Isolation, identification, structural elucidation, and toxicity prediction using (Q)-SAR models of two degradants: AQbD-driven LC method to determine the Roxadustat impurities

Roxadustat (RDT) is the first orally administrated HIF-prolyl hydroxylase inhibitor drug used to treat anemia caused by chronic kidney disease. Two unknown degradants were detected in photolytic and oxidative stress conditions during the forced degradation study of RDT and isolated them using the preparative HPLC. The structural characterization of these impurities was confirmed using ESI-LC-MS, NMR, and FT-IR spectroscopic analysis. To evaluate the toxicity of degradants, (Q)-SAR models such as Derek and Sarah model were utilized. The current study aims to develop a stability-indicating related substances quantification method in RDT along with degradant impurities by implementing AQbD principles. The Box-Behnken Design (BBD) was utilized to optimize the final analytical method conditions. p-values for the model and lack of fit were <0.05 and >0.05, respectively. The optimized CMPs are 30:70 (v/v) ACN: Methanol in mobile phase-B, 50:50 mobile phase-A & B in the initial gradient program, 0.95 mL/min of flow rate, and 40 degrees C as column oven temperature. Agilent Zorbax XDB-C8, (250 x 4.6) mm, 5 & mu;m analytical column was used to separate the desired components from the sample matrix peaks and themselves. The optimized method was validated in compliance with regulatory requirements. The recoveries for all the impurities ranged from 96.8% to 99.7%, with an%RSD <2.0, and the Pearson corre-lation coefficient(r) was> 0.998. The current method can be used in quality-control laboratories to quantify RDT impurities without any developmental trials.