CONCENTRATED COMPOSITIONS OF PROTEINS, THEIR PREPARATION AND USE THEREOF

Abstract

The invention relates to a method for producing a composition comprising reversible protein complexes (RPCs), the method comprising the steps of contacting a protein and a complexing agent in a buffer solution, wherein the complexing agent is dextran sulphate or chondroitin sulphate, and wherein the protein and the complexing agent have opposite net charges when comprised in the buffer solution; formation of RPCs between the protein and the complexing agent in the buffer solution; and obtaining a suspension comprising the RPCs. Provided herein are also compositions, including pharmaceutical compositions/ formulations comprising the reversible protein complexes (RPCs) of the invention, in particular as obtained by the method provided herein.

Claims

1. A method for producing a composition comprising reversible protein complexes (RPCs), the method comprising the steps of: a) contacting a protein and a complexing agent in a buffer solution, wherein the complexing agent is dextran sulphate or chondroitin sulphate, and wherein the protein and the complexing agent have opposite net charges when comprised in the buffer solution; b) formation of RPCs between the protein and the complexing agent in the buffer solution; and c) obtaining a suspension comprising the RPCs formed in step (b).

2. The method according to claim 1, wherein the complexing agent is dextran sulphate, in particular dextran sulphate with 40 kDa molecular weight.

3. The method according to claim 1 or 2, wherein the pH of the buffer solution is adjusted to be lower than the isoelectric point of the protein.

4. The method according to any one of claims 1 to 3, wherein the pH of the buffer solution is adjusted to 2 to 5 pH units below the isoelectric point of the protein, in particular 3 pH units below the isoelectric point of the protein.

5. The method according to claim 1 or 2, wherein the buffer solution has a pH ranging from 1 to 6, in particular wherein the buffer solution has a pH ranging from 3 to 6, in particular wherein the buffer solution has a pH ranging from 4.5 to 5.5.

6. The method according to any one of claims 1 to 5, wherein the buffer solution has an ionic strength ranging from 20 to 50 mM, in particular wherein the buffer solution has an ionic strength ranging from 20 to 30 mM.

7. The method according to any one of claims 1 to 6, wherein the buffer solution comprises histidine or citrate as buffering agent.

8. The method according to any one of claims 1 to 7, wherein the buffer solution comprising the protein and the complexing agent is obtained by mixing a first solution comprising the protein and a second solution comprising the complexing agent.

9. The method according to claim 8, wherein the first solution comprising the protein and/or the second solution comprising the complexing agent comprises a buffering agent.

10. The method according to any one of claims 1 to 9, wherein the protein and the complexing agent are contacted at a mole-charge ratio ranging from 1:0.2 to 1:2, in particular wherein the protein and the complexing agent are contacted at a mole-charge ratio ranging from 1:0.2 to 1:1.

11. The method according to any one of claims 1 to 10, wherein the protein is contacted with the complexing agent in the buffer solution at a protein concentration ranging from 1-40 mg/mL, in particular from 1-5 mg/mL.

12. The method according to any one of claims 1 to 11, wherein the protein is an antibody, a growth factor, a hormone, a cytokine, an enzyme, or a fragment and/or fusion protein of any of the foregoing.

13. The method according to claim 12, wherein the antibody is an antibody, in particular wherein the antibody is a monoclonal antibody, a polyclonal antibody, a chimeric antibody, a multispecific antibody, an antibody fusion protein, an antibody-drug-conjugate or an antibody fragment.

14. The method according to any one of claims 1 to 13, wherein the complexing agent has a negative net charge when comprised in the buffer solution.

15. The method according to any one of claims 1 to 14, wherein the complexing agent comprises a hydrophobic moiety.

16. The method according to any one of claims 1 to 15, wherein the composition comprising the RPCs comprises at least one excipient.

17. The method according to claim 16, wherein the at least one excipient is added to the composition before and/or after the formation of the RPCs.

18. The method according to claim 17 or 18, wherein the at least one excipient is a stabilizer and/or a surfactant.

19. The method according to any one of claims 1 to 18, wherein the method comprises a further step of exchanging the liquid fraction of the suspension comprising the RPCs.

20. The method according to claim 19, wherein the liquid fraction of the suspension comprising the RPCs is exchanged by centrifugation of the suspension comprising the RPCs and resuspension of the sedimented RPCs in a buffer solution or water.

21. The method according to claim 19, wherein the liquid fraction of the suspension comprising the RPCs is exchanged by dialysis of the suspension comprising the RPCs against a buffer solution or water.

22. The method according to any one of claims 1 to 21, wherein the method comprises a further step of enriching the RPCs in the suspension to obtain an enriched RPC suspension.

23. The method according to claim 22, wherein enriching the RPCs in the suspension comprises the steps of: a) centrifuging the suspension comprising the RPCs to obtain a supernatant and a precipitate comprising an enriched RPC suspension; and b) removing the supernatant from the precipitate to obtain an enriched RPC suspension.

24. The method according to claim 22 or 23, wherein the liquid fraction of the enriched RPC suspension is at least in part replaced with a non-aqueous solvent during the enrichment step.

25. The method according to claim 24, wherein the non-aqueous solvent is triacetin, diethylene glycol monoethyl ether or ethyl oleate.

26. The method according to any one of claims 1 to 25, wherein the method comprises a further step of lyophilizing the suspension comprising the RPCs or the enriched RPC suspension to obtain a lyophilisate.

27. The method according to claim 26, wherein at least one cryoprotectant is added to the suspension comprising the RPCs or the enriched RPC suspension before the lyophilisation step.

28. The method according to claim 27, wherein the at least one cryoprotectant is selected from a group consisting of sugars, amino acids, methylamines, lyotropic salts, polyols, propylene glycol, polyethylene glycol and pluronics.

29. The method according to any one of claims 26 to 28, wherein the protein concentration of the suspension comprising the RPCs or the enriched RPC suspension is adjusted to 10 to 100 mg/mL, in particular to 40 to 80 mg/mL, prior to the lyophilisation step.

30. The method according to any one of claims 1 to 25, wherein the method comprises a further step of spray drying the suspension comprising the RPCs or the enriched RPC suspension to obtain a spray dried powder.

31. The method according to claim 30, wherein the protein concentration of the suspension comprising the RPCs or the enriched RPC suspension is adjusted to 1 to 10 mg/mL, in particular to 1 to 5 mg/mL, prior to the spray drying step.

32. The method according to claim 30 or 31, wherein the liquid fraction of the suspension comprising the RPCs or the enriched RPC suspension is exchanged prior to the spray drying step.

33. The method according to claim 32, wherein exchanging the liquid fraction of the suspension comprising the RPCs or the enriched RPC suspension reduces the concentration of at least one buffering agent, complexing agent and/or excipient in the suspension.

34. The method according to claim 33, wherein the suspension comprising the RPCs or the enriched RPC suspension is substantially free of buffering agent after exchanging the liquid fraction of the suspension.

35. The method according to claim 33 or 34, wherein the liquid fraction of the suspension is exchanged before the spray-drying step to obtain a mole-charge ratio between the protein and the complexing agent between 1:0.2 to 1:1, in particular between 1:0.4 to 1:0.8.

36. The method according to any one of claims 30 to 35, wherein spray drying is performed at an inlet temperature 115° C. and/or an outlet temperature of 48° C.

37. The method according to any one of claims 30 to 36, wherein spray drying is performed at a feed rate of 17 mL/min.

38. The method according to any one of claims 30 to 37, wherein the method comprises a further step of resuspending the spray dried powder in a non-aqueous solvent (NAS) to obtain an RPC-NAS suspension.

39. The method according to claim 38, wherein the non-aqueous solvent is at least one selected from a group consisting of: diethylene glycol monoethyl ether, ethyl oleate, triacetin, isosorbide dimethyl ether and glycofurol.

40. The method according to claim 39, wherein the spray dried powder is resuspended to obtain a RPC-NAS suspension with a protein concentration ranging from 50 to 300 mg/mL, in particular ranging from 100 - 250 mg/mL.

41. A composition comprising reversible protein complexes (RPCs), wherein the composition is obtained by the method according to any one of claims 1 to 40.

42. A composition comprising reversible protein complexes (RPCs), wherein the RPCs comprise a protein and a complexing agent, and wherein the complexing agent is dextran sulphate or chondroitin sulphate.

43. The composition according to claim 42, wherein the complexing agent is dextran sulphate, in particular dextran sulphate with 40 kDa molecular weight.

44. The composition according to claim 42 or 43, wherein the protein has a positive net charge when comprised in the RPCs.

45. The composition according to claim 44, wherein the protein is an antibody, a growth factor, a hormone, a cytokine, an enzyme, or a fragment and/or fusion protein of any of the foregoing.

46. The composition according to claim 45, wherein the antibody is a monoclonal antibody, a polyclonal antibody, a chimeric antibody, a multispecific antibody, an antibody fusion protein, an antibody-drug-conjugate or an antibody fragment.

47. The composition according to any one of claims 42 to 46, wherein the complexing agent has a negative charge when comprised in the RPCs.

48. The composition according to any one of claims 42 to 47, wherein the complexing agent comprises a hydrophobic moiety.

49. The composition according to any one of claims 42 to 48, wherein the composition comprises at least one excipient.

50. The composition according to claim 49, wherein the at least one excipient is a stabilizer and/or a surfactant.

51. The composition according to any one of claims 42 to 50, wherein the protein has a higher melting temperature when comprised in the RPC compared to the uncomplexed protein.

52. The composition according to any one of claims 42 to 51, wherein the RPCs comprising the protein and the complexing agent dissociate at physiological pH and ionic strength.

53. The composition according to any one of claims 42 to 51, wherein the RPCs comprising the protein and the complexing agent dissociate in 10 mM to 100 mM PBS (pH 7.4, 137 mM NaCl) when diluted to a protein concentration of 0.1 to 10 mg/mL.

54. The composition according to any one of claims 42 to 53, wherein the composition is a suspension.

55. The composition according to claim 54, wherein the suspension is obtained with the method according to any one of claims 1 to 25.

56. The composition according to claims 54 or 55, wherein the protein concentration in the suspension ranges from 50 to 250 mg/mL, in particular wherein the protein concentration in the suspension ranges from 100 to 200 mg/mL.

57. The composition according to any one of claims 54 to 56, wherein the suspension comprises uncomplexed complexing agent.

58. The composition according to any one of claims 54 to 57, wherein the RPCs comprised in the suspension have a mean particle size ranging from 5 to 20 .Math.m, in particular wherein the RPCs comprised in the suspension have a mean particle size ranging from 6 to 12 .Math.m.

59. The composition according to any one of claims 54 to 57, wherein the RPCs comprised in the suspension have a mean particle size ranging from 100 to 4000 nm, in particular wherein the RPCs comprised in the suspension have a mean particle size ranging from 150 to 2000 nm.

60. The composition according to any one of claims 54 to 57, wherein the RPCs comprised in the suspension have a mean particle size ranging from 0.1 to 20 .Math.m, in particular wherein the RPCs comprised in the suspension have a mean particle size ranging from 0.1 to 12 .Math.m.

61. The composition according to any one of claims 54 to 60, wherein the suspension is injectable through a 26G needle.

62. The composition according to any one of claims 54 to 61, wherein the suspension is stable for at least 4 weeks at 4° C. and/or 25° C.

63. The composition according to any one of claims 54 to 62, wherein the suspension has a viscosity ranging from 2 to 20 cP, in particular ranging from 3 to 15 cP, when measured at 20° C.

64. The composition according to any one of claims 54 to 63, wherein the pH of the suspension is lower than the isoelectric point of the protein.

65. The composition according to any one of claims 54 to 64, wherein the pH of the suspension is 1 to 3 pH units lower than the isoelectric point of the protein, in particular wherein the pH of the suspension is 2 pH units lower than the isoelectric point of the protein.

66. The composition according to any one of claims 54 to 64, wherein the pH of the suspension ranges from 1 to 6, in particular wherein the pH of the suspension ranges from 4.5 to 5.5.

67. The composition according to any one of claims 54 to 66, wherein the suspension comprises a buffering agent.

68. The composition according to claim 67, wherein the buffering agent is histidine or citrate.

69. The composition according to any one of claims 54 to 68, wherein the suspension has an ionic strength ranging from 20 to 50 mM, in particular wherein the suspension has an ionic strength ranging from 20 to 30 mM.

70. The composition according to any one of claims 54 to 66, wherein the suspension is substantially free of buffering agents.

71. The composition according to any one of claims 54 to 70, wherein the suspension further comprises a non-aqueous solvent.

72. The composition according to claim 71, wherein the non-aqueous solvent is diethylene glycol monoethyl ether, triacetin or ethyl oleate.

73. The composition according to any one of claims 42 to 53, wherein the composition is a lyophilisate.

74. The composition according to claim 73, wherein the lyophilisate is obtained with the method according to any one of claims 26 to 29.

75. The composition according to claim 73 or 74, wherein the lyophilisate comprises a buffering agent.

76. The composition according to claim 75, wherein the buffering agent is histidine or citrate.

77. The composition according to any one of claims 73 to 76 wherein the lyophilisate comprises at least one cryoprotectant.

78. The composition according to claim 77, wherein the at least one cryoprotectant is selected from a group consisting of: sugars, amino acids, methylamines, lyotropic salts, polyols, propylene glycol, polyethylene glycol and pluronics.

79. The composition according to any one of claims 73 to 78, wherein the lyophilisate is stable for at least 4 weeks at 40° C.

80. The composition according to any one of claims 73 to 79, wherein the lyophilisate is reconstituted in a liquid to a protein concentration ranging from 50 to 250 mg/mL, in particular wherein the lyophilisate is reconstituted in a liquid to a protein concentration ranging from 100 to 200 mg/mL.

81. The composition according to claim 80, wherein the liquid is PBS.

82. The composition according to claim 80 or 81, wherein the resuspended lyophilisate has a viscosity ranging from 2 to 20 cP, in particular ranging from 10 to 20 cP.

83. The composition according to any one of claims 42 to 53, wherein the composition is a spray dried powder.

84. The composition according to claim 83, wherein the protein content of the spray dried powder is at least 40% by weight (w/w), at least 50% by weight (w/w), at least 60% by weight (w/w).

85. The composition according to claim 83 or 84, wherein the spray dried powder is obtained with the method according to any one of claims 30 to 37.

86. The composition according to any one of claims 83 to 85, wherein the spray dried powder comprises a buffering agent.

87. The composition according to claim 86, wherein the buffering agent is histidine or citrate.

88. The composition according to any one of claims 83 to 85, wherein the spray dried powder is substantially free of buffering agents.

89. The composition according to any one of claims 83 to 88, wherein the RPCs comprised in the spray dried powder have a mean particle size ranging from 5 to 50 .Math.m, in particular ranging from 10 to 40 .Math.m, in particular ranging from 20 to 35 .Math.m.

90. The composition according to any one of claims 83 to 89, wherein the spray dried powder is re-suspended in a liquid to a protein concentration in the suspension ranging from 50 to 300 mg/mL, in particular wherein the spray dried powder is re-suspended in a liquid to a protein concentration in the suspension ranging from 100 to 250 mg/mL.

91. The composition according to claim 90, wherein the liquid is a non-aqueous solvent.

92. The composition according to claim 91, wherein the non-aqueous solvent is at least one selected from a group consisting of: diethylene glycol monoethyl ether, ethyl oleate, triacetin, isosorbide dimethyl ester and glycofurol, preferably diethylene glycol monoethyl ether, ethyl oleate or triacetin.

93. The composition according to any one of claims 90 to 92, wherein the reconstituted spray dried powder has a viscosity ranging from 10 to 100 cP, in particular ranging from 20 to 80 cP.

94. A pharmaceutical formulation comprising the composition according to any one of claims 41 to 93.

95. The pharmaceutical formulation according to claim 94, wherein the pharmaceutical formulation comprises the suspension according to any one of claims 54 to 72, the reconstituted lyophilisate according to any one of claims 80 to 82, or the re-suspended spray dried powder according to any one of claims 90 to 93.

96. The pharmaceutical formulation according to claims 94 or 95 for use as a medicament.

97. The pharmaceutical formulation according to any one of claims 94 to 96 for use in the treatment of an autoimmune disease, an immune dysregulation disease, carcinoma, sarcoma, glioma, melanoma, lymphoma, leukemia, chronic lymphocytic leukemia, follicular lymphoma, diffuse large B cell lymphoma, multiple myeloma, non-Hodgkin’s lymphoma, Alzheimer’s disease, type 1 or type 2 diabetes, amyloidosis, or atherosclerosis.

98. The pharmaceutical formulation for use according to claim 97, wherein the pharmaceutical formulation is administered subcutaneously, intramuscularly, transdermally, ocullarly, such as subconjunctivally, intracamerally, intravitreally, subretinally, or suprachoroidally, to the brain, such as intralumbarly, intrathecally, or intraventricularly, intra-articularly, or by inhalation.

99. Use of the pharmaceutical formulation according to claims 94 or 95 for the treatment of a disease selected from the group consisting of autoimmune disease, immune dysregulation disease, carcinoma, sarcoma, glioma, melanoma, lymphoma, leukemia, chronic lymphocytic leukemia, follicular lymphoma, diffuse large B cell lymphoma, multiple myeloma, non-Hodgkin’s lymphoma, Alzheimer’s disease, type 1 or type 2 diabetes, amyloidosis, and atherosclerosis.

100. Use of the pharmaceutical formulation according to claims 94 or 95 in the preparation of a medicament for the treatment of a disease selected from the group consisting of autoimmune disease, immune dysregulation disease, carcinoma, sarcoma, glioma, melanoma, lymphoma, leukemia, chronic lymphocytic leukemia, follicular lymphoma, diffuse large B cell lymphoma, multiple myeloma, non-Hodgkin’s lymphoma, Alzheimer’s disease, type 1 or type 2 diabetes, amyloidosis, and atherosclerosis.

101. A method of treating a subject suffering from a disease selected from the group consisting of an autoimmune disease, an immune dysregulation disease, carcinoma, sarcoma, glioma, melanoma, lymphoma, leukemia, chronic lymphocytic leukemia, follicular lymphoma, diffuse large B cell lymphoma, multiple myeloma, non-Hodgkin’s lymphoma, Alzheimer’s disease, type 1 or type 2 diabetes, amyloidosis, and atherosclerosis, the method comprising the steps of (a) producing the pharmaceutical formulation according claims 94 or 95; and (b) administering the pharmaceutical formulation to a subject in need thereof.

102. The method according to claim 101, wherein the pharmaceutical composition is administered subcutaneously, intramuscularly or transdermally, in particular wherein the pharmaceutical composition is administered subcutaneously.

103. A method of subcutaneous, intramuscular or transdermal administration of a pharmaceutical formulation, the method comprising the steps of (a) producing the pharmaceutical formulation according to claims 94 or 95; and (b) administering the pharmaceutical formulation to a subject by subcutaneous, intramuscular or transdermal delivery.

Description

BRIEF DISCRIPION OF FIGURES

[0519] FIG. 1: Visual aspect of RPC suspension obtained after mixing VEGF-Ang2 and dextran sulfate solutions at 1:1 mole-charge ratio in histidine buffer 20 mM pH 5.0.

[0520] FIG. 2: Protein melting temperature before (naked protein) and after complexation (RPC) and dissociation.

[0521] FIG. 3: DSC thermogram showing the melting temperature of a, VA2 in a complexed form (Tm=144.2° C.) and b, dextran sulfate sodium salt (Tm=171.9° C.) present in RPC formulations.

[0522] FIG. 4: Visual aspect of RPC suspension at 200 mg/mL showing the paste-like aspect of the formulation. a, spatula kept face up; b, spatula turned face down.

[0523] FIG. 5: Visual aspect of spray dried VA2 RPC.

[0524] FIG. 6: SEM images of VA2 RPC particles after spray drying of the suspensions in histidine buffer and in Ultra pure water (MilliQ water).

[0525] FIG. 7: VA2 RPC spray dried powder suspended in NAS solvents. EO, ethyl oleate; IDME, isosorbide dimethyl ether.

[0526] FIG. 8: Percentages of complexation and dissociation of the different protein formats with dextran sulfate (DS) at 1:1 mole-charge ratio.

[0527] FIG. 9 : Percentage complexation and dissociation of RPC using VEGF-Ang2 as a protein model and different complexing agents.* Complexation performed at pH 4.0. ** Dissociation performed with PBS 100 mM.

[0528] FIG. 10: Percentages of complexation and dissociation of VA2 and DS at different protein concentrations.

[0529] FIG. 11: RPC particle size obtained after complexation with DS at different protein concentrations.

[0530] FIG. 12: Percentage complexation of VA2 and DS in histidine buffer 20 mM at different pH; and their corresponding percentage dissociation in PBS.

[0531] FIG. 13: Percentage complexation of VA2 and DS in histidine buffer pH 5.0 at different ionic strengths; and their corresponding percentage dissociation in PBS.

[0532] FIG. 14: Percentage complexation and dissociation of VA2 with DS in presence of different additives. FIG. 15: Visual aspect of the VA2 RPC in a, histidine buffer (control); b, in presence of different additives (sucrose, polysobate 20, poloxamer 188 or mixture of those); c, in ultra pure water (MilliQ water) (no complexation).

[0533] FIG. 16: Visual aspect of RPC formulations (F1-F4) and VA2 control DP solution after 4 weeks storage at different temperatures. Note hard-cake formation, gel-aspect and shrinkage of RPC suspensions after 4 weeks at 25 or 40° C. A, front view; B, bottom view.

[0534] FIG. 17: Visual aspect of VA2 RPC 60 mg/mL a, before lyophilization; b and c, after lyophilization (b, front view; c, bottom view); d, after reconstitution of the lyophilized cake in PBS (120 mg/mL) stored 4-weeks at 5° C.

[0535] FIG. 18: Visual aspect of RPC formulation after 4 weeks storage at different temperatures. Ctrl - control (in histidine buffer, no buffer exchange), C — centrifuged (in MilliQ, buffer exchange by centrifugation), D - dialyzed (in MilliQ, buffer exchange by dialysis), A - front view, B - bottom view before vortexing, C — bottom view after vortexing.

[0536] FIG. 19: Comparison of visual aspect of the samples - “hard-cake” has formed in control sample (left), whereas particles in dialysed sample remained dispersed (right).

[0537] FIGS. 20A-20B: Comparison of particle size for RPC that had buffer exchanged to ultrapure water by means of centrifugation (FIG. 20A) or dialysis (FIG. 20B). The first results were obtained by laser diffraction measuring techniques (FIG. 20A), whereas the second by dynamic light scattering (FIG. 20B).

EXAMPLES

[0538] Aspects of the present invention are additionally described by way of the following illustrative non-limiting examples that provide a better understanding of embodiments of the present invention and of its many advantages. The following examples are included to demonstrate preferred embodiments of the invention. It should be appreciated by those of skill in the art that the techniques disclosed in the examples which follow represent techniques used in the present invention to function well in the practice of the invention, and thus can be considered to constitute preferred modes for its practice. However, those of skill in the art should appreciate, in light of the present disclosure that many changes can be made in the specific embodiments which are disclosed and still obtain a like or similar result without departing from the spirit and scope of the invention.

Example 1: Development of Reversible Protein Complex (RPC) Formulations

1.1 Materials

[0539] Therapeutic proteins including monoclonal antibodies (mAbs), bispecific crossmAb, cytokine fusion mAb and DutaFab, were provided by F.Hoffmann-La Roche AG (Basel, Switzerland). Histidine-HCl and L-Histidine base were obtained from Ajinomoto (Osaka, Japan), citric acid and Trisodium citrate from Merck (Darmstadt, Germany), polysorbate 20 from Croda (East Yorkshire, UK), poloxamer 188 from BASF (Ludwigshafen, Germany) and sucrose from Pfanstiehl (Zug, Switzerland). Dextran sulfate sodium salt (DS), Sodium dodecyl sulfate (SDS), Chondroitin sulfate (CS), Sodium taurocholate hydrate (ST), Triacetin, Diethylene glycol monoethyl ether (Transcutol®), Isosorbide dimethyl ether, Tetraglycol (Glycofurol), Ethyl oleate, PBS tablets and PVDF filters were purchased from Sigma-Aldrich (Buchs, Switzerland). Slide-A-Lyzer Dialysis Cassettes (MWCO 10 K) were obtained from Thermo Scientific. MilliQ water (resistivity > 18 MQ cm) was prepared using a Merck Millipore MilliQ water purification system (Darmstadt, Germany). All solvents used were from an analytical grade.

1.2 Methods

1.2.1 Preparation of Protein Solutions and Complexing Agent Solutions

[0540] The different protein solutions in histidine buffer (20 mM, pH 5.3-5.8), containing mainly sucrose, surfactants and/or sodium chloride; were dialyzed prior complexation to exchange the buffer with fresh histidine buffer (20 mM, pH 5.0). Dialysis was run during 2h at room temperature then over night at 5 ± 3° C. The dialyzed protein was further diluted to 5 mg/mL in histidine buffer (20 mM, pH 5.0).

[0541] Complexing agent solutions (Dextran sulfate sodium salt, Sodium dodecyl sulfate, Chondroitin sulfate, Sodium taurocholate hydrate) were prepared at 50 mg/mL in histidine buffer (20 mM, pH 5.0).

1.2.2 Protein Content

[0542] Protein concentration was measured by UV absorbance at 280 nm using a spectrophotometer (NanoDrop One.sup.c, ThermoFisher Scientific). In order to determine the protein concentration, RPC suspension was first dissociated using PBS 10 mM, then 4 .Math.L were placed on the instrument pedestal for quantification.

1.2.3 Formation of Reversible Protein Complexes

[0543] Proteins charge was calculated from their amino acid sequences, then the corresponding charge per mole was determined for each protein. The mole-charge was also determined for each of the complexing agents. Reversible protein complexes were prepared by mixing the protein solution (5 mg/mL) with the complexing agent solution (50 mg/mL) at 1:1 mole-charge ratio, aiming for a 100% charge neutralization. Total neutralization of the protein charge by the complexing agent leads to precipitation of the protein and formation of a whitish protein-particulate suspension.

[0544] Percentage complexation was determined after centrifugation of 1 mL sample of the RPC suspension (10 000 rpm, 5 min) and quantifying the amount of protein remaining in the supernatant using UV spectrometry (nanoDrop One.sup.c, Thermo Scientific) according to the following equation:

[00002]$Complexation\left(\%\right)=\frac{initialprot.conc-Prot.conc.insupernatant}{Initialproteinconcentration}$

1.2.4 Dissociation of Reversible Protein Complexes

[0545] Reversible protein complexes were dissociated following pH increase by diluting the RPC suspension to 1 mg/mL final protein concentration in PBS 10 mM, pH 7.4. Proteins being uncharged at pH 7.4, charge interactions between the proteins and the complexing agents decrease leading to dissociation of the complexes.

[0546] Protein concentration following dissociation was determined by UV and the percentage dissociation was calculated according the following equation:

[00003]$Dissociation\left(\%\right)=\frac{conc.proteindissociated}{conc.proteincomplexed}$

1.2.5 Particle Size

[0547] RPC particle size distribution was measured using a laser diffraction analyzer (Partica LA-960, HORIBA). RPC suspension was loaded in the sample bath of the instrument containing Ultra pure water (MilliQ water) to a concentration that allows 70-95% transmittance then the measurement was performed under circulation mode. Mean particle size values are reported. Particle size distribution of the spray dried RPC powder was also evaluated using SEM.

1.2.6 Zeta-Potential

[0548] Particle surface charge was evaluated using Malvern Zetasizer Nano ZS. Particle surface charge of RPC was evaluated in both histidine buffer and Ultra pure water (MilliQ water) media. Since complexation only occurs under specific conditions, complexation was first performed in histidine buffer then dialysis was run against Ultra pure water (MilliQ water) to remove buffer ions. RPC suspensions in both media were diluted to 0.1 mg/mL in their corresponding media then samples (~0.8 mL) were loaded into the zeta potential cells for analysis.

1.2.7 Viscosity

[0549] Viscosity measurements were performed using a rheometer (Physica MCR 301, Anton Paar) equipped with a cone-plate geometry. The viscosity of RPC formulations was determined by placing 80 .Math.L of the sample in the center of the plate. The method used consisted of 3 steps; 120 s for sample equilibration to 20° C. in the first step, followed by a second step where 1000 s.sup.-1 shear rate was applied for 10 s, then a last step where 1000 s.sup.-1 shear rate was applied for 5 s.

1.2.8 Protein Stability

[0550] Protein stability was evaluated before complexation and after complexation and dissociation. Protein purity was monitored using size exclusion chromatography (SEC) and protein charge was monitored using ion exchange chromatography (IEC).

1.2.9 Protein Melting Temperature

[0551] Melting temperature of reversible protein complexes was measured in both liquid form (suspension) using nanoDSF and solid form (spray-dried powder) by DSC.

1.2.9.1 Melting Temperature by nanoDSF

[0552] Protein melting temperature was measured before complexation as naked protein, after complexation as RPC suspension, and after dissociation of RPC in PBS 10 mM. 30 .Math.L of each solution (~ 1.2 mg/mL) were placed in a 384 well microplate, transferred into a 48 capillary array (~10 .Math.L) then introduced in the instrument (Prometheus, nanoTemper). The method consists on applying a temperature ramp from 25 to 95° C. with a 0.5° C./min heating rate. Excitation power was set at 28%, protein fluorescence intensities were recorded (ratio 350/330 nm) and melting temperature (transition midpoint T.sub.m, 50% unfolded protein) was recorded.

1.2.9.2 Melting Temperature by DSC

[0553] Protein melting temperature as RPC powder was measured after complexation and spray drying using DSC (Q2000, TA instruments). The sample was placed in the instrument (~ 50 mg) then the method consisted of an equilibration step at 25° C. followed by a ramp of 5° C./min to -5° C. and an isothermal step for 5 min, then a ramp of 2° C./min to 250° C. and an isothermal step for 5 min and finally a ramp of 5° C./min to 25° C.

1.3 Results

[0554] Mixing the protein (VEGF-Ang2) solution with the complexing agent (dextran sulfate) solution in histidine buffer (20 mM pH 5.5) at 1:1 mole-charge ratio resulted in the formation of a whitish suspension of protein particles (FIG. 1)

[0555] Mean particle size of the RPC suspension measured by laser diffraction was 10.8 ± 0.8 .Math.m. Particles sedimentation was observed when formulation was left to stand, however, particles were easily resuspended after simple agitation.

[0556] Surface charge of the RPC particles in histidine buffer was -13.9±0.6 mV; after buffer exchange (wash) with Ultra pure water (MilliQ water) using dialysis, the surface charge was -43.6±0.5 mV.

[0557] Stability of the protein evaluated by SEC showed no significant difference before and after complexation and dissociation in PBS with percentages monomer of 95.2% and 95.3%, respectively. Same observation using IEC with percentages main peak of 67.2% and 66.7%, respectively before and after complexation and dissociation in PBS.

[0558] Protein melting temperature (Tm) was determined in the control naked form (uncomplexed), after complexation as a suspension form and after dissociation of RPC in PBS. Results showed no significant difference in the Tm of the naked control protein (67.5° C.) and after complexation and dissociation of the RPC (68.4° C.) confirming the results observed by SEC and IEC concluding that complexation-dissociation process does not affect the stability of the protein (FIG. 2). Interestingly, the complexed protein (RPC) does not show a clear inflection point suggesting that the Tm of the protein in the complexed form shifted to temperatures higher than 95° C. (limit of the nanoDSF temperature range), which indicates a higher stability compared to the naked protein.

[0559] DSC was further used to determine the exact Tm of the protein in the complexed form (FIG. 3). Results showed a shift of the protein Tm from 67.5° C. to 144.2° C. when it is complexed with DS, demonstrating a higher stability of the protein in the RPC form.

Example 2: Post-Processing Steps of Reversible Protein Complex Formulations

[0560] RPC concept was used to develop highly concentrated protein formulations. Two approaches were evaluated; the first consists on up-concentration of the RPC suspension in histidine buffer to a series of concentrations ranging from 60 mg/mL up to 200 mg/mL. The second approach consists on re-suspending the spray dried RPC particles in a non-aqueous solvent.

2.1 Up-Concentration

[0561] RPC suspensions were up-concentrated by centrifugation (Eppendorf Centrifuge 5810 R). RPC suspension (5 mg/mL) was filled into falcon tubes (50 mL) containing magnetic stirrers then centrifuged (3900 rpm, 15 min, 5° C.) for up-concentration. Supernatant was discarded (100% complexation) then depot were homogenized and pooled into one falcon tube. Intermediate concentration was determined by dissociating 10 .Math.L suspension in 990 .Math.L PBS 10 mM pH 7.4. Concentration was adjusted to 120 mg/mL by further centrifugation or dilution with histidine buffer.

[0562] For higher concentrations (up to 200 mg/mL), high-speed centrifuge (Beckman Coulter Optima L-90K Prep Ultracentrifuge) was used for up-concentration. RPC suspension was placed in a high-speed centrifuge tube (Beckman Coulter, 70 mL) containing magnetic stirrer then centrifuged at 10 000 rpm during 10 min at 5° C. Supernatant was removed then depot was homogenized by vortex, 10 .Math.L were dissociated in 990 .Math.L PBS to evaluate the intermediate concentration. Centrifugation was carried on until reaching final target concentration of 200 mg/mL.

[0563] Up-concentration of RPC suspension up to 200 mg/mL resulted in a very dense white suspension with a paste-like aspect (FIG. 4). Nevertheless, this formulation was injectable through a 26G needle. RPC suspensions at 180 and 160 mg/mL had a similar paste-like aspect, whereas RPC suspensions become more liquid at around 120 mg/mL, and RPC suspensions at 90 and 60 mg/mL were liquid.

[0564] Viscosity measurement of the different up-concentrated suspensions showed a shear thinning effect (viscosity decrease upon application of a constant shear rate). This feature renders the injectability of the formulation easier compared to a Newtonian solution. However, the viscosity measurement method used was developed for liquid solutions and was not appropriate for the paste-like RPC formulation.

2.2 Spray Drying

[0565] Spray drying was performed using Büchi Mini Spray Dryer B-290. Inlet temperature was set at 115° C. (outlet ~ 48° C.) and nitrogen aspiration was set at 100%. The feed rate (peristaltic pump) was set at 17 mL/min. RPC formulation was kept under stirring to avoid particle sedimentation during spray drying process.

[0566] Spray drying of VA2 RPC formulations at 5 mg/mL (F1-F3) resulted in a fine, white powder (FIG. 5).

[0567] In order to optimize the protein content in the spray dried RPC powder and the stability of the protein following the spray drying process of RPC suspensions, different formulation compositions (F1 to F3) were evaluated (Table 1). RPC formulations were prepared as mentioned in Example 1, then the corresponding amounts of sucrose and polysorbate 20 were added to F1. Since the inventors targeted the highest protein content in the final powder, the inventors used the minimal excipients possible to limit their contribution to the final powder content while still ensuring protein stability during spray drying process. Given that the components of histidine buffer highly contribute to the final solid content of the RPC powder, histidine buffer was exchanged with ultra pure water (MilliQ water) using dialysis following complexation (F2 and F3). The corresponding amounts of sucrose and polysorbate 20 were then added to F3.

TABLE-US-00001 Composition of RPC solutions at 5 mg/mL to be spray dried. Excipient Medium Sucrose (mg/mL) Polysorbate 20 (mg/mL) F1 Histidine buffer 2.05 0.40 F2 Ultra pure water (MilliQ water) - - F3 Ultra pure water (MilliQ water) 1.00 0.20

[0568] Following spray drying, protein content in the RPC powder was determined by UV after dissociation of a specific amount in PBS, and using the following the equation:

[00004]$Proteincontent\left(\%\right)=\frac{ActualVA2conc.}{TheoriticalVA2conc.}$

[0569] Protein stability was evaluated by SEC and IEC and protein melting temperature was determined by DSC in RPC as powder form and by nanoDSF after solubilisation in PBS.

[0570] Scanning electron microscope (SEM) analysis of F1 and F3 particles obtained after spray drying showed loose, round and dispersed particles in VA2 RPC washed with Ultra pure water (MilliQ water) (F3, FIG. 6), while particles in VA2 RPC in histidine buffer were more agglomerated forming clusters and fused-like particles (F1, FIG. 6). Whether this difference in particle shape is due to the different media used or to the spray drying process itself is to be further investigated.

[0571] Analysis of VA2 RPC particles in histidine buffer versus Ultra pure water (MilliQ water) before and after spray drying showed an increase in the particle size after spray drying, from ~8 .Math.m to ~30 .Math.m (Table 2), probably due to the adsorption of the excipients added (sucrose and PS20) on the RPC particles and possible cluster formation during the spray drying process.

TABLE-US-00002 VA2 RPC particle size in histidine buffer versus Ultra pure water (MilliQ water) before and after spray drying. Formulation Mean (.Math.m) Median (.Math.m) F1 -Before SD 7.2 6.4 F1 - After SD 30.6 20.1 F3 - Before SD 8.7 7.3 F3 - After SD 33.5 8.2

[0572] As expected, protein content in the spray dried powder varied in the formulations (F1-F3) according to the amount of the buffer salts and the excipients added (Table 3).

TABLE-US-00003 Protein content and stability in VA2 RPC spray dried powder. VA2:DS Medium Excipients VA2 cont. (%) SEC (% mono) IEC (% main peak) F1 1:1 Histidine buffer 20 mM pH 5.0 DS 0.84 mg/mL Sucrose 2.1 mg/mL PS20 0.4 mg/mL 43.1 97.1 65.2 F2 1:1 Ultra pure water (MilliQ water) DS 0.84 mg/mL 70.2 94.6 66.8 F3 1:0.6 Ultra pure water (MilliQ water) DS 0.56 mg/mL Sucrose 1 mg/mL PS20 0.2 mg/mL 63.0 95.5 66.1

[0573] Protein content in F1 was 43.1%, with a high contribution of histidine buffer salts to the solid content of the spray dried powder. Removal of the excipient and washing out histidine buffer with MQ water in F2 increased the protein content up to 70.2%. Decreasing the protein:DS mole-charge ratio to 1:0.6, followed by a wash out of the histidine buffer salts and a decrease of the amount of excipients added in F3 led to a final protein content of 63.0%.

[0574] Protein stability during spray drying process was assessed by SEC and IEC. No significant change was observed in the chemical stability (IEC) in all formulations. However, an increase in the high molecular weight species (HMWS) was observed by SEC, especially in F2, probably due to the absence of histidine buffer and excipients to protect the RPC during the spray drying process. Indeed, in F3, adding half excipients could decrease the HMW species. Hence, F3 seems to be a good compromise to optimize the protein content in the final spray dried powder and ensure the protein stability during the spray drying process.

2.3 Resuspension of Spray Dried RPCs

[0575] A series of apolar solvents were tested as resuspension media including ethyl oleate, triacetin, Transcutol (diethylene glycol monoethyl ether), Glycofurol, and Isosorbide dimethyl ether. RPC spray dried powder was incorporated in the corresponding volume of the solvent to reach 100, 150, 200 and 250 mg/mL then homogenized.

[0576] RPC dissociation after up-concentration or resuspension in non-aqueous media was evaluated after dilution in PBS. Melting temperature of the spray dried RPC powder was evaluated by DSC and using nanoDSF after dissociation in PBS. Protein stability after up-concentration or resuspension in non-aqueous media was evaluated by SEC and IEC. Particle size and viscosity of the high-concentration formulations were evaluated as mentioned in Example 1.

[0577] Suspension of VA2 RPC spray dried powder in NAS resulted in white suspensions (FIG. 7).

[0578] After dissolving 110 mg of RPC spray dried powder in 0.25 mL PBS, the final volume was 0.29 mL. The contribution of the solid content is 0.04 mL (~14% of the total volume). This volume was taken into account when preparing the VA2 RPC-NAS suspensions at different concentrations (ex. spray dried powder was dissolved in 0.26 mL NAS for a total volume of 0.3 mL). The target and actual concentrations of VA2 in NAS are listed in Table 4.

TABLE-US-00004 Target and actual protein concentrations in NAS, their corresponding viscosity and stability (SEC and IEC). NAS Target Cone (mg/mL) Actual Cone (mg/mL) SD (mg/mL) Viscosity (cP) SEC (% Mono) IEC (% Main p.) Diethylene glycol monoethyl ether 180 196.7 11.8 20 230 238.3 6.2 - 93.4 64.7 280 258.3 2.4 30 Isosorbide dimethyl ether 180 186.7 8.5 - 230 220.0 4.1 28 89.6 61.9 280 231.7 4.7 - Glycofurol 180 188.3 4.7 - 230 210.0 7.1 60 88.1 61.7 280 241.7 4.7 - Ethyl oleate 180 157.3 6.8 - 230 223.3 20.5 50 94.9 64.2 280 237.3 1.9 - Triacetin 180 183.3 6.2 - 230 193.3 6.2 70 95.0 64.2 280 283.3 8.5 100 PBS 200 213 3.2 275 95.5 66.2

[0579] Dissociation of VA2 RPC SD powder in PBS 10 mM at 213 mg/mL resulted in very viscous solution (275 cP, Table 4). Suspension of VA2 RPC SD powder in NAS resulted in lower viscosity compared to PBS ranging from 20 to 70 cP for concentrations ranging from 193 to 223 mg/mL (Table 4). Protein stability evaluation after a short incubation time with the NAS showed an increase in the HMWS within Glycofurol and IDME. Protein aggregation was also observed within Triacetin, Transcutol and EO but at a lower extent (Table 4). Results from IEC also showed higher stability within Triacetin, Transcutol and EO compared to Glycofurol and IDME. Even though there is a reduction in the viscosity, injectability of the solutions was not possible through a 26G needle due to a large particle size.

Example 3: Universality of the Concept

3.1 Universality of the Concept Using Different Proteins

[0580] Feasibility of RPC was evaluated using different protein formats (mAbs, bispecific crossmAb, cytokine fusion mAb, DutaFab) and dextran sulfate as the complexing agent at 1:1 mole-charge ratio.

[0581] The inventors have used dextran sulfate with 40 kDa molecular weight as the complexing agent. It is reported that DS has an average of two negative charges per monomer, corresponding to 240 negative charges per mole polymer. The number of positive charges of each protein was calculated from the amino acid sequence and are summarized in Table 5. Accordingly, the determined protein:DS complexation weight ratios corresponding to a 1:1 mole-charge ratio between each protein and DS are summarized in Table 5.

TABLE-US-00005 Weight ratios between proteins and the complexing agent, dextran sulfate (DS), corresponding to 1:1 mole-charge ratio, used to prepare reversible protein complexes. Proteins MW Total (+) Weight ratio (kDa) charges Protein DS VEGF-Ang 2 146 146 1 0.167 aSyn-mAb 145 152 1 0.175 FAP-OX40 195 188 1 0.160 VEGF-PDGF DutaFab 48 52 1 0.181 Bevacizumab 149 150 1 0.168 Pertuzumab 148 148 1 0.167 Gantenerumab 146 156 1 0.178

[0582] The study showed that RPC formation was possible with the different proteins tested corresponding to different protein formats (mAbs, bispecific crossmAbs, fusion mAbs, DutaFabs) with percentage complexation ranging from 96.1±1.4% to 99.5±0.3% (FIG. 8). The reversibility of the concept was also confirmed with regards to the different protein formats with percentage dissociation ranging from 96.5±1.6% to 105.4±0.8% in PBS 10 mM. Both complexation and dissociation were instantaneous at room temperature. RPC concept is hence applicable to a range a wide range of biologics.

[0583] Protein complexation with DS followed by dissociation in PBS 10 mM did not affect the stability of the protein. SEC and IEC results showed no significant difference between the percentages of monomers and main peaks obtained with the different proteins tested before and after complexation and dissociation (Table 6).

TABLE-US-00006 Percentage monomer by SEC and main peak by IEC obtained with the different proteins before complexation and after complexation-dissociation. * Analytical methods not available. Proteins % Monomer by SEC % Main Peak by IEC Before complexation After dissociation Before complexation After dissociation VEGF-Ang 2 95.2 95.3 67.2 66.7 aSyn-mAb 99.0 98.6 * * FAP-OX40 * * * * VEGF-PDGF DutaFab 99.1 98.8 88.7 86.9 Bevacizumab 97.2 97.5 71.3 69.8 Pertuzumab 99.4 98.6 62.1 61.5 Gantenerumab 97.7 98.4 49.0 49.1

3.2 Universality of the Concept Using Different Complexing Agents

[0584] Feasibility of the RPC was also evaluated using different complexing agents (CA) including dextran sulfate (DS), chondroitin sulfate (CS), SDS and sodium taurocholate (ST), with a bispecific cross mAb (VEGF-Ang2) as the protein model. Protein solution was prepared at 5 mg/mL and complexing agents’ solutions were prepared at 50 mg/mL in histidine buffer 20 mM pH 5.0, complexation was performed at different prot:CA mole-charge ratios.

[0585] The optimized Protein (VEGF-Ang2) to the complexing agent’s mole-charge ratios and the corresponding volume ratios are summarized in Table 7. Typically, with VEGF-Ang2 and DS, the inventors previously assessed 100 % complexation and dissociation at 1:1 mole-charge ratio. The ratio optimization showed that the same complexation-dissociation efficiency could be reached even at 1:0.6 ratio, corresponding to the minimal DS to be added for a total complexation of the protein. This suggests that the theoretical charge calculation method based on the protein sequence overestimated the number of positive charges available for complexation by DS, thus the actual amount of DS needed to provide 100% charge neutralization is lower than calculated. It is important to note that the inventors were considering the total number of positive charges distributed over the protein and not the protein net charge, as the protein net charge would underestimate the number of positive charges by charge addition.

TABLE-US-00007 Optimized mole-charge and weight ratio between VEGF-Ang2 and different complexing agents (CA) used to prepare reversible protein complexes (RPC). Complexing agent (CA) Mole-charge ratio Weight ratio Protein CA Protein CA Chondroitin sulfate (CS) 1 0.2 1 0.1 Dextran sulfate (DS) 1 0.6 1 0.1 Sodium dodecyl sulfate (SDS) 1 0.7 1 0.2 Sodium taurocholate (ST) 1 4.0 1 2.0

[0586] When adding DS, SDS and CS to the protein solution, complexation occurs at pH 5.0 (~ 100%). When using ST, complexation occurs only after adjusting the pH of the buffer solution to 4.0; moreover, 3 hours incubation time are needed to reach 100% complexation (FIG. 9).

[0587] Dissociation of the RPC formed with DS or CS is instant and complete. Dissociation of RPC formed using SDS and ST require more time; the use of PBS with a higher ionic strength (100 mM) accelerated the dissociation process (96.1% and 89.4%, respectively for SDS and ST, FIG. 5).

[0588] SEC analysis showed a good stability of protein after complexation and dissociation when using DS and CS. However, ST and even more, SDS, significantly degraded the protein. These results were also confirmed by IEC (Table 8).

TABLE-US-00008 Percentage monomer by SEC and main peak by IEC of a bispecific mAb obtained before complexation and after complexation-dissociation using different complexing agents. Complexing agent (CA) % Monomer by SEC % Main Peak by IEC Before complexation After dissociation Before complexation After dissociation Chondroitin sulfate 97.3 96.7 66.1 65.2 Dextran sulfate 95.2 66.7 Sodium dodecyl sulfate 21.1 61.1 Sodium taurocholate 90.1 62.9

Example 4: Robustness of the Concept

[0589] Robustness of the RPC formation was evaluated along a range of protein concentration, buffer type, pH and ionic strength to determine the optimal conditions for RPC formation.

4.1 Effect of Protein Concentration

[0590] RPC formation was evaluated using protein concentrations ranging from 1 to 100 mg/mL in histidine buffer. VEGF-Ang2 dialyzed stock solution (130 mg/mL) was diluted to 100, 50, 40, 30, 25, 20, 5 and 1 mg/mL in histidine buffer 20 mM pH 5.0. The corresponding volume of DS (50 mg/mL) was added to 1 mL protein solution of every concentration at 1:1 mole charge ratio (1:0.167 weight ratio). Percentage complexation was evaluated after centrifugation and percentage dissociation was evaluated in PBS 10 mM.

[0591] Percentages of complexation and dissociation of VA2 and DS were determined at different protein concentrations (FIG. 10). Results showed that complexation occurs at any protein concentration, from 1 mg/mL to at least 100 mg/mL (highest concentration evaluated). Although, percentage complexation ranged from 96.7±2.4% to 99.5±0.0% at protein concentration ranging from 1 mg/mL to 40 mg/mL; complexation at concentrations higher than 40 mg/mL are limited because of the high viscosity associated with the high protein concentration, hindering the complexing agent from spreading over the solution and reaching every protein molecule to achieve a homogenous protein complexation.

[0592] Percentages of dissociation of the complexes formed at protein concentration from 1 mg/mL to 40 mg/mL ranged from 98.1±2.3% to 82.3±5.4%, with lower dissociation percentages at higher concentrations. Thus, the optimal protein concentration range for complete complexation and dissociation is defined to be from 1 mg/mL to 5 mg/mL (FIG. 10).

[0593] Initial protein concentration used for complexation was also found to have an effect on the final RPC particle size distribution (Table 9 and FIG. 11).

TABLE-US-00009 RPC particle size after complexation with DS at different protein concentrations. Protein concentration Mean (.Math.m) Median (.Math.m) SD (um) 1 mg/mL 7.8 7.5 2.1 5 mg/mL 11.1 8.0 13.8 20 mg/mL 21.6 19.4 10.3

4.2 Effect of Buffer Strength and pH on Complexation

[0594] RPC formation was evaluated using a range of histidine buffers with ionic strengths ranging from 5 to 50 mM and pH ranging from 1 to 7; and within citrate buffer 10-20 mM pH 5.0 at 5 mg/mL protein concentration and 1:1 mole-charge ratio with DS. RPC formation was also evaluated using ultra pure water (MilliQ water) as medium. Protein stock solution was dialyzed against ultra pure water (MilliQ water) then diluted to 5 mg/mL in ultra pure water (MilliQ water). The complexation was prepared by mixing the protein solution with the DS (50 mg/mL in ultra pure water (MilliQ water)) solution at 1:1 mole-charge ratio. Percentage complexation was evaluated after centrifugation and percentage dissociation was evaluated in PBS 10 mM.

[0595] Percentage complexation was evaluated within histidine buffer 20 mM at different pH ranging from 1 to 7 (FIG. 12). Their corresponding percentage dissociation were determined in PBS.

[0596] Percentages complexation ranged from 98.6±1.0% to 99.2±0.9% from pH 1 to 5.5, 93.5±0.2% at pH 6 and 1.3±0.4% at pH 7. The corresponding percentages dissociation ranged from 98.1±4.6 % to 109.2±3.8% (FIG. 12). Thus, the optimal buffer pH range for complexation is from 4.5 to 5.5 in order to ensure complete complexation and protein stability (strong acidic pH may degrade protein).

[0597] On the other hand, percentage complexation was evaluated in histidine buffer pH 5 at different ionic strengths ranging from 5 mM to 50 mM (FIG. 13). Their corresponding percentage dissociation were determined in PBS.

[0598] Percentages complexation ranged from 98.6±1.0% to 99.9±0.1% for ionic strengths ranging from 20 mM to 50 mM, 82.2±0.2% at 10 mM and 66.3±0.3% at 5 mM. The corresponding percentages dissociation ranged from 98.9±1.4% to 109.9±0.9% (FIG. 13). Although, RPC formulation with histidine buffer 50 mM showed a precipitation of the particles followed by the formation of a gel-like depot. Thus, the optimal buffer ionic strength range for complexation is from 20 mM to 30 mM in order to ensure complete complexation and formulation stability.

[0599] The inventors mainly used histidine buffer in this study, although, complexation also occurs using other buffers including citrate buffer 20 mM pH 5 with a percentage complexation of 97.2±2.0% and a percentage dissociation of 108.0±1.1%.

4.3 Effect of Buffer Strength, pH and Volume on Dissociation

[0600] Complex dissociation was evaluated using different media including PBS 10 mM (pH 7.4, NaCl 137 mM), phosphate buffer 10 mM (pH 7.4) and histidine buffer 20 mM containing saline (pH 5.0, NaCl 137 mM). RPC suspension was diluted in the corresponding media to 1 mg/mL, vortexed then dissociation was evaluated by visual inspection of the samples; when the solution is clear, protein content was measured by UV to determine the percentage of dissociation.

[0601] PBS (10 mM pH 7.4) volume needed for total instantaneous dissociation of RPC 5 mg/mL was evaluated at different RPC:PBS dilution ratios; 1:4, 1:2, 1:1.5 and 1:1 by mixing 100, 100, 200, 250 .Math.L RPC with respectively 400, 300, 300, 250 .Math.L PBS 10 mM. Percentage disscociation and pH of the final solutions were determined.

[0602] Different buffer ionic strengths and pH were tested to dissociate RPC particles. RPC particles dissociate instantly and completely within PBS resulting in a clear solution. RPC dilution in phosphate buffer 10 mM pH 7.4 resulted in a turbid solution due to incomplete dissociation of the RPC particles (~ 50% dissociation); addition of NaCl enabled total dissociation of the RPC particles. Dissociation of RPC particles in histidine buffer 20 mM pH 5.0 containing saline was even lower (~17%), where addition of NaCl also enabled total dissociation of the RPC particles. Thus, the best buffer for complete RPC dissociation is PBS 10 mM pH 7.4 (100 mM can also be used in some cases to accelerate the dissociation). In terms of PBS volume, a ratio PBS:RPC of at least 2:1 is required for total dissociation of RPC particles, resulting in pH increases to at least pH 6.5 in the formulation for total dissociation.

4.4 Effect of Excipients

[0603] The effect of generic excipients used in protein formulations on the RPC formation was evaluated by adding sucrose and/or surfactants to the formulation buffer. Sucrose, polysorbate 20 and/or poloxamer 188 were added to the protein solution diluted to 5 mg/mL in histidine buffer 20 mM pH 5.0 (formulations F1 to F6, Table 10) prior complexation with DS (1:1 mole-charge ratio) then the corresponding percentages of complexation and dissociation were determined.

TABLE-US-00010 Additional excipients to RPC formulations in histidine buffer. Excipient F1 F2 F3 F4 F5 F6 Sucrose - 240 mM - - 240 mM 240 mM Polysorbate 20 - - 0.05% - 0.05% - Poloxamer 188 - - - 0.05% - 0.05%

[0604] The presence of the additives (sucrose, polysorbate 20 and/or poloxamer 188) did not affect the RPC formation. Percentage complexation ranged from 99.2 to 99.6%. Additives did not affect RPC dissociation in PBS neither, with percentages ranging from 92.3% to 97.5% (FIG. 14,FIG. 15).

[0605] Complexation of VA2 with DS in ultra pure water (MilliQ water) instead of histidine buffer was not possible. Buffer is necessary for RPC formation (FIG. 15).

Example 5: Short-Term Stability Study

[0606] A short-term study was conducted to evaluate the stability of the protein in RPC formulation (120 mg/mL) at different temperatures, in presence or absence of additives (sucrose, Poloxamer 188, polysorbate 20). RPC formulations (F1-F4) were formulated in histidine buffer and compared to formulation F5 where histidine buffer was exchanged with ultra pure water (MilliQ water) by dialysis.

[0607] Lyophilization of RPC suspension at 60 mg/mL (F6) was successful and resulted in a homogenous cake that was easily reconstituted (and dissociated) to a final concentration of 120 mg/mL in PBS resulting in a transparent solution (FIG. 17).

[0608] In order to evaluate the feasibility of freeze drying and reconstitution of RPC suspensions, F6 was up concentrated to 60 mg/mL, 2 mL were filled into 6-mL vials then samples were placed in the lyophilizer for freeze drying. F6 is meant to be resonstituted in 0.85 mL PBS for a final concentration of 120 mg/mL VA2, 240 mM sucrose and 0.05% PS 20.

[0609] Stability of the protein in RPC formulations was compared to the standard protein solution (F7; drug product liquid solution 120 mg/mL).

[0610] 0.5 mL of formulations F1-F5 and F7 were filled into 2-mL glass vials then stored with the lyophilized samples in the stability chambers (5, 25 and 40° C.) for four weeks. Composition of the different formulations is listed in Table 11.

[0611] Stability of the formulations was monitored at different time points in terms of visual aspect, protein content (UV), physical and chemical stability (SEC, IEC) and viscosity.

TABLE-US-00011 Composition of the different formulations prepared for the stability study. Formulation Buffer Sucrose (mM) NaCl (mM) DS (mM) Methionine (mM) Polysorbate 20 (% w/v) Poloxamer 188 (% w/v) F1 Histidine 20 mM pH 5.0 - - 0.5 - - - F2 240 - 0.5 - - - F3 240 - 0.5 - 0.05 - F4 240 - 0.5 - - 0.05 F6 120 - 0.25 - 0.025 - F5 Ultra pure water (MilliQ water) - - 0.5 - - - F7 Histidine acetate 20 mM pH 5.8 160 25 - 7 0.04 -

[0612] RPC formulations up-concentrated to 120 mg/mL (F1-F4) were liquid suspensions at initial time and no change was observed after 4-weeks storage of the different formulations at 5° C. However, after 4-weeks storage at 25° C. or 40° C., few visual changes were observed including the formation of a hard-cake, which in some formulations formed a shrinked pellet with or without a gel-aspect (FIG. 16). These hard cakes are irreversible, a magnetic stirrer and a vortex are needed to resuspend the RPC particles and reconstitute the suspension. Another interesting observation was during dissociation in PBS. After 4-weeks storage at 5° C. and 25° C., RPC particles dissociated completely; however, dissociation of RPC particles in formulations stored at 40° C. was partial forming a turbid solution upon dilution in PBS.

[0613] Interestingly, the hard cakes observed in F1-F4 formulations after 4-weeks storage et 25° C. and 40° C. were not observed in F5.

5.1 Protein Content Recovery After Dissociation

[0614] Protein concentration measured in the different RPC suspensions (F1-F5) following dissociation in PBS showed 92.7% to 107.0% recovery of the protein content after 4-weeks storage at 5° C. and 25° C. However, after 4-weeks storage at 40° C., protein content recovery ranged from 8.7% to 16.8%, resulting from incomplete dissociation of RPC particles (Table 12).

[0615] Total protein content was recovered following reconstitution of RPC lyophilisates (F6) in PBS after 4-weeks storage at 5, 25 and 40° C. with percentage recovery ranging from 101.1±1.6 to 105.0±0.4 (Table 12). Lyophilisation of RPC suspension could solve the incomplete dissociation observed after storage of RPC suspension at 40° C.

[0616] Protein content in F7 remains stable along the storage period at all temperatures (Table 12).

TABLE-US-00012 Initial protein concentration and protein content recovery after 4-weeks storage of VA2 RPC formulations (F1-F5) and VA2 control solution (F6) at different temperatures. Initial 4 weeks 5° C. 4 weeks 25° C. 4 weeks 40° C. Conc. mg/mL Conc. Recovery mg/mL % Conc. Recovery mg/mL % Conc. Recovery mg/mL % F1 127.0 ± 2.2 124.7±1.2 98.2±0.9 117.7±1.2 92.7±0.9 11.0±0.8 8.7±0.6 F2 121.7 ± 1.2 124.3±2.1 102.2±1.7 122.3±2.9 100.5±2.4 20.0±4.2 16.4±3.5 F3 122.7±0.9 131.3±1.7 107.0±1.4 128.7±5.8 104.9±4.7 14.7±0.5 12.0±0.5 F4 128.0±0.8 136.7±1.7 106.8±1.3 135.7±2.1 106.0±1.6 16.3±1.7 12.7±1.3 F5 115.3±0.5 113.0±0.8 98.0±0.8 108.0±2.2 93.6±1.5 19.3±3.1 16.8±2.7 F6 126.3±1.2 131.0±0.8 103.7±1.0 127.7±0.9 101.1±1.6 132.7±0.9 105.0±0.4 F7 112.7±0.5 118.0±1.4 - 118.7±1.2 - 118.0±2.4 -

5.2 Protein Stability

[0617] SEC and IEC analysis of the protein in RPC formulations (F1-F4) after 4-weeks storage showed a good stability at 5° C. following dissociation in PBS. After 4-weeks storage at 25° C., 1.9% to 2.5% loss in the monomer was observed by SEC and 5.1% to 5.7% loss in the main peak was observed by IEC. After 4-weeks storage at 40° C., 22.0% to 28.8% loss in the monomer was observed by SEC and 41.6% to 45.7% loss in the main peak was observed by IEC (Table 13).

[0618] SEC analysis clearly showed that VA2 RPC suspension is not stable in ultra pure water (MilliQ water) (F5) as 8.7% loss in the monomer was observed after 4-weeks at 25° C. and 55.4% at 40° C.

[0619] No significant difference was observed before and after freeze drying process (F6) with percentages monomer obtained by SEC analysis of 96.5% and 96.8%, respectively. Follow up of the percentage monomer in RPC lyophilizate showed stable monomer after at least 4-weeks at 5° C. and 25° C.; 1.5% loss in the monomer was observed at 40° C. Similarly, no significant difference was observed by IEC analysis before and after freeze drying process with percentages main peak of 66.1% and 66.3%, respectively. Follow up of the percentage main peak in RPC lyophilizate showed a stable main peak after 4-weeks at 5° C., 1.7% loss at 25° C. and 5.4% loss at 40° C.

[0620] Protein percentage of monomer by SEC in control formulation (F7) remains stable after 4-weeks storage at all temperatures; percentage main peak by IEC showed good stability at 5° C. after 4-weeks storage, 2.4% and 22.9% loss in the main peak at 25° C. and 40° C., respectively.

5.3 Viscosity

[0621] Viscosity measurements showed no significant difference between the viscosities of RPC suspensions in histidine buffer (F1-F4); the different excipients seem not to contribute significantly to the final RPC viscosity. Overall, viscosity of VA2 formulations was lower in RPC suspensions compared to the control drug product solution (F7) with 10-12 cP versus 20 cP, respectively. There is a clear drop in the viscosity after exchanging histidine buffer with ultra pure water (MilliQ water) (F5), from 10 cP to 4 cP. After reconstitution of the lyophilized RPC suspension ion PBS (F6), viscosity increased slightly compared to RPC suspension (16 cP), however, still inferior to the control (F7) (Table 13).

TABLE-US-00013 Initial viscosity and mean percentage monomer by SEC and percentage main peak by IEC after 4-weeks storage of VA2 RPC and control formulations (120 mg/mL) at different temperatures. Dosage form Viscosity (cP) SEC (% monomer) IEC (% main peak) Initial 4-weeks Initial 4-weeks 5° C. 25° C. 40° C. 5° C. 25° C. 40° C. F1 RPC suspension Histidine buffer 10 95.1 96.0 92.7 66.2 66.2 68.1 61.1 20.5 F2 11 95.2 96.1 93.3 73.2 66.4 68.2 61.1 24.8 F3 12 95.1 96.3 92.9 71.0 66.8 68.1 61.1 23.2 F4 11 95.4 96.4 92.9 69.8 66.5 68.3 61.1 23.7 F5 RPC suspension Ultra pure water (MilliQ water) 4 95.6 95.1 86.4 39.7 * * * * F6 RPC suspension Lyophilized 16** 96.8 96.8 96.5 95.3 66.3 66.0 64.6 60.8 F7 Control solution 20 96.0 96.7 96.4 96.0 67.9 68.2 65.5 44.9 *Analysis not performed.** After reconstitution in PBS.

Example 6: Dialysis of RPCs

[0622] In previous examples “hard-cake” formation has been observed upon storage of certain RPC samples at 25° C. and 40° C. This phenomenon affects visual aspect of the samples (the sample can appear “shrunken”) and complete resuspension of the particles is no longer possible, even if agitation is applied. Exchanging initial sample buffer (20 mM histidine buffer pH 5) with identical but fresh buffer or ultrapure water has resulted in improved sample behaviour at equal storage conditions. Buffer exchange has been performed by means of centrifugation where the protein-polymer complexes (RPC particles) had been sedimented, the supernatant was removed and the complexes in precipitate were then resuspended in fresh buffer or ultrapure water. Analyses of the samples have shown comparable particle size of RPC complexes in fresh histidine buffer and ultrapure water (i.e. in .Math.m range) and lower physical stability of dissociated protein in ultrapure water when analyzed by size exclusion chromatography.

[0623] The aim of the subsequent experiments was to explore alternative methods for buffer exchange (e.g. dialysis) and further investigate as well as confirm improved visual aspect of samples treated in this manner.

Methods

[0624] VEGF-Ang 2 (VA2) protein stock solution was prepared in 20 mM histidine buffer (pH 5) at a concentration of 5 mg/mL. For formation of protein-polymer complexes, 50 mg/mL complexing polymer (dextran sulfate sodium salt) solution in 20 mM histidine buffer (pH 5) was gradually added to the protein at constant mixing on a magnetic stirrer.

[0625] The prepared samples were always assessed for the % of complexation which was found to be > 98%.

[0626] In the next step, complexed solution was split into 3 parts with equal volumes - one part was dialysed against ultrapure water, another against fresh histidine buffer and the latter served as control and was not further manipulated. The dialysis was carried out using dialysis cassettes or tubings with MWCO of 10-100 kDa.

[0627] After dialysis, the samples were recovered and the protein concentration was adjusted to 120 mg/mL by means of centrifugation. The precipitate containing protein-polymer complexes was collected and the supernatant discarded.

[0628] The obtained samples were filled in 2 mL glass vials, stoppered, crimped and stored at different conditions (i.e. 5° C., 25° C., 30° C. or 40° C.).

[0629] At predefined time points they were characterized with regard to visual aspect and particle size prior dissociation of protein-polymer complexes in 10 mM phosphate buffer saline (pH 7.4) by means of laser diffraction or dynamic light scattering techniques (depending on the expected particle size). After dissociation, size and charge variants of the protein were determined by size exclusion and ion exchange chromatography respectively.

Results

[0630] “Hard-cake” formation has not been observed for dialysed samples (either dialysed against ultrapure water or fresh histidine buffer) when stored for 1 month at any of the storage conditions. For samples dialysed against ultrapure water, the suspension appeared more liquid and lower amounts of particles have sedimented, whereas samples dialysed against histidine buffer resulted as more “pasty” and dense. Full resuspension was possible by agitation. In contrast to that, “hard-cake” formation has been detected in control samples stored at 25° C. and 40° C. (FIG. 18 and FIG. 19).

[0631] Surprisingly, for samples dialysed against ultrapure water, the particle size measured has been significantly lower than for samples in histidine buffer (control or dialysed against fresh buffer). For the first time, it has been in nm range, whereas for the latter (as previously measured and reported) in .Math.m range (FIGS. 20A-20B). Additionally, the particle size of RPC when buffer was exchanged to ultrapure water by means of centrifugation has also been in .Math.m range, which may imply that buffer exchange by means of dialysis results in better “rinsing” of the particles and more efficient buffer exchange.

[0632] Incomplete dissociation of complexes in samples stored at 40° C. was still observed, meaning full recovery of complexed protein was not possible (Table 14).

[0633] Stability of samples dialysed against ultrapure water was lower in comparison to the control (Table 14).

TABLE-US-00014 Percentage monomer by SEC and main peak by IEC obtained for RPC suspension control and RPC suspension dialyzed against ultra pure water (MilliQ). Dosage form SEC (% monomer) IEC (% main peak) Initial 4-weeks Initial 4-weeks 5° C. 25° C. 40° C. 5° C. 25° C. 40° C. RPC suspension Control 97.5 97.6 96.5 * 73.5 73.9 66.1 * RPC suspension MilliQ 97.2 97.9 91.3 * 72.6 73.3 60.7 * * Analysis not performed 49962115vl