EXPERIMENTAL UHDR SET-UP USING A LOW ENERGY PROTON BEAM FOR IN-VITRO IRRADIATION

Recent investigations in vivo showed that short pulses of photons and electrons with ultra high dose rate (> 40 Gy/s) are less harmful for healthy tissues and as efficient as conventional dose-rate irradiation to inhibit tumor growth. In the case of protons however, FLASH effects have not been studied much in part due to the limited availability of facilities that can achieve such ultra high dose rates.
In order to investigate the biological mechanism behind FLASH effect with proton beam, we developed two UHDR systems allowing us to control the exposure time of cells to the high current beam on the 2 MV ALTAÏS particle accelerator at LARN laboratory. The first one relies on electromagnetic deflection, which allows us to pulse the beam and control both pulse width and interval between pulses. A real-time beam profile read-out using a scintillator and a CCD camera allows us to observe the tomography of the beam and ensure its homogeneity. The dose-rate is given by an optimized home-made faraday cup.
The second system is continuous UHDR. Beam diagnostic tools are similar but the exposure is made continuous by mounting cell samples on a precision motor and sweeping them in front of the proton beam. Using this configuration, the whole sample cannot be irradiated at the same time, mimicking what is actually done in PBS in the clinic. The exposure time is finely tuned by adjusting the motor speed. The irradiation set up for CONV-PT is already available at LARN and will be used for comparison.

The complete system has been validated using unlaminated ETB3 and HD-V2 Gafchromic films. The pulse generator allows us to obtain multiple pulses with adjustable width from a minimum of 10 μs up to several seconds. The smallest time interval between pulses is 10 µs. Mean dose rates from 40 to 1000 Gy/s are easily obtained.
Two complete systems are proposed to allow in vitro sample irradiation with either continuous or pulsed low energy proton beam. Parameters can be probed independently to decipher what could be the main actor of FLASH effect. On one hand, the pulsed setup could reproduce the pulse macrostructure of clinical machines. One the other hand, the continuous setup can be used to mimic pencil beam scanning effects. Biological experiments using 3D cultures of HCT-116 and T98G are ongoing to validate our FLASH beam.