HSPiP, Computational Modeling, and QbD-Assisted Optimized Method Validation of 5-Fluorouracil for Transdermal Products.

Abstract

This study addressed the simplest and most efficient HPLC (high-performance liquid chromatography) method for the estimation of 5-fluorouracil (5-FU) from rat blood plasma by implementing the Hansen solubility parameters (HSP), computation prediction program, and QbD (quality by design) tool. The mobile phase selection was based on the HSP predictions and experimental data. The Taguchi model identified seven variables (preoptimization) to screen two factors (mobile phase ratio as A and column temperature as B) at three levels as input parameters in "CCD (central composite design)" optimization (retention time as Y1 and peak area as Y2). The stability study (freeze-thaw cycle and short- and long-term stability) was conducted in the rat plasma. Results showed that HSPiP-based HSP values and computational model-based predictions were well simulated with the experimental solubility data. Acetonitrile (ACN) was relatively suitable over methanol as evidenced by the experimental solubility value, HSP predicted parameters (δh of 5-FU - δh of ACN = 8.3-8.3 = 0 as high interactive solvent whereas δh of 5-FU - δh of methanol = 8.3-21.7 = -13.4), and instrumental conditions. CCD-based dependent variables (Y1 and Y2) exhibited the best fit of the model as evidenced by a high value of combined desirability (0.978). The most robust method was adopted at A = 96:4 and B = 40 °C to get earlier Y1 and high Y2 as evidenced by high desirability (D) = 0.978 (quadratic model with p < 0.0023). The estimated values of LLOD and LLOQ were found to be 0.11 and 0.36 μg/mL, respectively with an accuracy range of 94.4-98.7%. Thus, the adopted method was the most robust, reliable, and reproducible methodology for pharmacokinetic parameters after the transdermal application of formulations in the rat.