Rapid throughput solubility screening assay development and its applications in preformulation

Drug solubility is an important physicochemical property for orally administered drugs because it can significantly influence drug dissolution and absorption profiles. Investigative drugs need to have solubility profiles established as early as possible because poor solubility may result in incomplete dissolution and low bioavailability. The limited quantities of drug and time in early product development, often delay such evaluation, resulting in unexpected challenges in later stage. This dissertation describes a new solubility screening method based on the 96-well plate format that minimizes drug material requirements while returning accurate solubility information timely. The method was first applied to estimate the salt solubility parameter, KSP. High concentration drugs were initially dissolved in DMSO, and robotically distributed into 96-well plates. DMSO was evaporated, and drugs were equilibrated in various acids at different concentrations to yield final total drug concentrations of about 2.5 mM. All wells were filtered, and supernatants from wells with precipitates were subjected to rapid gradient HPLC analysis. An iterative procedure was employed to calculate apparent KSP values based on mass and charge balance equations. A correlation coefficient of greater than 0.975 for 8 drugs totaling 16 salts was reported. Reproducibility was better than 10%. The dynamic range of the assay could be expanded by adjusting drug and counter-ion concentrations. The method was then used to evaluate the solubilities of Biopharmaceutic Classification Scheme (BCS) class II drugs in animal and simulated human GI fluids. The goal was to establish correlates between in vitro and in vivo drug solubility in the GI tract. Four BCS class II drugs with solubilities ranging from a few ug/mL to over 100 ug/mL were evaluated in simulated GI fluid in both fasted and fed conditions. Method reproducibility was within 15% relative standard deviation. Solubility values from the screening method correlated well with literature values. Effects of commonly used pharmaceutical solubilizers (PEG 400, Tween 80, hydroxypropyl-?cyclodextrin) on danazol, mefenamic acid, phenytoin and griseofulvin in simulated GI fluids containing bile salts (which are also solubilizers) were also assessed. At typical concentrations (>5%), PEG 400's solubilization predominated over bile salt. Likewise, Tween 80 (surfactant) was more effective than bile salts. Addition of hydroxypropyl-?-cyclodextrin actually reduced solubility, this was attributed to competitive binding from bile salt monomers and drug molecules to hydroxypropyl-?-cyclodextrin. We believe that this 96-well format-screening assay is a powerful tool for pre-formulation researchers.