ESTRO 2025 - Abstract Book

S2738

Physics - Dose prediction, optimisation and applications of photon and electron planning

ESTRO 2025

936

Poster Discussion Intensity Modulated Proton Therapy for pelvic RT based on normal tissue complication probability estimates of persistent GI toxicity in PARCER trial. Subhajit Panda 1 , Anouk Corbeau 2,3 , Ankita Gupta 1 , Sofia Spampinato 2 , Prachi Mittal 4 , Mayuri Charnalia 1 , Nilesh Ranjan 4 , Prachi sawant 4 , A. Chithambara Prabu 1 , Prakash Shinde 1 , Jeevanshu Jain 1 , Lalit Chaudhari 1 , Mischa Hoogeman 2,5 , Remi Nout 2 , Supriya Chopra 1 1 Department of Radiation Oncology, Advanced Centre for Treatment, Research and Education in Cancer (ACTREC), Tata Memorial Centre, Homi Bhabha Institute, Navi Mumbai, India. 2 Department of Radiotherapy, Erasmus MC Cancer Institute,University Medical Centre Rotterdam, Rotterdam, Netherlands. 3 Department of Radiation Oncology, Leiden University Medical Center, Leiden, Netherlands. 4 Department of Radiation Oncology, Tata Memorial Hospital,Tata Memorial Centre, Homi Bhabha National Institute, Mumbai, India. 5 Department of Medical Physics & Informatics, Holland PTC, Delft, Netherlands Purpose/Objective: Intensity Modulated Proton Therapy (IMPT) is being investigated for gynaecological cancers in clinical trials. However, a robust Normal Tissue Complication Probability (NTCP) based model for patient selection model is not available. We recently developed an NTCP model for persistent gastrointestinal (GI) toxicity, measured as GI-C MOSES 1 , using data from the phase-III PARCER trial in postoperative cervical cancer. The purpose of this work was to pilot ΔNTCP based proton therapy selection for gynecological cancers. Material/Methods: Ten patients from PARCER IMRT arm, who did not meet trial-specific bowel dose constraints, were selected. Clinical IMRT (Tomotherapy) plans were available for these patients. IMPT plans were generated in RayStation (version 12A) for a Pencil Beam Scanning (PBS) proton unit (Proteus 235, IBA). Robust optimization was performed with 5 mm setup and ±3.5% range uncertainty. Plans were evaluated across 28 scenarios. Each plan delivered 50Gy in 25fractions. PTV D95 and CTV D95 in the worst-case scenario were assessed for IMRT and IMPT, respectively. Dose-volume parameters for the bowel bag (BB) and individual loops of small and large bowel (total bowel, TB) were compared between IMRT and IMPT (paired t-test).An NTCP model for persistent late GI toxicity, including BB V30Gy and V40Gy as input parameters, was previously developed using individual patient and treatment characteristics. Finally, ΔNTCP between IMRT and IMPT was estimated using this NTCP model. The proportion of patients with ΔNTCP between 5 and 10% and >10% was determined. Results: All IMRT and IMPT plans achieved adequate target coverage with PTV D95 >95% for IMRT and CTV D95 >98% for IMPT. For BB, IMRT had significantly higher V15Gy (1130cm 3 vs. 499cm 3 , p<0.001), V30Gy (684cm 3 vs. 402cm 3 , p<0.001), and V40Gy (456cm3 vs. 336cm3, p<0.001) compared to IMPT (shown in Figure 1). IMRT also showed higher TB V15Gy (541cm3 vs. 227cm3, p<0.001), V30Gy (305cm3 vs 165cm3, p<0.001), and V40Gy (186cm3 vs 141cm3, p<0.001) than IMPT. ΔNTCP of persistent GI toxicity for IMRT and IMPT are depicted in Figure 2. ΔNTCP was 5-10% in 7/10 patients and ≥10% in 2/10 patients. The median ΔNTCP was 7.9% (range 1.7-16.1%).