Characterization, selection, and development of an orally dosed drug polymorph from an enantiotropically related system.

Solid-state characterization methods are used to study a dimorphic pharmaceutical compound and select a form for development. Polymorph screening found that [4-(4-chloro-3-fluorophenyl)-2-[4-(methyloxy)phenyl]-1,3-thiazol-5-yl] acetic acid can crystallize into two non-solvated polymorphs designated Forms 1 and 2. Physical methods including vibrational spectroscopy, X-ray powder diffraction, solid-state NMR (SSNMR), thermal analysis, and gravimetric water vapor sorption are used to fully characterize the two polymorphs. Temperature-dependent competitive ripening experiments and solubility measurements indicated that the polymorphs in this system exhibit enantiotropy with a thermodynamic transition temperature of 35+/-3 degrees C. This complicates the selection of a polymorph to progress in drug development. Both forms had undesirable qualities; however, a particular drawback of Form 1 was found in its tendency to convert to Form 2 upon milling. Combining this effect and the desired formulation approach with physical property results led to a rationale for the choice of Form 2 for further development. Because this form is thermodynamically metastable at room temperature, analytical approaches were developed to ensure its exclusive presence, including a quantitative infrared spectroscopic method for drug substance and (13)C and (19)F solid-state NMR limit tests for the undesired form in drug product at drug loads of 8.3% (w/w).

Enantiotropically-related polymorphs of {4-(4-chloro-3-fluorophenyl)-2-[4-(methyloxy)phenyl]-1,3-thiazol-5-yl} acetic acid: crystal structures and multinuclear solid-state NMR.

Single crystal X-ray diffraction (SCXRD), powder X-ray diffraction (PXRD), and solid-state NMR (SSNMR) techniques are used to analyze the structures of two nonsolvated polymorphs of {4-(4-chloro-3-fluorophenyl)-2-[4-(methyloxy)phenyl]-1,3-thiazol-5-yl} acetic acid. These polymorphs are enantiotropically-related with a thermodynamic transition temperature of 35 +/- 3 degrees C. The crystal structure of Form 1, which is thermodynamically more stable at lower temperatures, was determined by SCXRD. The crystal structure of Form 2 was determined using PXRD structure solution methods that were assisted using two types of SSNMR experiments, dipolar connectivity experiments and chemical shift measurements. These experiments determined certain aspects of local conformation and intermolecular packing in Form 2 in comparison to Form 1, and provided qualitative knowledge that assisted in obtaining the best possible powder structure solution from the X-ray data. NMR chemical shifts for 1H, 13C, 15N, and 19F nuclei in Forms 1 and 2 are sensitive to hydrogen-bonding behavior, molecular conformation, and aromatic pi-stacking interactions. Density functional theory (DFT) geometry optimizations were used in tandem with Rietveld refinement and NMR chemical shielding calculations to improve and verify the Form 2 structure. The energy balance of the system and other properties relevant to drug development are predicted and discussed.

A study of variable hydration states in topotecan hydrochloride.

Topotecan hydrochloride, a pharmaceutical compound developed as a treatment for cancer, exhibits variable hydration states in a crystalline solid form chosen for manufacturing. This variability requires additional controls for successful development, and presents a characterization and detection challenge for analytical methods. In this study, overall water content was determined by Karl Fischer titration and thermogravimetric analysis (TGA) on topotecan HCl equilibrated at different relative humidity levels. These results, when combined with information obtained from dynamic water vapor sorption and differential scanning calorimetry (DSC), indicate that this form of topotecan HCl contains 3 mol of water integral to the crystalline structure and up to two additional moles of water depending on the relative humidity. Powder X-ray diffraction experiments did not detect significant differences in topotecan HCl samples equilibrated at trihydrate and pentahydrate states, and showed that the crystal lattice dimensions are not affected unless the form is dried below the trihydrate state. This behavior is typical of crystal structures with channels that can accommodate additional loosely bound water. To study the role of the loosely bound water in the crystal structure in more detail, solid-state (13)C and (15)N nuclear magnetic resonance (NMR) were used to examine the differences between the hydration states. Both the trihydrate and pentahydrate states yielded similar solid-state NMR spectra, consistent with the lack of change in the crystal lattice. However, minor but readily detectable differences in the (13)C spectra are observed with changes in water content. Interpretation of this data suggests that the loosely bound channel water is hydrogen-bonding to specific portions of the topotecan parent molecule. Topotecan HCl trihydrate was hydrated with D(2)O vapor to confirm the nature and location of the channel water using (13)C and (2)H solid-state NMR. Despite the detectable association of the channel water with hydrogen bonding sites on the topotecan molecule, (2)H quadrupolar echo experiments indicate that the channel water is highly mobile at room temperature and at -60 degrees C.

Structural analysis of polymorphism and solvation in tranilast.

Five polymorphic forms of tranilast were characterized by thermal, diffractometric, and spectroscopic techniques. The crystal structures of the most stable anhydrous form (Form I), a chloroform solvate, and a dichloromethane solvate were determined from single-crystal X-ray analysis. Two additional anhydrous forms of tranilast (Forms II and III) were also studied, but were not amenable to SCXRD. All five forms were also analyzed using solid-state nuclear magnetic resonance, Fourier transform infrared, and Fourier transform-Raman spectroscopy, and thermal methods. From the trends observed in the crystal structures and the spectral data, some conclusions can be made about hydrogen bonding, molecular conformation, and crystal packing differences in the polymorphs and solvates. Form II was found to be a spectroscopically distinctive polymorph that is probably missing an important intramolecular hydrogen bond coupled with a conformational change. In contrast, Form III was found to be more similar to the crystallographically characterized forms, and is more likely a packing and hydrogen-bonding polymorph with a weakened intermolecular hydrogen-bonding interaction relative to the other forms. From a pharmaceutical development perspective, it is shown that although the anhydrous forms of tranilast have similar thermal properties, they can be reliably distinguished by spectroscopic methods.