2016 年 42 巻 5 号 p. 305-316
So-called pharmacometrics were used for information creation and decision-making in the drug development and drug approval process. However, for individualized drug therapy in clinical settings, pharmacometrics were not fully utilized. Therefore, our research group has defined, as “clinical pharmacometrics”, clinical contribution to decision-making in individualized drug therapy, especially with dosage optimization support and adverse effects management, by pharmacist-led quantitative analysis, evaluation and prediction using mathematical statistical methods. We show here some examples of practicing clinical pharmacometrics. In the case of supporting meropenem dosage optimization, we developed a population pharmacokinetic model and then performed random simulation to predict the efficacy of dosages (the drug exposure time above the minimum inhibitory concentration) by employing a probabilistic approach considering the variation of the patient and bacteria. The optimized meropenem dosage achieved favorable outcome. In the case of managing adverse effects of voriconazole, we developed a population pharmacogenomic-pharmacokinetic model and a logistic pharmacodynamic model, and then performed simulation to predict the safety of dosages (the hepatotoxic effects). These results optimized the voriconazole dosage according to the body weight and CYP2C19 genotype of the patient, considering the hepatotoxicity probability. As in these examples, for optimal drug therapy in individual patients, pharmacists need to practice clinical pharmacometrics by repeating the cycle of processes: clinical data measurement in the medical setting, acquisition of quantitative pharmaceutical knowledge, and provision of therapeutic benefits to patients.