Dose rate and dose robustness for proton transmission FLASH-RT treatment in lung cancer

PURPOSES: To evaluate the plan quality and robustness of both dose and dose rate of proton pencil beam scanning (PBS) transmission FLASH delivery in lung cancer treatment. METHODS AND MATERIALS: An in-house FLASH planning platform was used to optimize 10 lung cancer patients previously consecutively treated with proton stereotactic body radiation therapy (SBRT) to receive 3 and 5 transmission beams (Trx-3fds and Trx-5fds, respectively) to 34 Gy in a single fraction. Perturbation scenarios (n=12) for setup and range uncertainties (5 mm and 3.5%) were introduced, and dose-volume histogram and dose-rate-volume histogram bands were generated. Conventional proton SBRT clinical plans were used as a reference. RTOG 0915 dose metrics and 40 Gy/s dose rate coverage (V(40Gy/s)) were used to assess the dose and dose rate robustness. RESULTS: Trx-5fds yields a comparable iCTV D(2%) of 105.3%, whereas Trx-3fds resulted in inferior D(2%) of 111.9% to the clinical SBRT plans with D(2%) of 105.6% (p<0.05). Both Trx-5fds and Trx-3fds plans had slightly worse dose metrics to organs at risk than SBRT plans. Trx-5fds achieved superior dosimetry robustness for iCTV, esophagus, and spinal cord doses than both Trx-3fds and conventional SBRT plans. There was no significant difference in dose rate robustness for V(40Gy/s) coverage between Trx-3fds and Trx-5fds. Dose rate distribution has similar distributions to the dose when perturbation exists. CONCLUSION: Transmission plans yield overall modestly inferior plan quality compared to the conventional proton SBRT plans but provide improved robustness and the potential for a toxicity-sparing FLASH effect. By using more beams (5- versus 3-field), both dose and dose rate robustness for transmission plans can be achieved.