IL139053A - Pharmaceutical composition comprising factor VIII and neutral colloidal particles - Google Patents

Description

n np ni-nann Pharmaceutical composition comprising factor VIII and neutral colloidal particles Opperbas Holding B.V.

C. 128687 1 139053/2 FIELD OF THE INVENTION The present invention relates to a stable pharmaceutical formulation for the slow release of coagulation promoting substances for the treatment of blood coagulation disorders.

BACKGROUND OF THE B FJNTnON Hemophilia A is one of the most frequently occurring inherited coagulation disorders. Patients with hemophilia A are prone to frequent hemorrhages as a result of one or more rmsfunctions of the coagulation system. One of the causes of hemophilia is a shortage of Factor VHI (FVffl) in the blood. This problem can be treated with Factor VM concentrates. However, in about 15% of the patients the occurrence results of Factor VUI neulializing antibodies, so-called inhibitors, whereby a therapy with Factor VUI concentrates is hardly possible.

Two basic approaches have been described in the literature to ' protect FVEH from inactivation by inhibitors.

WO/80/01456 to Hemker discloses a pharmaceutical composition suitable for oral administration comprising FVm incorporated within liposomes of 0.5-1.0 microns formed from phospholipids. The phospholipids have a net charge, and the FVH. is incorporated between the layers of the liposome. It is claimed that FVffl levels in the plasma remained above about 5% of the normal value for a period of 50 hours.

US 4,348,384 to Horikoshi states that a composition as described in Hemker was prepared, but did not give satisfactory results. Therefore, Horikoshi incorporates a protease inhibitor into the liposome together with FVTfl, in order to protect it from proteolysis. 3% of the normal plasma levels of FVTfi were obtained over a period of 6 hours.

US 5,013,556 to Woodle discloses a liposome composition for use in delivering various drugs via the bloodstream. . The liposome contains between 1-20 mole percent of an amphipathic lipid derivatized with a polyalkylether. Here also, the drug compound is entrapped within the liposome. These liposome compositions are available commercially under the name of Stealth® vesicles (SUV's, small unilamellar vesicles comprised of phospholipid and polyethylene glycol (PEG) covalently bound to phospholipid).

A further problem with this approach is that liposomes having a large diameter have a short half-life. Therefore, the liposomes must be downsized under high pressure, which , can affect protein activities as in . coagulation factors V and VTfl.

In a second approach, Barrowcliffe, T.W., et al. (1983) J. Lab.

Clin. Med. 101 :34-43 teaches that mixing FVffl with phospholipid extracted from human and/or animal brain imparts significant protection to the FVTfi in vitro. In this approach, the phospholipid is bound to the FVffl rather than encapsulating it. emball-Cook, G. and Barrowcliffe, T.W. (1992) Thromb. Res. 67:57-71, teaches that a negatively-charged phospholipid surface is necessary for FVffl binding. Negatively charged phosphatidyl serine and phophatidic acid were found to be highly active in binding to FVffl, while phosphatidyl choline was inactive. However, negatively-charged 139053/2 phospholipids are toxic, and those derived from brain tissue may carry pathogenic agents.

EP 689,428 discloses a liposome composition comprising liposomes having an outer surface layer of hydrophilic polymer chains. A polypeptide or polysaccharide effector molecule is covalently attached to the distal ends of the polymer chains by activation of the lipid anchor prior to effector coupling.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a pharmaceutical composition comprising a protein or polypeptide for therapeutic treatment In particular, it is an object of the present invention to provide a pharmaceutical composition comprising FVIH for the treatment of blood coagulation disorders.

It is a further object of the invention to provide FVTH in a form having an extended half-life in the bloodstream.

It is a still further object of the invention to provide a use of a colloidal particle in the preparation of a pharmaceutical composition for use in a method for treating patients suffering from blood coagulation disorders, particularly hemophilia, and most particularly those having FVIII inhibitors.

In one aspect of the present invention there is provided a pharmaceutical composition for parenteral . admkiistration comprising a therapeutically effective amount of coagulation factor Vm (FVEE) and substantially neutral colloidal particles, the particles comprising 1-20 mole ■percent of an amphipathic lipid derivatized with a biocompatible hydrophilic polymer, the polymer carrying substantially no net charge, wherein the FVIII is not encapsulated in the colloidal particles.

The present invention is based on the surprising and unexpected finding that neutral phospholipids derivatized with a bio-compatible hydrophilic polymer can be used to bind FVEI and protect it from inhibitors in the bloodstream. This provides a significant advantage over the prior art compositions, since the phospholipids used are synthetic and non-toxic, and can therefore be used in vivo for therapeutic treatment. Furthermore, the liposome does not encapsulate the FVffi so that smaller sized liposomes can be used which have a longer half-life in vivo, since they are not removed by the reticuloendothelial system (RES). As will be described below in greater detail, FVTfl interacts non-covalently with the polymer chains on the external surface of the liposomes, and no chemical reaction is carried out to activate the polymer chains, unlike the composition disclosed in EP 689,428.

In the present specification, the terms "substantially neutral" and "substantially no net charge" mean neither positively nor negatively charged. However, a very low measured charge within experimental error of zero is included within the meaning of the above terms.

The term "therapeutically effective amount" is to be understood as referring to an amount of FVffi which results in a level of FVITI in the bloodstream having a desired therapeutic effect. Such an amount can be experimentally determined by administering compositions comprising different amounts of FVTTI and measuring the level in the blood at various times after adrninistration.

The amphipat ic lipid used to prepare the colloidal particles is preferably a phospholipid, and may be obtained from either natural or synthetic sources. A most preferred phospholipid is phosphatidylcholine, most preferably egg-phosphatidylcholine.

The biocompatible hydrophilic polymer may include polymers from the polyalkylether, polylactic or polyglycolic acid families. Preferably, the polymer is polyethylene glycol (PEG). The purpose of the polymer is to sterically stabilize the SUVs, thus preventing fusion of the vesicles in vitro, and allowing the vesicles to escape adsorption by the RES in vivo. The 139053/2 polymer will preferably have a molecular weight of between about 1000 to about 5000 daltons, more preferably approximately 2000 daltons.

The colloidal particles will preferably have a mean particle diameter of between 0.05 to 0.4 microns, most preferably about 0.1 microns. This is to increase their circulation time in vivo and prevent their adsorption by the RES. The amphipathic lipid comprises 1 to 20 mole % of the particles, preferably approximately 1-5%, most preferably 5%.

A variety of known coupling reactions may be used for preparing vesicle forming lipids derivatized with hydrophilic polymers. For example, a polymer (such as PEG) may be derivatized to a lipid such as phosphan^ylethanolamine (PE) through a cyanuric chloride group. Alternatively, a capped PEG may be activated with a carbonyl diimidazole coupling reagent, to form an activated imidazole compound. Other reactions are well known and are listed, e.g. in the aforementioned U.S. 5,013,556, whose contents are incorporated herein by reference.

The FVTH used in the composition of the invention is commercially available. It may be from a natural human source, or, preferably, it may be recombinantly prepared. Recombinant FVHI is commercially available, for example, Antihemophilic Factor (Recombinant), rFVni-SQ (Pharmacia), and Kogenate, Miles Inc., Pharmaceutical Division, Elkhart, IN, U.S A., among other suppliers.

The composition of the invention is administered parenterally, preferably iv. The prior art compositions were intended for oral use only, due to side effects caused during injection by the liposome composition. The composition of the invention, on the other hand, is not toxic by injection, apparently due to the lack of charge, among other causes. Amounts of up to 0.5gm/Kg body weight of colloidal particles according to the invention have been injected without detectable toxic symptoms. The dose is expected to be 6 139053/2 in the approximate range of 25-75 i.u./Kg. body weight. The particle to FVIII ratio (w/unit FVIII) will preferably be between 0.1 mg/unit and 10 mg/unit, and most preferably, approximately 1 mg/unit. Although the free form of FVIII :C has a half-life of less than 2 hours (FVIII measured by clotting activity) in mice, FVIII administered in the composition of the invention is expected to be effective for at least 24 hours, which is the period of effective activity of the coagulation promoting compound. The composition of the invention is expected to be effective in "on demand" and prophylactic treatment of hemophilia patients, and particularly those patients who have developed FVIII inhibitor antibodies.

The effectiveness of FVTJI contained in the composition of the invention may be determined by a chromogenic assay which determines FVTJI activity by two consecutive steps: (1) the FVHI-dependent conversion of Factor X to Factor Xa in a coagulation-factor reagent composed of purified components, and (2) the enzymatic cleavage of a chromogenic Factor Xa substrate to yield a chromophore which can be quantified spectrophotometrically. Under appropriate assay conditions, there exists a linear relationship between the rate of Factor Xa formation and the FVHI concentration. In addition, FVffl activity may be determined by a one-stage clotting assay. This assay deterrnines FVTJI activity by the conversion of prothrombin to thrombin, which subsequently cleaves fibrinogen to form a clot composed of fibrin. FVHI activity in hemophillic mice may also be determined by measuring the survival of the mice following a tail cut In a further aspect of the invention, there is provided a pharmaceutical composition for parenteral administration comprising a therapeutically effective amount of a protein or polypeptide and substantially neutral colloidal particles, said particles comprising 1-20 mole percent of an amphipathic lipid derivatized with a biocompatible hydrophilic polymer, said polymer carrying substantially no net charge, wherein said 01286871X23-01 7 139053/2 protein or polypeptide is selected from the group consisting of: (a) proteins or polypeptides capable of externally binding said colloidal particles; and (b) proteins of polypeptides capable of binding polyethylene glycol (PEG), and wherein said protein or polypeptide is not encapsulated in said colloidal particles.

The term 'proteins or polypeptides capable of externally binding said colloidal particle " includes proteins and polypeptides which, similarly to FVffl, bind to membranes comprising phosphatidylcholine: phosphau ylserine (PC:PS) (see Haemostasis and Thrombosis. Arthur L. Bloom and Duncan P. Thomas (eds) (1987) Churchill Livingstone, pg. 179-180). Non-lirriiting examples of such proteins are coagulation factors such as prothrombin, Factor X and Factor V.

The term ^proteins or polypeptides capable of binding polyethylene glycol" includes proteins and polypeptides which bind to PEG or derivatives of PEG by any non-covalent mechanism, such as . ionic interactions, hydrophobic interactions, hydrogen bonds and Van der aals attractions (Arakawa, T. and Timashef£ S.N. (1985) Biochemistry 24:6756-6762; Lee, J.C. and Lee, LJL.Y. (1981) J. Biol. Chem. 226:625-631).

Passages of the description which are not within the scope of the claims do not constitute part of the invention.

BRIEF DESCRIPTION OF THE DRAWING: In order to understand the invention and to see how it may be carried out in practice, preferred embodiments will now be described, by way of non-liniiting example only, with reference to the accompanying drawing which illustrates survival of hemophilic mice injected with FVm following a tail cut at various time periods post-injection. 01286871Y23-01 DETAILED DESCRIPTION OF A PREFERRED EMBODIMENT Examples 1-3 Methods and Materials 1. Egg phosphatidylcholine (E-PC) liposomes A tert-butanol solution of egg phosphatidylcholine (E- PC) was prepared by dissolving 2.0 gr. E-PC, 1.9 mg -tocopherol and fluorescein-labeled phosphatidylethanolamine (1 : 1000 lipid molar ratio) in 18ml tert-butanol.

The organic solvent was removed from the lipidic mixture by lyophilization and the lipids reconstituted in water to 10 % w/v. The obtained liposomes were reduced in size by extruding them through a series of polycarbonate (PC) filters (0.4 um, 0.2 μχη, 0.1 u and 0.05 um) using the Liposofast-Basic or Liposofast-50 extruder (Avestin) to obtain liposomes of an average size of 0.1 μτη. 2. Egg phosphatidylcholine/polyethyleneglycol-phosphatidyl ethanolamine (E-PC/PEG-PE) liposomes . A tert-butanol solution of egg phosphatidylcholine (E- PC) and polyethyleneglycol-phosphatidyl ethanolamine (PEG-PE) was prepared by mixing 0.73 gr. E-PC, 0.185 gr. PEG-PE, 0.86 mg - a-tocopherol and fluorescein labeled phosphatidyl emanolamine (1:1000 lipid molar ratio) in 18ml tert-butanol.

The organic solvent was removed from the lipidic mixture by lyophilization and then reconstituted in water to 10 % w/v. The obtained liposomes were reduced in size as described in 1 above to obtain liposomes of an average size of 0.1 um. s 3. Egg phosphatidylcholine-phosphatidyl glycerol (E-PC/PG) liposomes A tert-butanol solution of egg phosphatidylcholine (E- PC) and phosphatidyl glycerol (PG) was prepared by mixing 0.822 gr. E-PC, 0.0924 5 gr. PG, 0.86 mg a-tocopherol and phosphatidylethanolamine fluorescein labeled (1 : 1000 lipid molar ratio) in 18ml tert-butanol.

The organic solvent was removed from the lipidic mixture by lyophilization and then reconstituted in water to 10 % w/v. The obtained liposomes were reduced in size as described in 1 above to obtain liposomes of 10 an average size of 0.1 um. 4. Reconstitution of the human recombinant factor Vffl: Kogenate (rFVTfl formulated with human albumin, Bayer) lots 70K026 and 70K027, were used in the following examples. One vial 15 containing about 500 IU of FVTU activity was reconstituted with 2 ml water and allowed to solubilize. 200 μΐ aliquots were frozen at -20°C until use.

For the preparation of albumin-depleted Kogenate, lot # 70K027 was used. 10 vials of Kogenate were reconstituted in 20 ml water and chromatographed on a hydrophilic silica gel (3-10 um beads). Fractions of 20 10ml were collected and the protein and FVTILAg activities were monitored. A 50 % recovery in FVTJLAg activity was found in one peak of fractions 4-6 and another of fractions 8-14. Since the protein assay gave a peak at fractions 9-12, fractions 4-6 were pooled, aliquoted and lyophilized for further use. 5. Hemophilic mice prepared as described in Bi, L., et al. (1995) Nature Genetics 10:119-121, were used. 6. FVIILAg activity was determined using a FVffl chromogenic assay commercially available from Dade AG, Dudingen, Switzerland. 7. Preparation of composition and injection to hemophiliac mice A liposomal aliquot was mixed with a predetermined volume of FVIII to obtain a FVIILAg activity of 5-10 IU/ml and rolled at RT to achieve homogeneity. 8. Groups of 5-10 hemophiliac mice were injected IV bolus through the tail vein, with 200 or 400 μΐ of the mixture. The mice were bled from the eye at regular time intervals (lh, 4hs, 8hs, 24hs, 32hs and 48hs) and the FVIILAg activities in the plasma were followed. 9. The pharmacokinetics of FVIII was detennined from the results by using the RSTRIP computation software to obtain the initial FVTfLC activity (Ao) and the half-life time ( ) of the factor in the mice blood circulation.

Results 1. Effect of Lipid composition on the Half-Life of Factor VUL Liposomes of 0.05 um comprising E-PC, E-PC/PEG-PE and E-PC/PG were prepared, mixed with Kogenate in a 72:1 lipid to protein (w/w) ratio and injected into hemophiliac mice. As a control, Kogenate was diluted in saline and injected into the mice in the same manner as the liposomal mixtures. The ph-umacokinetic parameters were determined as described above, and the results are summarized in the following table: Table # 1 : Effect of lipid composition on the half-life of FVIII * Ao = initial concentration of FVHI:C It can be seen from the table that liposomes containing E-PC /PEG-PE were the most effective since both the initial FVTII activity and the half-life time were higher for this composition than for Kogenate or Kogenate-liposome mixtures where the liposomes were composed of E-PC/PG or E-PC only.

Moreover, 40% of the mice injected with free FVIII and 100% of the mice injected with FVIII /PC complex did not exhibit any recovery of FVUI chromogenic activity, while only 10% of the mice injected with FVIII/PC+PEG exhibited the same phenomena 60 min. after injection. 2. Effect of Lipid/Protein Ratio on the Half-Life of Factor VEX Various lipid to protein ratios in the liposome composition were obtained by mixing various aliquots of liposomes of 0.05um comprising E-PC/PEG-PE with Kogenate. These were injected into hemophiliac mice. As a control, Kogenate was diluted in saline and injected into the mice in the same manner as the liposomal mixtures. The phannacokinetic parameters were determined as described above, and the results are summarized in the following table: Table #2: Effect of lipid to protein ratio on the half-life of F VIII It can be seen from Table #2 that increasing the lipid/protein ratio increases the half-life time of FVffl in the blood circulation in the hemophiliac mice. The differences in the initial FVHI.'C activities appear not to be related to the lipid/protein ratio. 3. Effect of different Factor Vffi sources SUVs of 0.05μτη were prepared containing E-PC and PEG-PE (94:6 mol %), mixed with FVHI concentrates from various sources (Kogenate, Baxter and Omrixate) in a 72:1 lipid to protein ratio and injected into hemophiliac mice. As a control, each FVffi concentrate from the various sources was diluted in saline and injected into the mice in the same manner as the liposomal mixtures. The pharmacokinetic parameters were determined as described above, and the results are summarized in the following table: Table #3: Effect of factor FVIII source on the half-life of F VIII Mixtures containing liposomes and FVIII from Baxter or Omrixate increased the half-life of the factor by 20%, when compared with the pharmacokinetic values of the free factor, as can be seen from the above table. The half-life of factor FVIII from Kogenate, mixed with E-PC/PEG-PE liposomes was twice as long as compared with the free factor form.

Example 4 Methods and materials 1. Liposome preparation: Liposomes were prepared as follows: Egg phosphatidyl choline (EPC) and distearoyl phosphatidyl-ethanolamine methyl polyethylene glycol 2000 (DSPE-PEG 2000) were weighed to a ratio of 80:20 w/w (5% molar ratio of DSPE-PEG 2000), respectively, dissolved to 10% w/v in tert-buthanol (Reidel-de Haen), and the solution was lyophilized. The obtained dry lipid powder was resuspended to 10% w/v in a buffer. containing 130 mM NaCl, 10 mM sodium citrate, 1 mM CaCh pH 7.0 to form liposomes. The liposomes were filtered in an extruder apparatus (Avestin) through polycarbonate filters 1.2 μτη, 0.2 μτη and 0.1 μτη in size to form liposomes of 120-140 nm in size. 2. Liposome quality control Quality control of the liposomes included: 1) Size distribution measured by sub-micron particle analyzer (N4 plus, Coulter Electronics). 2) Phospholipid determination (by phosphorus). 3) Chemical stability of the lipids by TLC.

The tests were performed as described in: Barenholz, Y. and Amselem, S. (1993) in Liposome Technology, 2nd edition, Vol. I (Gregoriadis, G., ed.), CRC Press, Boca Rayton, Fl, pp.527-616. 3. Formulation of FVHI and liposomes Kogenate (Lot no. 70K027, 620 IU) or New Kogenate (500 IU) was dissolved in 1 ml or 2 ml of ¾O. The rFVffl-SQ concentrate was dissolved in liposome solution. Factor VTfl was formulated with liposomes by mixing FVTfl concentrate with the liposomes for about 1 hour at room temperature. The ratio of lipids to FVITI units was about 1 mg lipids/1 unit Fvm. 4. Injection into hemophilic mice and bleeding procedures Factor Vffl and FVm formulated with liposomes were injected - into the tail vein of hemophilic mice. The injected dose was 3 units/mouse for Kogenate (2 separate experiments) and New Kogenate and 4 units/mouse for rFVni-SQ. The mice were bled into citrate tubes at 10 minutes after the injection and at about 4, 19 and 27 hours post-injection.

. Measurement of FVIII concentrate in mouse plasma Human FVIII concentrate in mouse plasma was measured using a chromogenic assay (Chromogenix) according to the manufacturer's instructions, and by one stage clotting assay (using Stago reagents and ST4 clotting machine) according to the manufacturer's instructions. 6. Pharmacokinetics analysis Pharmacokinetics parameters were analyzed using a computer program (RSTRIP, MicroMath Inc.). 7. Survival of hemophilic mice following a tail cut Mice were injected with free rFVffi-SQ or liposome formulated rFVIII-SQ (4 units/mouse, 11 mice in each group. At 20 hours post injection 2 cm of the tail were cut. Tails of the surviving mice were cut again at 28, 44, 52, 69, 88 and 140 hours post-injection (2 mm each time).

Results The results of FVTfl activity at each time point post-injection and the pharmacokinetic parameters [FVIII half-life (HL) and the area under the curve (AUC)] of 4 different experiments are summarized in Tables 4-7B. In Tables 4 and 5, 3 units of Kogenate /mouse were injected; In Table 6, 3 units of New Kogenate /mouse were injected; and in Tables 7 A, 7B, and in Fig. 1, 4 units of rFVITI-SQ /mouse were injected.

Table 4: Factor VIII activity (u/ml measured by a chromogenic assay) and phaimacokinetic parameters following injection of human FVIII into hemophilic mice Injected T = T = 4.5 T= 19 T = 27 Area Half Material hours hours hours under life min. the (HL) curve (h) (AUC) * (TU*h/ ml) Kogenate Average 2.878 0.450 ± 0.015 ± 0.0016 7.218 1.619 (n=7) (u/ml) ± + 0.392 0.0043 ±0.004 SD 0.571 Injected T = T = 3.6 T= 18.1 T Material hours hours 26.1 min. hours Kogenate Average 2.951 1.121 ± 0.023 ± 0.014 ± 10.9.69 2.460 (u/ml) ± + 0.337 0.003 0.0018 liposome SD 0.333 s (n=7) Table 5: Factor VIII activity (u/ml measured by a chromogenic assay) and pharmacokinetic parameters following injection of human FVIII into hemophilic mice Injected T = T T= T Area Half Material 3.334 19.334 26.3 under life min. hours hours hours the (HL) curve (h) (AUC) (IU*h/ ml) ogenate Average 3.686 0.960 ± 0.0128 0.0025 9.314 1.632 (n=8) (u/ml) ± ± 0.469 ± 0.007 ± 0.006 SD 0.674 Injected T = T = 3.5 T= 19.5 T Material hours hours 26.5 min. hours Kogenate Average 3.618 1.571 ± 0.032 ± 0.012 ± 15.059 2.771 + (u/ml) ± + 0.137 0.009 0.009 liposome SD 0.982 s (n=8) Table 6: Factor VIII activity (u/ml measured by a chromogenic assay) and pharmacokinetic parameters following injection of human FVTII into hemophilic mice Injected T = T = 2.5 T= 17 T = 26 Area Half Material hours hours hours under life min. the (HL) curve 00 (Aue> (IU*h/ ml) New Average 1.841 0.120 ± 0.004 ± 0.0015 1.910 0.592 Kogenate (u/ml) ± ± 0.036 0.003 ± 0.003 (n=4) SD 0.643 Injected T = T T= T Material 2.666 17.166 26.166 min. hours hours hours New Average 2.393 0.352 ± 0.011 ± 0.008 ± 3.544 0.904 Kogenate (u/ml) ± + 0.131 0.0019 0.001 + SD 0.243 liposome s (n=6) Table 7A: Factor VIII activity (u/ml measured by a chromogenic assay) and pharmacokinetic parameters following injection of human FVIII into hemophilic mice Injected T = T T= Area Half Material 4.166 20.166 under life min. hours hours the (HL) curve (h) (AUC) (IU*h/ ml) RVin Average 3.937 0.444 ± 0 7.9 1.270 (n-10) (u/ml) ± + 0.131 SD 0.449 Injected T = T = 4.5 T= 20.5 Material hours hours min.

Rvm + Average 3.828 0.555 ± 0.005 ± 9.249 1.555 liposome (u/ml) ± ± 0.198 0.008 s (n=l l) SD 1.08 Table 7B: Factor VIII activity (u/ml measured by a one stage clotting assay) and pharmacokinetic parameters following injection of human FVIII into hemophilic mice In addition, the half-life of human FVIII in each mouse was calculated and the FVffi half-lives in all the experimental groups were statistically compared to each other by a student t-test The statistical analysis indicates that in all 4 experiments human FVTfl half-lives in the groups that received liposome-formulated FVHI were higher and significantly different (p<0.055) from human FVTil half-lives in the groups that received free FVIII: HL of Kogenate versus HL of Kogenate + liposomes (table 4) p=0.054; HL of Kogenate versus AUC of Kogenate + liposomes (table 5) p=0.031; HL of New Kogenate versus HL of New Kogenate + liposomes (table 6) p=0.0085; HL of rFVni-SQ versus HL of rFVTfl-SQ + liposomes (table 7A) p=0.0045; HL of rFVm-SQ versus HL of rFVIII-SQ + liposomes (table 7B) p=0.022).

These results indicate that the formulation of Kogenate, New Kogenate or rFVTII-SQ with liposomes significantly increases the factor half-life (HL) and the area under the curve (AUC) of FVIII in hemophilic mice (factor of 1.6-2.0 for HL and AUC).

Survival of the mice is illustrated in Fig. 1. The results of the tail cut experiment indicate that the liposome formulated FVIII is biologically active longer than free FVIII, and therefore can protect hemophilic patients for a longer period of time.

Example 5 Effectiveness of F VTA composition in patients with inhibitors units of FVIII (Kogenate) were incubated for one hour at room temperature with 120 run liposomes (15 mg lipids) containing EPC:DSPE-PEG2000 (95:5 mole %). Then, 1 unit of free FVIII (Kogenate) or 1 unit of liposome formulated FVIII were incubated for 2 hours at 37 °C with various dilutions of a serum from a hemophilia patient who had developed inhibitors (anti FVIII antibodies). After the incubation, the activity of factor VIII was measured by a criminogenic assay.

The results are summarized in table 8: Table 8: Activity (units/ml) of factor VIII in the presence of FVIII inhibitors It can clearly be seen from this experiment that adrninistration of FVTTI together with the colloidal particles is effective in protecting the FVIII from serum inhibitors.

Claims (19)

- 23 - 139053/2 CLAIMS:

1. A pharmaceutical composition for parenteral administration comprising a therapeutically effective amount of coagulation factor VIII (FVIII) and substantially neutral colloidal particles, said particles comprising 1-20 mole percent of an amphipathic lipid derivatized with a biocompatible hydrophilic polymer, said polymer carrying substantially no net charge, wherein said FVIII is not encapsulated in said colloidal particles.

2. The pharmaceutical composition of Claim 1 wherein the colloidal particle has a mean particle diameter of between 0.05 to 0.4 microns.

3. The pharmaceutical composition of Claim 2 wherein the colloidal particle has a mean particle diameter of approximately 0.1 microns.

4. The pharmaceutical composition of Claim 1 wherein said amphipathic lipid is a phospholipid from natural or synthetic sources.

5. The pharmaceutical composition of Claim 4 wherein said amphipathic lipid is phosphatidyl ethanolamine derivatized with PEG.

6. The pharmaceutical composition of Claim 1 wherein said biocompatible hydrophilic polymer is selected polyalkylether, polylactic and polyglycolic acid families.

7. The pharmaceutical composition of Claim 6 wherein said biocompatible hydrophilic polymer is polyethylene glycol.

8. The pharmaceutical composition of Claim 7 wherein the polyethylene glycol has a molecular weight of between 1000 to 5000 daltons.

9. The pharmaceutical composition of Claim 8 wherein the polyethylene glycol has a molecular weight of approximately 2000 daltons.

10. The pharmaceutical composition of Claim 1 wherein the FVIII is from a natural source.

11. The pharmaceutical composition of Claim 1 wherein the FVIII is recombinantly prepared.

12. The pharmaceutical composition of Claim 1 wherein the particle to FVIII ratio (w/unit FVIII) is between 0.1 mg/unit and 10 mg/unit. - 24 - 139053/2

13. The pharmaceutical composition of Claim 12 wherein the particle to FVffi ratio (w/unit FVIII) is approximately 1 mg/unit.

14. The pharmaceutical composition of Claim 1 wherein said amphipathic lipid comprises phosphatidylcholine.

15. The pharmaceutical composition of Claim 1 comprising phosphatidylcholine and phosphatidyl ethanolamine derivatized with PEG.

16. Use of a colloidal particle in the preparation of a pharmaceutical composition for parenteral administration for treatment of a patient suffering from hemophilia comprising a therapeutically effective amount of coagulation factor VIII (FVIII) and substantially neutral colloidal particles, said particles comprising 1-20 mole percent of an amphipathic lipid derivatized with a biocompatible hydrophilic polymer, said polymer carrying substantially no net charge, wherein said FVIII is not encapsulated in said colloidal particles.

17. A use according to Claim 16 wherein said patient has developed FVIII inhibitor antibodies.

18. A pharmaceutical composition for parenteral administration comprising a therapeutically effective amount of a protein or polypeptide and substantially neutral colloidal particles, said particles comprising 1-20 mole percent of an amphipathic lipid derivatized with a biocompatible hydrophilic polymer, said polymer carrying substantially no net charge, wherein said protein or polypeptide is selected from: (a) proteins or polypeptides capable of externally binding said colloidal particles; and (b) proteins or polypeptides capable of binding polyethylene glycol, and wherein said protein or polypeptide is not encapsulated in said colloidal particles.

19. Use of a colloidal particle in the preparation of a pharmaceutical composition for parenteral administration comprising a therapeutically effective amount of a protein or polypeptide and substantially neutral colloidal particles, said particles comprising 1-20 mole percent of an amphipathic lipid derivatized with a - 25 - 139053/2 biocompatible hydrophilic polymer, said polymer carrying substantially no net charge. wherein said protein or polypeptide is selected from: (a) proteins or polypeptides capable of externally binding said colloidal particles; and (b) proteins or polypeptides capable of binding polyethylene glycol, and wherein said protein or polypeptide is not encapsulated in said colloidal particles. For the Applicants,