Machine stability and dosimetry for ultra-high dose rate FLASH radiotherapy human clinical protocol

Abstract

Background

The FLASH effect, induced by ultra-high dose rate (UHDR) irradiations, offers the potential to spare normal tissue while effectively treating tumors. It is important to achieve precise and accurate dose delivery and to establish reliable detector systems, particularly for clinical trials needed to help the clinical transfer of FLASH-Radiotherapy (FLASH-RT). However, the use of monitoring chambers with UHDR beams is presently limited, leading to the reliance on passive dosimetry and machine stability.

Purpose

This study aimed to investigate the energy and output stability of a UHDR Mobetron (IntraOp, USA) and to compare it with its conventional dose rate (CDR) mode. Furthermore, we assessed the dosimetric accuracy of a human clinical protocol for FLASH-RT.

Methods

Over a 26-month duration, we assessed the short- and long-term stability of the output and energy of the Mobetron system. Daily checks were conducted for 9 MeV CDR and UHDR. In parallel, the IMPulse clinical trial involving patients with skin metastases from melanoma was initiated. Prescription doses ranging from 22 to 28 Gy were administered. Pre-, post-, and in vivo dosimetry using alanine and thermoluminescent dosimeters (TLDs) was performed and compared to the prescription doses.

Results

Short-term output fluctuations remained below 0.6 % and 1 % for 9 MeV CDR and UHDR, respectively. Long-term output fluctuations were within 2 % and the long-term energy fluctuations were below 2 mm (R₅₀) for both modes. The delivered doses of the IMPulse trial showed deviations below 4 % compared to prescription doses for all patients.

Conclusions

The Mobetron system demonstrated favorable short- and long-term stability. There was a good agreement between the prescribed and the measured dose for the clinical IMPulse trial. The stability of this UHDR machine allows us to effectively conduct human clinical protocols as well as preclinical experiments, even in the absence of a real-time monitoring system.