Background and Aims

Background: The beam stability and the precision of the dose delivery are essential to guarantee the safety and quality of proton FLASH therapy in ultra-high dose rates (UHDRs).

Aim: A proton pencil-beam scanning pattern was designed and irradiated on a high spatiotemporal resolution 2D strip ionization chamber array (SICA) with an in-house designed phantom as an all-in-one pre-treatment QA for proton FLASH therapy.

Methods

Methods: A PBS plan with a single field was designed with a 5x5 cm² uniform field at the center for dose and dose rate consistency. Asymmetrical lines on the perimeter are used to test positioning accuracy in discrete spot mode, varied speeds in continuous scanning mode, and range verification. The 250 MeV proton beam was irradiated on the SICA with the Varian ProBeam system at 100nA and 215nA nozzle currents. A reference Advanced Markus ionization chamber was placed downstream to the SCIA. The QA items, including the spot positioning, spot scanning speed, absolute dose, and average dose rates were recorded for trend analysis.

Results

Results: The overall delivered spot positionings are within 0.2mm compared with the planned positions. The low and high spot scanning speeds are within 1 mm/ms deviation to the requested specification of the beam control system. The output variation of the central field was within ±5% compared with the average and ±2.5% after normalized with the reference. More than 79.6% and 97.8% of the measurement points reached the FLASH threshold (40 Gy/s) at 100 nA and 215 nA in the 5x5 cm²region defined by a dose threshold of 80% of the center dose, respectively.

Conclusions

Conclusions: The pre-treatment QA program for FLASH therapy in UHDR with the designed pattern irradiated on a SICA can be performed in single irradiation. The SICA is a reliable tool to fulfill a QA solution for proton PBS FLASH radiotherapy.

Background and Aims

DNA-damaging treatments such as radiotherapy (RT) became promising to improve the efficacy of immune checkpoint inhibitors by enhancing tumor immunogenicity. However, accompanying treatment-related detrimental events in normal tissues had been the dark side of the combination regimen, and had posed a major obstacle for patients potentially amenable for radioimmunotherapy, thus presenting new challenges to the dose delivery mode of clinical RT. In the present study, ultra-high dose rate FLASH X-ray irradiation was applied to counteract intestinal toxicity in radioimmunotherapy.

Methods

In the present study, the high-energy X-ray with ultra-high dose rate (110-120 Gy/s) was applied in the single pulse mode. Programmed cell death ligand-1 (PD-L1) knockout (KO) mice undergoing conventional dose rate (CONV) or FLASH X-ray irradiation were undertaken as animal models to simulate the concurrent administration of RT with checkpoint blockade.

Results

In the context of PD-L1 blockade, FLASH X-ray minimized mouse enteritis by alleviating CD8⁺ T cell-mediated deleterious immune response compared with CONV irradiation. Mechanistically, FLASH irradiation was less efficient than CONV X-ray in eliciting cytoplasmic double-stranded DNA (dsDNA) and in activating the cGAS-STING signaling pathway in the intestinal crypts, resulting in the suppression of the cascade feedback consisting of CD8⁺ T cell chemotaxis and gasdermin E-mediated intestinal pyroptosis in the case of PD-L1 blocking. Meanwhile, FLASH X-ray was as competent as CONV RT in boosting the anti-tumor immune response initiated by cGAS activation and achieved equal tumor control in metastasis burdens when combined with anti-PD-L1 administration.

Conclusions

Together, we proposed the ‘DNA integrity’ hypothesis to elucidate the mechanism of the FLASH sparing effect, suggesting that ‘relatively intact DNA integrity’ within intestinal cells during the ‘instantaneous’ (approximately 120 ms) IR exposure was critical for FLASH X-ray in sparing PD-L1 deficient mice from detrimental enteritis. However, the inherent genomic instability of tumor cells makes them still sensitive to FLASH radiation.

Background and Aims

There is a need for beam monitoring and feedback control systems for FLASH that are dose-rate independent and capable of controlling the beam at the level of individual pulses. We aim to use a beam current monitor (BCT) to measure dose in real-time, validate time structure of the beam, and build a feedback control loop.

Methods

We modified a 3D printed collimator for pre-clinical anatomy-specific irradiation of mice to accommodate a 55 mm diameter BCT positioned surrounding the beam path. The signal from the BCT was digitized on a fast oscilloscope. We determined the effect of the modified collimator with BCT on the entrance beam profile. The linearity of the BCT was established by varying dose per pulse in the range 0.2 to 2.3 Gy/pulse. The pulse width on the clinical linac was varied, and this change was measured using the BCT. Finally, the ability of the BCT to measure the variability of dose per pulse and pulse width due to a mistuned automatic frequency control (AFC) system was demonstrated.

Results

The BCT had minimal effect on the entrance profile and exhibited excellent linearity with dose per pulse up to 2.5 Gy/Pulse (R² =0.99). The peak beam current for the 17 MeV FLASH beam was measured to be ~10 mA for a 2Gy/pulse output. Manual tuning of AFC showed the presence of a ramp-up in dose per pulse and pulse width. During the ramp-up period, the pulse width was observed to be as small as 0.5 us and the pulse output was 1/4^th of the stable output.

Conclusions

We demonstrated real-time per pulse readout of electron beam charge, correlated with entrance dose. Going forward, we will perform a failure modes and safety analysis of the control system and assess long-term stability with the goal of supporting a human clinical trial.

Background and Aims

Our previous studies demonstrated that single high dose FLASH total abdominal irradiation (FLASH-TAI) produced significantly reduced mortality from gastrointestinal syndrome, preserved epithelial integrity, and spared cell death in crypt base columnar cells compared to conventional dose rate total abdominal conventional irradiation (CONV-TAI) in C57BL/6J mice. Here, we compare FLASH-TAI with CONV-TAI in additional mouse strains with and without defects in DNA double strand break (DSB) repair.

Methods

FLASH-TAI and CONV-TAI were performed in S129 mice with and without a knockout in Fanca (and thus deficiency in homologous recombination DNA DSB repair) at 14 and 16 Gy; in BALB/c mice which have reduced expression of DNA-dependent protein kinase catalytic subunit (DNA-PKcs) (and thus deficiency in nonhomologous end joining DNA DSB repair) at 10 and 12 Gy; and in C57BL/6J mice at 12 Gy. We used a clinical linear accelerator configured to produce a ~16-17 MeV electron beam collimated to deliver a uniform dose throughout the mouse abdominal cavity in both FLASH (216 Gy/s) and conventional (0.079 Gy/s) dose rates. Regenerating crypts were counted in 3-5 hematoxylin & eosin stained mid-jejunal circumferences per mouse (4-8 mice/group) 96 hours post irradiation.

Results

While Fanca wild type mice had increased regenerating crypts compared to Fanca knockout after both FLASH-TAI and CONV-TAI, both Fanca wild type and Fanca knockout mice had increased numbers of regenerating crypts following FLASH-TAI compared to CONV-TAI. BALB/c mice had ~5-fold fewer regenerating crypts after TAI compared to C57BL/6J mice. However, BALB/c had significantly more regenerating crypts after FLASH-TAI compared to CONV-TAI.

Conclusions

These data indicate that FLASH spares crypt regeneration compared to CONV independent of FANCA-dependent and DNA-PKcs dependent DNA DSB repair processes. Future work includes an analysis of the kinetics of DNA DSB repair in these strains of mice.