Two chemometric assisted spectrophotometric models were applied for the quantitative analysis of velpatasvir and sofosbuvir in their newly FDA approved pharmaceutical dosage form. The UV absorption spectra of velpatasvir and sofosbuvir showed certain degree of overlap which exhibited degree of difficulty for the choice of certain method provides simultaneous quantitative analysis of the cited drugs. Artificial neural networks and genetic algorithm artificial neural networks were the suitable model for the quantitative analysis of velpatasvir and sofosbuvir in their binary mixture. Experimental design and building the calibration set for the binary mixture were achieved to implement the described models. The proposed models were optimized with the aid of five-levels, two factors experimental design. Spectral region of 380-400 nm was rejected which resulted in 181 variables. GA reduced absorbance matrix to 72 and 36 variables for velpatasvir and sofosbuvir respectively. The models succeeded to estimate the studied drugs with acceptable values of root mean square error of calibration and root mean square error of prediction. The developed models were successfully applied to the quantitative analysis of the two drugs in Epclusa® tablets. The results were statistically compared with another published quantitative analytical method with no significant difference by applying Student t-test and variance ratio F-test.
Keywords: ANN; GAANN; Sofosbuvir; Spectrophotometry; Velpatasvir.
Copyright © 2021. Published by Elsevier B.V.