OPTIMIZED CANCER CHEMOTHERAPY VIA PHARMACODYNAMIC MODELS

点击翻译按钮获取中文摘要

DESCRIPTION: (Applicant's Abstract) Determination of anticancer drug doses is normally based on the maximum tolerated dose (MTD) obtained in Phase I trials. As the drug is subsequently used in Phase II and Phase III trials, all patients receive the same dose, normalized to body weight or body surface area, dictated by the MTD, thus, preventing the use of a dosing regimen tailored to an individual. In this scenario, many patients will be either underdosed or overdosed. This application is aimed at rectifying these deficiencies through the use of population pharmacokinetic (PK) and pharmacodynamic (PD) models. Such models can characterize a drug's PK and PD properties in an individual based upon these properties in the population and the individual's covariates, or patient specific factors (for example, age, sex, renal function) that can influence the drug's PK and PD. The population-based models can provide a quantitative platform to design individual patient dosing regimens to achieve a desired PK or PD endpoint. One of the novel aspects of this application is to model the time-dependent nature of PD responses that can be subsequently applied to the design of optimized drug dosing regimens. Different PD modeling techniques will be used to characterize drug-induced myelosuppression (MYLS), a key dose-limiting toxicity for many anticancer drugs. Two large clinical databases, (1, an existing Johns Hopkins Oncology Center database, and 2, an ongoing Fox Chase Cancer Center database) will provide an extensive set of PK and PD (i.e. MYLS) data including patient covariates to evaluate PD models for topotecan's (TPT's) MYLS, both as a single agent and in combination. Each type of PD model will account for the time- dissociation between measured TPT's plasma concentrations and MYLS, as well as intrasubject and intersubject variability. The predictive performance of each PD modeling strategy will be compared from numerous index datasets constructed from the two large databases, and further undergo rigorous bootstrap validation analyses. These procedures will indicate optimal modeling techniques for MYLS, and quantitative methods to design individual patient dosing regimens that may improve drug therapy.