Combined modeling and experimental approaches for the rational design of intratumoral drug delivery systems

Conventional chemotherapy can provide only limited palliative treatment against primary and secondary hepatic tumors. Surgical resection is the only curative option. However, only a small number of patients with hepatic tumors are candidates for surgery. Recent research suggests that radiofrequency (RF) thermal ablation is an effective, minimally invasive technique for treating malignant hepatic tumors. However, tumor recurrence at the margin of treatment is often observed resulting from an inability to adequately kill the hepatic parenchyma adjacent to the treated tumors. The goal of this research is to develop a supplemental local drug therapy following RF ablation to eliminate the remaining cancer cells.; Considering the doubling time of tumor cells, the duration of cell cycle and the narrow therapeutic windows of anticancer drugs, an ideal drug delivery device should be able to deliver drugs to tumors to stop further tumor growth and maintain drug concentration to ensure drug efficacy. To achieve this goal, this work made an ambitious attempt to rationally design the release kinetics of a local drug delivery device, based on mathematical modeling of the drug transport processes in tissue environment. Polymer millirods with dual release kinetics (initial burst followed by sustained release) were proposed to be the ideal design for local cancer treatment. To validate this design, dual release polymer millirods, together with two other control millirods (burst and sustained release), were designed and fabricated. In vivo drug pharmacokinetics of three types of millirods was investigated and compared in animal models. The results demonstrated the advantage of dual-release millirods in providing the optimal drug pharmacokinetics at the ablation boundary, and the value of mathematical modeling in design of dose formulations. Wound healing response of the RF ablated liver tissue was also analyzed. Results showed that improvement of the mathematical model may be necessary for the future long term drug release studies.