Note: Descriptions are shown in the official language in which they were submitted.
5S57
The present invention relates to formulations for
the controlled release in vivo of therapeutic agents. In
another aspect, it relates to vesicles and, particularly,
to phospholipid vesicles.
Numerous conditions in both man and lower animals
are responsive to drugs or other therapeutic agents adminis-
tered ln vivo. To be useful, these agents must be adminis-
tered at a dosage level which is high enough to cause the
desired effect, and that dosage level must be maintained
for a sufficient period of time to achieve that effect.
Many such therapeutic agents are routinely administered by
intra-muscular injection at dosage levels calculated to
produce a concentration of the agent in the circulatory
system which is effective. Thereafter, injections are
repeated as necessary to maintain the -therapeutically
effective level.
It is often the case that the therapeutically effec-
tive level of the agent in circulation is main-tained only a
short time after injection because of breakdown of the agent
by the host's defensive mechanisms against foreign substances.
~5557
60724-1519
~oreover, the agent itself may have intolerable side effects,
even lethal ones, if administered in amounts which substantial-
ly exceed the therapeutically useful level. Therefore, pro-
longing the effective concentration of the agent in the body by
increasing the dosage is always limited by the amount of
toxicity. Even in cases where toxi.city is low, the size of an
injection can be limited by the size of the bolus which can
be administered safely.
In view of such problems, efforts have been made to
develop delivery systems for therapeutic agents which may be
administered _ vivo and which, after administration, gradually,
release the agent into its environment in order to prolong the
interval over which the effective concentration of the agent is
maintained in that environment. In this way, the interval
between administrations of the agent can be increased and, in
some instances, the need for further administration can be
eliminated.
One such approach to obtaining prolonged or sustained
release has been to encapsulate the therapeutic agent in a
"vesicle". As used herein, the term vesicle refers to a
micellular particle which is usually spherical in form and
which is frequently obtained from a lipid which forms a bi-
layered membrane and is referred to as a "liposome". ~ethods
for making such vesicles are well known in the art. Typcially,
they are prepared from a phospholipid such as distearoyl phos-
phatidylcholine or lecithin, and may include other materials
such as positively or negatively charged compounds. Vesicles
made from phospholipids are commonly
60724-1519
referred to simply as "phospholipid vesicles". Depending on
the techniques for its preparation, a vesicle may form as a
simple bilayered shell (a unilamellar vesicle) or it may form
in multiple layers (multilamellar vesicle).
After administration, typcially as a suspension in
physiological saline, the vesicles gradually release the encap-
sulated therapeutic agent which then displays its expected
effect. Mowever, prior to its release, the agent exhibits no
pharmacokinetic properties and is protected by the vesicle from
metabolic degradation or other attack by the host's defense
mechanisms against foreign substances. Accordingly, the agent
can be safely administered in an encapsulated form in dosages
which are high enough that, if directly given the host, could
result in severe side effects or even death.
The time interval over which an effective concentra-
tion of the therapeutic agent is maintained after administra-
tion as a vesicle encapsulant is generally thought to be a
function of the rate at which it is released from the vesicle
and the rate at which it is absorbed from the point of admin-
istration after release. Since the former may vary withvesicle structure and the latter by reason of the properties of
the agent, the interval over which the useful concentration of
an agent in circulation is maintained can vary widely. General-
ly, the rate of release from the vesicle is though-t to be cont-
rolling for most compounds and sustained release of encapsu-
lated drugs over a period of 6-8 hours is a common observation.
See F.J.T. Fields, (1981) Liposomes: From Physical Structure
to Therapeutic Applications; Research Monographs in
~24SS57
Cell and Tissue Physioloqy, Vol. 7, C.G. ~night, ed., Elsevier/
North Holland, N.Y., p. 51ff and R.W. Stevenson et al, Diabeto-
ogia, 19, 217 (1980~. However, intramuscular injections in
mice of vesicle encapsulated interferon resulted in localized
levels of interferon which, after three days, were equivalent
to levels observed over 2-4 hours ~ter injection of free inter-
feron. D.A. Eppstein et al, J. Virol., 41, 575. This longer
time of sustained release likely reflects the lower mobility
of this biomacromolecule from the injection site.
Notwithstanding the advance in sustained release which
has been achieved using vesicles as encapsulants, further
improvements in sustained release composi~ions are desirable to
reduce still further the interval between administrations of
the therapeutic agents. Even a 6-8 hour period of sustained
release makes out-patient treatment difficult, if not impossible,
and longer intervals would reduce the workload of hospital or
other medical personnel, not to mention reducing the patient's
discomfort. Furthermore, although by vesicle encapsulation the
period over which an effective concentration of therapeutic
agents could be maintained is extended, no effective means,:~o
control the rate of release results from encapsulation itself.
Accordingly, there remains as yet unmet, a desire for sustained
release formulations of therapeutic agents which extend even
further the interval over which an effective concentration of
the agent is maintained.
SU~lARY OF THE INVENTION
According to the present invention, a process and a
composition are provided for controlling the rate at which a
therapeutic encapsulated in a vesicle is released from the
vesicle after _ vivo administration. This is achieved by
lZ~iS~
724-1519
suspending the vesicles, which encapsulate a solution of the
therapeutic agent, in a solution containing sufficient solute
that its osmolarity, relative to that of the solution within
the vesicles, is at least substantially isotonic, that is,
at least approximately 25% of the osmolarity inside the
vesicles and is of greater osmolarity than physiological
saline. The suspension can then be administered parenterally,
for example by subcutaneous or intramuscular injection.
The rate of release of the therapeutic agent from
the vesicles after administration is a function of the initial
osmotic pressure. Thus, as the osmolarity of the suspending
solution becomes less hypotonic, relative to the solution with-
in the vesicles, the rate of release of the therapeutic agent
is slowed. Slowest releases are obtained when the suspending
solution approaches an isotonic, or even hypertonic, relation-
ship with respect to the solution within the vesicles. The
compositions of the present invention exhibit a longer inter-
val over which the sustained release of the agent is maintained
compared to agents encapsulated in vesicles and administered
as suspensions in physiological saline as described in the
prior art.
BRIEF DESCRIPTION OF THE FIGURES
... _ . . ... .
Figure 1 illustrates the effect of temperature on the
stability of vesicle formulations.
Figure 2 illustrates the effec-t of vesicle membrane
fluidity on sustained release of 2-PAMCl.
Figure 3 demonstrates the effect of vesicle compos-
ition on sustained release of 2-P~Cl.
Figure 4 illustrates the effect of cholesterol cont-
ent of phospholipid vesicles on sustained release of 2-PAMCl.
5S5~
724-1519
Figure 5 illustrates the effect of vesicle physical
structure on sustained release of 2-PAMCl.
Figure 6 illustrates the effect of suspending 2-PAMCl
loaded vesicles in a solution substantially isotonic to the
vesicles.
DET~ILED DESCRIPTION
As noted above, the present invention relates to
method for controlling the rate of release n vivo of a
therapeutic agent from a vesicle encapsulant by adjusting the
osmotic pressure between the solution of therapeutic agent
within the vesicle and the solution in which the vesicles are
suspended for parenteral administration.
Thus in an aspect, the present invention provides
a composition suitable for parenteral administration to animals
including human beings, said composition comprising a solution
of a therapeutic agent encapsulated in vesicles, the vesicles
being pharmaceutically acceptable and being suspended in a
solution containing sufficient solute to provide an osmolarity
which is at least about 25% of the osmolarity of the solution
within the vesicles and which is of greater osmolarity than
physiological saline.
In another aspect, the present invention provides a
process for producing such a composition. The process comprises
encapsulating a solution of the agent in pharmaceu-tically
acceptable vesicles and suspending the vesicles in a solution
containing sufficient solute to provide an osmolarity which is
at least 25~ of the osmolarity of the solution of the therapeu-
tic agent within the vesicles and which is of greater osmolari-
ty than physiological saline.
When such vesicles are parenterally administered to
" ~2~5S~
724-1519
a host, for example, intramuscularly, the interval over which
sustained release is maintained is substantially increased.
Although we do not wish to be bound byany particular
theory, the increase in the interval of sustained release
obtained as the osmolarity of the suspending solution is made
less hypotonic, relative to the solution within the vesicles,
may result from the fact that, so long as the concentration
of solute in the suspending solution is such that the solu-tion
is not hypotonic, that is the osmolarity approaches at least
a substantially isotonic relationship with the solution within
the vesicles, the osmotic pressure is not great enough to
cause the solvent of the suspending solution to migrate into
the vesicles which would increase the hydraulic pressure with-
in the vesicles and cause them to break down or release part
of their contents. After administration, however, the
concentration of liquid medium around the vesicles gradually
becomes more hypotonic with respect to the solution within the
vesicles, for example, by absorption of the solute from the
- suspending solution. As this occurs, the osmotic pressure
between the vesicles and the hypotonic liquid medium causes
liquid to migrate into the vesicles, causing release of the
therapeutic agent. By contrast, the prior art practice of
administering the therapeutic agent in vesicles suspended in
physiological saline which is already hypotonic to the vesicles
results in a much more rapid release of the vesicle contents.
Accordingly, the therapeutic agen-t is also more rapidly re-
leased. By adjusting the osmolarity between the solution of
therapeutic agent and the suspending solution, however, the
rate of release
-- 7 --
5557
can be varled giving a degree of control over this rate not
hitherto attained.
In the practice of the present invention, the thera-
peutic agent i~ dissolved in a suitable solvent, for example,
physiological saline, usually at or near the saturation point
in the case of agents of limited solubility, and encapsulated
in a suitable vesicle, Techn;ques to do this are well known in
the art and need not be described in detail here. Presently
preferred for use in the invention are multilamellar phospho-
lipid vesicles although unilamellar vesicles and vesicles ofother than phospholipid can be used, the basic essential
criterion being that the material of the vesicles be tolerable
by the host in the amounts to be administered.
The solution used for suspending the vesicles i~
preferably physiological saline ~-O.15M NaCl) to which has
been added a second solute to adjust the concentration of this
solution to a level that gives the desired rate of release.
This concentration will be adjusted ~o that the osmoloarity of
the suspending solution is substantially isotonic with the
solution within the vesicles, that is, the suspending solution
contains sufficient solute to provide an osmolarity that is
at least approximately 25%, preferably at least 40 to 50~
of the osmolarity ol the solution wi~thin the vesicles. The usually
desired result of the invention, i.e., lengthening the interval
of sustained rele2se, can be achieved if the osmolarity of
the suspending solution is within the range from substantially
isotonic to hypertonic with respect to the vesicles, with it
being especially ~esirable for the suspending solution to be
at least about 75~ to about 90~ of being isotonic. Since
solute is thus added to the suspending solution, it will be
clear that the suspending solution will have an osmolarity
greater than that of physiological saline.
~SSS7
60724-1519
Suitable for use as solutes in the suspending agents
are any solutes which are well tolerated by the host. Among
these may be mentioned the sugars such as dextrose and the
hexoses such as glucose and polypeptides which do not exhibit
significant biological effects. Presently preferred is glucose
as it is readily obtained as a sterile substance for admini-
stration to humans and is readily absorbed by the body.
Any of a wide variety of therapeutic agents may be
used as part of the invention. Among these may be mentioned
antibiotics, metabolic regulators, immune modulators, toxin anti-
dotes, etc. For example, the invention is well suited for the
controlled release of antidotes to cholinesterase inhibitors.
In order to demonstrate the advantages of the present
invention, there follows a description of experiments carried
out with vesicle encapsulated 2-pralidoxime chloride (2-PAMCl),
an agent which is a well known and thoroughly studied antidote
to toxic organophosphates which inhibit cholinesterase. Persons
exposed to such intoxicants in lethal amounts suffer cardiac
insufficiency or respiratory paralysis which results in death.
~ Agents such as 2-PAMCl reactivate cholinesterase if administered
in a timely fashion. However, dosages of 2-PAMCl high enough to
maintain the therapeutic level for a long period of time cannot
be administered because undesirable side effects, even death,
can result.
EXPERIMENTAL RESULTS
.
A. Materials
L- a-distearoyl phosphatidylcholine (DSPC),
L- a-dipalmitoyl phosphatidylcholine (DPPC) from Calbiochem, and
_g_
12~557
60724-1519
cholesterol (Chol), stearylamine (SA), and dicetylphosphate (DCP)
from Sigma Chemical Cornpany were used without further purifica-
tion to prepare vesicles. 2-Pyridinealdoxime (2-PAM), pralido-
xime chloride (2-PAMCl) and Iodomethane were purchased from
Aldrich Chemical Company and AG lx8 ion exchange resin was from
BioRad Laboratories (Richmond, CA). Ultrapure InC13 was purchased
from Ventron Corporation (Danvers, MA). [ H]-cholesterol oleate
(specific activity: 52 Ci/mole) and [14C]- Iodomethane (specific
activity: 10 Ci/mole) were purchased from New England Nuclear.
Carrier-free InC13 was purchased from Medi-Physics (Glendale,
CA) and purified according to the me-thod of Hwang and Mauk, Proc.
Natl. Acad. Sci. USA, 74, 4991 (1977)~ The ionophore A23187 was
from Eli Lilly and Co. Sprague-Dawley rats in the range of 200-
250 g were purchased from Charles River Laboratories and kept in
an AAALAC approved laboratory for one week before use in expe-
riments.
B. Methods
Preparation of Vesicles
Small unilamellar vesicles (SUV's) were prepared by
probe sonication of the lipid mixture in phosphate buffered saline
(PBS) containing either nitrilotriacetic acid (NTA) or 2-PAMC1.
See Mauk et al, Anal. Biochem., 94, 302, 307 (1979). A trace
amount of [3H] cholesterol oleate was included in the lipid mix-
ture as a marker for the lipid ~hase. Following sonication,
annealing, and low speed centrifugation, the NTA ex-ternal to the
vesicle was removed by passage of the preparation over a
Sephadex G-50 column, equilibrated with PBS.
--10--
~ r~r
~;~455S7
6072g~1519
Large unilamellar vesicles (LUV's) were prepared by
the reverse phase evaporation (REV) method described by Szoka
and Papahadjopoulos, Proc. Natl. Acad. Sci. USA, 75, 4194
(1978). REV vesicles are formed when an aqueous buffer con-
taining the material to be encapsulated is introduced into a
mixture of phospholipid and organic solvent, and the organic
solvent is subsequently removed by evaporation under reduced
pressure. The REV vesicles are then passed through a gel
permeation column to remove the solvent residue and the
unencapsulated drug.
Multilamellar vesicles (MLV's) were prepared by
stirring the dry lipid film with the material to be encap-
sulated. Free unencapsulated materials can be separated from
MLV encapsulated material by centrifugation at 12,000 xg. In
our preparation of MLV's for in v o injection, 40 mg of
DSPC:Chol (2:1 molar ratio) (or other compositions as indicated)
were stirred for 1/2 hour in a round bottom flask with 1-2 ml
of PBA containing 0.5 M 2-PAMCl, or 3M 2-PAMCl as indicated.
Synthesis of [ C] 2-PAMCl
Radiolabeled 2-PAMCl for use in the studies could not
be obtained from any source. Therefore, the radiolabeled drug
was synthesized. [14C] labeled 2-Pralidoxime Iodide (2-PAMI)
was first synthesized by refluxing 2-pyridine aldoxime (2-PAM)
with [14C]-methyl iodide in nitrobenzene for three hours at
75-80C. The reaction was then stopped, and the yellow preci-
pitate of 2-PAMI was filtered and recrystallized from methanol.
These yellow crystals of 2-PAMI were then dissolved in a minimal
amount of water and passed through an anionic exchange column
(BioRad AG lx8). The chloride salt of 2-PAM was isolated by
drying the solution with a rotary evaporator followed by
--11--
~2~5~
60724-1519
recrystallization from ethanol. Approximately 1.5 gm of pure
[14C] labeled 2-PAMCl was obtained with specific activity of
~Ci/mole. The chemical identity of this material was
confirmed by (1) the melting point of the compound which was
found to be 232-234C (literature value 235C) and (2) the
characteristic absorption of an acidic solution of 2-PAMCl at
approximately 295, 245, and 21~ nm; and ~3) co-migration during
thin layer chromatography of the l~C labeled compound with
99~ pùre 2-PAMCl.
In Vitro Vesicle Studies
The chemical structure of vesicles was altered by
varying the length of their phosphatidyl holine carbon chain,
cholesterol content, and surface charge. The stability of the
vesicle formulations ln vitro was studied by gamma-ray perturbed
angular correlation spectroscopy (PAC) as described by Hwang
and Mauk, Proc. Natl. Acad. Sci. USA, 74, 4991 (1977). Prior
to PAC measurements the vesicles were loaded with llInC13.
Typcially 1.0 mg of vesicle withtheionophore A23187 incorporated
within the bilayer was incubated with InC13 at 80C for
45 minutes. During incubation, the lllIn passed through the
ionophore and complexed with NTA inside the vesicles. The
remaining In outside the vesicle was subsequently complexed
to EDTA and separated from the loaded vesicles by chromatograph-
ing the mixture on a Sephadex G-50 column equilibrated with PBS.
The vesicles, now loaded with In-NTA, were then suspended
in a 1:1 solution of physiological saline and rat plasma. Gamma-
ray PAC spectroscopy was then used to monitor the structural
~integrity of vesicles by measuring the tumbling rate of lllIn 3.
-12-
,~
~ ,.
5557
60724-1519
In chelated to nitrilotriacetic acid exhibits
a fast tumbling rate. However, upon disruption of the vesicle,
the released lllIn+3 rapidly binds to macromolecules in the
surrounding medium and exhibits a markedly decreased tumhling
rate. Repeated PAC measurements of each vesicle formulation
were used to estimate vesicle stability based on the time course
for the release of In.
Similarly, identical vesicle formulations loaded with
[ C]-2-PAMCl were used to measure the release rate of 2-PAMCl.
Aliquots were periodically withdrawn from each preparation and
free 2-PAMCl separated from the microencapsulated drug by gel
filtration chromatography as described by Hwàng, Biochem, 8,
344 (1969). The amount of ~ 4C]-2-PA~lCl remaining within the
vesicles was measured by liquid scintillation counting.
In Vi~o Vesicle Studies
The rate of release of [ 4C]-2-PAMCl from the various
vesicle formulations was measured in vivo using male Sprague-
Dawley rats obtained from Charles River Inc. Dosages ranging
from 5-240mg/kg body weight (BW) of free and encapsulated
[ C]-2-PAMCl were injected into the thigh muscle. Individual
injection volumes never exceeded 0.15 ml. At scheduled times
after injection, the rats were either sacrificed or bled through
the eye orbit. Radiolabeled 2-PAMCl was measured in blood and
plasma by liquid scintillation counting using the New England
Nuclear procedure for counting labeled biological material. See
L.S.C. Note, #44, New England Nuclear Applications Laboratory,
Boston, MA.
-13-
.. ..
1~45S57 60724-1519
C. Results
_ . _
In Vitro Vesicle Stability
As indicated in the previous section, the rate of
In release from vesicles in the presence of plasma can be
monitored by PAC spectroscopy. Figure 1 shows the percent
llIn remaining encapsulated with seven vesicle formulations
varying in cholesterol concentration, carbon chain length, and
surface charge. With the exception of the negatively-charged
DCP vesicle, all vesicle formulations with 33 mole percent or
more cholesterol exhibited the same transition temperature as
monitored by 1 lIn release. The DCP formulation produced
vesicles with a transition temperature ( In release)
approximately 10C higher than other vesicle formulations
containing the same amount of cholesterol.
In Vivo Results
In the following results, the data are presented as
the concentration of 2-PAMCl as a function of time for various
vesicle formulations. The time dependence for a standard
in~ection of free 2-PAM Cl shows that the blood concentration
drops below therapeutic level (TL) in 2-3 hours (Figure 6A). As
will be shown, all vesicle formulations with encapsulated
2-PAMCl exhibited extended blood levels. Therapeutic levels of
drug were maintained typically 6-8 hours for those formulations
having only buffered saline as the suspending medium. Vesicles
suspended in a high osmolar medium showed dramatically longer
therapeutic levels.
The Effect of Altering the Chemical Composition of
the Vesicle Membrane
Altering membrane fluidity by changing lipid composi-
tion did not affect the extended release period achieved wi-th
-14-
",~,
~,
,~ . .
~ 557 60724-1519
all vesicle formulations (Figure 2). Similarly, altering the
lipid composition and charge of the vesicle membrane did not
significantly alter the release period either. (Figure 3).
Among the vesicle composition and charge variables
studied, the only factor affecting the release rate for 2-PAMCl
was cholesterol content. Increasing the cholesterol content of
vesicle membranes from 12.5 - 50.0 mole percent appeared to
slightly lengthen the time therapeutic levels of 2-PAMCl
remained in circulation (Figure 4).
The Effects of Altering the Physical Structure of Vesicles
The effect of vesicle structure on the extended
release of 2-PAMCl was examined using multilamellar vesicles
(MLV's) and large unilamellar vesicles (LUV-s) prepared by
reverse phase evaporation (REV) vesicles. As shown in Figure 5,
DSPC: Cholesterol vesicles possessing the two distinct struct-
ures exhibited essentially identical 2-PAMCl release properties.
Consequently MLV's containing 30 mole percent cholesterol were
used in the high osmolarity studies as described below.
The Effects of Altering Encapsulated Volume
and using High Osmolar Suspending Solu-tion
The above data show that there are comparatively
small changes in the time dependency of 2-PAMCl blood-levels for
the formulations tested. The results indicate that simple mani-
pulation of vesicle composition and morphology are not likely
to provide extended release beyond 6-8 hours. These results are
consistent with other published observations which show modest
extended release times.
-15-
12~5557
60724-1519
Presented below are results which show that blood
levels of 2-PAMCl can be dramatically extended by increasing
the osmolarity of the medium in which the vesicles are susp-
ended. Also, proportionately higher concentrations of drug
can be encapsulated without leakage from the vesicles.
The most effective vesicle formulations for extend-
ing the therapeutic plasma levels (4.0 ~g/ml plasma) in rats
was found to be a 2:1 DSPC: Cholesterol lipid mixture, formed
as MLV's and encapsulating a 3 molarsolution of[ C]-2-PAMCl
which was suspended in a isomolar glucose-physiological saline
solution 3M in glucose, after working to remove mother liquor.
In this case, the osmolarity of the suspending medium is about
60% of the osmolarity of the encapsulated drug. Intramuscular
injections (0.15 ml) of this vesicle formulation extended the
therapeutic plasma drug level from 2-1/4 hours, obtained with
the conventional saline formulation, (Figure 6A), to at least
40 hours (Figure 6B). Animals receiving the isomolar vesicle
formulation exhibited no toxic symptoms and their blood drug
levels never exceeded the 20 ~g/ml level achieved by control
animals receiving the standard 12 mg 2-PAMCl saline formu-
lation (Fi~ure 6A).
The extension of therapeutic blood levels is
related to the amount of drug encapsulated. All encapsulation
techniques extended the maintenance of therapeutic drug levels.
Doubling the encapsulating lipid material from 2.5 mg to 5.0
mg by doubling the quantity of lipid (and thereby increasing
encapsulated volume of drug solution) increased the time that
therapeutic levels were maintained from 2.5 to 7.0 hours (Figure
6, C,D). Similar:Ly increasing the amount of drug encapsulated
from a saline solution containing 60 mg of 2-PAMCl prevented
-16-
t~
1245S5~
60724-1519
all acute toxic symptoms while extending the therapeutic blood
titers to 15 hours (Figure 6 E). All animals receiving similar
60 mg injections of 2-PAMCl unencapsulated in saline solution,
or encapsulated using a technique which reduces the encapsula-
tion ef~iciency died within 30 minutes a~ter injection.
These data suggest that encapsulated drug acts as
a third compartment ~rom which its slow release lowers the
peak blood levels seen when equal, ~mencapsulated dosages
are injected.
-16a-
.~ . .
~45557
This delayed release prevents toxic blood levels from being
attained and conserves drug for later therapeutic use. All
encapsulation systems tested extended the time therapeutic
blood levels of drug were maintained from 2.25 hours to about
6.0 hours. However, the chemical structure of the lipid membrane
and the physical structure of the vesicle (whether MLV, SUV,
or L W) had little effect on extending therapeutic blood levels.
The foregoing demonstrates that suspension of vesicle
formulations containing a therapeutic agent in a solution that
approaches being isotonic or even hypertonic with respect -to the
vesicle greatly extends the interval over which sustained
release occurs. It further demonstrates that dose levels of
a therapeutic agent which are toxic when the agent is adminis-
tered alone or in less effective sustained release formulations
can be safely given using a composition of the present invention.
It will be understood by those skilled in the art
that the foregoing merely illustrates the presently preferred
embodiments of the invention and that modifications may be
made in order to accomplish specific ends which donot depart
fromlthe spirit of the present invention which is to be limited
only by the appended claims
