Background and Aims

Normal tissue-sparing property of FLASH-RT has been shown in various studies, including a dose-escalating trial with single-dose FLASH-RT (25-41Gy) in cat-patients. Results prompted us to design this prospective, randomized clinical phase-III-trial in cat-patients with spontaneous tumors, to compare single-high-dose FLASH-RT to a standard of care (SOC); with tumour control-rate at 1 year as primary endpoint (hypothesis= 95% with FLASH-RT versus 71% for SOC, alpha=0.05 and beta=0.2, 29 cats needed).

Methods

Ethic’s approval was obtained (ZH204/18) and cats with T1-T2 N0 carcinomas of the nasal planum were randomly assigned to 2 arms of electron radiation. Arm 1 used 10x4.8Gy (90%IDL), delivered in one week with a 6MeV linear accelerator, dose rate of 600MU/min. Arm 2 used 1x30Gy (89%IDL) with eRT6/Oriatron, delivered in 20ms using 3 pulses, instantaneous dose rate of 6.3x10⁶Gy/s (mean dose rate 1700Gy/s).

Results

While acute side effects were mild to moderate and similar in both arms, the trial was prematurely stopped due an excess of maxillary bone necrosis which occurred 9-12 months after RT in 3/7 cats treated with FLASH-RT (43%), as compared to 0/9 cats in SOC. Regarding the primary endpoint, all cats were free of tumor progression at 1 year in both arms, but one tumor progression occurred later in FLASH-RT arm. Overall survival rates were similar in both arms, 690 days for SOC and 680 days for FLASH.

Conclusions

When compared to SOC, 1x30Gy-FLASH was beyond the maximal tolerated dose, causing severe late toxicity without better tumor control.

Acknowledgments: Krebsliga, KFS-4438-02-2018: Phase III clinical trial on cat patients

Background and Aims

Ultra-high dose rate (UHDR) irradiation produces the FLASH effect (anti-tumor effect without normal tissue toxicity) at average dose rates above 100 Gy/s. Two physico-chemical scenarios were proposed as a mechanistic basis for the FLASH effect following water radiolysis: 1. Altered radical-radical reactions and 2. Depletion of O₂ by free radicals. To investigate these questions, we determined primary radiolytic yields (G°-values) of hydrogen peroxide and hydroxyl radicals. G(H₂O₂) was also determined in low O₂ condition (1%), intermediate levels (4%) and atmospheric conditions (21%) following homogenous phase. O₂ depletion was also measured.

Methods

Scavenging methods were used to estimate radiolytic yields of H₂O₂ in water samples. Hydroxyl radicals production was estimated using EPR spin trapping with DMPO. O₂ measurements were performed using OxyLite probe.

Results

When UHDR and CONV-irradiation were compared, similar primary yields of H₂O₂ were found and EPR measurements suggested no differences in HO· production. However, G(H₂O₂) was significantly lower after irradiation at UHDR in samples equilibrated at 4%. O₂ measurements resulted in similar but low depletion with both modalities at intermediate and atmospheric O₂ conditions, whereas at low O₂ level, oxygen depletion was lower at UHDR.

Conclusions

These observations suggest that initial radiation chemistry is similar in both modalities: Similar yields of radicals ‘‘escaping’ ’track recombination after ionization. However, irradiation at UHDR produces less H₂O₂ at intermediate O₂ levels following initial chemistry events, supporting occurrence of scenario 1. O₂ depletion hypothesis is not favored by results obtained in pure water.

Acknowledgement: The study is funded by FNS Synergia grant (FNS CRS II5_186369)

Background and Aims

Future RT devices using very-high energy electrons (VHEE) (50-250MeV) may produce suitable beams to treat deep-seated tumours conformally and at ultra-high dose rates (UHDR) capable of triggering the FLASH effect. The FLASH effect has been observed for large doses delivered with overall treatment times less than 200ms. Such treatment durations do not allow the use of a movable gantry and multiple fixed beam lines (FBL) become mandatory. This treatment planning study evaluates VHEE dose distributions in patients using a varying number of FBL with different energies and source-axis-distances (SAD). The minimum requirements for delivering conformal VHEE RT comparable to conventional VMAT plans and trade-offs between plan quality and number of beam lines are assessed.

Methods

We performed VHEE and VMAT treatment planning for multiple indications (glioblastoma, mediastinum, lung, prostate) using RayStation (research version) and compared the dosimetric quality of VHEE plans to VMAT while assessing the impact of arrangement and number of FBL, beam energies, and SAD.

Results

Most substantial coverage and conformity improvement is achieved when increasing the beam energy from 50 to 100MeV. Further improvement is obtained specifically for deep-seated targets (>10-15cm) when increasing energies further to 200MeV. While VHEE plans using 16 coplanar beams outperform VMAT plans, we found that VHEE plans using only 3-5 beams have DVH metrics that are comparable to VMAT plans. Beams with SAD>1m are preferable for treatments using few beams.

Conclusions

UHDR VHEE devices with only a few FBL may provide competitive dosimetric conformity that may be additionally enhanced by the FLASH effect.