Low Frequency Raman Spectroscopy for Evaluation of Crystallinity from Ritonavir Tablets

Description:
Polymorphism is the phenomenon for a compound to exist in more than one crystal form with different intermolecular arrangements. Polymorphs have distinct physicochemical properties such as solubility, dissolution, chemical stability, melting point and bioavailability, therefore crystalline form or structure of solids should be carefully evaluated to ensure safety and efficacy. A well-known instance of polymorphism after marketing of a drug is the case of ritonavir. Ritonavir is a protease inhibitor that is used for the treatment of Acquired Immunodeficiency Syndrome (AIDS) and coronavirus disease (COVID-19) under U.S. Food and Drug Administration (FDA) emergency use authorization as Paxlovid co-packaged with nirmatrelvir. Prior to its current formulation as an amorphous drug dispersion formed through hot melt extrusion process, ritonavir was formulated as a semisolid capsule which was eventually recalled due to issues with polymorphism. Here we report low frequency Raman (LF Raman) spectroscopy as a rapid and powerful method to detect and quantify two crystal forms of ritonavir in commercial ritonavir 100 mg tablets under stress testing. Three lots of reference listed drug (RLD) and three different generic formulations of ritonavir tablets were used in a four-week study under open dish accelerated stress testing conditions (40 ⁰C, 75% RH). Form I and Form II were identified by comparison with reference standards which demonstrated unique Raman features at 49 cm-1 and 30 cm-1, respectively. Based on LF Raman data, Form I and Form II ritonavir crystals were identified in generic ritonavir tablets starting within one week under accelerated stress conditions. In addition, a quantitative XRPD method was developed and validated to compare and confirm the LF Raman analysis. LF Raman based predictions for percentage crystallinity of Form I and Form II ritonavir were in agreement with the validated XRPD results, as root-mean-squared-error were found in the range of 7-10%.

Speaker: Sevim Hatipoglu - US Food and Drug Administration