A PHASE APPROPRIATE APPROACH TO RP-HPLC METHOD DEVELOPMENT FOR IMPURITIES ANALYSIS IN ACTIVE PHARMACEUTICAL INGREDIENTS VIA CONTINUOUS MANUFACTURING PROCESS UNDERSTANDING

Abstract

A concept of a systematic approach to the development of phase appropriate RP-HPLC methods for quantifying organic impurities in Active Pharmaceutical Ingredients (APIs) is presented. This dynamic and practical approach emphasizes the utilization of comprehensive chromatographic knowledge gained throughout the lifecycle of drug development based on continuous understanding of the API manufacturing process. At the beginning of a project, a chromatographic database can be built by scouting the relevant impurities and the API on carefully selected RP-HPLC columns, different organic modifiers and different pHs with only two gradient runs for each combination of RP-HPLC column, organic modifier, and pH. This comprehensive chromatographic database contains the experimental data and the predicted separations by chromatographic simulation software (e.g., DryLab®). Once this chromatographic database is established, an appropriate RP-HPLC method can be selected from the database based on pre-defined chromatographic method performance criteria (method attributes). More importantly, the database can be updated for any new impurities or obsolete impurities that are no longer relevant and can be taken out of the database throughout the lifecycle of the project based on continuous manufacturing process understanding, which facilitates the selection and development of appropriate RP-HPLC methods to achieve pre-defined method goals at different phases of drug development. A group of impurities identified in the manufacturing of pazopanib hydrochloride (API of Votrient, a recently approved anti-cancer medicine by the U.S. FDA) were used as an example to simulate the method development process and demonstrate the validity of this approach. Phase appropriate key chromatographic method attributes are proposed for each of the defined phases of drug development.