A statistical experimental approach to cosolvent formulation of a water-insoluble drug

19-Nor-1 alpha, 25-dihydroxyvitamin D2, an analog of vitamin D2, is a nonpolar compound with limited solubility in water. An injectable solution was formulated using a cosolvent system consisting of water, ethanol, and propylene glycol. A statistical response surface approach was used to evaluate the effect of these three solvents on the solubility of the drug (25 degrees C) in the ternary cosolvent system. The data generated from five selected formulations were used to develop a multiple linear regression model that quantitatively defines the solubility of the drug as a function of the cosolvent composition. Close agreement was found between the experimental data and data calculated using the model. The capability of this model to predict drug solubility in cosolvent systems with various combinations of the three solvents was also verified.