Compositions in the form of an injectable aqueous solution including at least human insulin A21G and a glucagon suppressor with prandial action

Abstract

A composition in the form of an injectable aqueous solution, with pH from 3.5 to 4.4, including at least human insulin A21G and at least one glucagon suppressor with prandial action. In an embodiment, the glucagon suppressor with prandial action is selected from an amylin analog or an amylin receptor agonist or a GLP-1 analog or a GLP-1 receptor agonist (GLP-1 RA). In an embodiment, the glucagon suppressor with prandial action is an amylin analog or an amylin receptor agonist. In an embodiment, the glucagon suppressor peptide with prandial action is pramlintide. Also, a method for obtaining human insulin A21G, includes at least one step of reacting human insulin A21G, B31R, B32R (insulin glargine) with rat carboxypeptidase B at an insulin/carboxypeptidase ratio from 500 to 2000, at a pH from 7.5 to 8.5 and a temperature from 20 to 30° C. for 10 to 20 hours.

Claims

1. A composition in the form of an injectable aqueous solution, the pH of which is from 3.5 to 4.4, including at least human insulin A21G referred to as regular in a range from 100 to 300 U/ml and pramlintide at a concentration of between 0.4 to 3.0 mg/ml.

2. A composition according to claim 1, wherein the concentration of human insulin A21G is in a range from 100 to 200 U/ml.

3. The composition according to claim 1, wherein the concentration of human insulin A21G is 100 U/ml.

4. The composition according to claim 1, wherein the concentration of human insulin A21G is 200 U/ml.

5. The composition according to claim 1, wherein the concentration of pramlintide ranges from 0.5 to 1.5 mg/ml.

6. The composition according to claim 1, wherein the concentration of pramlintide ranges from 0.6 to 1 mg/ml per 100 U/ml of human insulin A21G.

7. The composition according to claim 1, wherein the pH of the solution is from 3.8 to 4.2.

8. The composition according to claim 1, wherein the pH of the solution is 4.0.

9. The composition according to claim 1, wherein the composition further includes a zinc salt.

10. The composition according to claim 1, wherein the composition further includes m-cresol.

11. The composition according to claim 1, wherein the composition further includes a surfactant selected from the group consisting of Poloxamer 188, Polysorbate 20, and Polysorbate 80.

12. The composition according to claim 1, wherein the composition further includes a poloxamer 188 excipient.

13. The composition according to claim 1, wherein the composition further includes methionine.

14. The composition according to claim 1, wherein the composition is used in a diabetes treatment method, wherein the composition is administered in a bolus before meals.

15. The composition according to claim 1, wherein the composition is used in a diabetes treatment method, wherein the composition is administered to improve control of postprandial glycemia.

16. The composition according to claim 1, wherein the composition is used in a diabetes treatment method, wherein the composition is administered to improve control of postprandial glycemia and to decrease the adverse effects of pramlintide.

17. The composition according to claim 1, wherein the composition is used in a diabetes treatment method, wherein the composition enables a decrease in food consumption induced by insulin.

18. A solid composition comprising at least human insulin A21G referred to as regular and pramlintide in a ratio from 33 to 167 U insulin /mg of pramlintide.

19. A method for treating a diabetic patient in need of an insulin formulation, comprising administering by injection to said patient a composition comprising at least human insulin A21G referred to as regular in a concentration range from 100 to 300 U/ml and pramlintide at a concentration range from 0.4 to 3 mg/ml, wherein the pH of the composition is in a range from 3.5 to 4.4.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1: Graphic determination of the fibrillation lag time.

(2) FIG. 1 is a representation of the graphic determination of the fibrillation lag time based on a virtual example. On the abscissa, the time in minutes appears, and, on the ordinate, the fluorescence ThT appears in arbitrary units (a.u.), and LT stands for lag time.

(3) FIG. 2: Concentrations of pramlintide and insulin in the plasma after administration of formulation A21-8 (mean±standard deviation).

(4) The squares represent the concentration of insulin, and the triangles represent the concentration of pramlintide.

(5) On the abscissa, the time in minutes after injection appears; on the ordinate on the left, the baseline-corrected insulin concentration in pmol/L appears, and, on the ordinate, the baseline-corrected pramlintide concentration in pmol/L appears.

(6) FIG. 3: Glycemia after administration of formulation A21-8 (mean±standard deviation).

(7) On the abscissa, the time in minutes after injection appears, and, on the ordinate, the glycemia in % of the baseline level appears.

(8) FIG. 4: Concentration of pramlintide after administration of formulation A21-9 (curve drawn with squares) and PRAM (curve drawn with triangles) (mean±standard deviation).

(9) On the abscissa, the time in minutes after injection appears, and, on the ordinate, the baseline-corrected pramlintide concentrations appear (pre-dose concentrations subtracted individually), in pmol/L.

EXAMPLES

Example 1. Preparation of Human Insulin A21G

(10) 5 g of insulin glargine (Gan & Lee Pharmaceuticals) are mixed with the enzyme carboxypeptidase B (Reference 08039852001; Sigma-Aldrich) at pH 8.0 (pH adjusted by addition of Tris buffer), and the mixture is allowed to stand for 17 hours at 25° C., the insulin glargine concentration being approximately 4 mg/mL. The enzyme/glargine ratio is 1/500. The mixture is then purified by liquid chromatography, dialyzed against hydrochloric acid 0.01N and then lyophilized. The result is human insulin A21G with a purity of 98% and a yield of approximately 90%. The molecular weight of the insulin measured by mass spectrometry (Maldi-Tof) is 5752 Da. The human insulin A21G can also be obtained using the recombinant technology as described by Kohn et al. (Peptides 2007, 28, 935-948).

Example 2. Compositions of Prandial Insulins and of Pramlintide, Exenatide or Lixisenatide at Acidic pH

Preparation of a Solution of Human Insulin A21G 100 U/mL (3.5 mg/mL) and of Pramlintide 1 mg/mL Containing m-Cresol (25 mM), Glycerol (184 mM) and Zinc Chloride (300 μM) at Acidic pH from of 3.5 or 4.0

(11) A concentrated solution of excipients (m-cresol, glycerol) is added to a concentrated solution of human insulin A21G (300 U/mL at pH 3.5). A concentrated solution of pramlintide (Ambiopharm) (10 mg/mL at pH 4) and a concentrated solution of zinc chloride are added to this concentrated solution of human insulin A21G and of excipients so as to obtain the intended final composition. The final pH, namely 3.5 or 4.0, is adjusted to the desired value by addition of an aqueous solution of NaOH or of HCl. The solution obtained is clear and homogeneous; it is subjected to a 0.22 μm filtration and stored in glass cartridges (1 mL of solution per cartridge).

Preparation of a Solution of Human Insulin A21G 100 U/mL and of Pramlintide 0.6 mg/mL Containing m-Cresol (25 mM), Glycerol (184 mM) and Zinc Chloride (300 μM) at Acidic pH of 4.0

(12) This solution is prepared in the same manner as the solution presented above.

Preparation of a Solution of Human Insulin A21G 100 U/mL and of Pramlintide 1 mg/mL Containing m-Cresol (25 mM) and Glycerol (184 mM) at Acidic pH of 4.0

(13) A concentrated solution of excipients (m-cresol, glycerol) is added to a concentrated solution of human insulin A21G (800 U/mL at pH 3.5). A concentrated solution of pramlintide (10 mg/mL at pH 4) is added to this concentrated solution of human insulin and of excipients so as to obtain the intended final composition. The final pH, namely 4.0, is adjusted to the desired value by addition of an aqueous solution of NaOH or of HCl. The solution obtained is clear and homogeneous; it is subjected to a 0.22 μm filtration and stored in glass cartridges (1 mL of solution per cartridge).

Preparation of a Solution of Human Insulin A21G 100 U/mL and of Pramlintide 1 mg/mL Containing m-Cresol (25 mM), Glycerol (184 mM) and Tween 20 (10 μg/mL) at pH 4

(14) A concentrated solution of excipients (m-cresol, glycerol) is added to a concentrated solution of human insulin A21G (300 U/mL at pH 3.5). A concentrated solution of pramlintide (10 mg/mL at pH 4) and a concentrated solution of Tween 20 are added to this concentrated solution of human insulin A21G and of excipients so as to obtain the intended final composition. The final pH is adjusted to the desired value by addition of an aqueous solution of NaOH or of HCl. The solution obtained is clear and homogeneous; it is subjected to a 0.22 μm filtration and stored in glass cartridges (1 mL of solution per cartridge).

Preparation of a Solution of Human Insulin A21G 100 U/mL and of Pramlintide 1 mg/mL Containing m-Cresol (25 mM), Glycerol (184 mM), Zinc Chloride (300 μM) and Tween 20 (10 μg/mL) at pH 4

(15) A concentrated solution of excipients (m-cresol, glycerol) is added to a concentrated solution of human insulin A21G (300 U/mL at pH 3.5). A concentrated solution of pramlintide (10 mg/mL at pH 4), a concentrated solution of zinc chloride and concentrated solution of Tween 20 are added to this concentrated solution of human insulin A21G and of excipients so as to obtain the intended final composition. The final pH is adjusted to the desired value by addition of an aqueous solution of NaOH or of HCl. The solution obtained is clear and homogeneous; it is subjected to a 0.22 μm filtration and stored in glass cartridges (1 mL of solution per cartridge).

Preparation of a Solution of Human Insulin A21G 100 U/mL and of Exenatide 50 μg/mL Containing m-Cresol (25 mM), Glycerol (184 mM) and Zinc Chloride (300 μM) at Acidic pH

(16) A concentrated solution of excipients (m-cresol, glycerol) is added to a concentrated solution of human insulin A21G (300 U/mL at pH 3.5). A concentrated solution of exenatide (Bachem) (10.5 mg/mL at pH 4) and a concentrated solution of zinc chloride are added to this concentrated solution of human insulin A21G and of excipients so as to obtain the intended final composition. The final pH 4.0 is adjusted to the desired value by addition of an aqueous solution of NaOH or of HCl. The solution obtained is clear and homogeneous; it is subjected to a 0.22 μm filtration and stored in glass cartridges (1 mL of solution per cartridge).

Preparation of a Solution of Human Insulin A21G 100 U/mL and of Lixisenatide 100 μg/mL Containing m-Cresol (25 mM), Glycerol (184 mM) and Zinc Chloride (300 μM) at Acidic pH

(17) A concentrated solution of excipients (m-cresol, glycerol) is added to a concentrated solution of human insulin A21G (230 U/mL at pH 3.5). A concentrated solution of lixisenatide (Ambiopharm) (10.5 mg/mL at pH 4) and a concentrated solution of zinc chloride are added to this concentrated solution of human insulin A21G and of excipients so as to obtain the intended final composition. The final pH 4.0 is adjusted to the desired value by addition of an aqueous solution of NaOH or of HCl. The solution obtained is clear and homogeneous; it is subjected to a 0.22 μm filtration and stored in glass cartridges (1 mL of solution per cartridge).

Preparation of a Solution of Human Insulin A21G 100 U/mL, of Exenatide 50 μg/mL and of Pramlintide 0.6 mg/mL Containing m-Cresol (25 mM), Glycerol (184 mM) and Tween 20 (10 μg/mL) at Acidic pH of 4.0

(18) A concentrated solution of excipients (m-cresol, glycerol) is added to a concentrated solution of human insulin A21G (300 U/mL at pH 3.5). A concentrated solution of pramlintide (Ambiopharm) (10 mg/mL at pH 4), a concentrated solution of exenatide (Bachem) (10.5 mg/mL at pH 4) and a concentrated solution of Tween 20 are added to this concentrated solution of human insulin A21G and of excipients so as to obtain the intended final composition. The final pH 4.0 is adjusted to the desired value by addition of an aqueous solution of NaOH or of HCl. The solution obtained is clear and homogeneous; it is subjected to a 0.22 μm filtration and stored in glass cartridges (1 mL of solution per cartridge).

Preparation of a Solution of Human Insulin 100 U/mL and of Pramlintide 1 mg/mL Containing m-Cresol (25 mM), Glycerol (184 mM) and Zinc Chloride (300 μM) at Acidic pH of 3.5 or 4.0

(19) A concentrated solution of excipients (m-cresol, glycerol) is added to a concentrated solution of human insulin (Amphastar Pharmaceuticals) (800 U/mL at pH 3.5). A concentrated solution of pramlintide (10 mg/mL at pH 4) and a concentrated solution of zinc chloride are added to this concentrated solution of human insulin and of excipients so as to obtain the intended final composition. The final pH, namely 3.5 or 4.0, is adjusted to the desired value by addition of an aqueous solution of NaOH or of HCl. The solution obtained is clear and homogeneous; it is subjected to a 0.22 μm filtration and stored in glass cartridges (1 mL of solution per cartridge).

Preparation of a Solution of Insulin Aspart 100 U/mL and of Pramlintide 1 mg/mL Containing m-Cresol (25 mM), Glycerol (184 mM) and Zinc Chloride (300 μM) at Acidic pH of 3.5 or 4.0

(20) A concentrated solution of excipients (m-cresol, glycerol) is added to a concentrated solution of insulin aspart (HEC Pharmaceuticals) (500 U/mL at pH 3). A concentrated solution of pramlintide (10 mg/mL at pH 4) and a concentrated solution of zinc chloride are added to this concentrated solution of insulin aspart and of excipients so as to obtain the intended final composition. The final pH, namely 3.5 or 4.0, is adjusted to the desired value by addition of an aqueous solution of NaOH or of HCl. The solution adjusted to pH 4.0 is turbid after the pH adjustment. The solution adjusted to pH 3.5 is clear. It is subjected to a 0.22 μm filtration and stored in glass cartridges (1 mL of solution per cartridge).

Preparation of a Solution of Insulin Lispro 100 U/mL and of Pramlintide 1 mg/mL Containing m-Cresol (25 mM), Glycerol (184 mM) and Zinc Chloride (300 μM) at Acidic pH of 3.5 or 4.0

(21) A concentrated solution of excipients (m-cresol, glycerol) is added to a concentrated solution of insulin lispro (Gan & Lee Pharmaceuticals) (650 U/mL at pH 3). A concentrated solution of pramlintide (10 mg/mL at pH 4) and a concentrated solution of zinc chloride are added to this concentrated solution of insulin lispro and of excipients so as to obtain the intended final composition. The final pH, namely 3.5 or 4.0, is adjusted to the desired value by addition of an aqueous solution of NaOH or of HCl. The solution obtained is clear and homogeneous; it is subjected to a 0.22 μm filtration and stored in glass cartridges (1 mL of solution per cartridge).

Preparation of a Solution of Insulin Glulisine 100 U/mL and of Pramlintide 1 mg/mL Containing the Excipients of the Commercial Product Apidra® (29 mM of m-Cresol, 50 mM of Tris, 86 mM of Zinc Chloride and 8.15 μM of Tween 20) at Acidic pH of 3.0, 3.5 or 4.0

(22) The pH of the commercial solution of insulin glulisine, Apidra®, is adjusted to pH 2.5 by addition of an aqueous solution of HCl. This solution is added to pramlintide in the form of a powder so as to obtain a solution containing 100 U/mL of insulin and 1 mg/mL of pramlintide. The final pH is adjusted to the desired value by addition of an aqueous solution of NaOH or of HCl. The solutions adjusted to pH 3.5 and 4.0 are turbid after the pH adjustment. The solution adjusted to pH 3.0 is clear. It is subjected to a 0.22 μm filtration and stored in glass cartridges (1 mL of solution per cartridge). After a few hours of storage, the solution is turbid and heterogeneous.

Preparation of a Solution of Pramlintide 1 mg/mL Containing m-Cresol (20 mM), Mannitol (236 mM) and Acetic Acid/Sodium Acetate Buffer (30 mM) at pH 4.0

(23) A concentrated solution of pramlintide at 10 mg/mL is prepared by dissolution of pramlintide in the form of a powder purchased from Ambiopharm. This solution is added to a concentrated solution of excipients (m-cresol, mannitol, acetic acid/sodium acetate buffer) so as to obtain the intended final composition. The final pH is adjusted to 4.0±0.2 by addition of NaOH/HCl.

Preparation of a Solution of Human Insulin A21G 100 U/mL and of Pramlintide 0.6 mg/mL Containing m-Cresol (25 mM), Glycerol (184 mM), Acetic Acid/Sodium Acetate Buffer (18 mM) and Tween 20 (8 μM) at pH 4

(24) Concentrated solutions of glycerol and m-cresol are added to a concentrated solution of human insulin A21G in an acetic acid/sodium acetate buffer at pH 4 (300 U/mL at pH 4). A concentrated solution of pramlintide (Ambiopharm) (10 mg/mL at pH 4) and a concentrated solution of Tween 20 are finally added to this concentrated solution of human insulin A21G and of excipients so as to obtain the intended final composition. The final pH is adjusted to the desired value by addition of an aqueous solution of NaOH or of HCl. The solution obtained is clear and homogeneous; it is subjected to a 0.22 μm filtration.

(25) The compositions prepared above are presented in table 1 below:

(26) TABLE-US-00001 TABLE 1 Compositions of insulin and/or of glucagon suppressors Insulin type Pramlintide Zinc Tween 20 Other Compositions (U/mL) (mg/mL) pH (μM) (μg/mL) GLP-1 RA excipients A21-1 Human insulin 1.0 4.0 300 0 0 m-cresol (25 mM) A21G (100) glycerol (184 mM) A21-2 Human insulin 1.0 4.0 0 0 0 m-cresol (25 mM) A21G (100) glycerol (184 mM) A21-3 Human insulin 1.0 3.5 300 0 0 m-cresol (25 mM) A21G (100) glycerol (184 mM) A21-4 Human insulin 1.0 4.0 0 10 0 m-cresol (25 mM) A21G (100) glycerol (184 mM) A21-5 Human insulin 1.0 4.0 300 10 0 m-cresol (25 mM) A21G (100) glycerol (184 mM) A21-6 Human insulin 0 4.0 300 0 exenatide m-cresol (25 mM) A21G (100) 50 μg/mL glycerol (184 mM) A21-7 Human insulin 0 4.0 300 0 lixisenatide m-cresol (25 mM) A21G (100) 100 μg/mL glycerol (184 mM) A21-8 Human insulin 0.6 4.0 300 0 0 m-cresol (25 mM) A21G (100) glycerol (184 mM) RHI-1 Human insulin 1.0 4.0 300 0 0 m-cresol (25 mM) (100) glycerol (184 mM) RHI-2 Human insulin 1.0 3.5 300 0 0 m-cresol (25 mM) (100) glycerol (184 mM) ASP-1 Insulin aspart 1.0 4.0 300 0 0 m-cresol (25 mM) (100) glycerol (184 mM) ASP-2 Insulin aspart 1.0 3.5 300 0 0 m-cresol (25 mM) (100) glycerol (184 mM) GLU-1 Insulin glulisine 1.0 4.0 300 10 0 m-cresol (29 mM) (100) Tris (50 mM) NaCl (86 mM) GLU-2 Insulin glulisine 1.0 3.5 10 0 m-cresol (29 mM (100) Tris (50 mM) NaCl (86 mM) GLU-3 Insulin glulisine 1.0 3.0 10 0 m-cresol (29 mM) (100) Tris (50 mM) NaCl (86 mM) LIS-1 Insulin lispro 1.0 4.0 300 0 0 m-cresol (25 mM) (100) glycerol (184 mM) LIS-2 Insulin lispro 1.0 3.5 300 0 0 m-cresol (25 mM) (100) glycerol (184 mM) PRAM — 1.0 4.0 — — 0 m-cresol (20 mM) mannitol (236 mM) acetate (30 mM) A21-9 Human insulin 0.6 4.0 — 10 0 m-cresol (25 mM) A21G (100) glycerol (184 mM) acetate (18 mM) A21-10 Human insulin 0.6 4.0 — 10 exenatide m-cresol (25 mM) A21G (100) 50 μg/mL glycerol (184 mM)

Example 3. Study of the Compatibility of the Prandial Insulins with Pramlintide at Acidic pH

(27) Visual Appearance of the Solutions of Insulin and of Pramlintide at Acidic pH.

(28) The observation is carried out at ambient temperature after 2 to 3 hours of stabilization of the solution stored in cartridges. Table 2 presents the visual appearance of solutions of insulin and of pramlintide described above.

(29) TABLE-US-00002 TABLE 2 Visual appearance of the solutions of insulin and of pramlintide. Composition Visual appearance A21-1 Clear A21-2 Clear A21-3 Clear RHI-1 Clear RHI-2 Clear ASP-1 Turbid ASP-2 Clear GLU-1 Turbid GLU-2 Turbid GLU-3 Turbid LIS-1 Clear LIS-2 Clear

(30) Among the insulins evaluated, only human insulin, insulin lispro and human insulin A21G enable to obtain a homogeneous and clear formulation with pramlintide at pH 4, demonstrating the solubility of the species. The insulins aspart and glulisine are not suitable for obtaining a clear formulation with pramlintide at pH 4.0.

Example 4. Study of the Fibrillation Lag Time

(31) Principle

(32) The poor stability of a peptide can lead to the formation of amyloid fibrils which are defined as ordered macromolecular structures. These fibrils may lead to the formation of a gel within the sample.

(33) The test of monitoring the fluorescence of thioflavin T (ThT) is used to analyze the physical stability of the formulations. Thioflavin T is a small probe molecule which has a characteristic fluorescence signature when it binds to amyloid fibrils (Naiki et al. (1989) Anal. BioChem. 177, 244-249; LeVine (1999) Methods. Enzymol. 309, 274-284).

(34) This method enables to monitor the formation of fibrils at low ThT concentrations within undiluted formulations. This monitoring is carried out under accelerated stability conditions: under stirring and at 37° C.

Experimental Conditions

(35) The samples were prepared immediately before the start of the measurement. The preparation of each composition is described in the associated example. Thioflavin T is added to the composition from a concentrated stock solution so as to induce a negligible dilution of the composition. The thioflavin T concentration in the composition is 40 μM. A volume of 150 μL of the composition was introduced into a well of a 96-well plate. Each composition was analyzed in triplicate within the same plate. The plate was sealed with transparent film to prevent evaporation of the composition.

(36) This plate was then placed in the enclosure of a plate reader (EnVision 2104 Multilabel, Perkin Elmer). The temperature is set at 37° C., and a lateral agitation at 960 rpm with an amplitude of 1 mm is imposed.

(37) A reading of the fluorescence intensity in each well versus time is carried out with an excitation wavelength of 442 nm and an emission wavelength of 482 nm.

(38) The fibrillation process manifests itself by a strong increase in fluorescence after a delay referred to as lag time.

(39) For each well, this delay is determined graphically as the intersection between the baseline of the fluorescence signal and the slope of the fluorescence curve versus time, which is determined during the initial strong increase in fluorescence, as shown in FIG. 1. The value of the lag time plotted corresponds to the average of the lag times of 3 wells.

(40) The clear solutions of pramlintide and insulin at pH 3.5 and 4.0 of the preceding example are then subjected to the fibrillation test in the presence of ThT.

(41) The lag time reported in Table 3 corresponds to the average of 3 measurements; the uncertainty interval corresponds to the standard deviation between these 3 results.

(42) TABLE-US-00003 TABLE 3 Lag times of the solutions of insulin and pramlintide. Composition Lag time (h) A21-1 13.7 +/− 0.8 A21-2 10.4 +/− 1.8 A21-3 15.6 +/− 5.3 RHI-1  5.0 +/− 0.7 RHI-2 1.7 +/− 0  ASP-2 2.0 +/− 0  LIS-1  1.8 +/− 0.1 LIS-2  4.3 +/− 0.4

(43) Unexpectedly, the formulations containing human insulin A21G have fibrillation lag times that are much longer than those of the commercial insulins tested at pH 3.5 or at pH 4.0, in particular longer than the rapid insulin analogs, insulin lispro and insulin aspart.

Example 5. Study of the Fibrillation Lag Time in the Presence of Tween 20

(44) In Table 4, the lag times of solutions of human insulin A21G and pramlintide at pH 4 in the presence of Tween 20 are presented.

(45) TABLE-US-00004 TABLE 4 Lag time of the solutions of human insulin A21G and pramlintide in the presence of Tween 20. Composition Lag time (h) A21-4 38.1 +/− 8.8 A21-5 48.3 +/− 9.2

(46) The physical stability is thus improved in the presence of Tween 20 at 10 μg/mL

Example 6. Physical Stability of the Formulations at 30° C. Under Static Conditions

(47) Glass cartridges filled with 1 mL of composition are placed in an oven maintained at 30° C. These cartridges are inspected visually in order to detect the appearance of visible particles or of turbidity. This inspection is carried out according to the recommendations of the European Pharmacopoeia (EP 2.9.20): the cartridges are subjected to an illumination of at least 2000 lux and are observed on a white background and on a black background. These results are in agreement with the US pharmacopoeia (USP <790>).

(48) TABLE-US-00005 TABLE 5 Physical stability of the solutions of insulin and pramlintide at 30° C. under static conditions. Composition Visual appearance after 4 weeks A21-4 Clear A21-5 Clear A21-2 Clear A21-1 Clear A21-3 Clear ASP-2 Turbid

(49) Insulin aspart formulated with pramlintide at pH 3.5 is less stable than human insulin A21G formulated with pramlintide at pH 3.5 or 4.0.

Example 7. Study of the Physical Stability of Human Insulin A21G with Exenatide and Lixisenatide

(50) Formulations A21-6 and A21-7 are placed in cartridges and then kept at 30° C. for 4 weeks. The fibrillation lag times are measured for extemporaneously prepared formulations and presented in table 6.

(51) TABLE-US-00006 TABLE 6 Lag time and physical stability at 30° C. under static conditions of the solutions of human insulin A21G and exenatide or lixisenatide. Visual appearance after Composition Lag time (h) 4 weeks at 30° C. A21-6 18.1 +/− 5.6 Clear A21-7 33.0 +/− 9.7 Clear

Example 8. Chemical Stability of a Formulation of Human Insulin A21G and Pramlintide

(52) All the formulations are at pH 4.0 or pH 3.5 and contain 100 U/mL of human insulin A21G, 1 mg/mL of pramlintide, 25 mM of m-cresol and 184 mM of glycerol. The formulations are stored in glass cartridges and kept at 30° C. under static conditions. Human insulin A21G and pramlintide are assayed by reverse phase liquid chromatography (HPLC). The measurements are presented in table 7.

(53) TABLE-US-00007 TABLE 7 Change of the concentrations of insulin (a) in U/mL and of pramlintide (b) in mg/mL Concentration at Concentration the initial time after 4 weeks Composition (a) (b) (a) (b) A21-1 105 1.05 103 1.02 A21-2 105 1.02 106 1.03 A21-3 101 1.12 105 1.03 RHI-1 101 1.6 76 1.03 RHI-2 99 1.02 62 1.03 LIS-1 105 1.07 89 1.03 LIS-2 106 1.05 82 1.04 ASP-2 104 1.05 55 1.02

(54) The formulations containing human insulin A21G and pramlintide exhibit good chemical stability after 4 weeks at 30° C. The formulations of pramlintide with commercial insulins undergo rapid degradation at pH 4.0 and even more rapid degradation at pH 3.5.

Example 9. Pharmacokinetic and Pharmacodynamic Studies in Dogs

(55) Pharmacokinetic and pharmacodynamic study in dogs of the composition consisting of human insulin A21G (100 U/mL, that is to say 3.5 mg/mL) and pramlintide (0.6 mg/mL). The tested formulation is at pH 4.0 and contains 25 mM of m-cresol and 184 mM of glycerol (Formulation A21-8).

(56) Four animals which had fasted for approximately 18 hours received injections by subcutaneous administration in the neck at the dose of 0.2 U/kg of insulin and 0.12 μg/kg of pramlintide. In the hour preceding the injection, one or more blood samples are drawn in order to determine the basal level of glucose, of insulin and of pramlintide. Blood samples are then drawn during the 5 hours after administration of the formulation. The glycemia is determined by means of a glucometer. The levels of insulin and of pramlintide in the plasma are determined by an ELISA test.

(57) The pharmacokinetic parameters of formulation A21-8 are estimated based on baseline-corrected insulin and pramlintide concentrations in the plasma. A standard non-compartmental analysis is carried out with the aid of the software Phoenix WinNonlin (version 7, Certara). The values of the parameters (mean±standard deviation) are reported in tables 8 and 9 below:

(58) TABLE-US-00008 TABLE 8 PK parameters of total insulin analog Tmax insulin Cmax insulin AUC0-last insulin Formulation (min) (pmol/L) (min*pmol/L) A21-8 38 ± 26 218 ± 141 16165 ± 4160

(59) TABLE-US-00009 TABLE 9 PK parameters of pramlintide Tmax Cmax AUC0-last pramlintide pramlintide pramlintide Formulation (min) (pmol/L) (min*pmol/L) A21-8 20 ± 8 104 ± 39 5121 ± 2961

(60) The mean pharmacokinetic (PK) profiles of total insulin (squares) and of pramlintide (triangles) in the plasma are presented in FIG. 2.

(61) The mean glycemia profiles expressed as percentages of the baseline level are represented in FIG. 3.

(62) It is observed that pramlintide and human insulin A21G both have prandial absorption kinetics giving rise to an early hypoglycemic activity followed by a return to a level close to baseline glycemia after 5 hours post-administration. These pharmacokinetic and pharmacodynamic results clearly indicate that Formulation A21-8 is compatible with use at meal time.

Example 10: Pharmacokinetic Studies of Pramlintide in Pigs

(63) Pharmacokinetic study in pigs of the composition consisting of human insulin A21G (3.5 mg/mL equivalent to 100 U/mL of insulin) and pramlintide (0.6 mg/mL).

(64) Domestic pigs weighing approximately 50 kg, catheterized beforehand in the jugular, were fasted for 2.5 hours before the start of the experiment. During the hour preceding the injection of insulin, 3 blood samples were drawn to determine the baseline level of glucose and of insulin.

(65) The injection of the formulations of human insulin A21G combined with pramlintide (A21-9) or of pramlintide (PRAM) at the dose of 0.2 U of insulin/kg and 1.2 μg of pramlintide/kg is performed subcutaneously in the flank of the animal with the aid of an insulin pen (Novo, Sanofi or Eli Lilly) equipped with a 31 G needle.

(66) In order to determine the concentrations of pramlintide in the plasma, blood samples are drawn at the following times: 4, 8, 12, 16, 20, 30, 40, 50, 60, 70, 80, 100, 120, 150 and 180 minutes. After each drawing, the catheter is rinsed with a dilute heparin solution.

(67) Pharmacokinetic results of the solution of human insulin A21G and pramlintide A21-9 and of the solution of pramlintide PRAM in pigs

(68) The results of 3 studies carried out on the same cohort of pigs are pooled to compare the pharmacokinetics of pramlintide between formulation A21-9 and formulation PRAM. The pharmacokinetic parameters of formulations A21-9 and PRAM are estimated based on the baseline-corrected pramlintide concentrations in the plasma. A standard non-compartmental analysis is carried out with the aid of the software Phoenix WinNonlin (version 7, Certara). The values of the parameters (mean±standard deviation) are plotted in the following table.

(69) TABLE-US-00010 TABLE 10 PK parameters of pramlintide of compositions A21-9 and PRAM t.sub.max AUC.sub.0-30 min AUC.sub.0-t Peptides pramlintide pramlintide pramlintide Formulation (mg/mL) N (min) (min*pmol/L) (min*pmol/L) A21-9 human insulin 34 42.1 ± 22.4 2461 ± 1467 13642 ± 5934 A21G (3.5) pramlintide (0.6) PRAM Pramlintide (0.6) 34 23.2 ± 15.0 4357 ± 3127 14806 ± 7872 Value p Comparison A21-9 34 0.0004 0.0007 0.8018 (significant if versus PRAM p < 0.05)

(70) Where t.sub.max=time necessary to observe the maximum plasma concentration; AUC.sub.0-30min=area under the curve of the plasma concentration versus time between 0 and 30 min after injection; AUC.sub.0-t=area under the curve of the plasma concentrations versus time between 0 and the last quantifiable concentration after injection

(71) The pharmacokinetic results of pramlintide obtained with formulations A21-9 and PRAM are presented in FIG. 4. The analysis of these profiles and of the parameters indicates that the combination of human insulin A21G and of pramlintide (formulation A21-9, curve plotted with squares) leads a significant slowing of the absorption of pramlintide compared to pramlintide alone (formulation PRAM, curve plotted with triangles). Formulation A21-9 leads to a plasma peak (t.sub.max) which is significantly delayed (approximately 18 min, p<0.05) and to an early plasma exposure to pramlintide (AUC.sub.0-30min) which is significantly decreased (approximately 43%, p<0.05) in comparison to formulation PRAM. On the other hand, the total plasma exposure to pramlintide (AUC.sub.0-t) appears to be similar between the two formulations, suggesting comparable bioavailabilities.

Example 11. Study of Food Consumption in Rats after Injection of Control Compositions and after Injection of Compositions Including Human Insulin A21G and/or Pramlintide

(72) This study was carried out on a population of 40 at least 6 week old male Sprague Dawley rats.

(73) The rats had free access to food and water, except for a 6 hour fasting period preceding the subcutaneous injection of the compositions described in the table below.

(74) TABLE-US-00011 TABLE 11 Compositions injected in the rats and number of rats treated Composition Control Humulin ® PRAM A21-9 [Pramlintide] — —  1 0.6 (mg/mL) Dose of Pramlintide — — 60 60 (μg/kg) Insulin type — Human — Human A21G [Insulin] (U/mL) 100 100 Insulin dose — 10 — 10 (U/kg) Number of rats 10 10 10 10

(75) The control composition is a saline solution, that is to say an aqueous solution containing 150 mM of NaCl.

(76) The composition Humulin® R is a commercial solution of human insulin marketed by ELI LILLY. This product is a human insulin at 100 U/mL. The excipients of Humulin® R are glycerol, meta-cresol, sodium hydroxide and hydrochloric acid for pH adjustment (pH 7.0-7.8) and water.

(77) At t0, immediately after the injection, the food is distributed (approximately 100 g par rat). The food consumption (cumulative mean) is measured one, two and three hours after t0, or t+1 h, t+2 h and t+3 h.

(78) The results are presented in the following table:

(79) TABLE-US-00012 TABLE 12 food consumption 1, 2 and 3 hours after injection Compositions Control Humulin PRAM A21-9 Food consumption at t + 1 h (g) 3.8 4.7 1.5 3.2 Food consumption at t + 2 h (g) 4.4 5.3 3.3 4.4 Food consumption at t + 3 h (g) 5.9 6.9 4.2 5.3

(80) These results show that the composition A21-9 combining insulin A21G and pramlintide enables not only to decrease the food consumption induced by the injection of insulin, but also to limit the food consumption to a level less than or equal to the level of the control group that received an injection of Control composition (saline solution).