Room: Davidson Ballroom A
Purpose: To investigate the potential for dose-averaged linear energy transfer (LET) reduction of novel planning approaches for brain cases with suspected treatment related toxicity following proton therapy.
Methods: The planned clinical single-field uniform dose distributions, with fractionation doses of 1.8 Gy(RBE) to the target (RBE=1.1), were carefully dose mimicked using an independent treatment planning system. Monte Carlo computed physical dose and LET were used to calculate the relative biological effectiveness-weighted dose (Dá´¿á´®á´±) using two variable RBE-models, with assumed Î±/Î² ratios of 3 or 10 Gy for the targets and 2 Gy for the normal tissues. Resulting dose distributions and normal tissue complication probabilities (NTCPs) were evaluated for the critical structures. Subsequently, three intensity-modulated proton plans were generated with the objective to have similar or better physical dose distribution as the clinical plans, while lowering RBE and LET in critical structures using: (1) original beam arrangements with track-end objectives (penalized protons stopping in critical structures), (2) alternative beam arrangements, (3) alternative beam arrangements and track-end objectives.
Results: The mimicked plans were equivalent to the clinical plans, fulfilling the tolerance doses (RBE=1.1). Applying variable RBE-models resulted in a predicted increase of 7-12 Gy(RBE) in near-max doses for the structures with observed toxicity. Tolerance doses were exceeded and NTCPs increased from <1% (RBE=1.1) to 2-15% (variable RBE-models). The alternative plans were similar or better than the clinical plans in terms of physical dose, while generally allowing for substantial Dá´¿á´®á´±, LET, and NTCP reductions in the critical structures assuming the variable RBE-models.
Conclusion: Although direct causality cannot be established, the analysis indicates that enhanced RBE due to high LET could be a potential cause of the observed toxicities. Future planning strategies should include carefully prepared beam arrangements and preferably optimization objectives allowing for LET reduction in critical structures, without compromising the target dose.
Funding Support, Disclosures, and Conflict of Interest: Jakob Oeden is part-time employed as an industrial PhD student at RaySearch Laboratories. Erik Traneus is full-time employed at RaySearch Laboratories.