single intralymph injection of as few as 1
X 104 cellular cancer vaccine cells, prepared using the method of
the invention, can induce an effective immune response against
parenteral tumor cell challenge and can cure established tumors.
The therapeutic efficacy of this method of administration appears
CA 022~8082 1998-12-lO
W O 97/47271 rCT~US97/10238
equivalent to that achieved using 100-times more cells
administered subcutaneously.
In a specific embodiment of the method of the invention, Bi-
MAbs that react with CD28 are used as a bispecific bridging
substance. As briefly discussed in Section 2, supra, B7
interacts with both CD28 and CTLA-4 on T cells. Under certain
circumstances, the B7-CTLA-4 interaction generates a negative
signal which prevents T cell activation. Thus, by immunizing
with cells coated with Bi-MAbs specific for CD28, the
interactions between B7 on such cells and CThA-4 (or other T cell
activation down-regulating molecules) is minimized and/or by-
passed. In addition, Bi-MAbs reactive with other costimulatory T
cell surface molecules (i.e., CD2, CD48) may also be used in the
practice of the method of the invention. Furthermore, more
powerful costimulation may be achieved by using a multiplicity of
Bi-MAbs having specificity for various T cell costimulators.
Bridging substances may be prepared using well known
technologies. As shown in the examples which follow, Bi-MAbs may
be used effectively as the bridging substance. Such Bi-MAbs may
be generated using methods well known in the art, such as, for
example, those described in Section 6.2., infra, and as described
in the references cited therein. Bi-MAbs containing multiple T
cell costimulatory molecule binding sites may be prepared by
chemical linkage in order to provide a means for generating
multiple costimulatory activation circuits.
In addition to Bi-MAbs, molecules engineered to contain
functional binding sites specific for both the antigen(s) of the
CA 022~8082 1998-12-10
WO97/47271 PCT~S97/10238
26
target cell(s) and the T cell costimulatory molecule(s) may be
used as the bridging substance in the practice of the method of
the invention. Such molecules may, for example, comprise
proteins, other macromolecules, and polymers engineered to
contain the desired binding sites, and may be prepared by using
genetic engineering technologies, synthetic technologies, or by
chemical linkage of component polypeptides, polymers, and/or
- other macromolecules. The binding site components of such
bridging substances may comprise, for example, Fab2 antibody
fragments, antibody binding sites, natural or engineered ligands,
or other factors reactive with the target cell antigen(s) and T
cell costimulatory molecule(s).
The bridging substance may be administered to a patient in
vivo using a pharmaceutically acceptable carrier or a variety of
drug delivery systems well known in the art. As an example, for
cancer immunotherapy, a combination of the cytokines TNF-~ and
IFN-y and an anti-CD28 Bi-MAb may be formulated within a
controlled release preparation which is administered to the
patient by directly injecting the preparation into the lymph
nodes or into the tumor itself.
The method of the invention is particularly useful in
treating cancers. This aspect of the invention is more fully
described by way of the examples presented in Section 6., infra.
The data presented in the examples indicate that strongly
immunogenic tumor cells can be generated using the two-step
method of the invention, comprising (l) in vitro treatment of
autologous tumor cells with a combination of y- interferon (IFN-
CA 022~8082 1998-12-10
WO97/47271 PCT~S97/10238
y) and tumor necrosis factor-~ ~TNF-~), and (2) pre-incubation
with a Bi-MAb specific for both antigen on tumor cells and CD28
on T cells. The resulting modified tumor cells are able to act
as a cellular vaccine that elicits CTL-mediated immunity which
can both prevent and cure established tumors.
In particular, the studies described in the examples which
follow show that cytokine-treated, anti-CD28 Bi-MAb-armed
hepatoma cells induce protective immunity against parental tumor
cell challenge and, moreover, cure established gross hepatomas in
mice. In addition, the studies described in Section 6.7., infra,
show that the method of the invention also induces protective
immunity against lymphoma and colon carcinoma.
Different routes of administration, or combinations thereof,
may be preferred when treating different cancers or other
diseases using the cellular vaccines of the invention. As
illustrated by the study briefly described in Section 6.4.,
immunization with cytokine-treated (and un-armed) cells followed
by intravenous administration of an anti-CD28 Bi-MAb induces some
anti-tumor immunity in the hepatoma model system.
In comparison, as shown by the results of the studies
described in Sections 6.5. and 6.6., infra, the administration of
cells treated with cytokines in vitro and armed with Bi-MAbs in
vitro induces uniform tumor immunity and cures established
hepatomas. It is possible that insufficient localization of the
Bi-MAbs to the tumor tissue following intravenous injection in
the former case is responsible for the difference in therapeutic
efficacy.
CA 022~8082 l998-l2-lO
W O 97/47271 PCT~US97/10238
28
Individuals may be immunized against a variety of diseases
with cytokine-treated, anti-CD28 Bi-MAb-armed autologous cells,
thereby providing the individual's immune system with a signal
sufficient to activate T cells and confer protective immunity.
Similarly, individuals may be treated for a variety of diseases
by administering cytokine-treated, anti-CD28 Bi-MAb-armed
autologous cells of the disease or lesion, thereby providing the
- individual's immune system with a signal sufficient to activate T
cells and induce a cytotoxic T lymphocyte response. In both
cases, BI-MAbs with a specificity for other T cell costimulatory
molecules may be used to arm the treated cells with a means to
physically bridge to T cells in vivo.
In addition to MHC class I, ICAM-l, ICAM-2 and VCAM-l
molecules, treatment with cytokines may enhance tumor antigen
processing by tumor cells, and may induce the expression of other
cèll surface molecules essential for T cell activation. The
combination of cytokine treated hepa 1-6 cells and immobilized
anti-CD28 MAb fails to stimulate splenic T cells in vitro or to
induce anti-tumor immunity in vivo in applicant's model system.
This suggests that the signal delivered by the interaction
between CD28 and anti-CD28 MAb is not sufficient in itself to
induce T cell activation.
In contrast, a strongly immunogenic response is obtained
when cytokine treated tumor cells armed with anti-CD28 Bi-MAb in
vitro interact with CD28 on T cells, indicating that a physical
bridging function is an important component of the activation
process.
CA 022~8082 1998-12-10
Wo97/47271 PCT~S97tlO238
29
In addition, the observation that cytokine treated, B7
transfected hepa 1-6 cells were not able to activate splenic T
cells in vitro (Section 6.2., infra) is consistent with the
recent finding that B7 may interact with CTLA-4 to deliver a
negative regulatory signal, and provides a strong rationale for
using anti-CD28 Bi-MAbs to physically link the antigen pre~enting
cell specifically to CD28 molecules for T cell activation.
The invention provides an effective alternative to gene
transfer and tumor:APC engineering for the development of
cellular vaccines. In particular embodiments of the invention,
described in the following examples, the attachment of Bi-MAbs to
tumor or other target cells takes place in vitro. Accordingly,
the antigens on such cells need not be unique to those cells.
Essentially, any antigen may be targeted. Bi-MAbs can be
produced by linking anti-CD28 MAbs to Mabs that recognize any
antigen expressed on the tumor or other target cell, including
antigens which are also expressed on large populations of cells
in the individual to be treated (i.e., lymphocytes). This
approach may be particularly useful in situations where Mabs to
tumor specific antigens are not available.
EXAMPLE 6.1
'. nK I N~: INl~v T~PRESSION OF AD~T~'~ION
~Nn M~C MOT~CUT~ ON HEPA 1-6 ~T.~r.T..~ IN VTT~O
2~
Hepa l-6 is a chemically induced hepatoma originating in a
C57BL/6 mouse ~G. J. Darlington et al., 1980, J. Natl. Cancer
Inst. 64: 809). Cells derived from this tumor grow rapidly and
form subcutaneous nodules in syngeneic animals. Hepa 1-6 cells
CA 022~8082 l998-l2-lO
W O 97/47271 PCT~US97/10238
lack both MHC class I and B7 molecules on their cell surfaces and
do not induce a host immune response even when transfected with
genes encoding the B7-1 or B7-2 molecule.
The conditions and cytokines most optimal for the
amplification of activation signals on hepa 1-6 cells were
determined as follows. One ml of Hepa 1-6 cells was plated into
24 well tissue culture plates at a concentration of 2 x 106 cells
/ ml and incubated with either IFN-y, 100 U, or TNF-~, 50 U, or a
combination of IFN-y and TNF-~ at these same concentrations in
complete RPMI-1640 medium supplemented with 10 ~ fetal calf
serum, 2mM glutamine, lx non-essential amino acid and 1 mM sodium
- pyruvate for 48 hr at 37~C. Hepa 1-6 cells incubated similarly
but in medium alone were used as control.
Cells were washed with phosphate-buffered saline (PBS)
and stained with rat monoclonal antibodies to mouse MHC class I
~Ml/42), MHC class II (M5/114), CD44 (KM81) (ATCC), ICAM-1
- ~HA58), ICAM-2 (3C4) and VCAM-1 (51-lOC9) ( PharMingen, San
Diego, California). To stain for mouse B7-1 (CD80) and B7-2
(CD86), we used CTLA4-Ig, a soluble fusion protein containing the
variable domain of the human CTLA4 protein and the hinge, CH2,
and CH3 domains of the human IgG1 constant region (Y. J. Guo et
al., 1994, Science 263: 518; C. Caux et al., 1994, J. Exp. Med.
180: 1841; E. Murphy et al., 1994, J. Exp. Med. 180: 223; R.
Seder et al., 1994, J. Exp. Med. 179: 299; B. Blazar et al.,
1994, Blood 83: 3815; K. Hathcoc~ et al., 1993, Science 262: 905;
P Linsley et al., 1991, J. Exp. Med. 1~: 561).
Cells were lncubated with the antibodies or chimeric
CA 022~8082 1998-12-10
W097/47271 PCT~S97/10238
protein for 40 minutes on ice. A rat antibody to mouse CD3
(YCD3) and a soluble human CD44-Ig chimeric protein were used as
a negative controls. Cells were washed three times. Fluorescent
isothiocyanate ~FITC)-conjugated goat antibody to rat Ig or FITC-
labeled rabbit antibody to human Ig was added for an additional
40 minutes on ice. Samples were then washed, fixed and analyzed
in a FACScan (Becton Dickinson, San Jose, California). The mean
fluorescent intensity in the negative control group (medium
alone) was at background level except for ICAM-l.
Hepa 1-6 cells incubated with a combination of interferon
(IFN-y) and tumor necrosis factor (TNF-~) showed expression of
MHC class I, intercellular adhesion molecule 2 (ICAM-2) and
vascular adhesion molecule 1 (VCAM-l), and showed significantly
enhanced expression of intercellular adhesion molecule 1 (ICAM-l)
(FIG. l). This pattern of expression was maintained for more
than three days in vitro after removing cytokines from the
medium. However, the cytokine treated hepa 1-6 cells (CT-hepa l-
6) were still able to form tumors in syngeneic animals. It was
assumed that this may be because the cells were deficient in
providing a CD28 mediated costimulatory signal due to absence in
expression of B7.
EXAMPLE 6.2
STTMULATION ~Nn PROLTFT~'T~ATTON OF SYr~
sPTT~NTC T_~T~r-T~ ~N~r~ BY CYTORTN~ TT~T~'~T~n
~T~PA 1-6 ~T'T-T--'~ ANl-~ ANTI-CD28 B;-Ml~b TN VTTRO
The following example demonstrates that cytokine activated
Hepa 1-6 cells in combination with Bi-MAbs to CD28 and tumor cell
CA 022~8082 1998-12-1o
W O 97/47271 PCT~US97/10238
antigens stimulate proliferation of splenic T cells in ~itro,
indicating that such Bi-MAbs can provide a CD28 costimulatory
signal. Three Bi-MAbs, CD28:gp55, CD28:gp95, and CD28:yp210,
each with one binding specificity for the CD28 molecule on T
cells and a second binding specificity for one of three
glycoproteins expressed on tumor cell surfaces, were prepared and
used as follows.
- For preparation of Mabs and Bi-MAbs, Wistar rats were
;~ml1n;zed with 2 x 107 Hepa 1-6 cells in CFA. Following three
additional boosts with the same cells in ICFA over an 8 week
period, spleen cells from immunized rats were fused with YB2/0
rat myelomas as previously described (J. Alan & T. Robin, in:
Immunochemistry in Practice, Chapter 2 ~Blackwell, New York, 2d
ed. 1988)). More than twenty Ig-producing hybridomas were
selected by immunofluorescent staining. Three antibodies reacted
with hepa 1-6 cells by flow cytometry analysis. These Mabs
separately recognized a 55 Kd, 95 Kd or 210 Kd glycoprotein
expressed on most tumor cells as determined by
immunoprecipitation. The Mabs were designated as anti-gp55,
anti-gp95 and anti-gp210 respectively. Anti-mouse CD28 Mabs were
generated by first immunizing Wistar rats with a mouse T cell
hybridoma cell line expressing high levels of CD28 antigen on the
cell surface. After cell fusion, hybridomas producing anti-CD28
MAb were selected with immunofluorscent analysis by FACScan.
Anti-CD28 Mabs were further characterized and confirmed by
immunoprecipitation and T cell proli~eration and IL-2 production
assays. Hybridoma producing anti-mouse CD18 used to generate
CA 022~8082 1998-12-10
WO97/47271 PCT~S97/10238
CD18:gp55 Bi-MAb was purchased from ATCC. All Mabs used in these
experiments were purified by passage of ascites from nude mice
over a protein G column.
Bi-MAbs were produced from these Mabs as previously
described (J. A. MacLean et al., 1993, J. Immunol. 150: 1619; L.
K. Gilliland et al., 1988, Proc. Natl. Acad. Sci. USA ~: 7719;
C. Bode et al., 1989, J. Biol. Chem. 264: 944).
- Normal splenic T cells were purified by nylon wool column.
Purified T cells (5 x 106/ well) were co-cultured in complete
~PMI-1640 medium at 37~C for 96 hours with 5 x 105 irradiated
(5000 roentgens) cytokine treated ~as described in Section 6.1.,
supra) or untreated hepa 1-6 cells in the presence or absence of
CD28:gp55, CD28:gp95 or CD28:gp210 Bi-MAb. A CD18:gp55 Bi-MAb
that bridges CD18 on T cells to gp55 on tumor cells and a mixture
of parental CD28 plus gp55 Mabs were used as controls. Cytokine
treated or untreated hepa 1-6 cells transfected with B7 gene and
expressing high level of B7 molecules on cell surfaces were also
used as a control. The percentage of CD3+CD8+CD25+ T cells (mean
V SD) was determined by three color analysis in FACScan using Cy-
ChromTM labeled anti-CD3, PE-labeled anti-CD8 and FITC-conjugated
anti-CD25 antibodies (PharMingen, San Diego, California).
Each of the Bi-MAbs was tested both in vitro and in vivo for
its ability in combination with cytokine treatment of hepa 1-6
cells to activate tumor specific CTLs. Mouse splenic cells were
cocultured with either cytokine treated or untreated hepa 1-6
cells in the presence of purified anti-CD28 Bi-MAb or control
antibody, 50 mg/ml each, at 37OC for 9 days.
CA 022~8082 1998-12-10
WO97/47271 PCT~S97/10238
34
The results presented in FIG. 2 indicate that the
combination of cytokine treated tumor cells and any one of the
three anti-CD28 Bi-MAbs significantly stimulated splenic T cell
proliferation.
No stimulation was obtained in the absence of either anti-
CD28 Bi-MAbs or the cytokine treated autologous tumor cells.
Interestingly, hepa 1-6 cells transfected with, and expressing
- high levels of B7, were not effective in stimulating naive T
cells in vitro when treated with the cytokines in similar manner.
The majority of lymphocytes generated by this approach were
CD3+CD8+CD25+ T cells.
EXAMPLE 6.3
IN VTTRO ~:Y16~ O~lCITY OF CTLs GFNF~Z~TED BY
CYT~TN~ T~T~n HEPA 1-6 TUMOR ~r-T-~
IN C~MRINATTON WITH ANTI-CD28 Bi-MAbs
Cytotoxicity of CTLs generated by in vitro priming of naive
splenic T cells with cytokine treated hepa 1-6 cells in
combination with anti-CD28 Bi-MAbs or control MAb was established
as follows. Nylon wool-enriched naive splenic T cells (5 x
106/well) were first stimulated in vitro by incubation with 5 x
105 irradiated (5000 roentgens) cytokine treated hepa 1-6 cells
(as described in Section 6.1., supra) in combination with anti-
CD28 Bi-MAbs, control antibodies or irradiated B7+ hepa 1-6 alone
at 37~C for 9 days. y-irradiated naive splenic cells (5 x 1o6)
were added into cultures as feeder cells. At the 3rd and 6th day
after stimulation, 2 ml of complete RPMI-1640 medium containing
recombinant human IL-2 (20 U) were added into each culture well
CA 022~8082 1998-12-10
WO97/47271 PCT~S97/10238
separately.
Cytotoxicity of CTLs toward syngeneic, allogenic tumor cells
and a NK sensitive YAC-l cell line was determined in a standard 4
hour 5lCr release assay. Results from three experiments are
shown in FIG. 3 as the percentage of 5lCr release at l:20 E:T
ratio as measure of tumor cell lysis (mean VSD). These results
~how that CTLs generated as described herein have cytolytic
- activity specific to autologous tumor cells.
EXAMP~E 6.4
~:~r~ OF TNTRAv~O~!Y AnMTN~ ~PI) ~NTT-CD2B Ri-~hs
ON T~NO~T~T~'NTCTTY OF ~.~k I N~: TT~T~n ~T~'PA 1- 6 ~T.'T.T..~
Mice injected with cytokine treated hepa 1-6 cells followed
by intravenous administration of anti-CD28:gp55 Bi-MAb at a dose
of l00 ug on day l, 2 and 4, experienced delayed tumor formation
and 40 percent of these mice (8/20) had tumor regression.
Animals injected with a combination of parental hepa 1-6 cells
and anti-CD28 Bi-MAb or a combination of CT-hepa 1-6 cells and
control antibody all developed tumors and died within 60 days
after the inoculation of tumor cells. Immunohistological studies
at 24, 48, and 72 hr after injection of the Bi-MAbs showed that
the Bi-MAbs were widely distributed in lungs, liver and kidney in
addition to tumor tissue.
EXAMPLE 6.5
IN~ I ON OF PRO~ L~v~: Ir---UNI~LY IN VTVO WIT~ ~yLO~KI N K
T~T~TRn RT'PA 1-6 TUMOR ~T-T--~ ARMED WTTH ANTT-CD~8 Bi-M~hs
Tumor cells were first treated in vitro with a combination
CA 022~8082 1998-12-10
WO97/47271 PCT~S97/10238
36
of IFN-y, 100 U, and TNF-~, 50 U, in RPMI-1640 medium with 10~
fetal calf serum at 37~C, 5~ CO2 for 48 hours. Cells were then
washed with Phosphate-Buffered Saline, PH 7.4 (PBS) x 3 at 20~C
and incubated with anti-CD28 Bi-MAbs at a concentration of 50
ug/ml on ice for 45 min as described in Section 6.1., supra.
After an additional incubation in an equal volume of 30~
polyethylene glycol (PEG) in RPMI-1640 for 60 minutes at 4~C, the
cells were washed x3 as described above and suspended in a final
concentration of 1-2 x 107 / ml PBS. To arm cells with Bi-MAbs
or Mabs, cytokine treated or untreated parental tumor cells were
pre-incubated with respective antibodies for 45 minutes.
Five groups of C57BL/6 mice, 5 per group, were immunized
subcutaneously with 1 x 106 cytokine treated hepa 1-6 cells armed
with CD28:gp55, CD28:gp95, CD28:gp 210 or a control Bi-MAb
CD18:gp55, or with 1 x 106 untreated hepa 1-6 cells armed with
CD28:gp55 Bi-MAb. After two weeks, mice in each of the groups
were challenged with 2.5 x 106 parental hepa 1-6 cells injected
subcutaneously.
Cytokine treated hepa 1-6 tumor cells (CT-hepa 1-6) pre-
incubated/armed with anti-CD28 Bi-MAbs completely lost their
ability to form tumors in syngeneic mice, whereas cytokine
treated hepa 1-6 cells pre-incubated/armed with control
antibodies retained their tumor forming capacity. The results of
the immunization experiment are shown in FIG. 4. Mice immunized
with CT-hepa 1-6 cells armed with each of the CD28/tumor antigen
Bi-MAbs developed protective immunity against challenge with
parental tumor cells, and all of these animals remained tumor-
CA 022~8082 1998-12-10
WO97t47271 PCT~S97/10238
free for 120 days after such challenge (FIG. 4). In contrast,
all mice injected subcutaneously with either untreated Hepa 1-6
cells armed with anti-CD28 Bi-MAbs or CT-hepa 1-6 cells armed
with the control antibody CD18:gp55 developed tumors and died
within 50 days following challenge with the parental hepa 1-6
cells (FIG.4). This experiment was repeated twice with
comparable results.
EXAMP~E 6.6
APpT.TCATTON OF 1~! M~ n OF 1~ NV~-Nl lON TO T~ UP!PATOMA CANCER
~';Y~-l'lSM: C~E7P! OF T~.~TART.T.~:~n HEPAOTt)M~-~ WIT~ Y-LC~K 1~1': Tl~ TT'n,
B~ h_A~2M~n ~PA 1-6 r~TT--~
To establish that immunization with CT-hepa 1-6 cells armed
with anti-CD28 Bi-MAbs can cure established hepatomas, the
following three studies were performed, each of which indicates
that the therapeutic administration of Bi-MAb-armed, cytokine
treated hepa 1-6 cells is an effective therapy for hepatoma.
In the first study, forty mice were inoculated
subcutaneously with 2 x 106 wild type hepa 1-6 cells. Fourteen
days later, after the development of microscopic tumors, the mice
were divided into four groups of ten each. The groups were
treated subcutaneously with 2 x 106 cytokine treated or untreated
hepa 1-6 cell armed with either the CD28:gp55 Bi-MAb or a control
CD18:gp55 Bi-MAb.
The results of this study are shown in FIG. 5. All mice in
the group treated with CT-Hepa 1-6 cells armed with anti-CD28 Bi-
MAb survived for more than 100 days (FIG 5). In contrast, all
mice in the group treated with control Bi-MAb-armed/cytokine
CA 022~8082 1998-12-10
WO97/47271 PCT~S97/10238
38
treated hepa 1-6 cells, and all mice in the groups treated with
untreated hepa 1-6 cells armed with the CD28:gp55 Bi-MAb or the
CD18:gp55 control Bi-MAb, died within about 40 days (FIG. 5).
This experiment was repeated twice with comparable results.
In the Qecond study, five group~ of five mice each were
injected subcutaneously with 1 x Io6 hepa 1-6 cells. After four
weeks, mice bearing tumors 6-8 mm (in the greatest dimension)
were injected subcutaneously with 1 x 106 cytokine treated hepa
1-6 cells armed with the CD28:gp55, CD28:gp95 or CD28:gp210 Bi-
MAbs, respectively, and then injected subcutaneously with a boost
of the Bi-MAb-armed/cytokine treated hepa 1-6 cells at the same
dose 7 days thereafter. Cells pre-incubated with CD18:gp55 or
mixed with both anti-CD28 and anti-gp55 parental Mabs ~CD28+gp55)
at a concentration of 50 ug each were used as controls. Tumor
size was periodically measured.
The results of this study are shown in FIG. 6. Hepatomas in
the mice treated with the CD28:gp55, CD28:gp95 and CD28:gp210 Bi-
MAbs regressed to undetectable size within about 40 days (FIG.
6). In contrast, hepatomas in the mice treated with the CT-hepa
1-6 cells pre-incubated with either control Bi-MAb or MAb more
than doubled in size within the same period, and all of these
mice within that period (FIG. 6). This experiment was repeated
three times with comparable results.
In the third study, y-irradiated Bi-MAb armed tumor cells
were used as the vaccine. Three groups of five mice were
inoculated subcutaneously with 1 x 106 hepa 1-6 cells. After two
weeks, mice were then injected subcutaneously either with 1 x 106
CA 022~8082 1998-12-lo
W O 97/47271 PCT~US97/10238
39
y-irradiated, cytokine treated hepa 1-6 cells armed with
CD28:gp55 Bi-MAb, or with a combination of y-irradiated, cytokine
treated hepa 1-6 cells plus a mixture of parental anti-CD28 and
anti-gp55 Mabs, or with 1 x 106 y-irradiated hepa 1-6 alone.
The results of this study, shown in FIG. 7, are similar to
the results obtained in the second study described above. Only
the mice treated with the y-irradiated, cytokine treated hepa 1-6
cells armed with CD28:gp55 Bi-MAb survived (for more than 100
days). All mice in the other two groups died within 40 days of
treatment.
Upon ex~m-n~tion, tumor tissue from mice first injected with
the parental tumor cells and then given cytokine treated
autologous tumor cells pre-incubated with anti-CD28 Bi-MAb showed
marked inflammatory responses with abundant lymphocyte
infiltration. In accordance with in vitro stimulation data, the
majority of the infiltrating lymphocytes were CD3+CD8+ CD25+ T
cells, as determined by immunofluorescent staining of tissue
sections with rat anti-mouse CD3, CD8 and CD25 MAbs. There was
no local immune response in mice injected either with untreated
tumor cells armed with anti-CD28 Bi-MAb or with cytokine treated
tumor cells armed with control Bi-MAbs.
To further investigate if the induced immunity was mediated
by CTLs, mice were depleted of CD8+ T cells by antibody treatment
before or after immunization. Depletion of CD8+ T cells either
before or after immunization abrogated the ability of the
cellular vaccine to elicit anti-tumor immunity in vivo.
-
CA 022~8082 1998-12-10
W O 97147271 PCTAUS97/10238
EXAMPLE 6.7
APPLICATTON OF THE METHOD OF THE lNVhNLl ON TO Al~DITIONAL CA~CER
SYST~M~: lN~ oN OF PROTECTIVE TM~U~l-LY Ar~TNsT LYNPHOMA ANn
C~TON CARCTNOMA USTNG ~l~klN~: T~ATRn. Bi-MAb-ARMED C~NCER r~T-T-~
The immunogenicity of cytokine treated, autologous tumor
cells pre-incubated with anti-CD28 Bi-MAbs was tested in two
additional cancer systems, EL-4 lymphoma and SMCC-1 colon
carclnoma .
The EL-4 lymphoma becomes immunogenic when transfected with
the B7 gene. In contrast, SMCC-1 colon carcinoma remains non-
immunogenic even after transfection with the B7 gene. Both of
these cell lines grow rapidly and develop subcutaneous tumors in
syngenic C57 BL/6 mice (see, for example, Li et al., 1996, J.
Exp. Med. 180: 211). Both of these cell lines express the gp55
antigen on their cell surfaces. Accordingly, the anti-CD28:gp55
Bi-MAb described in the previous examples was also used in these
studies.
Three groups of mice were immunized subcutaneously with 1 x
106 cytokine treated, CD28:gp55 Bi-MAb-armed, hepa 1-6, EL-4 or
SMCC-1 tumor cells respectively.
After two weeks, mice immunized with the modified hepa 1-6
cells were divided into three groups and challenged by a
subcutaneous injection with 1 x 106 hepa 1-6 cells, SMCC-1 cells
or EL-4 cells respectively.
The mice immunized with the cytokine treated, CD28:gp55 Bi-
MAb-armed SMCC-1 cells were divided into two groups. One group
was challenged by subcutaneous inoculation with 1 x 106 SMCC-1
cells. The other group was challenged by subcutaneous
CA 022~8082 1998-12-10
WO97/47271 PCT~S97/10238
inoculations of EL-4 cells and 1 x 106 hepa 1-6 cel].s into the
- left and right flanks of mice, respectively.
Similarly, the mice immunized with the cytokine treated,
CD28:gp55 Bi-MAb-armed EL-4 cells were divided into two groups
which were challenged with either EL-4 cells alone or with both
SMCC-l and hepa 1-6 cells.
The results of this study, presented in Table I below, were
repeated twice with identical results.
TABLE I
Specificity of the immune responses elicited by
cytokine treated and anti-CD28 Bi-MAb armed tumor cell~.
IMMUNIZATION CELLS ~UAT.T~NGE CELLS NUMBER OF MICE
WITH TUMORS
Bi-MAb-CT Hepa 1-6Hepa 1-6 0 of 10
Bi-MAb-CT Hepa 1-6SMCC-l 6 of 6
Bi-MAb-CT Hepa 1-6 EL-4 5 of 5
Bi-MAb-CT SMCC-l SMCC-l 0 of 6
Bi-MAb-CT SMCC-lEL-4 + Hepa 1-6 6 of 6
Bi-MAb-CT EL-4 EL-4 0 of 6
Bi-MAb-CT EL-4SMCC-l + Hepa 1-6 6 of 6
Immunization with cytokine treated EL-4 (CT-EL-4) or
cytokine treated SMCC-l ~CT-SMCC-l) tumor cells armed with the
anti-CD28:gp55 Bi-MAb elicited anti-tumor immunity against
autologous parental tumor in all animals (Table 1).
The elicited immunity in all three experimental groups was
tumor-specific. For example, immunization with CT-hepa 1-6 cells
armed with anti-CD28 Bi-MAb did not inhibit growth of syngeneic
CA 022~8082 1998-12-10
WO97/47271 PCT~S97/10238
42
EL-4 or SMCC-l tumors in vivo (Table l). Interestingly, in the
absence of treatment with cytokines, the parental EL-4 cells
expressed high levels of MHC class I at cell surfaces, yet when
pre-incubated with anti-CD28:gp55 Bi-MAb, they were still not
able to induce protective immunity. In addition, CTLs from mice
immunized with cytokine treated autologous tumor cells armed with
anti-CD28 Bi-MAbs specifically lysed the parental tumor cells but
~ not other tumor cells in vitro.
EX~MPLE 6.8
IN~ ON OF CEL~UT.~ nhUN1-LY BY TmMORS ~M~n WITH ~nU~TIvALENrr
RRTnGE MOT.~CUr~
The above data showed that SMCC-l, EL-4 and Hepal-6 cells
became more immunogenic after treatment in vi tro with a
combination of cytokines and anti-CD28 bispecific monoclonal
antibody. The modified tumor cells were able to elicit
anti-tumor specific immunity that were both preventive and
curative.
The following data showed that EL-4 and SMCC-l cells were
also effective for eliciting preventive and curative antitumor
immunity in syngenic animals without cytokine treatment when
these cells were precoated with two different bispecific
monoclonal antibodies (bi-Mabs), one specific for CD28 and
another specific for 4-lBB.
The cells were coated in vitro with anti-gp55:anti-CD28 and
anti-gpll5:anti-4-lBB bi-Mabs in a concentration of 50 ug/ml on
ice for 45 min. After fixed with PEG and washed for three times,
CA 02258082 1998-12-10
WO97/47271 PCT~S97/10238
the tumor cells coated with the two different bi-Mabs were
subcutaneously injected into syngenic animals at different doses.
Two weeks later, the immunized animals were challenged with
parental tumor cells and tumor formation rate was observed (Table
II).
Table II.Comparison of the efficacy of immunogenic tumor cells
modified with different process in ~itro on eliciting preventive
antitumor immunity in ~ivo
Imm-ln;zatinn EQ~ Challenae Tumor Formation
Irradiated EL-4 Wt lx106 1x106 EL-4 Wt 100~
Irradiated CT-EL-4Wt lxl06 1x106 EL-4 Wt 100%
Irradiated EL-4 coated CD28BiMab lx106lx106 EL-4 Wt S0
Irradiated CT-EL-4 coated CD28BiMab
1x1061x106 EL-4 Wt 0~
lxlOslx106EL-4 Wt 20%
5x104lxl06 EL-4 Wt 60~ -
Irradiated EL-4 Wt coated
anti-CD28&anti-4-lBB BiMabs lx106lx106 EL-4 Wt 0%
lxlOs lx106 EL-4 Wt 0%
5xlo4 lxlo6 EL-4 Wt 10%
In curative experiments, syngenic animals were first
inoculated subcutaneously with 2xlO6parental tumor cells. After
two to four weeks, the tumor bearing animals were injected with
the modified tumor cells. The mean survival time was monitored
(Table III).
CA 02258082 1998-12-10
WO 97/47271 PCT/US97/10238
44
T~ble I~:. CompArison of the efficacy of ~ ~gen~c tumor v~cc~ns~ armed with
~ono~ralent or multivalent bi-~pecific Mab~ on eliciting curati~re antitumor
i.. unity in vivo
I~mors Vaccin~tion Dose Animal S1~rvival (~)*
SMCC-1 SMCC-lWt 1x106 0
SMCC-l CT-SMCC-l+CD28 BiMabs lX106 100
5x105 30~
SMCC-1 SMCC-l~CD28 & 4-lBB BiMabs lx106 100t
5x105 80
SMCC-1 SMCC-l+CD28& 4-lBB BiMabs
(multivalent BiMabs) 1x106 100
5x105 100
lx105 60
* 60 day survival rate after tumor vaccine treatment.
EX~MPLE 6.9
20~!~p~ TIoN O~ t~T~T~uLAR T~ . -UN I ~l'Y A~ATNsT VIRUS INFECTED CELI-S
Primary liver cells were obtained from clinical biopsy
laboratory and cultured in hepatocellular media. Autologous
peripheral blood lymphocytes were obtained from same patients
during operation and cultured in complete RPMI-1640 medium
supplemented with 5~ human AB serum and 5~ fetal calf serum, 20
iu/ml rh-IL-2.
Liver cells were infected by ElB deleted Adenovirus as
reported previously. Virus infection was confirmed by RT-PCR and
histological e~m;n~tion.
Infected liver cells were then treated in vitro with (1)
cytokines alone, (2~ BiMabs alone, or (3) cytokine + bispecific
Mabs. The modified virus-infected liver cells were irradiated at
CA 02258082 l998-l2-lO
W O 97/47271 PCT~US97/10238
a dose of 5000R and were then co-cultured with autologous PBL in
complete RPMI-1640 medium as reported previously.
The cytotoxicity of the CTLs generated by the unmodified
and the modified liver cells was determined with a standard 4 h
slCr release assay.
Table IV.Generation of cytotoxic T cell~ in vitro by stimulating
autologous peripheral blood lymphocytes with virus-infected human
fetal liver cells pretreated with a combination of INF-y and
TNF-a and coated with anti-CD28 bispecific monoclonal antibody
T~rget cells Treatm~nt Effectors E:T Ratio Cytoto~;citv
Liver cells None PBL 1:50 -3
Liver cells Cytokines PBL 1:50 ~3
Liver cells BiMabs PBL 1:50 ~3
Liver cells Cyto+BiMab PBL 1:50 ~3%
Ad-Liver cells None PBL 1:50 ~3
Ad-Liver cells Cytokines PBL 1:50 ~5
Ad-Liver cells BiMabs PBL 1:50 ~13
Ad-Liver cells Cyto+biMabs PBL 1:50 ~20%
Ad-Liver cells biMabs (Multivalent*)
PBL 1:50 ~40
Ad-Liver cells Cyto+biMabs(Multivalent*)
PBL 1:50 ~40~
~Anti-Gpll5:Anti-CD28 bispecific monoclonal antibody was generated by
chemical linking several anti-CD28 and anti-Gpll5 monoclonal antibodies
together and purified by sequential affinity columns, one specific binding
gp-115 and another for CD28 monoclonal antibody specifically.
The present invention is not to be limited in scope by
the embodiments disclosed herein, which are intended as single
CA 022~8082 1998-12-lo
W O 97/47271 PCTrUS97/10238
46
illustrations of individual aspects of the invention, and any
which are functionally equivalent are within the scope of the
invention. Various modifications of the invention, in
addition to those shown and described herein, will become
apparent to those skilled in the art from the foregoing
description, and are similarly intended to fall within the
scope of the invention. For example, Guo et al., Nature
Me~'c;ne, vol. 4:1-5, (April, 1997) provide examples and
references.
All publications referenced are incorporated by
reference herein, including drawings, nucleic acid sequences
and amino acid sequences listed in each publications. All the
compounds disclosed and referred to in the publications
mentioned above are incorporated by reference herein,
including those compounds disclosed and referred to in
articles cited by the publications mentioned above.
Other embodiments of this invention are disclosed in the
following claims.
