Quality Risk Assessment and DoE – Practiced Validated Stability-Indicating Chromatographic Method for Quantification of Rivaroxaban in Bulk and Tablet Dosage Form

A systematic DoE and Analytical Quality by Design (AQbD) approach was utilized for the development and validation of a novel stability indicating high-performance thin-layer chromatographic (HPTLC) method for Rivaroxaban (RBN) estimation in bulk and marketed formulation. A D-optimal design was used to screen the effect of solvents, volume of solvents, time from spotting to development and time for development to scanning. ANOVA results and Pareto chart revealed that toluene, methanol, water and saturation time had an impact on retention time. The critical method and material attributes were further screened by Box-Behnken design (BBD) to achieve optimal chromatographic condition. A stress degradation study was carried out and structure of major alkaline degradant was elaborated. According to the design space, a control strategy was used with toluene: methanol: water (6:2:2) and the saturation time was 15 min. A retention factor (RF) of 0.59 +/- 0.05 was achieved for RBN using chromatographic plate precoated with silica gel at detection wavelength 282 nm with optimized conditions. The linear calibration curve was achieved in the concentration range of 200-1,200 ng/band with r2 > 0.998 sug-gesting good coordination between analyte concentration and peak areas. The quadratic model was demonstrated as the best fit model and no interaction was noted between CMAs. The optimized HPTLC method was validated critically as stated in International Conference on Harmonization (ICH) Q2 (R1) guideline and implemented successfully for stress degradation study of RBN. The developed HPTLC method obtained through AQbD application was potentially able to resolve all degradants of RBN achieved through forced degradation study. The obtained results demonstrate that a scientific AQbD approach implementation in HPTLC method development and stress degradation study dras-tically minimizes the number of trials in experiments, ultimately time and cost of analysis could be minimized.