ESTRO 2024 - Abstract Book

S4725

Physics - Optimisation, algorithms and applications for ion beam treatment planning

ESTR0 2024

[1] Balgobind BV, Visser J, Grehn M, et al. Refining critical structure contouring in STereotactic Arrhythmia Radioablation (STAR): Benchmark results and consensus guidelines from the STOPSTORM.eu consortium [published online ahead of print, 2023 Oct 10]. Radiother Oncol. 2023;189:109949. doi:10.1016/j.radonc.2023.109949

3047

Digital Poster

Origins – The Harvard Cyclotron Large-Field Proton Patients Reconsidered

Hanne M Kooy 1 , Nicolas Depauw 1 , Stefano Gianolini 1 , Judith A Adams 1 , Miles Wagner 2 , Benjamin M Clasie 1 , Norbert J Liebsch 1

1 Massuchesetts General Hospital, Radiation Oncology, Boston, USA. 2 Mevion LTD, R&D, Littleton, USA

Purpose/Objective:

The Harvard Cyclotron (HCL) treated patients with SOBP fields up to 160 MeV from 1975 to 2001. These patients were treated up to modern standards, far exceeding contemporaneous standards. Their outcomes are the foundation for the motivation and practice of proton radiotherapy. The cohort is unique because of their quality is on par to modern treatments with historical and documented follow-up. No other historical patient cohort serves as an on-par comparison to modern treatments. The cohort is characterized by particular emphasis on complex and high-dose disease sites. The extensive analysis for these patients is based on basic algorithms for proton dose calculations. We wish to establish the adequacy and accuracy of the patient dosimetry in comparison to the modern standard of Monte Carlo calculations and, only now, competitive photon intensity modulated treatments. Finally, we aim to revisit foundational studies based on the original dosimetry compared to modern dosimetry.

Material/Methods:

We converted the original data from proprietary format to DICOM RT data: CT, structure, plan (with beams) and dose. The 3D treatment planning system code, Rx, was perused for interpretation of this data. Rx computational processes are documented in a DICOM conformance statement necessary for computational interpretation. Of particular interest is the absolute dose prediction. The dose-calculation method was limited and recalculation with Monte Carlo will improve accuracy and may yield new insights. We describe the HCL planning and treatment protocols. Treatment planning is 3D and included 1 photon fraction / week. Dosimetry model data is reconstructed from original measurements. Of particular necessity is to model the field-calibration protocol in the Monte Carlo to align original dose data with new dose data.

Results:

We salvaged 1,200 patients. The patient population comprises soft tissue sarcoma, H&N neoplasms including chondrosarcoma and CNS. The large cohort of eye patients treated at the HCL are not included. Dose recomputations were done with a fast in-house GPU Monte Carlo system dubbed GMC that was commissioned with the original HCL