Delayed release composition for peroral administration

Abstract

According to the invention there is provided a pharmaceutical composition comprising N-butyloxycarbonyl-3-(4-imidazol-1-ylmethylphenyl)-5-iso-butylthiophene-2-sulfonamide (C21), or a pharmaceutically-acceptable salt thereof, in which composition the C21 or salt thereof is protected by the presence of a coating comprising an enteric substance. Preferred dosage forms comprise capsules in which C21 or salt thereof is presented in the form of a dry powder mixture or a suspension of particles of C21 in a solvent in which it is insoluble. Such dosage forms find utility in the treatment of lung diseases, such as idiopathic pulmonary fibrosis, sarcoidosis and respiratory virus-induced tissue damage.

Claims

1. A pharmaceutical dosage form that is suitable for peroral administration to the gastrointestinal tract, which dosage form comprises: a pharmaceutical composition comprising N-butyloxycarbonyl-3-(4-imidazol-1-ylmethylphenyl)-5-iso-butylthiophene-2-sulfonamide, or a pharmaceutically-acceptable salt thereof, in which composition the N-butyloxycarbonyl-3-(4-imidazol-1-ylmethylphenyl)-5-iso-butylthiophene-2-sulfonamide or salt thereof is the sole active agent and is protected by the presence of a coating comprising an enteric substance.

2. The dosage form as claimed in claim 1, wherein the enteric substance is polyvinyl acetate phthalate or a methacrylic acid copolymer.

3. The dosage form as claimed in claim 1, wherein the final dosage form comprises an enterically-coated pill, tablet, capsule or film.

4. The dosage form as claimed in claim 3, wherein the final dosage form is an enterically-coated capsule.

5. The dosage form as claimed in claim 1, wherein N-butyloxycarbonyl-3-(4-imidazol-1-ylmethylphenyl)-5-iso-butylthiophene-2-sulfonamide or the salt thereof is provided in the form of a powder, a simple mixture, granules, pellets, beads, a solution or a suspension.

6. The dosage form as claimed in claim 5, wherein the form is a simple powder mixture.

7. The dosage form as claimed in claim 6, wherein the capsule is a hard-shell, two-piece capsule.

8. The dosage form as claimed in claim 7, wherein the capsule comprises hydroxypropyl methylcellulose.

9. The dosage form as claimed in claim 5, wherein the form is a suspension of particles of N-butyloxycarbonyl-3-(4-imidazol-1-ylmethylphenyl)-5-iso-butylthiophene-2-sulfonamide or the salt thereof in a lipid carrier in which it is insoluble.

10. The dosage form as claimed in claim 9, wherein the capsule is a soft-shell, single-piece capsule.

11. The dosage form as claimed in claim 10, wherein the capsule comprises gelatin.

12. The dosage form as claimed in claim 1 wherein N-butyloxycarbonyl-3-(4-imidazol-1-ylmethylphenyl)-5-iso-butyl-thiophene-2-sulfonamide or pharmaceutically-acceptable salt thereof is provided in the form of particles having a weight- and/or a volume-based mean diameter that is no more than about 50 μm.

13. The dosage form as claimed in claim 1 that is essentially water-free.

14. The dosage form as claimed in claim 1 wherein the pharmaceutically-acceptable salt of N-butyloxycarbonyl-3-(4-imidazol-1-ylmethyl-phenyl)-5-iso-butylthiophene-2-sulfonamide is a sodium salt.

15. A process for the production of a dosage form as defined in claim 4, which comprises loading N-butyloxycarbonyl-3-(4-imidazol-1-ylmethylphenyl)-5-iso-butylthiophene-2-sulfonamide or a pharmaceutically-acceptable salt thereof into a capsule, which capsule is coated with an enteric substance.

16. A dosage form obtainable by the process as defined in claim 15.

17. A method of treatment of an interstitial lung disease, which method comprises the administration of a dosage form as defined in claim 1 to a patient in need of such treatment.

18. The method of treatment as claimed in claim 17, wherein the interstitial lung disease is idiopathic pulmonary fibrosis.

19. The method of treatment as claimed in claim 17, wherein the interstitial lung disease is sarcoidosis.

20. A pharmaceutical dosage from that is suitable for peroral administration to the gastrointestinal tract, which dosage form comprises: a pharmaceutical composition comprising N-butyloxycarbonyl-3-(4-imidazol-1-ylmethylphenyl)-5-iso-butylthiophene-2-sulfonamide, or a pharmaceutically-acceptable salt thereof, in which composition the N-butyloxycarbonyl-3-(4-imidazol-1-ylmethylphenyl)-5-iso-butylthiophene-2-sulfonamide or salt thereof is the sole active agent and is protected by the presence of a coating comprising an enteric substance, wherein the enteric substance is selected from the group consisting of cellulose acetate, cellulose acetate succinate, cellulose acetate phthalate, cellulose acetate tetrahydrophthalate, polyvinyl acetate phthalate, hydroxyethyl] ethyl cellulose phthalate, methacrylic acid copolymers, polymethacrylic acid/acrylic acid copolymers, styrol maleic acid copolymers, hydroxypropyl methyl cellulose phthalate, acrylic resins, cellulose acetate trimellitate, hydroxypropyl methylcellulose trimellitate, shellac, hydroxyethyl ethyl cellulose phthalate, carboxymethylcellulose and hydroxypropyl! methyl] cellulose acetate succinate.

21. A method of treatment of an interstitial lung disease, which method comprises orally administering the pharmaceutical dosage form of claim 20 to a patient in need of such treatment.

22. The method of treatment as claimed in claim 20, wherein the interstitial lung disease is idiopathic pulmonary fibrosis.

23. The method of treatment as claimed in claim 20, wherein the interstitial lung disease is sarcoidosis.

Description

EXAMPLES

Example 1

Observation of Food Effect in a Clinical Setting

(1) A Phase I clinical trial was conducted to evaluate the safety, tolerability, and pharmacokinetics of C21 in healthy male and female subjects. The study design allowed a gradual escalation of dose with intensive clinical data and PK monitoring to ensure the safety and wellbeing of subjects.

(2) The study was designed and conducted in accordance with the CRO's standard operating procedures (SOPs), which comply with the ethical principles laid down in the International Council for Harmonisation (ICH) Good Clinical Practice (GCP) guidance as required by the major regulatory authorities, and in accordance with the Declaration of Helsinki as amended by the 48th General Assembly in October 1996. C21 was found to be safe and generally well tolerated. No serious adverse events were recorded.

(3) In addition, the effect of food on the pharmacokinetics (PK) of C21 was investigated in an open label manner in 10 male and female subjects at a single oral dose of 75 mg.

(4) Subjects were to receive two single doses of 75 mg of C21 at least 3 days apart in a randomised fashion either in fasted or in fed state. A Food and Drug Administration (FDA)-recommended high-fat breakfast was used to investigate maximum effects.

(5) In the food effect part of the study, the subjects' ages ranged from 22 to 44 years with a median age of 36.0 years. Most of the subjects were Caucasian (9 (90.0% of) subjects) and male (8 (80.0% of) subjects). The mean (SD) BMI of overall subjects was 23.96 (2.223) kg/m.sup.2.

(6) C21 sodium salt as a pre-frozen oral solution (comprising 2.5 mg/mL C21 sodium salt sodium salt dissolved in an aqueous carbonate buffer) with a composition as shown in Table 1 below.

(7) TABLE-US-00001 TABLE 1 Quality references Components Function Quantity per mL standard Citric acid Buffer component 2 mg Ph. Eur. monohydrate Denatonium Flavouring agent 3 μg USP-NF benzoate Ethanol (96 per Solvent for Approx. 35 μL Ph. Eur. cent) denatonium benzoate Hydrochloric acid pH regulator q.s. to pH 2.0-3.5 Ph. Eur. Sodium hydroxide pH regulator q.s. to pH 2.0-3.5 Ph. Eur. Water, purified Solvent to 1.0 mL Ph. Eur.

(8) Plasma concentrations below the lower limit of quantification (LLOQ) were presented as below the lower limit of quantification (BLQ). Plasma concentrations for C21 were summarised by study part, dose cohort and nominal time point.

(9) Pharmacokinetic parameters were calculated by non-compartmental analysis methods from the concentration-time data using Phoenix® WinNonlin® (Version 8.0) or higher following these guidelines: Actual sampling times relative to dosing rather than nominal times were used in the calculation of all derived pharmacokinetic parameters. There was no imputation of missing data. Any subjects with missing concentration data were included in the PK analysis set provided that at least C.sub.max and AUC.sub.(0-t) could be reliably calculated.

(10) All BLQ values pre-dose and in the absorption phase prior to the first quantifiable concentration were substituted by zeros. Single BLQs that fell between two evaluable concentrations were substituted by missing, before the calculation of the PK variables. Consecutive BLQs that fell between evaluable concentrations were substituted by zero, before the calculation of the PK variables. Terminal BLQ values were disregarded.

(11) Bioequivalence of PK parameters from the open label food effect arm were determined by constructing 90% confidence intervals around the estimated difference between the Test and Reference treatments using a mixed effects model based on natural log transformed data. The mixed effects model was implemented using SAS Proc Mixed, with REML estimation method and Kenward-Roger degrees of freedom algorithm.

(12) Natural logarithm transformed AUC.sub.(0-inf) (if data permitted), AUC.sub.(0-24) and C.sub.max were analysed using a mixed effects model with sequence, period and treatment as fixed effects and subject within sequence as a random effect. Estimates of the mean differences (Test-Reference) and corresponding 90% confidence intervals were obtained from the model. The estimated mean differences and 90% confidence intervals for the differences were exponentiated to provide estimates of the ratio of geometric means (Test/Reference) and 90% confidence intervals for the ratios. C21 in the fasted state is the Reference treatment and C21 in the fed state is the Test treatment.

(13) The comparisons assessing the food effect (bioequivalence) were used to assess the effect of food on the rate and extent of absorption of C21 administered under fasted and fed conditions with estimates of the mean differences as well as corresponding 90% confidence intervals being presented.

(14) Pharmacokinetic parameter data for C21 are summarized descriptively in Table 2 below.

(15) TABLE-US-00002 TABLE 2 Parameter C21 (75 mg) Fed C21 (75 mg) (Unit) Statistic (N = 10) Fasted (N = 9) C.sub.max (ng/mL) n 10 9 Arithmetic Mean 269.0 1810 SD 58.92 668.2 CV % 21.9 36.9 Geometric Mean 263.2 1708 Geometric CV % 22.4 37.1 C.sub.av (ng/mL) n 10 9 Arithmetic Mean 33.0 77.4 SD 8.92 26.4 CV % 27.0 34.1 Geometric Mean 32.0 73.6 Geometric CV % 25.7 34.1 t.sub.max (h) n 10 9 Median 1.26 0.670 Minimum 0.67 0.33 Maximum 4.0 0.68 t.sub.1/2 (h) n 10 9 Arithmetic Mean 0.862 0.602 SD 0.242 0.235 CV % 28.1 39.0 Geometric Mean 0.835 0.569 Geometric CV % 26.6 35.4 AUC.sub.(0-6) n 10 9 (h*ng/mL) Arithmetic Mean 746.6 1853 SD 157.8 630.5 CV % 21.1 34.0 Geometric Mean 732.0 1764 Geometric CV % 21.1 34.0 AUC.sub.(0-12) n 10 9 (h*ng/mL) Arithmetic Mean 790.5 1857 SD 211.9 632.6 CV % 26.8 34.1 Geometric Mean 767.4 1767 Geometric CV % 25.5 34.1 AUC.sub.(0-24) n 10 9 (h*ng/mL) Arithmetic Mean 792.0 1857 SD 214.1 632.6 CV % 27.0 34.1 Geometric Mean 768.5 1767 Geometric CV% 25.7 34.1 AUC.sub.(0-t) n 10 9 (h*ng/mL) Arithmetic Mean 767.3 1841 SD 219.9 626.8 CV % 28.7 34.1 Geometric Mean 741.5 1752 Geometric CV % 27.6 34.2 AUC.sub.(0-inf) n 10 9 (h*ng/mL) Arithmetic Mean 792.0 1857 SD 214.2 632.6 CV % 27.0 34.1 Geometric Mean 768.5 1767 Geometric CV% 25.8 34.1 % AUC.sub.ex (%) n 10 9 Arithmetic Mean 3.5 0.89 SD 2.4 0.29 CV % 69.0 32.2 CL/F (L/h) n 10 9 Arithmetic Mean 100.3 44.52 SD 23.57 14.43 CV % 23.5 32.4 Geometric Mean 97.59 42.44 Geometric CV % 25.8 34.1 V.sub.z/F (L) n 10 9 Arithmetic Mean 120 36.6 SD 26.2 12.4 CV % 21.9 33.9 Geometric Mean 118 34.8 Geometric CV % 21.0 33.9 CV%: Coefficient of variation; SD: Standard deviation. N: The number of subjects included in the PK Analysis Set for each treatment.

(16) Following oral doses of 75 mg C21, peak plasma concentrations occurred at a median t.sub.max of 0.67 and 1.26 hours when administered under fasted and fed conditions, respectively. The geometric mean C.sub.max was 1708 ng/mL for 75 mg fasted and 263 ng/mL for 75 mg fed.

(17) Variability (geometric CV %) for C.sub.max was 37% and 22% for fasted and fed, respectively. The geometric mean AUC.sub.(0-12) was 1767 h*ng/mL and 767 h*ng/mL for 75 mg fasted and fed, respectively. All AUC parameters were comparable for each treatment and geometric CV % for AUC parameters ranged from 21% to 34% for 75 mg fasted and fed. The mean t.sub.1/2 was less than 1 hour for both fasted and fed treatments. Consistent with measured AUC values, CL/F was approximately 2-fold higher for the fed treatment and V.sub.z/F was 3-fold higher compared to the fasted treatment values.

(18) Statistical analysis of the food effect on C21 is summarized in Table 3.

(19) TABLE-US-00003 TABLE 3 Geometric PK Geometric Mean Parameter Mean Ratio (unit) Treatment (95% CI) (Fed:Fasted) 90% CI (%) AUC.sub.(0-inf) C21 (75 mg) 768.52 0.449 (41.04, (h*ng/mL) Fed (N = 10) (642.86, 918.75) 49.04) C21 (75 mg) 1713.18 Fasted (N = 9) (1431.01, 2050.98) AUC.sub.(0-24) C21 (75 mg) 768.48 0.449 (41.03, (h*ng/mL) Fed (N = 10) (642.85, 918.66) 49.04) C21 (75 mg) 1713.20 Fasted (N = 9) (1431.09, 2050.93) C.sub.max C21 (75 mg) 263.19 0.155 (12.08, (ng/mL) Fed (N = 10) (222.21, 311.72) 19.78) C21 (75 mg) 1702.66 Fasted (N = 9) (1422.88, 2037.44) (N = number of subjects with non-missing values. A subject dropped out before receiving treatment under fasted condition).

(20) The model is a mixed analysis of variance (ANOVA) model with a fixed term for sequence, period, treatment, and a random effect for subject within sequence.

(21) Statistical analysis of the effect of food on C21 PK parameters (see Table 3 above) showed there was a decrease in C.sub.max when C21 was given with food. The geometric mean ratio for C.sub.max was 0.16, and the 90% confidence interval fell below 1 (100%), indicating that the C.sub.max difference was statistically significant. Administration of 75 mg C21 with food decreased AUC.sub.(0-24) and AUC.sub.(0-inf); the geometric mean ratios were both 0.45 and the 90% confidence intervals fell below 1 (100%) indicating the differences in AUC values were significant.

Example 2

Dissolution Studies

(22) (A) 50.7 mg of C21 sodium salt (Ardena, Riga, Latvia) was added to 900 mL of 0.09M carbonate buffer (pH 8.95) at a temperature of 37±3° C. with stirring. The compound dissolved instantly. After 15 minutes of stirring, 2M acetic acid solution was added dropwise to provide a pH of 4.52. Evolution CO.sub.2 was noted. After stirring for 1 hour, the formation of small white particles was observed. After additional stirring for a further 1.5 hours, 1M NaOH was added to increase pH to 6.8. Stirring was continued for another 1.5 hours with no significant change in appearance (small white particles).

(23) (B) 51.2 mg of C21 sodium salt was added to 900 mL of acetate buffer (pH 4.4) at the same temperature with stirring. The added compound formed a thin slurry which was floated on the surface top. After stirring for 1 hour, 1M NaOH solution was added dropwise to increase the pH to 7.2. The slurry became thinner and pieces became smaller and goo-like. Stirring was continued for another 1.5 hours with no significant change in appearance (small goo-like particles).

(24) (C) 53.2 mg of C21 sodium salt was added to 900 mL 0.1M of HCl buffer (pH 1.0) at the same temperature with stirring. The added compound dissolved instantly. After 20 minutes of stirring, 1M NaOH solution was added dropwise to increase the pH to 4.5. After addition of the NaOH solution no precipitation was observed. After stirring for 2 hours formation solution still was clear.

(25) (D) 51.0 mg of C21 sodium salt was added to 900 mL of 0.1M citrate buffer (pH 4.42) at the same temperature with stirring. The added compound formed a thin slurry. It seemed that nothing had dissolved. After stirring for 7 hours at the same temperature with no significant change in appearance. Analysis by UPLC showed no decomposition of C21 had occurred at the end of the experiment.

(26) (E) 50.8 mg of C21 sodium salt was added to 900 mL of an acetate buffer (pH 4.49) at the same temperature. The added compound formed a thin slurry which floated on the surface. After stirring for 1 hour, a 1M NaOH solution was added dropwise to increase the pH to 6.8. The slurry became slightly thinner. Stirring was continued for another hour with no significant change in appearance. UPLC showed no decomposition of C21 had occurred at the end of the experiment.

(27) Taken together, these results show that C21's Zwitterion that is formed at intermediated pHs is, unexpectedly, insoluble. This explains the food effect seen in Example 1 above.

Example 3

Dosage Form of the Invention

(28) An excipient blend was prepared by weighing 21.4 g of colloidal silicon dioxide (Aerosil®; Evonik) into a weighing boat. 2033.8 g of mannitol (Pearlitol 50C, Roquette) was then weighed and approximately half of that amount was poured into to a 25L V-shell of a V-blender (Multiblender, Pharmatech, UK).

(29) The weighed amount of colloidal silicon dioxide was then added to the V-shell, followed by the remaining mannitol. The resultant mixture was blended for 10 minutes at 30 rpm.

(30) The excipient blend was then sieved through an 800 μm sieve, prior to blending for a further 20 minutes at 30 rpm.

(31) Half of the resultant excipient blend was weighed and re-added to the V-shell. 528 g of C21 sodium salt (Ardena, Riga, Latvia) was then added to the V-shell. The remaining excipient blend was then added to the V-shell, followed by blending for 10 minutes at 30 rpm.

(32) The resultant blend was then sieved through an 800 μm sieve, followed by blending for 20 minutes at 30 rpm.

(33) After the blend was prepared, blend uniformity was determined by weighing about 270 mg of blend sample into a 100 mL volumetric flask, adding 40 mL of MilliQ water and 20 minutes of sonication, adding 40 mL of methanol and sonicating for a further 20 minutes. After equilibrating to room temperature, 1.0 mL of the sample solution was added to a 10 mL volumetric flask. This was followed by diluting to the desired volume with methanol and mixing.

(34) The sample was filtered through a 0.45 μm PTFE membrane syringe filter, and the first 3 mL of the filtrate were discarded. The amount of C21 sodium salt was determined by UHPLC. The resulting solution should contain 0.1 mg/mL of C21 Na-salt (for 100% of the nominal sample concentration).

(35) The results are shown in Table 4 below.

(36) TABLE-US-00004 TABLE 4 Sample Assay (%, l.c.) 1 100.3 2 102.1 3 104.1 4 100.9 5 98.7 6 99.3 Mean 100.9 RSD 1.9

(37) After this, 26.1 g of magnesium stearate (Ligamed MF-2-V, Peter Greven, Germany) was sieved through an 800 μm sieve and added to the blend, following by final blending for 15 minutes at 15 rpm.

(38) The final composition is as set out in Table 5 below.

(39) TABLE-US-00005 TABLE 5 Composition Ingredient mg/capsule % w/w C21 sodium salt 52.8 20.24 mannitol (Pearlitol 500) 203.38 77.93 colloidal silicon dioxide (Aerosil 200) 2.14 0.82 magnesium stearate (Ligamed MF-2-V) 2.61 1.00

(40) Approximately 6,700 capsules were encapsulated using an MG Compact (MG2, Bologna, Italy) with dosators Size 0, in which the following settings were applied chamber—11 mm; compression—0 mm; powder layer: 30.0 mm.

(41) Weight sorting was done applying a 5% tolerance limit on the net fill weight of a capsule and was found to be 18.6%. After encapsulation the capsules were manually primary packaged in 100 mL high density polyethylene (HDPE) jars with child-resistant, tamper evident caps containing desiccant (56 capsules/jar). A total of 97 jars were produced and labelled for use in a clinical trial.

(42) About 600 of these capsules (obtained according to the procedure described in Example 3 above) were coated using a pan coater (4M8TriX Pan Coater with a 2L drum and a 0.7 mm diameter nozzle; ProCepT, Belgium) with an aqueous acrylic enteric coating system, which was applied as a 20% solution in water.

(43) 640 g of purified water was weighed into an 800 mL beaker and 160 g of pre-weighed Acryl-EZE® 93F19225 Clear (Colorcon) was then added gradually to the water with mechanical stirring for 20 minutes, placing the propeller stirrer in the centre and as close to the bottom of the vessel as possible, and forming a vigorous vortex to homogenize. The dispersion was passed through a 200 μm sieve prior to the coating process.

(44) In order to promote good movement of the capsules within the pan coater and to reach the proper volume to be loaded into the drum, 560 dummy capsules (size 0) were also added, with a different capsules colour (dark green) and fill weight (410 mg/capsule of mannitol (Pearlitol 160C)), to allow for separation afterwards by weigh sorting and visual inspection.

(45) A stepwise coating was performed with samples taken at predetermined times.

(46) The coating process was performed by filling the drum with the capsules, setting the inlet temperature to 40° C. Whilst the drum was turning, the capsules were heated to 30° C. before spraying the coating solution, at which point the final inlet temperature was set.

(47) At the end of the coating process, the heating system was turned off and the capsules allowed to dry under slow tumbling.

(48) The amount of sprayed solution was 740 g. The filling content per capsule of the composition is 260.93 mg per capsule. With an empty capsule weighing 96.1 mg and a total weight of 357 mg per filled capsule, the total weight per coated/filled capsule is 490.66 mg, which gives a coating amount per capsule of 133.66. This amounted to a weight of coating per capsule that was 139.08% of empty capsules and 37.44% of filled capsules.

(49) The capsules were submitted to a Ph. Eur. (10th edition) standard (2.9.1 apparatus B; with disks) two-stage disintegration test subjecting coated capsules (n=6) to: (a) pH 1.2 (0.1N HCl in water, made by mixing 250 mL of 0.2M NaCl, 425 mL of 0.2M HCl and 325 mL of purified water); and (b) pH 6.8 using a phosphate buffer (made by mixing 250 mL of 0.2M potassium hydrogen phosphate, 112 mL of 0.2M NaOH and 638 mL of purified water).

(50) The apparatus consisted of a basket-rack assembly, a device for raising and lowering the basket, a 1 L beaker and a thermostatic arrangement for heating the fluid at 37° C. (±2° C.). The basket-rack assembly was designed to contain 3 capsules in 3 different transparent cylinder tubes placed onto a stainless-steel screen, which allowed for entrance of the solution into the tubes.

(51) For these experiments cylindric transparent plastic discs, with five holes, were placed on top of the floated capsules to keep them inside the tube during the test (without the discs the capsules would float on the surface of the media).

(52) According to the Ph. Eur., capsules with a gastro-resistant shell should survive for 2 hours in acid medium without showing signs of disintegration or rupture permitting the escape of the content.

(53) After the 2 hours the basket-rack assembly was gently dried, and the coated capsules inspected visually in order to identify any sign of deformation or rupture.

(54) Subsequently the basket was transferred into a phosphate buffer solution pH 6.8. According to the Ph. Eur. specification, all of the capsules should disintegrate within 60 minutes.

(55) Visual inspection showed fully intact capsules in the acidic medium (a) and rapid capsule disintegration in the more basic medium (b) and, accordingly, that the capsules were successfully enterically-coated.

Example 4

Stability Study of the Dosage Form of the Invention

(56) Enterically-coated capsules obtained using the method described in Example 3 above were tested in a study to evaluate the stability in a clinical representative packaging at ICH (International Council of Harmonisation) storage conditions (i) 25° C. and 60% RH (long term storage condition) and (ii) 40° C. and 75% RH (accelerated storage condition).

(57) The final composition is as set out in Table 6 below.

(58) TABLE-US-00006 TABLE 6 Concentration (mg/capsule) Ingredients 50 mg strength C21 (free acid) 50 Mannitol (Pearliol ® 500) 203.38 Colloidal Silicium Dioxide 2.14 (Aerosil ®) Magnesium stearate 2.61 (Ligamed ® MF-2-V) Vcaps Plus, size 0, white opaque 1 piece Coating Layer (Acryl-EZE ®; 139.08 % of empty capsule)

(59) The stability results under different storage conditions and time are presented in Tables 7 (assay and chromatographic purity) and 8 (disintegration) below. Water content was measured using the Karl Fischer titration method. Impurity 1 is previously known for C21, while impurities 2 and 3 are novel. LOR stands for limit of reporting (i.e. 0.10%, I.c.).

(60) TABLE-US-00007 TABLE 7 Water C21 Impurities (%, l.c.) Time content (%, Sum Condition (months) (%, w/w) l.c.) 1 2 3 (≤2) Initial — 95.2 0.20 0.16 < LOR 0.36 25° C. and 3 — 97.8 0.22 0.15 < LOR 0.37 60% RH 4.5 <0.5 98.8 0.24 0.15 < LOR 0.39 6 <0.5 100.7 0.24 0.15 < LOR 0.39 40° C. and 3 — 95.4 0.60 0.14 < LOR 0.75 75% RH 4.5 <0.5 97.1 0.80 0.14 < LOR 0.94 6 <0.5 101.9 1.0 0.15 < LOR 1.2

(61) TABLE-US-00008 TABLE 8 Time Condition (months) Medium Mean (min:s) % RSD Initial 0.1M HCl pH 1.2 N/A N/A Phosphate buffer pH 6.8 33:55 <0.1 25° C. and 3 0.1M HCl pH 1.2 N/A N/A 60% RH Phosphate buffer pH 6.8 28:57 6.9 4.5 0.1M HCl pH 1.2 N/A N/A Phosphate buffer pH 6.8 26:39 7.5 6 0.1M HCl pH 1.2 N/A N/A Phosphate buffer pH 6.8 31:48 8.8 40° C. and 3 0.1M HCl pH 1.2 N/A N/A 75% RH Phosphate buffer pH 6.8 25:24 7.8 4.5 0.1M HCl pH 1.2 N/A N/A Phosphate buffer pH 6.8 25:53 7.5 6 0.1M HCl pH 1.2 N/A N/A Phosphate buffer pH 6.8 27:57 6.2

(62) All stability disintegration results were in line with the expectations, the enterically-coated capsules did not dissolve in acidic media and rapidly dissolved at pH 6.8. The assay was stable over the 6 months stability period at both storage conditions.