Comparison of a rational vs. high throughput approach for rapid salt screening and selection.

In recent years, high throughput (HT) screening has become the most widely used approach for early phase salt screening and selection in a drug discovery/development setting. The purpose of this study was to compare a rational approach for salt screening and selection to those results previously generated using a HT approach. The rational approach involved a much smaller number of initial trials (one salt synthesis attempt per counterion) that were selected based on a few strategic solubility determinations of the free form combined with a theoretical analysis of the ideal solvent solubility conditions for salt formation. Salt screening results for sertraline, tamoxifen, and trazodone using the rational approach were compared to those previously generated by HT screening. The rational approach produced similar results to HT screening, including identification of the commercially chosen salt forms, but with a fraction of the crystallization attempts. Moreover, the rational approach provided enough solid from the very initial crystallization of a salt for more thorough and reliable solid-state characterization and thus rapid decision-making. The crystallization techniques used in the rational approach mimic larger-scale process crystallization, allowing smoother technical transfer of the selected salt to the process chemist.

Automated approach to couple solubility with final pH and crystallinity for pharmaceutical discovery compounds.

The design and validation of a novel high-throughput system for thermodynamic solubility determination requiring only 5 mg of sample is described. The system uses a sintered nickel filter assembly to recover excess solids from saturated solutions for rapid crystallinity assessment via powder X-ray diffraction (PXRD). Moreover, the system measures the pH of filtrates to provide a final pH value with the solubility measurement. The limit of detection for the UV-vis plate reader used on this system is approximately 0.001 mg/ml, while the practical upper limit is approximately 3 mg/mL. The solubility measurements of 60 proprietary Pfizer compounds were used to validate the nickel filter assembly against a more conventional polyvinylidenedifluoride (PVDF) filter. Additionally, a comparison was made between a subset of 10 compounds run on the automated system and a more traditional shake-flask method employing HPLC analysis. In both cases, a favorable comparison was obtained.

Identifying the stable polymorph early in the drug discovery-development process.

The thermodynamically most stable polymorph under ambient conditions is almost without exception the most desirable crystalline form for development by a pharmaceutical company. It is, therefore, beneficial to discover and to characterize this polymorph at the earliest possible stage of development. A screen for discovering the stable polymorph of a pharmaceutical compound early in the drug discovery-development process is developed and described. In this screen, a small amount of compound is suspended in a diverse group of solvents for two weeks in an effort to crystallize the most stable polymorph. The solubility of the compound in each solvent utilized in the stable polymorph screen is also simultaneously determined using a simple gravimetric method. Ritonavir and an early development candidate (Pfizer compound A) are used as model compounds to demonstrate the utility of the screen for finding the stable polymorph early in the drug discovery-development process.

Rapid, small-scale determination of organic solvent solubility using a thermogravimetric analyzer.

A rapid gravimetric method for determining drug candidate solubility in organic solvents has been developed. The scale, speed, precision, and accuracy of the method make it ideal for solubility screening of pharmaceutical compounds during early development. The method utilizes a thermogravimetric analyzer to automate drying and weighing. Results for model compounds compare favorably with literature values.

Semi-empirical relationships between effective mobility, charge, and molecular weight of pharmaceuticals by pressure-assisted capillary electrophoresis: applications in drug discovery.

Relationships between effective mobility (m(eff)), calculated charge (Z(c)), and molecular weight (MW) are semi-empirically derived for pharmaceuticals using pressure-assisted capillary electrophoresis (PACE). We determined the m(eff) at 12 different pH points (2.0-11.4) of 66 pharmaceutical-like compounds ranging in MW from 79 to 825 g/mol. Plots of the observed m(eff) values versus Z(c)/MW(x ) (where x is a fractional coefficient) gave linear relationships. For anions, it was found that the best correlation (R(2) = 0.9666) exists when the fractional coefficient is equal to 0.4920, resulting in the equation m(eff) = 0.1853 (Z(c)/MW (0.4920)). For cations, the best linear relationship (R(2) = 0.9861) gave the equation m(eff) = 0.3888 (Z(c)/MW (0.6330)). The m(eff), Z(c)/MW(x) relationships were then applied to: (i) developing a technique for selecting an appropriate pH to achieve optimal separation of pharmaceuticals and (ii) determining the maximum charge of a molecule in the pH range of determination of negative log of the dissociation constants (pK(a)) by PACE, thus enabling the correct choice of model equation to be automated without structure analysis.