Probing the therapeutic window of proton minibeam radiotherapy using dose-response curves in a mouse model.

Purpose: Proton minibeam radiotherapy (pMBRT) has been observed in preclinical studies to spare normal tissues through its spatially fractionated dose profile. Translating pMBRT to clinical application requires quantification of its therapeutic gain, compared to conventional proton therapy. We compare pMBRT to conventional proton therapy in vivo, focusing on reducing damage to non-target tissues while ensuring the same uniform target dose to achieve equal tumor control.

Methods And Materials: We used a multislit collimator in an established mouse irradiation setup to deliver a uniform dose to the target while maintaining a high dose contrast in the entrance region. The right hind legs of 75 female C3H/HeNRj mice were irradiated with the highest dose contrast. Acute skin toxicity was recorded up to 25 days post-irradiation, using a seven-level scoring scheme (0.5 to 3.5) to quantify skin reaction following a well-established protocol. For tumor control comparison, we used CDF1 female mice with a C3H mouse mammary carcinoma subcutaneously implanted in the foot. Dose-response curves of the level of acute skin toxicity and tumor control were generated as a function of the planning target volume (PTV) dose for both conventional and pMBRT setups, allowing for direct comparison.

Results: pMBRT demonstrated significantly improved normal tissue sparing ability compared to conventional irradiation for same doses in the target. No incidence of higher levels (Score 2.5, 3.0 and 3.5) of toxicity was observed in the pMBRT group, in contrast to the higher toxicity often seen in mice treated with conventional modality at the same PTV dose. At the maximum deliverable dose, the incidence of skin toxicity was still too low to complete the dose-response curves for pMBRT. The estimated grid factor of < 0.65 (Score 1.5) and < 0.7 (Score 2) suggests a substantial enhanced tissue sparing potential with pMBRT. Both modalities show similar tumor control, with TCD50 of 46.9 Gy for conventional therapy and 45 Gy for pMBRT.

Conclusion: We present a comparison method to quantify the efficacy of pMBRT. The observed reduction in acute normal tissue toxicity for pMBRT, compared to conventional proton therapy for at the same PTV dose and maintaining similar tumor control, suggests that pMBRT may offer a substantial therapeutic gain.