ELSEVIER International Journal of Pharmaceutics 122 (1995) 17-25 international journal of pharmaceutics Solid-state characterization of zanoterone William L. Rocco a,., Cynthia Morphet b, Sharon M. Laughlin a a Sanofi Research Division, 1250 S. Collegeville Rd, Collegeville, PA 19426, USA b Shaklee Inc., 1992Alpine Way, Hayward, CA 94545, USA Received 1 June 1994; revised 12 November 1994; accepted 17 January 1995 Abstract Differential scanning calorimetry (DSC) data, in conjunction with Fourier transform infrared (FTIR) spec- troscopy and X-ray powder diffraction (XRPD), was used to discriminate various polymorphic forms of zanoterone. The solid-state degradation rate of form IV (hemihydrate A) was found to be greater than that of form III at 40 ° C/25% RH and 40 ° C/75% RH. At 40 ° C/75% RH, the rate of degradation was 4-fold higher for form IV vs form III. Mixtures of forms III and IV were observed following desolvation of the acetonitrile solvate at 40 ° C with humidity levels of 60 and 70% RH, while at 40 ° C and 20-40% RH pure form III was formed. At 60 ° C, a mixture of forms III and IV was observed at 70% RH but not at lower relative humidity. The concentration of each form of zanoterone in solution was measured as a function of time at room temperature in 3% sodium lauryl sulfate. Forms I, II, and V showed greater initial concentrations than form III, but the values decreased with time due to the precipitation of form VI (hemihydrate B). Therefore, concentrations measured for forms I, II, and V were not equilibrium solubilities for these forms. Forms IV and VI showed lower initial concentrations (than form III) which did not change significantly with time. Keywords: Polymorphism; Physical characterization; Solubility; Stability I. Introduction It has been shown that the varying properties of the different crystal forms of a compound can influence the bioavailability, stability, and proces- sibility of the drug product (Haleblian and Mc- Crone, 1969; Haleblian, 1975; Shefter, 1981; Byrn, 1982; York, 1983). Complete characterization of the different polymorphs is an essential part of a preformulation program. * Corresponding author. Tel. 610-983-7949. Several crystal forms of zanoterone exist; how- ever, form III is the desired polymorphic form used in the clinical drug product. Form III was chosen due to the high reproducibility of isolation in manufacture. During the manufacture of zan- oterone, the drug substance is first isolated as the acetonitrile solvate, then dried to remove the acetonitrile, thus converting the solvate to form III. This investigation was performed to provide an extensive study of the physical characteristics and to gain an understanding of the process pa- rameters which are critical to obtaining the de- sired quality of zanoterone: 0378-5173/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0378-5173(95 )00030-5
`
`Merck Exhibit 2160, Page 1
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`18 W.L. Rocco et al. / International Journal of Pharmaceutics 122 (1995) 17-25 H C °H ~C=-C-H Several polymorphic forms were characterized by thermal analytical techniques, Fourier trans- form infrared spectroscopy, and X-ray powder diffraction. This study examines the physical/ chemical stability and solubility of various poly- morphs and illustrates the effects of desolvation conditions on the resulting crystal form. 2. Materials and methods 2.1. Preparation of polyrnorphs Zanoterone was recrystallized from acetoni- trile by dissolving in solvent at 70°C and cooling to 0°C in an ice bath. The product crystallized and was isolated by filtration as the acetonitrile monosolvate form of zanoterone. Form I was obtained by heating form VI of zanoterone to 180°C for 30 min under nitrogen. Form II was obtained by recrystallization from ethanol and vacuum drying at 45 ° C. Form III was isolated by desolvating the acetonitrile solvate form at 80 ° C under vacuum. Samples of form IV (hemihydrate A) were initially isolated by sus- pending form III in acetonitrile, cooling to 0 ° C, filtering, and air drying under ambient condi- tions. Form IV was also observed in the desolva- tion study (below) by desolvating the acetonitrile solvate at 40 ° C/75% RH. Form V was obtained by recrystallization from methanol, followed by recrystallization from ace- tonitrile, and air drying at 80°C. Form VI (hemihydrate B) was isolated by recrystallization from ethanol/water (70:30% v/v) (23°C crystal- lization temperature). Forms I, II, and III were identified by Houtaling (unpublished data) in 1985. 2.2. Characterization techniques Differential scanning calorimetry (DSC) was performed on the Perkin Elmer DSC-7 system at 10°C/min heating rate under nitrogen purge. The samples were encapsulated in aluminum pans with pierced lids to allow for escape of volatiles. The system was calibrated with indium (m.p. 156.6 ° C) and tin (m.p. 231.9 ° C) prior to use. Thermal gravimetric analysis was performed on the Perkin Elmer TGA-7 system at 10 ° C/min with nitrogen purge gas. The accuracy of the instrument was verified with barium chloride di- hydrate (14.7% water) prior to use. Fourier transform infrared (FTIR) spectra were obtained with the Nicolet 730 system equipped with an MCT detector. The samples were prepared as potassium bromide disks at approx. 1% drug. For quantitative analysis, the total drug concentration was increased to 3%, data were converted to absorbance, and the ratio of absorbance, A(1642 cm- 1)/A(1581 cm- 1), was calculated. Plotting this ratio vs concentration allowed the determination of form IV concentra- tion in mixtures with form III. The absorbance values were corrected for baseline changes at each wavelength. Each concentration (2, 4, 6, 8, and 16%) was examined a minimum of three times and a linear regression performed on aver- age peak ratio as a function of concentration. X-ray powder diffraction (XRPD) patterns were obtained on the Scintag XDS system with Cu-a radiation and a liquid nitrogen cooled solid-state germanium detector. Samples were ground with a mortar and pestle, spread onto a zero background quartz plate, and scanned at a rate of 5°/min. The high-performance liquid chromatographic method utilized throughout this study consisted of a Waters 510 pump, a Waters WISP 710B autosampler, a Kratos 757 UV/Vis variable- wavelength detector, and a Waters 740 data mod- ule. The column was a 10/zm, 25 cm Whatman Partisil ODS-3, with a mobile phase of acetoni- trile/methanol/water/85% (w/w) phosphoric acid (400:300:300:2), at a flow rate of 1.0 ml/min. Eluting peaks were monitored at 240 nm. Two samples were prepared at each stability station, and duplicate injections were made for each sample. A 50 mg quantity of stressed drug substance was dissolved in 10.0 ml of acetonitrile and sonicated for 5 min to obtain a 5 mg/ml
`
`Merck Exhibit 2160, Page 2
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`W.L. Rocco et al. / International Journal of Pharmaceutics 122 (1995) 17-25 19 solution. This solution was injected onto the col- umn and the concentration of WIN 17782 H (the primary degradation product) was calculated from peak area ratios against an external stan- dard. The results were reported as a percentage of the total drug content of the sample. An aliquot of the original sample was diluted with acetoni- trile/water (60:40) to obtain a final concentration of 0.1 mg/ml. Dilution was necessary in the sam- ples prepared for the determination of zan- oterone to prevent column overload. This diluted sample was injected onto the column, and the concentration of drug substance was determined from peak area ratios against an external stan- dard. The solubilities of several crystal forms of zan- oterone were determined in aqueous sodium lau- ryl sulfate (3% by weight). The solubility values were not equilibrium values for forms I, II, and V due to apparent phase transformations. Approx. 8 mg of each sample was added to a screw-capped vial containing 2 ml of solvent. The samples were mixed on a laboratory rotator. An aliquot was filtered (0.45 ixm) at several time points and diluted 1:25 for HPLC analysis. The residual solid was recovered and analyzed by FTIR for several samples. 2.3. Physical~chemical stabifity studies Forms III and IV were investigated for chemi- cal/physical stability because these forms were the desired form and a crystalline impurity ob- served in production, respectively. The acetoni- trile solvate was included since it was used in production to generate form III. Samples of poly- morphic forms III, IV, and the acetonitrile sol- vate were held at 40°C (approx. 25% RH), 40 ° C/75% RH and 80 ° C/vacuum (the tempera- ture used in the manufacturing process). The samples were analyzed for degradation level (% WIN 17782) by HPLC, polymorphic form by FTIR, and water content by Karl-Fisher (Mettler DL 20) analysis. Surface area data were obtained with the Quantasorb (Quantachrome Inc.). Mois- ture sorption studies were performed on the VTI MB 300W system (VTI Corporation), cycling samples from 40 ° C/25% RH to 40 ° C/75% RH. The acetonitrile solvate also was held at 40 and 60 ° C over a range of relative humidity values to determine the effect of humidity and tempera- ture on the polymorphic form that is obtained after desolvation. The FTIR spectra of these samples were used to determine polymorphic form and also qualitatively to assess whether des- olvation was complete. 3. Results and discussion 3.1. Polymorph characterization 3.1.1. DSC The DSC scans for polymorphic forms I-VI and the acetonitrile solvate are shown in Fig. 1. The form I scan shows a single melting peak at approx. 212°C and a heat of fusion of 60.6 J/g. The scan for form II demonstrated an en- dothermic peak at approx. 195°C due to melting. The melting peak appeared to be preceded by a weak shoulder possibly indicative of a low level of form III. The XRPD data appear to confirm this hypothesis (section 3.1.3). The DSC scan for form III, the polymorphic form used in the clinical formulation, shows two significant endotherms. The endotherms were ob- served at 190°C due to simultaneous melting and polymorphic transformation of form III to form I and at 212°C due to the melting transition of form I. Since the peak at 190 ° C was broader and of lower energy (28 J/g) than a typical melting peak, it was apparent that the transition was more complex than simple melting. The scan for form IV (hemihydrate A) showed a broad endotherm at approx. 70-100 ° C followed
`
`Merck Exhibit 2160, Page 3
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`20 W..L. Rocco et al. / International Journal of Pharmaceutics 122 (1995) 17-25 Form I Form II Form III Form IV ~ ) Form V -% Form VI onitrile Solvate Temperature *C Fig. 1. Zanoterone DSC data for forms I-VI and acetonitrile solvate. Perkin Elmer DSC-7; scan rate, 10 ° C/rain. by endotherms at approx. 183 and 211°C. The broad endotherm appeared to be related to two separate events: the loss of moisture and the subsequent transformation of form IV to form III. The endotherms observed at 183 and 211°C were due to the transformation of form III to form I and the melting of form I, respectively. It should be noted that Karl-Fisher analysis of form IV samples indicated water levels of 1.7-2.0%, approximately equal to the theoretical value for a hemihydrate (2.1%). The DSC scan for form V showed multiple peaks characteristic of melting/recrystalliza- tion/remelting phenomena. The scan showed an endotherm at 195°C due to melting of form V, followed by an exotherm due to recrystallization. A second melting peak was observed at 209 ° C, presumably due to melting of form I. The DSC scan for form VI (hemihydrate B) shows an endothermic peak from 120 to 140°C followed by peaks at approx. 195°C (weak) and 212°C (sharp). Thermal gravimetric analysis showed a weight loss of about 2.2%, approxi- mately equal to the theoretical value for a hemi- hydrate of 2.1%. The data for water content was confirmed by Karl-Fisher analysis. It is important to note the differences in temperature for the peaks due to moisture loss for forms IV and VI. Form IV shows a DSC peak at 70-100°C while form VI demonstrates a peak from 120-140 ° C. Clearly the moisture in form VI is more tightly bound. The initial peak of the acetonitrile solvate sample, due to desolvation, is followed by peaks characteristic of polymorphic form III. Thermal gravimetric analysis showed that the desolvation was associated with the presence of 8.7% of ace- tonitrile. This value agrees well with the theoreti- cal value (9.0%) for a monosolvate of acetonitrile. The DSC data illustrated differences in the thermal behavior of crystal forms I-VI and the acetonitrile solvate. The presence of solvate for- mation was evident in the initial peaks in forms IV, VI, and the acetonitrile solvate. However, since a single technique is insufficient to investi- gate polymorphic behavior, these samples were further examined by FTIR and X-ray powder diffraction.
`
`Merck Exhibit 2160, Page 4
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`W.L. Rocco et aL / International Journal of Pharmaceutics 122 (1995) 17-25 21 sFo.. V~SForm II~ ~o0 3f2o ~4~0 steo ~ao z~0o zSz0 z6~o lfso i~eo HRVENUMBER Fig. 2. Zanoterone FTIR data for forms I-VI and acetonitrile solvate. Nicolet 730 FTIR; ]% potassium bromide pellet. 3.1.2. FTIR spectroscopy The FTIR spectra of forms I-VI of zan- oterone are shown in Fig. 2. Each spectrum was unique and can be differentiated in several re- gions. In the wavenumber region between 3400 and 3700 cm -1, where absorbance due to the hydroxyl bond was expected, the polymorphs show broad peaks typical of hydrogen bonding. A sharper hydroxyl peak was observed in the form VI sample at approx. 3600 cm-1. In the region of 3250-3400 cm -1, a different absorbance due to the alkyne hydrogen was observed for each poly- morph. The region between 2000 and 2500 cm -1 showed a weak peak for each polymorph at 2100-2110 cm -1. Absorbance in this region is typical of a carbon-carbon triple bond. The peak appeared to be more intense for form II. The acetonitrile solvate shows a unique peak at 2254 cm -1 due to the presence of a nitrile linkage. This peak was useful in qualitatively as- sessing the presence of a partial solvate. The double bond region (1450-1650 cm -1) showed a strong sharp peak at approx. 1642 cm-] in the form IV spectrum. This peak was not observed in the spectrum of polymorph III and was therefore used in quantitative studies. The ability to quantitate form IV in mixtures (section 3.2) with form III was determined by preparing mixtures at 2, 4, 6, 8, and 16% form IV in form III and plotting the peak ratio, A(1642 cm-~)/A(1581 cm-I), vs form IV concentration. The absorbance at each wavenumber was cor- rected for baseline changes. A linear correlation was observed with a correlation coefficient of 0.993. The FTIR data corroborated the DSC results, illustrating differences in the spectrum for each form. The hydroxyl, alkyne, and carbon-carbon double bond regions were particularly useful in distinguishing the different forms. X-ray powder diffraction patterns (below) confirm the DSC, and FTIR conclusions. 3.1.3. XRPD The XRPD patterns for forms I-VI and the acetonitrile solvate are shown in Fig. 3. The pat- terns were unique and indicated that the samples were predominantly different polymorphic forms. The pattern for the form II sample showed sev- eral weak peaks (12.0, 14.5 ° ) consistent with the form III pattern. These data suggested the pres- ence of form III at a low level, thus corroborating the DSC results. The pattern observed for form Intensity /| _ Form V ,,-_ ,,,,,~/~lh A ormv IJl ! I, _, n / AcetonitrileSolvate / , Jill I i I ,~ i . | Degrees (2-daeta) Fig. 3. Zanoterone XRPD patterns for forms I-VI and ace- tonitrile solvate. Scinta XDS system; 5 ° (20)/min.
`
`Merck Exhibit 2160, Page 5
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`22 W.L. Rocco et aL / International Journal of Pharmaceutics 122 (1995) 17-25 Form VI 3. 2' F~rnl "-'0-" Form II Form III 1'o 2o 3o "nine (~r) Fig. 4. Zanoterone solubility data. Solvent, 3% sodium lauryl sulfate; temperature, 23 ° C; HPLC analysis. IV may also suggest the presence of some form III. 3.1.4. Solubility The concentration in solution for forms I-VI of zanoterone was determined in 3% sodium lauryl sulfate as a function of time. Sodium lauryl sulfate was utilized to enhance the aqueous solu- bility and allow differentiation of the various forms. Two separate samples of form III were analyzed at 4 and 24 h to determine reproducibil- ity. The replicate values agreed within 2-3%. The data (Fig. 4) showed significant differences be- tween the samples. Forms I, II, and V, despite having higher melting points than form III, ap- peared to have higher initial concentrations in solution which then decreased with time. Form II was analyzed at 48 h to determine whether the solubility would eventually decrease with time. A concentration of 2.28 mg/ml was observed, indi- cating a concentration decrease. Since the con- centrations of forms I, II, and V decreased with time it was apparent that the initial concentra- tions were not equilibrium solubilities. After 24 h, the concentrations of the form I and form V samples had decreased to a concentration similar to that of form IV. However, these samples ap- pear to be still decreasing at 24 h and are super- saturated with respect to form VI. Since the FTIR spectrum of the residual solid was consis- tent with form VI (see below), one would expect the concentration for these forms to decrease to that of form VI eventually. It should be noted that a similar observation was made with stanozolol (Rocco, 1994). The concentration of form I of stanozolol in 2% SLS was observed to be higher than that of the monohydrate after 24 h despite showing conversion to the monohydrate by FTIR. The form IV (hemihydrate A) sample had a lower solubility (1.7 mg/ml) than form III (2.0 mg/ml), while the form VI (hemihydrate B) sam- ple was the lowest of all the samples at 1.5 mg/ml. Lower solubility and dissolution rate val- ues for the hydrated form of a compound are common; this behavior has been observed with numerous compounds including succinylsulfathia- zole (Shefter and Higuchi, 1963), theophylline (Shefter and Higuchi, 1963; De Smidt et al., 1986), and stanozolol, also a steroid (Rocco, 1994). The existence of multiple polymorphic forms for a particular hydrated state was also observed for amiloride HC1 (Jozwiakowski et al., 1993). Equiv- alent solubilities were reported for the two hy- drated polymorphs. Samples of residual solid from the solubility studies for forms I, III, IV, and V were isolated at the end of the experiment and analyzed by FTIR to determine if polymorphic changes oc- curred during the experiment. The spectra of some (form I, form V) of these samples appeared to indicate conversion to form VI, while others (form III, form IV) were consistent with the spec- trum of the sample prior to solubility analysis. The spectra of the residual solid in the form I and form V samples closely matched that of form VI. It is not surprising that the form I and V samples had converted to form VI, since form VI appears to be the most stable (least soluble). The form II sample was not analyzed by FTIR due to poor recovery. However, observation by polarized light microscopy indicated the presence of long needle-shaped crystals, in contrast to the starting material which consisted of irregular shaped crystals; this suggests the conversion to a hydrated form had initiated, since a needle-like habit was also observed for form VI.
`
`Merck Exhibit 2160, Page 6
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`W.L. Rocco et al. / International Journal of Pharmaceutics 122 (1995) 17-25 23 0.20 0.15 0.10" 0.05' 0.00 ,11 y = 1.0921e-3 + 1.1767e-2x R^2 = 1.000 ~ [] F=m., 5 10 Time (days) Fig. 5. Zanoterone solid-state stability at 40 ° C/75% relative humidity. % WIN 17782 increase vs time for form III vs form IV. 3.2. Sofid-state stabifity study-physical/chemical stability Samples of form III, form IV, and the acetoni- trile solvate were held at 40°C (approx. 25% RH), 40 ° C/75% RH, and 80 ° C to determine the effect on chemical degradation (% WIN 17782 formation) and polymorphic form. It should be noted that WIN 17782 is a potent estrogen and therefore the drug substance specification for % WIN 17782 was set at a very low value (0.1%). The data for % WIN 17782 are plotted in terms of increase (rather than actual value) for samples of forms III and IV held at 40 ° C/75% RH in Fig. 5. Each point is the average increase for two separate samples. The data showed a significantly higher rate of degradation for form IV vs form III. Linear regression of the data indicated that form IV degrades approx. 4-fold faster than form III under these conditions. Sur- face area analysis demonstrated that the average specific surface area was 0.6 m2/g for both forms, indicating that the reaction rate differences can- not be explained by surface area differences. The data for samples held at 40°C (approx. 25% RH) (Fig. 6) also suggested a higher degra- dation rate for form IV vs form III, but the increases were less substantial for these samples when compared with the 40 ° C/75% RH data. The greater rates of degradation for form IV may or may not be related to the increased mois- ture levels for form IV. The moisture levels ob- served for form IV were about 1.8% at both 40 ° C (approx. 25% RH) and 40 ° C/75% RH. In con- trast, the water levels for form III were 0.2% or less. Water participates in the reaction of zan- oterone to form WIN 17782 by attack at the methane sulfonyl linkage. It is possible that the methane sulfonyl linkage is more vulnerable in the form IV crystal structure and therefore form IV would degrade more rapidly than form III. It is interesting to note that both the form III and form IV samples degraded faster at 40 ° C/75% RH than at 40 ° C/25% RH despite having similar moisture levels under the two con- ditions (as measured by Karl-Fisher analysis). Ahlneck and Zografi (1990) discussed the ability of very small amounts of increased moisture to cause significant changes in chemical and physi- cal properties through modification of molecular mobility. In order to test the theory that very small increases in the amount of water may be present at 40 °C/75% vs 40°C/25% RH, sam- ples were analyzed on the VTI moisture sorption 0.05 0.04 7 • 0.03 0.02 0.01 0.00 /X / Form lU / . - -A- Form IV .r" / / / J ¢ i ¢ 4 4 ¢ 4 d j / / J / ,4 / . .A'"" D. , 5 10 15 Time ~days) Fig. 6. Zanoterone solid-state stability at 40 ° C/25% relative humidity. % WIN 17782 increase vs time for form III vs form IV.
`
`Merck Exhibit 2160, Page 7
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`24 W..L. Rocco et al. / International Journal of Pharmaceutics 122 (1995) 17-25 t~to (r F- m~ \ 7 Day' 2 Day Initial Form IV N Form IV Stressed 40°C/75% Rtl Initial ACN Sol,ate ~00 3~20 34~0 3160 2~80 2~00 E)EO E6~Q If60 I~BO NRVENUMBER Fig. 7. Zanoterone physical stability for form IV stressed at 80 ° C. Nicolet 730 FTIR; 1% potassium bromide pellet. system. Form III showed a reversible uptake of 0.12% moisture when cycled from 40°C/25% RH to 40 °C/75% RH and back to 40°C/25% RH. Form IV showed a similar behavior, indicat- ing that the mechanisms proposed by Ahlneck and Zografi may be a logical explanation for this phenomenon. The form III samples held at 80°C (vacuum) were more chemically stable than the 40 ° C/75% RH station. The form IV samples initially showed degradation at 80 ° C, however, a change in form was observed in the form IV samples (below) and the rate of degradation decreased. This change in form during the 80 ° C stressing prevents meaning- ful interpretation of degradation for the different forms. ~oo 3";'zo 34~o ~16o z6eo z6oo zSzo ~ ;]8o NIRVENUHBER Fig. 9. Zanoterone acetonitrile solvate held at 40°C/75% relative humidity. Nicolet 730 FTIR; 1% potassium bromide pellet. The form IV samples partially or completely transformed to form III with time at 80 ° C. Fig. 7 shows the FTIR spectra of a form IV sample held at 80 ° C as a function of time. The spectra clearly indicated a change for form IV under these con- ditions. Form III was stable with respect to poly- morphic form at 80 ° C. The acetonitrile solvate was observed to con- vert to form III at 40°C (approx. 25% RH) (Fig. 8) and 80°C but to form IV at 40 ° C/75% RH (Fig. 9). The form IV sample contained approx. 2% water. This observation indicated that the humidity should be held at a lower value in order to obtain form III. This subject is investigated in detail below. eo mm // Form III / Z~O ]__~__~~ / Stressed 40°C i--~ mm ~ lnitiat ~ ACN Solvate ~1oo a~zo 34~.o 3tso z~eO Zgoo Z~20 Z6~,O 1~60 l~.eO NRVENLIMBER Fig. 8. Zanoterone acetonitrile solvate held at 40°C/25% relative humidity. Nicolet 730 FTIR; 1% potassium bromide pellet. 3.3. Desolvation of acetonitrile solvate Samples of the acetonitrile solvate of zan- oterone were held at 40°C and 60°C at various humidity levels to determine the effect on poly- morphic form. FTIR analysis after 3 days showed all samples to be completely free of acetonitrile. The data indicated that at 40°C form III was obtained at lower humidity values (20-40% RH) while at higher humidity values (60-70% RH) significant levels of form IV with form III were observed. FTIR analysis indicated that at 40°C/60% RH approx. 15% form IV was ob- served after 3 days, while at 40°C/70% RH approx. 47% form IV was observed after 3 days.
`
`Merck Exhibit 2160, Page 8
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`W.L. Rocco et al. / International Journal of Pharmaceutics 122 (1995) 17-25 25 At 60 ° C, the form IV levels were reduced. At 60% RH (3 days) there was not a measurable peak (< 2%) in the spectra due to form IV. At 60 °C/70% RH the form IV level was approx. 10% after 3 days, significantly lower than the level of form IV (47%) at 40 ° C/70% RH. At 80 ° C, acetonitrile solvate samples were de- solvated at approx. 70% and 95% RH to deter- mine if form III would be obtained as a pure polymorph. FTIR analysis showed the absence of form IV, indicating the presence of pure form III. The concentration in solution for forms I-VI was determined at room temperature in 3% sodium lauryl sulfate. Forms I, II, and V, despite having higher melting points than form III, were found to have higher initial concentrations (than form III) which eventually decreased with time due to the precipitation of form VI. Form III had a solubility which was higher than the form IV and form VI samples after 24 h. References 4. Conclusions Zanoterone exists in multiple crystalline forms with unique physical and spectroscopic proper- ties. These forms appeared to have unique FTIR spectra and X-ray diffraction patterns, useful in identifying each solid phase. FTIR was found to be useful in determining the polymorphic purity of form III when contaminated with form IV. The higher degradation rate of form IV repre- sents an example of the potential effects of crys- tal form on degradation rate. Thus, during the course of preformulation studies, if multiple poly- morphs are discovered, it is important to deter- mine the degradation rate of each form. The isolation of the desired polymorphic form (form III) was shown to occur under conditions of lower relative humidity at 40-60 ° C. The isolation of form lII appeared to be independent of hu- midity at 80 ° C. This study clearly illustrated the effect of humidity on the isolation of form III and indicated that, in general, humidity should be carefully controlled during drying or desolvation. Ahlneck, C. and Zografi, G., The molecular basis of moisture effects on the physical and chemical stability of drugs in the solid state. Int. J. Pharm., 62 (1990) 87-95. Byrn, S., Solid Stale Chemistry of Drugs, Academic Press. New York, 1982. De Smidt, J., Fokkens, J., Grussels, H. and Crommelin, D., Dissolution of theophylline monohydrate and anhydrous theophylline in buffer solutions. Z Pharm. Sci., 52 (1986) 497-501. Haleblian, J., Characterization of habits and crystalline modi- fications of solids and their pharmaceutical applications. J. Pharm Sci., 64 (1975) 1269-1288. Haleblian, J. and McCrone, W., Pharmaceutical applications of polymorphism. 3. Pharm. Sci., 58 (1969) 911-929. Jozwiakowski, M., Williams, S. and Hathaway, R., Relative physical stability of the solid forms of amiloride HCI. Int. J. Pharm., 91 (1993) 195-207. Rocco, W., Solid state characterization of stanozolol. Drug Det:. Ind. Pharm., 20 (1994) 1831-1849. Shefter, E., Solubility by solid state manipulation. Techniques of Solubilization of Drugs, Dekker. New York, 1981, Ch. 5. Shefter, E. and Higuchi, T., Dissolution behavior of crys- talline solvated and nonsolvated forms of pharmaceuticals. J. Pharm. Sci., 52 (1963) 781-791. York, P., Solid-state properties of powders in the formulation and processing of solid dosage forms. Int. J. Pharm., 14 (1983) 1-28.
`
`Merck Exhibit 2160, Page 9
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`