Session Item

Dose rate verification in FLASH radiotherapy is essential for validating the dosimetric conditions that trigger the FLASH effect. Dose rate validation with good time resolution is particularly important for pencil beam scanning (PBS) where the instantaneous dose rate in a point varies on a millisecond time scale as the beam scans across the spot pattern. In this study, a scintillator-based detector system was used for time-resolved dose rate measurements in proton PBS irradiations. The dose rate response of the system was characterized, and the calibrated system was used to measure the instantaneous dose rate in pre-clinical proton FLASH studies with mice.

The detector system was based on four sub-millimeter fiber-coupled ZnSe:O scintillators. It measured the instantaneous dose rate at 50kHz, sufficient to provide the dose rate of each spot individually. As a first step, the detector system was calibrated. The detectors were placed on an Advanced Markus ionization chamber (IC) inside a water bath in the isocenter plane (Fig. 1A). A 52mm-by-52mm spot pattern with 2mm spot spacing was delivered using a 250MeV proton beam (Fig. 1B). The dose rate was altered by changing the requested nozzle beam current from 5nA to 215nA with 10nA steps. A calibration curve (Fig. 1C) was produced by plotting the mean detector signal [V] for each spot and nozzle current against an absolute dose rate calculated from the IC measurement [Gy/s]. The signal and instantaneous dose rate depended on the distance to the spot and the beam current. The integral dose from the IC was recalculated into a dose rate per spot by assuming each spot contributed to the total dose by a weight depending on the detector-to-spot distance and the spot duration. Moreover, the signal was used for locating the detector positions within the spot pattern and measuring the spot widths. The calibration was then applied to in vivo measurements during mouse experiments, in order to provide the instantaneous dose rate. The measured dose rates were compared to the dose rate determined from the spot size and the MU-rate found in machine log files, which was assumed to be the ground truth.

For each detector, the instantaneous dose rate as function of measured detector signal was modelled with a third-degree polynomial function (Fig. 1C). The calibrated system enabled instantaneous dose rate measurements for each delivered spot during mouse irradiations (Fig. 2). This dose rate was in good agreement with the ground truth dose rate with a root-mean-square difference of 13Gy/s across a wide range of instantaneous dose rates up to 1000Gy/s (Fig. 2B).

A detector system for direct measurements of the instantaneous dose rate was developed and calibrated in a wide range of dose rates. The detector system was successfully applied in vivo in mouse irradiations, enabling time-resolved dose rate measurements in pre-clinical proton FLASH studies on a sub-spot level.