A modeling tool for the personalization of pharmacokinetic predictions

A method to apply pharmacokinetic models to assist physicians in therapeutic drug monitoring is proposed. The practice of therapeutic drug monitoring is required for drugs characterized by a narrow therapeutic index, which consequently present toxicity concerns. The proposed method employs a physiologically based pharmacokinetic (PBPK) model to determine an initial assessment of the pharmacokinetics (PK) of a specific patient. To further increase the precision of this prediction, the method uses two experimental datasets: (i) the PK data from a group of reference subjects, and (ii) limited drug blood concentration measures of the specific patient under study. By combining the available information, it is possible to assess the precision of the initial model prediction and determine a correction factor to improve it. The resulting patient-specific PBPK model produces encouraging results as there is a concrete reduction in prediction errors of the individualized PK with respect to experimental data.