UC Irvine

PURPOSE: The capability of ultrahigh dose rate FLASH radiation therapy to generate the FLASH effect has opened the possibility to enhance the therapeutic index of radiation therapy. The contribution of the immune response has frequently been hypothesized to account for a certain fraction of the antitumor efficacy and tumor kill of FLASH but has yet to be rigorously evaluated. METHODS AND MATERIALS: To investigate the immune response as a potentially important mechanism of the antitumor effect of FLASH, various murine tumor models were grafted either subcutaneously or orthotopically into immunocompetent mice or in moderately and severely immunocompromised mice. Mice were locally irradiated with single dose (20 Gy) or hypofractionated regimens (3 × 8 or 2 × 6 Gy) using FLASH (≥2000 Gy/s) and conventional (CONV) dose rates (0.1 Gy/s), with/without anti-CTLA-4. Tumor growth was monitored over time and immune profiling performed. RESULTS: FLASH and CONV 20 Gy were isoeffective in delaying tumor growth in immunocompetent and moderately immunodeficient hosts and increased tumor doubling time to >14 days versus >7 days in control animals. Similar observations were obtained with a hypofractionated scheme, regardless of the microenvironment (subcutaneous flank vs ortho lungs). Interestingly, in profoundly immunocompromised mice, 20 Gy FLASH retained antitumor activity and significantly increased tumor doubling time to >14 days versus >8 days in control animals, suggesting a possible antitumor mechanism independent of the immune response. Analysis of the tumor microenvironment showed similar immune profiles after both irradiation modalities with significant decrease of lymphoid cells by ∼40% and a corresponding increase of myeloid cells. In addition, FLASH and CONV did not increase transforming growth factor-β1 levels in tumors compared with unirradiated control animals. Furthermore, when a complete and long-lasting antitumor response was obtained (>140 days), both modalities of irradiation were able to generate a long-term immunologic memory response. CONCLUSIONS: The present results clearly document that the tumor responses across multiple immunocompetent and immunodeficient mouse models are largely dose rate independent and simultaneously contradict a major role of the immune response in the antitumor efficacy of FLASH. Therefore, our study indicates that FLASH is as potent as CONV in modulating antitumor immune response and can be used as an immunomodulatory agent.