Polymorphic Characterization and Bioavailability of Nomegestrol Acetate①

CHEN Xiao-Feng LI Peng WANG Hui-Ping XU Juan NING Li-Feng

Polymorphic Characterization and Bioavailability of Nomegestrol Acetate①

CHEN Xiao-Feng LI Peng WANG Hui-Ping XU Juan NING Li-Feng②

(100081)

Polymorphscreening is currently one of the most important strategies of innova- tors and generic companies from both pharmaceutical and intellectual property rights perspectives. Different polymorphs may have varying physicochemical properties which influence the bioavailability. The purpose of this study was to investigate the crystal structures and physicochemical properties of Nomegestrol acetate (NOMAC) polymorphs. Forms I and II(dioxane solvate) were isolated and prepared by systemic crystallization screening in this study, and theforms are reported for the first time. A structural analysis and comparison of all the forms are presented. This study was also the first time to apply a rapid and feasible ultra-high- performance-liquid chromatography (UHPLC)-electrospray ionization (ESI)-tandem mass spectrometry (MS) method to determine plasma levels of NOMAC within 3.0 mins. And this study demonstrated that the optimal crystal Form I displayed higher bioavailability than API indicating that Form Icould be an alternative solid form that needs further research.

nomegestrol acetate, crystalline structure, solubility, bioavailability;

1 INTRODUCTION

The importance of pharmaceutical polymorphism has been increasingly recognized in the scientific community, industry and by regulatory agencies over the past decades[1]. Polymorphism is the ability of one compound to exist in more than one crystal structures in the solid state[2, 3]. Polymorphs have different crystalline structures, physical and chemi- cal attributes, such as compressibility, melting point, crystal habit, color, density, dissolution rate and solubility. These differences may affect the pharma- ceutical processing, the stability of drug product and the bioavailability, so the therapeutic efficacy of the drug[4-6]. Crystalline solids that involve the inclusion or incorporation of solvent molecules in the crystal lattice are known as solvates. There are marketed drug products that contain solvates such as darunavir ethanolate (Prezista), indinavir sulfate ethanolate (Crixivan), and warfarin sodium isopropanol solvate (Coumadin)[7, 8].

The most common polymorph screening methods include crystallization, melted substance, vapor and solutions by cooling or evaporating the solvent, addition of an antisolvent to the solution[9-11], slurrying, spray drying, sublimation, grinding and thermal desolvation of solvates[12]. The outcome of a crystallization can be affected by factors such as the solvent (viscosity and polarity), the rate of supersaturation (cooling, adding or evaporation rate), or initial solution concentration[13]. With the deve- lopment of polymorphic research, many new methods have emerged, such as high throughput screening (HTS)[14], microchannel crystallization[15], macromolecules as additives[16],. The polymorph screening methods have greatly promoted the deve- lopment of polymorphic research.

Polymorphism can be characterized using various analytical techniques such as powder X-ray diffrac- tion (PXRD), single-crystal X-ray diffraction (SXRD), differential scanning calorimetry (DSC), thermo gravimetric analysis (TGA), dynamic vapor sorption (DVS), optical and electron microscopy, infrared (IR), near-IR (NIR), Raman and solid-state nuclear magnetic resonance (ssNMR) spectro- scopy[17-21].

Nomegestrol acetate (NOMAC, Fig. 1) is a highly selective progestogen derived from 19-norproges- terone and is structurally related to the naturally occurring progesterone. It has been classified as a “pure” progestogen because it has strong affinity for the human progesterone receptor, strong antigona- dotropic activity and moderate antiandrogenic activity, and is devoid of any estrogen, androgen, glucocorticoid or mineralocorticoid[22-25]. According to the Biopharmaceutics Classification System (BCS), NOMAC belongs to BCS class II com- pounds with very low solubility (0.3～6 μg/mL in water) across physiological pH range. Therefore, increasing the solubility of NOMAC via polymor- phism and consequently improving its bioavaila- bility is of interest for the development of new dosage forms of NOMAC.

Fig. 1. Schematic representation of Nomegestrol acetate (NOMAC) molecule

In this study, two crystalline forms of NOMAC, a solvent-free form (Form I) and a dioxane solvate (Form II. dioxane is a class 2 solvent in the guidelines of FDA and its use should be limited in pharmaceutical products), were prepared using a variety of screened crystallization methods. The crystal structures of both forms were investigated thoroughly using a variety of relevant techniques, and the biopharmaceutical profiles were further investigated.

2 EXPERIMENTAL

2.1 Materials and reagents Chemicals and reagents

NOMAC, the pharmaceutical ingredient (API), was purchased from NEWCHEM Pharmaceuticals Co., Ltd (Verona, Italy, batch number: 20160513). Its chemical purity was more than 99.0% mass fractions, which was determined using high-performance- liquid chromatography (HPLC). All the solvents used for recrystallization were of analytical reagent grade (Fisher Chemical).

Forms I and II of NOMAC were prepared in our laboratory, while the HPLC grade methanol, acetoni- trile and water were obtained from Thermo Fisher Scientific (Waltham, MA, USA). Terfenadine (internal standard IS, 98% purity) was purchased from LIANHUAN Pharmaceuticals Co., Ltd (Jiangsu, China, batch number: 20170603).

Animals

Female Sprague-Dawley rats (weighting 230～250 g) were purchased from Sino-British Experi- ment Animal (Shanghai,China). The animals were treated in accordance with protocols approved by the Laboratory Animal Ethics Committee at the Na- tional Institute of Planned Parenthood Research. All animals were housed five per cage under a 12-h light-dark cycle with free access to food and tap water. Each animal was weighed once weekly and fed a standard rat chow.

2.2 Preparation of the polymorphs

The following different crystallization methods were used to comprehensively screen for polymor- phism screening, cooling, crystallization, evapora- tive crystallization, antisolvent addition and slurring, and two crystalline forms were finally obtained.

Preparation of Form I: To a 10 mL glass vial, 50 mg NOMAC and 4 mL 2-butanone were added. The mixture was slowly heated to 70 ℃ (1 ℃·min–1), and stirred at this temperature until the solid was dissolved. Then the reaction solution was filtered, and the filtrate solution was slowly cooled to 25 ℃ (1 ℃·min–1) in a cold bath. After that the solution was stirred for approximately 2 h at this temperature, filtered and finally vacuum-dried at room tempera- ture.

Preparation of Form II: To a 10 mL glass vial, 50 mg of NOMAC and 3 mL 1,4-dioxane were added. The mixture was slowly heated to 70 ℃ (1 ℃·min–1) and stirred at this temperature until the solid was dissolved, then the reaction solution was filtered.The filtrate solution was cooled down to 25 ℃ (1 ℃·min–1) in a cold bath and then the solution was stirred for approximately 2 h at this temperature, filtered and then vacuum-dried at room temperature.

2.3 Structural determination by X-ray diffraction for single crystals

Single-crystal XR diffraction data were collected by using a Rigaku AFC-10/Saturn 724-CCD diffrac- tometer equipped with graphite-monochromatized MoKa radiation (0.71073 Å) system up to a 2 h limit of 50.0 ℃at room temperature (25 ℃). Indexing and scaling of the data were performed using DENZO and SCALEPACK. The structure was solved by direct methods and expanded using difference Fourier techniques with Shelxs-97 and refined on F2by successive full-matrix least-squares techniques for the non-hydrogen atoms. Anisotropic displacement parameters were employed for non-H atoms and H atoms were treated isotropically with Uiso= 1.2 (for those attached to aromatic carbon and N atoms) or 1.5 times (for those bonded to methyl carbons) the Ueqof the parent atoms. All H atoms were located at the expected positions and refined using a riding model. CIF files were obtained from Shelxs-97. The simulated powder patterns were calculated from the single-crystal data using Mer- cury 3.1[26, 27].

2.4 Powder XRDanalysis

The XRD spectra were recorded using a BRUKER D8 Advance diffractometer (Bruker AXS GmbH, Karlsruhe, Germany) system with CuKaradiation (λ = 1.5406 Å) over an interval of 5～90 °/2θ. The measurement conditions were as follows: target: Cu; filter: Ni; voltage: 40 kV; current: 40 mA; time constant: 0.1 s; angular step: 0.016°, and detector: NaI (Tl) scintillation detector.

2.5 Thermogravimetry-differential scanning calorimeter

TGA-DSC was conducted using a TGA/DSC 3+ equipment (Mettler-Toledo, Switzerland) under a flow of nitrogen (20 mL·min–1) at a scan rate of 10 °C·min–1from 25 to 300 °C. The pan was made of aluminium oxide. The balance was calibrated by the standard weight. And the temperature was calibrated by the melt points of five different metals.

2.6 Infrared spectroscopy with attenuated total reflectance by Fourier transform (FTIR-ATR)

A Spectrum RX I FTIR spectrometer (Perkin Elmer, UK) was employed in the KBr diffuse- reflectance mode (sample concentration 2 mg in 20 mg of KBr) for collecting the IR spectra of samples. Dry KBr (50 mg) was finely ground in mortar, and the sample (1～2 mg) was subsequently added and gently mixed in order to avoid the trituration of crystals. A manual press was used to form the pellet. The spectra were measured over the range of 4000～450 cm–1. Data were analyzed using the Spectrum software.

2.7 Dissolution rate studies

We studied the dissolution curves of two crystal formsand NOMAC materal (API) using a Chinese pharmacopoeia method. Solid samples were sieved using a Gilson mesh sieve (No. 80) to obtain powders of uniform particle size, thenwere added in pH 1.0 and 0.2% sodium dodecyl sulfate (SDS) buffer at 37 °C, respectively.

The experiment was performed using the dissolu- tion apparatus (FADT-800RC, TiandaTianFa. Ltd., China) at the rotation speed set at 100 rpm, and samples (1 mL) were collected at 5, 10, 15, 20, 30, 45, 60, 80, 100, 120 and 180 min. Then the solutin was filtered through a 0.45 μm membrane filter (KeYiLong Ltd, China) and the solution concentra- tions were measured using an HPLC system (Agilent-1100 Agilent Ltd., USA) with the UV/Vis detector set at 275 nm. A Discovery C18 HPLC column (4.6 × 250 mm, 5 μm) was used with the column oven kept at 25 °C. The mobile phase consisted of an acetonitrile-phosphate buffer (30 mM, pH 6.4, 20:80, v/v). The flow rate of the mobile phase was 1 ml/min, and the injection volume was 10 μL. For data acquisition and processing, the LC solution software was used.

2.8 Pharmacokinetic study of new polymorphs

Dawley rats (200～250 g) were randomly divided into two groups and fasted overnight prior to administration. Following single oral dosing of either NOMAC (0.5 mg, group 1) or Form I (0.5 mg, group 2), blood samples (200 µL) were collected pre-dosing and at 0.25, 0.5, 1.0, 2.0, 4.0, 6.0, 8.0, 24.0, 36.0 and 48.0 h post-dosing. Plasma was harvested via centrifugation and stored at −20 °C before analysis.

To an aliquot of 50.0 µL rat plasma in a 10 mL glass tube, 20.0 µL of the internal standard (1.00 ng·ml-1terfenadine), 300 µL of NaOH aqueous (0.1 M), and 3.0 mL of diethyl ether were added. The obtained mixture was vortexed for 5 min, and then centrifuged at 4500 rpm for 10 min. The supernatant was transferred to another glass tube and evaporated to dryness at 40 °C under a gentle stream of air. The residue was reconstituted in 100 µL of methanol- water (55/45, v/v). A 40 µL aliquot of the recon- stituted solution was introduced into the LC-MS/MS system. Pharmacokinetic parameters for polymorphs were calculated using the DAS 2.0.1 (Mathematical Pharmacology Professional Committee of China, Shanghai, China) software using a non-compart- mental analysis. Data were expressed as the means ± SD.

3 RESULTS AND DISCUSSION

3.1 Characterization of the polymorphs

In order to obtain more details of the polymorphic structure information at the atomic level, high quality single crystals of polymorphic Forms I and II were submitted to single-crystalX-ray diffractionanalysis. The molecular structure of the title compound, with atom labelling scheme drawn at 50% probability displacement ellipsoid, is depicted in Fig. 2, and the corresponding packing diagram is shown in Figs. 3 and 4. For better comparison, a summary of the conditions for the data collection and structurerefinement parameters is given as follows. Form I crystallizes in the monoclinic system, space group21with= 2,= 7.0416(14),= 7.6838(15),= 18.498(4) Å,= 90°,= 995.3(5) Å3,D= 1.231 g·cm–3, formula C23H30O4,= 0.083 mm–1, the final= 0.0578 and= 0.0.0518 with> 2(). Another polymorphic Form II belongs to the orthorhombic system, space group212121with= 4,= 7.8457(16),= 14.039(3),= 22.1944(4) Å,= 90°,= 244.58(8) Å3,D= 1.246 g·cm3, formula C27H38O6,= 0.087 mm-1, the final= 0.0537 and= 0.1152 with> 2(). The crystals of Forms I and II are both blocky but crystallize in different space groups. Form I exists in monoclinic chiral space group21, and each asymmetric unit contains six NOMAC molecules. Form II belongs to the orthorhombic chiral212121space group with eight NOMAC and eight dioxane molecules in the asymmetric unit. The conformational differences from the rings and the rotating of the single bond of the side-chain substituents are the main causes of the bimolecular phenomena. Form I adopts a head-to-head packing configuration and forms a one-dimensional infinite chain running along theaxis with the intermolecular hydrogen bonding interaction. So far only one crystal structure of NOMAC has been reported in the Cambridge Crystallographic Data Centre (CCDC 949571), but it has different cell parameters with Form I. For example, they not only have different internal spatial structures, but also pack in different modes, too, indicating their different crystal structures. Form IIadopts a head-to-hail packing configuration, with NOMAC and solvent molecules (1:1) forming a one-dimensional infinite chain along theaxis. Form II belongs to solvate, because it is composed of dioxane. Form II molecules connect the dioxane molecule through no-classical hydrogen bonds like C(27A)–H(27A)–O(3). Although ring C shows chair conformation, the other rings adopt envelope conformation in the two polymorphs, respectively. But the twist degrees of the rings are not the same. The dihedral angles between C(13)–C(14)–C(16) and C(10)–C(12)–C(13) of FormsI and II are 128.20° and 126.89°, respectively. The key parameters that describe doubtlessly the different conformers are the torsional angles of the molecular backbone. The most relevant torsion angles responsible for this polymorphic conformation of NOMAC are compared in Table 1. In Form I, atoms C(20) and O(3) locate at each side of the plane formed by atom C(14)～C(18), and the torsion angle of C(20)–C(18)–O(3)–C(22) is 58.47°. In contrast to Form II, the torsion angle of C(4)–C(5)–O(2)–C(2) is 52.23°. The number of molecules in the asym- metric unit of the two polymorphs is also different (Fig. 3). From the overlay diagrams, it is distinctive that the conformational differences arise because of the conformers of rings and the rotation of side- chain substituents (-OCOCH3). The differences of these angles confirm the existence of two confor- mational polymorphs.

Fig. 2. Molecular view of the compound, showing 50% probability displacement ellipsoids and atom labeling scheme

Fig. 3. Packing structure of Form I

Table 1. Relevant Torsion Angles (°) of Form II in Comparison with Those in Form I

Two-dimensional packing diagrams of Forms I and II are shown in Figs. 3 and 4. The packing occurs only by non-classical hydrogen bonds. Form I packs in a fold line while Form II in a wave. Detailed parameters of main hydrogen-bonding interactions with symmetry codes are listed in Table 2. The inclusion of solvents plays a vital role in forming different crystalline forms. The dioxane molecules introduced resulted in the torsional angle difference in two forms, which induced a variation in the arrangement of NOMAC molecules in the unit cell. It is the dioxane molecule that induces the molecules to bend and pack differently in the forms I and II.

Fig. 4. Packing structure of Form II

Table 2. Hydrogen Bonds of Nomegestrol Acetate (NOMAC) Forms (Å), where D = Donor and A = Acceptor

Symmetry codes in FormsIand II :a2–, 1/2+, 1–;b2–, –1/2+, 1–;c1–, –1/2+, 1–;d–1+,,;e–, 1/2+, 2–;f1/2–, 1–, –1/2+;g1–, –1/2+, 1/2–;h3/2–, 1–, –1/2+;i1/2–, 1–, –1/2+;j1–, 1/2+, 1/2–;k3/2–, 1–, 1/2+;l1/2+, 3/2–, 1–;m1–, 1/2+, 3/2–;n–1+,,

In Form I, the shortest contact for non-classical hydrogen bond is C(17)H(17A)×××O(2) (2.38 Å) (i = 1,1/2+, 1). Additionally, the oxygen of carbonyl group acts as an acceptor in another nonclassical hydrogen bond of C(15)H(15)···O(3). There are no hydrogenbonding interactions on the oxygen of hydroxyl group. In Form II, an asym- metric unit consists of eight NOMAC and eight dioxane molecules. Hydrogen-bonding interactions in Form II give rise to two-dimensional networks by fostering the non-classical hydrogen bond chains, such as C(1)H(1B)×××O(5) (i = 1/2, 1,1/2+), C(25)H(25B)···O(4) (i = 1/2+, 3/2, 1) and C(26)H(26A)···O(6) (i = 1, 1/2+, 3/2).

3.2 Powder XR-Danalysis (PXRD)

PXRD is always used as method for the identi- fication of polymorphs. The PXRD results of the two novel polymorphs and the API are illustrated in Fig.5.Form I showed the characteristic peaksat 2= 12.4, 12.5, 13.6, 14.9, 19.5, 24.2, 28.8, 33.9, 38.9 and 43.9°, whereas Form II exhibited them at 2=10.2,13.3, 15.1, 17.3, 19.2, and 20.5°,indicating the pre- sence of two distinctive polymorphs. The PXRD patternsmeasured from the powder samples were also in good agreement withthose calculated from the single-crystal structures.

Fig. 5. Experimental powder X-ray diffraction (PXRD) patterns of Nomegestrol acetate NOMAC

3.3 Thermal analysis

We observed no noticeable weight loss before the compound decomposition occurred, which indicated that Form I was a solvent-free crystalline form. The melting point of Form Iwas observed as a pro- nounced endothermic peak with the extrapolated onset temperature of 181.42 ℃ and an associated heat of absorption of49.76 J·g1. In Form II, the weight loss occurred before its decomposition, showing that Form II contained solvent in the crystalline form. The dehydration (1 mol of dioxane per mol of NOMAC) of the sample showed to be a solvate by the diffraction techniques (= 61 and 116 ℃; Δ= 19.32%, Δ–1= 19.2%) (Fig. 6), probably because the noclassical weak hydrogen bond force is weak between NOMAC and the di- oxane molecule. Form II melted with a pronounced endothermic peak, with the extrapolated onset tem- perature of 182.27 ℃ and the associated heat of absorption of45.20 J·g1.

Fig. 6. Thermal analysis of Forms I and II(a).Thermogravimetric (TGA) curves of Forms I and II(b)

3.4 Infrared spectroscopy with attenuated total reflectance by Fourier transform (FTIR-ATR)

Both polymorphs of NOMAC can be easily identified and assigned by their IR spectra, as shown in Fig. 7. Prominent differences in the IR spectra can be found in the region above 800 cm1which reflects the X-H out-of-plane bending vibrations.Form I is solvent-free modification and its molecule of cry- stalline formed hydrogen bonds with carbon or oxygen atoms in NOMAC molecule. There was a strong and sharp band at 1659 cm1contributed to the C=O stretching asymmetrical deformation vibration. Form II is a dioxane solvate form, in which a strong and sharp band at 1117 cm1was contributed to the C-O-C stretching asymmetrical deformation vibration. The pinnacle at 1714 cm1is contributed to C=O stretching asymmetrical defor- mation vibration. The results revealed the two solid forms are different from each other by the analyses of FTIR spectra.

Fig. 7. IR spectra for polymorphs of E2 with the scanning region of 4000～450 cm-1

3.5 Dissolution rate studies

The dissolution profiles of Forms I and II were analyzed using the dissolution experiments perfor- med in different buffers at 37 ℃. The samples collected at the pre-set time were filtered prior to HPLC analysis and the residual solids after the dissolution experiments were identified using PXRD. The PXRD patterns measured from the solid residues were in good agreement with those of the original forms, indicating that the polymorphs maintained their form during the dissolution experiment. The purity of NOMAC was determined using the area normalization method.As shown in Fig. 8, the concentrations of the two forms in buffers increase rapidly at first,then it approached equi- librium slowly with increasing the time. Due to the strong hydrogen-bond interaction between dio- xane and drug molecule, the equilibrium dissolution rate of Form II was found to be lower than Form I.

Fig. 8. Dissolution profile for NOMAC, Forms I and II in pH 1.0 and 0.2% sodium dodecyl sulfate (SDS) buffer

3.6 Pharmacokinetic study

NOMAC was quantified using UHPLC-MS/MS. Typical chromatograms of blank plasma, the lowest limit of quantification (LLOQ) sample, and the test sample is shown in Fig. 9. Under optimized chro- matographic conditions, the retention time of the analyte and IS was 2.23 and 1.49 min, respectively, and the total run time was maintained at 3.0 min. The test results demonstrated that there was no significant endogenous interference in the deter- mination of the compound.

Because the toxic solvent involved in Form II,onlyForm I and the NOMAC material (API) were utilized to carry out in the pharmacokinetic experi- ments. The mean plasma concentration versus time plot for Forms I and API after oral administration is shown in Fig. 10. It was found that Form I has higher absorption than NOMAC material (API).After peak, plasma concentrations of the two forms began to decline, but remained at a high level, which formed a plateau. Plasma concentration of the two forms can be measured at a higher level till 8 h, a trace level till 36 h, and almost undetectable after 48 h. Overall, the Cmaxand AUC of Form I were appro- ximately 0.2 times those of the NOMAC material (API).

Fig. 9. Multiple reaction (MRM) chromatogramsI (nomegestrol acetate [NOMAC] and internal standard [IS]: (A) Drug-free plasma, (B) Plasma spiked with 2.5 ng/mL of pharmacokinetic parameters (NOMAC) and 5 μg/mL IS, and (C) Plasma samples 1 h after oral administration to a rat with 5 mg/kg for NOMAC Form I

Fig. 10. Mean plasma concentration time profile of Form I and NOMAC in rat afteroral administration

4 CONCLUSION

Two crystal forms of NOMAC were identified and prepared for the first time during the polymorph screening. And their structures were determined by single-crystal X-ray diffraction. The results of the powder dissolution experiments showed that Form I dissolved apparently faster and had a little higher equilibrium solubility than the NOMAC material (API). Therefore, Form I is the promising candidate for improving the solubility and dissolution rate of NOMAC. Furthermore, we validated a UHPLC-MS quantitative method to rapidly analyze NOMAC in the rat plasma within 3 mins. And the studies revealed that Form I improved the AUC and Cin an animal model compared with the NOMAC material (API).

Conflicts of Interest: The authors declare no conflict of interest.

(1) Haleblian, J.;McCrone, W. Pharmaceutical applications of polymorphism.. 1969, 58,911929.

(2) Byrn, S. R.; Pfeiffer, R. R.; Stephenson, G.;Grant, D. J.;Gleason, W. B.Solid-statepharmaceutical chemistry.. 1994, 6, 11481153.

(3) Vippagunta, S. R.; Brittain, H. G.; Grant, D. W. Crystalline solids.. 2001, 48, 326.

(4) Mcgregor, L.; Rychkov, D. A.; Coster, P. L.; Day,S.; Drebushchak, V. A.; Achkasov, A. F.;Nichol, G. S.; Pulham, C. R.; Boldyreva, E. V. A new polymorph of metacetamol..2015, 17, 61836192.

(5) Sanphui, P.; Goud, N. R.; Khandavilli, U. B.; Bhanoth, S.; Nanqia, A. New polymorphs of curcumin.. 2011, 47, 50135015.

(6) Tong, H. Y.; Chow, A. S.;Chan, H. M.; Chow, A. H.;Wan, Y. K.; Williams, L. D.; Shek, F. L.; Chan, C. K. Process-induced phase transformation of berberine chloride hydrates.. 2010, 99, 19421954.

(7) Lee, A. Y.; Erdemir, D.; Myerson, A. S. Crystal polymorphism in chemical process development.. 2011, 2, 259280.

(8) Byrn, S.; Pfeiffer, R.; Ganey, M.; Hoiberg, C.; Poochikian, G. Pharmaceutical solids: a strategic approach to regulatory considerations.. 1995, 12, 945954.

(9) Morissette, S. L.; Almarsson, Ö.; Peterson, M. L.; Remenar, J. F.; Read, M. J.; Lemmo, A. V.; Ellis, S.; Cima, M. J.; Gardner, C. R. High-throughput crystallization: polymorphs, salts, co-crystals and solvates of pharmaceutical solids.. 2004, 56, 275300.

(10) Aaltonen, J.; Alleso, M.; Mirza, S.; Koradia, V.; Gordon, K. C.; Rantanen, J. Solid form screening–a review.. 2009, 71, 2337.

(11) Newman, A. Specialized solid form screening techniques.. 2013, 17, 457471.

(12) Chieng, N.; Rades, T.; Aaltonen, J. An overview of recent studies on the analysis of pharmaceutical polymorphs.. 2011, 55, 618644.

(13) Getsoian, A.; Lodaya, R. M.; Blackburn, A. C. One-solvent polymorph screen of carbamazepine.. 2008, 348, 39.

(14) Carter, P. W.;Ward, M. D.Directing polymorph selectivity during nucleation of anthranilic acid on molecular substrates.. 1994, 116, 769781.

(15) Scottl, C.; Leonard, J. C.; Jeanette, T. D.A metastable polymorph of metformin hydrochloride: isolation and characterization using capillary crystallization and thermal microscopy techniques.. 2004, 4, 441449..

(16) Lang, M.; Grzesiak, L.; Matzger, A. J. The use of polymer hetero nuclei for crystalline polymorph selection.. 2002, 50, 1483414835.

(17) Aaltonen, J.; Allesr, M.; Mirza, S.; Koradia, V.; Gordon, K.; Rantanen, C. J. Solid form screening a review.. 2009, 71, 2337.

(18) Bugay, D. E. Characterization of the solid-state: spectroscopic techniques.. 2001, 48, 12.

(19) Stephenson, G. A.; Forbes, R. A.; Reutzel, S. M. Characterization of the solid state: quantitative issues.. 2001, 48, 6790.

(20) Newman, A. W.; Byrn, S. R. Solid-state analysis of the active pharmaceutical ingredient in drug products.2003, 8, 898905.

(21) Shan, B.; Kakumanu, V. K.; Bansal, A. K. Analytical techniques for quantification of amorphous/crystalline phases in pharmaceutical solids.. 2006, 95, 16411665.

(22) Sitruk-Ware, R. New progestogens for contraceptive use.2006, 12, 169178.

(23) Mueck, A. O.; Sitru-ware, R. Nomegestol acetate, a novel progestogen for oral contraception.2011, 76, 531539.

(24) Abdel-Aleem, H.; Abol-Oyoun, E. M.; Shaaban, M. M. The use of Nomegestrol acetate subdermal contraceptive implant, uniplant, during lactation.1996, 54, 281286.

(25) Situk-ware, R. Pharmacological profile of progestins.. 2004, 47, 277283.

(26) Dolomanov, O. V.;Bourhis, L. J.;Gildea, R. J.; Howard, J. K.; Puschmann, H. J. A complete structure solution, refinement and analysis program.. 2009, 42, 339341.

(27) Sheldric, G. M. A short history of SHELX.. 2008, 64, 112122.

27 June 2019;

7January 2020 (CCDC 1576901 for formI and 1561687 for FormII)

①This research was supported by the National Key R&D Program of China (No. 2016YFC1000901), andthe innovation fund of National Research Institute for Family Planning (No. KYS [2019] GJM02)

. Associate professor, majoring in medical chemistry. E-mail:saintcxf2017@163.com

10.14102/j.cnki.0254-5861.2011-2514