Evaluation of revised TRS398 dosimetry protocol for pencil beam scanning proton therapy systems

Purpose.The International Atomic Energy Agency revised the TRS398 Code of Practice in 2024, incorporating dosimetry protocol for pencil beam scanning (PBS) beam delivery system with updated beam quality correction factors (kQ,Qo) for the majority of ionization chambers used in proton beam therapy. This study aims to evaluate the revised TRS398 protocol for dosimetry in an established isochronous cyclotron-based PBS proton therapy system and to quantify its impact on absorbed dose determination.Methods.The reference dosimetry measurements were performed using three plane parallel ionization chambers (IBA PPC05, IBA PPC40, PTW Roos) and one cylindrical chamber (IBA FC65-G) in a water phantom on a Proteus Plus PBS proton therapy system. Uniform single-energy scanned fields (10 × 10 cm²) were delivered across 33 clinical proton beam energies ranging from 70.18 MeV to 226.2 MeV. Absorbed dose to water was calculated usingkQ,Qofrom both the previous and revised TRS398 protocols. Ion recombination (ks) and polarity (kpol) correction factors were determined and evaluated for all chambers for 17 proton energies.Results.The absorbed dose measurements demonstrated good inter-chamber agreement, with a maximum deviation of 2.72% at the lowest energy. Implementation of the revisedkQ,Qovalues resulted in mean absorbed dose differences of -1.07%, -0.66%, -0.50%, and -1.77% for the PPC05, PPC40, Roos, and FC65-G chambers, respectively, compared to the previous TRS398. The combined relative uncertainty in absorbed dose determination using the revised TRS398 was estimated to be 1.6%.Conclusion.This study assesses the impact of the revised TRS398 dosimetry protocol on absorbed dose determination in PBS proton beams. Experimental measurements using four dosimeters in an isochronous cyclotron-based system show that the revised TRS398 provides a robust framework for PBS reference dosimetry, with the effect of updatedkQ,Qovalues remaining within the combined measurement uncertainty.