Independent Reproduction of the FLASH Effect on the Gastrointestinal Tract: A Multi-Institutional Comparative Study

Abstract

FLASH radiation therapy (RT) is a promising new paradigm in radiation oncology. However, a major question that remains is the robustness and reproducibility of the FLASH effect when different irradiators are used on animals or patients with different genetic backgrounds, diets, and microbiomes, all of which can influence the effects of radiation on normal tissues. To address questions of rigor and reproducibility across different centers, we analyzed independent data sets from The University of Texas MD Anderson Cancer Center and from Lausanne University (CHUV). Both centers investigated acute effects after total abdominal irradiation to C57BL/6 animals delivered by the FLASH Mobetron system. The two centers used similar beam parameters but otherwise conducted the studies independently. The FLASH-enabled animal survival and intestinal crypt regeneration after irradiation were comparable between the two centers. These findings, together with previously published data using a converted linear accelerator, show that a robust and reproducible FLASH effect can be induced as long as the same set of irradiation parameters are used.

Redefining FLASH RT: the impact of mean dose rate and dose per pulse in the gastrointestinal tract

“…The current study is the first, to our knowledge, to demonstrate that high dose per pulse (DPP) and ultra-high dose rate (UHDR) conditions independently impact the magnitude of normal tissue toxicity with FLASH irradiation. The total dose delivered to the abdomen in this study was selected based on previous findings that 11 Gy produced similar toxicity between standard FLASH conditions (1-2 Gy/pulse, 100-200 Gy/s) and CONV (low DPP and low mean dose rate); that 12 Gy was the "middle ground" for a significant observable difference; and that 14 Gy was a lethal dose in which the regenerating crypt number in the conventional dose rate (CDR) group typically fell below 5 per jejunal circumference (10). The absence of difference in GI sparing between CONV and FLASH at 11 Gy was reproduced when similar DPPs were used (10).…”

“…A custom-made collimator was produced and seated at the exit window of the Mobetron, as described elsewhere, (10,25) to allow sedation and reproducible setup placement for full and consistent total abdominal irradiation to a field size of 40 x 40 mm 2 (Fig. 1).…”

“…Dosimetric calibration was done with dose-rate-independent Gafchromic EBT3 film paired with beam current transformers, the latter to allow real-time readout of the individual radiation beam parameters for each delivery (Table S1). During the FLASH irradiations, the beam current transformers were used to monitor the dose delivery and to log all radiation beam parameters on a per-pulse basis for each individual mouse (10,(26)(27)(28)(29). All CONV irradiations were done at a set MDR of 0.3 Gy/s and a DPP of 10 mGy.…”

“…Although the FLASH effect has been reported in different organ systems (e.g., brain, lung, gastrointestinal [GI] tract, skin) and animal species (e.g. rodents, human, swine, feline) (3,(6)(7)(8)(9)(10)(11)(12)(13)(14), some studies have reported either no effect or increased detrimental effects on normal tissue from UHDR relative to CDR irradiation (15)(16)(17)(18)(19). Furthermore, even in reports showing increased normal tissue sparing after FLASH relative to CONV, the magnitude of the relative amount of sparing was inconsistent between studies and between institutions (4,6,17).…”

“…Rather, other aspects of the radiation beam including radiation source (x-rays, protons, electrons, heavy ions), dose per pulse (DPP), pulse repetition frequency (PRF), pulse width (PW), total delivery time, irradiation volume, and intrapulse dose rate (IDR; the dose rate within a pulse) may all contribute to the FLASH effect (4,20,21). Indeed, when matching beam parameters were used, the FLASH effect has been shown to be reproducible between institutions using the same biological assay (10,11).…”