ESTRO 2024 - Abstract Book

S4693

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

ESTR0 2024

1. RT conv : Conventional plan without accounting for hyperthermia. Robust uniform EQD2 of 62 Gy and 50 Gy in 25 fractions to the GTV and the CTV, respectively ( α / β =10 Gy). Non-robust EQD2 objectives were used for the OARs ( α / β =3 Gy). 2. RT HT : Identical CTV objective and OAR objectives as RT conv but with incorporation of thermal radiosensitization in the OAR objectives. A robust uniform EQD2 objective of 75 Gy accounting for thermal radiosensitization and direct thermal cytotoxicity was used for the GTV. Weekly one-hour hyperthermia sessions delivered 30 min after the corresponding radiotherapy fraction was assumed (i.e. n HT =5, t =1 h, t int =0.5 h).

Subsequently, all plans were evaluated using EQD2 for 25 radiotherapy fractions with and without accounting for the five hyperthermia sessions.

Results:

The near-minimum, median and near-maximum GTV temperatures (T 98%, T 50% and T 2% ) were 41.3°C, 42.5°C and 43.7°C for case 1, and 40.0°C, 41.5°C and 43.4°C for case 2. This was achieved without exceeding 45°C for the normal tissues. Figure 1 shows the 2D distributions of temperature and EQD2 for all plans including hyperthermia for case 1, and Figure 2 shows the corresponding DVHs for GTV (with and without hyperthermia), femoral heads and bladder (with hyperthermia). The uniform GTV EQD2 of 62 Gy for all RT conv plans without hyperthermia was enhanced with about 4 6 Gy on average when accounting for hyperthermia (RT conv +HT). The GTV homogeneity index (HI=(EQD2 2% −EQD2 98% )/EQD2 50% ) for RT conv of about 3-4% were worsen to about 8% for RT conv +HT (Figure 2a). For RT HT , GTV EQD2 averages of approximately 68 Gy and 70 Gy were obtained for case 1 and 2, respectively, with HI≈8% (similar HI:s as RT conv +HT). The GTV average EQD2 was enhanced to a uniform EQD2 of 75 Gy when adding hyperthermia (RT HT +HT), in accordance with the planning objective (Figure 2a). The HI:s were also improved to about 3-4% (similar HI:s as RT conv without hyperthermia). This uniform EQD2 escalation was achieved by employing the thermoradiotherapy optimization framework, which optimized radiation doses on a voxel-by-voxel basis, considering the temperature within each voxel. Noticeable, this average escalation of around 7-8 Gy EQD2 for RT HT +HT compared to RT conv +HT caused only minor differences in the OAR doses (Figure 2b). However, the proton plans substantially reduced the EQD2 to all relevant OARs compared to the photon plans for both cases (indicated for spinal cord/bladder in Figure 1 and femoral heads/bladder in Figure 2b for case 1).