Tsukasa Sasaki, Research Specialist, Research Group 1, Analysis and Evaluation Research Institute, Pharmaceutical Technology Division, Daiichi Sankyo Co., Ltd.
Reverse-phase HPLC is an important analytical measurement method for estimating the efficacy and safety of small molecule drugs. In general, the main components of drugs are mixtures containing a certain amount of impurities, but by using reverse-phase HPLC, it is possible to estimate the amount of each component. Therefore, the search and optimization of HPLC conditions that separate each component with good reproducibility is a very important process in quality evaluation research. The main factors that affect separation include the pH of the buffer used in the eluent, the type of organic solvent, the analytical column temperature, and the flow rate of the eluent. When the target is an ionic compound, the pH and the type of organic solvent in particular have a large impact on the success or failure of separation. However, in reverse-phase HPLC, the variation in the time it takes for each component to pass through the analytical column (retention time) is not linear with factors such as pH and organic solvent, making it very difficult to predict and evaluate the success or failure of separation. Therefore, the speaker attempted an approach to estimate the success or failure of separation within a certain range of eluent conditions from retention time predictions using neural network modeling.
