Reoptimization of Multi-Criteria Proton Therapy Plans Including Linear Energy Transfer and Relative Biological Effectiveness

Proton therapy is a cutting-edge cancer treatment that delivers targeted radiation to tumors with little dose to surrounding healthy tissues. Multi-criteria optimization (MCO) can be employed to balance target coverage with healthy tissue sparing. Despite the potential of proton therapy, the biological effectiveness of proton radiation is still widely overlooked during treatment planning.



In this thesis, mathematical methods to incorporate the biological effect of protons into an MCO planning framework are developed. The resulting framework allows the prioritization of clinically established dose objectives and reoptimization with respect to the more experimental objectives related to the biological effect in a single step. This is achieved by transforming the resulting bilevel MCO problem into a single level problem through adjustments in the domination cone. The developed algorithm is demonstrated in two use cases: one incorporating linear energy transfer (LET) as a physical surrogate for the biological effect, and one to obtain robustness against variations in the relative biological effectiveness parameter.