Multi-Modality Optimal Radiation Treatments

University Of Washington, Seattle WA

Investigators

Abstract

Cancer is one of the leading causes of death in the United States, affecting many families. Various radiation treatment modalities, such as photons, protons, and neutrons, are currently used in an effort to eradicate cancer cells while preserving the functionality of healthy tissue. Each modality has its own physical and radiobiological strengths as well as weaknesses. Using more than one modality in treating cancer is motivated by the idea of exploiting the advantages of different radiation types to overcome the weaknesses that any single modality suffers, and eventually improving the efficacy of cancer therapy. Current efforts on finding an optimal, multi-modality treatment paradigm for a particular cancer type are mostly empirical, and heuristic searches result in a long and costly period of trial-and-error. This award supports research to develop an optimization tool that systematically searches for optimal, multi-modality radiotherapy schemes, which can then be tailored to particular cancer types as needed. The optimal treatment protocols will provide guidance in designing trials of new effective and efficient radiation treatments. If successful, this will lead to longer survival and increase quality of life for cancer patients. The multi-disciplinary project team will include students and trainees in different areas including industrial engineering, applied mathematics, and medical physics, and positively impact education in science and engineering. This research utilizes the stochastic control formalism, where the system consists of the tumor and normal tissue subject to the modality-specific uncertainty and tumor-specific radiobiological parameters. The time-dependent beam intensities of multiple radiation types are used as controls, and they determine the spatiotemporal dose distribution during a treatment course. Optimal treatment consists of the maximum biologically-effective-dose or equivalent uniform dose feasible for a given patient's physical and radiobiological condition, which is positively correlated with clinical outcomes. The research team will construct the general optimization framework, apply to specific cancer types using phantom studies, perform numerical simulations to quantify the potential benefits of an optimal solution over the conventional treatment schemes that are currently used in practice, and come up with treatment paradigms that could be tested in future clinical trials.