A novel treatment planning method via scissor beams for uniform‐target‐dose proton GRID with peak‐valley‐dose‐ratio optimization

Abstract Background Proton spatially fractionated RT (SFRT) can potentially synergize the unique advantages of using proton Bragg peak and SFRT peak‐valley dose ratio (PVDR) to reduce the radiation‐induced damage for normal tissues. Uniform‐target‐dose (UTD) proton GRID is a proton SFRT modality that can be clinically desirable and conveniently adopted since its UTD resembles target dose distribution in conventional proton RT (CONV). However, UTD proton GRID is not used clinically, which is likely due to the lack of an effective treatment planning method. Purpose This work will develop a novel treatment planning method using scissor beams (SB) for UTD proton GRID, with the joint optimization of PVDR and dose objectives. Methods The SB method for spatial dose modulation in normal tissues with UTD has two steps: (1) a primary beam (PB) is halved with interleaved beamlets, to generate spatial dose modulation in normal tissues; (2) a complementary beam (CB) is added to fill in previously valley‐dose positions in the target to generate UTD, while the CB is angled slightly from the PB, to maintain spatial dose modulation in normal tissues. A treatment planning method with PVDR optimization via the joint total variation and L1 (TVL1) regularization is developed to jointly optimize PVDR and dose objectives. The plan optimization solution is obtained using an iterative convex relaxation algorithm. Results The new methods SB and SB‐TVL1 were validated in comparison with CONV. Compared to CONV of relatively homogeneous dose distribution, SB had modulated spatial dose pattern in normal tissues with UTD and comparable plan quality. Compared to SB, SB‐TVL1 further maximized PVDR, with comparable dose‐volume parameters. Conclusions A novel SB method is proposed that can generate modulated spatial dose pattern in normal tissues to achieve UTD proton GRID. A treatment planning method with PVDR optimization capability via TVL1 regularization is developed that can jointly optimize PVDR and dose objectives for proton GRID.