S2144
Physics - Inter-fraction motion management and daily adaptive radiotherapy
ESTRO 2026
Oncol 2025;10:101874. Keywords: automated planning, dose prediction, deep learning
Digital Poster Highlight 2864 Weekly MRI-guided target adaptation for HPV- positive oropharyngeal cancer: in-silico comparison of classic and GTV-shrinking adaptation strategies Suzanne P.M. de Vette 1 , Nanna M. Sijtsema 1 , Ilse G. van Bruggen 1 , Daniel C. MacRae 1 , Luuk van der Hoek 1 , Hendrike Neh 1 , Minke J. Brinkman-Akker 1 , Cem Dede 2 , Zayne Belal 2,3 , Mohamed Naser 2 , Johannes A. Langendijk 1 , Clifton D. Fuller 2 , Lisanne V. van Dijk 1 1 Radiotherapy, University Medical Center Groningen, University of Groningen, Groningen, Netherlands. 2 Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, USA. 3 Radiation Oncology, University of Pennsylvania Perelman School of Medicine, Philadelphia, USA Purpose/Objective: HPV-positive oropharyngeal cancer patients have favorable survival outcomes, but often experience side-effects after radiotherapy. In these patients, the MR-ADAPTOR trial (NCT03224000) investigated the safety of adapting the high-dose GTV/CTV guided by the visible tumor on weekly MRIs, as HPV-positive tumors often shrink significantly during treatment. This approach may offer superior sparing of organ-of- interest (OOI) compared to classic adaptation, which irradiates the pre-treatment GTV regardless of shrinkage. Our in-silico study builds on MR-ADAPTOR by comparing treatment plans using its adaptive strategy versus classic adaptation for both VMAT and IMPT, assessing OOI mean dose and NTCP differences. Material/Methods: This study included 28 patients from the experimental arm of the MR-ADAPTOR trial (accrual 2018-2024). During treatment, patients underwent weekly CT and MR scans, and treatment plans were adapted whenever the axial MR images revealed a reduction in GTV compared to the previous week. New VMAT and IMPT treatment plans were generated using DL-based automated planning models to ensure consistent, comparable plan quality. Plans delivered 70 Gy to the tumor and affected lymph nodes, and 54.25 Gy to elective lymph nodes. Two adaptive strategies were compared (Figure 1): a conventional approach (Adaptclassic), and one incorporating tumor shrinkage (Adaptshrink, the MR-ADAPTOR approach). This allows for direct comparison of tumor shrinkage on the dosimetric parameters, independent of other anatomical changes. Mean OOI doses and physician-
Conclusion: This study evaluated the feasibility of a CBCT-based, dose prediction–driven automated workflow for ART in breast cancer. The approach demonstrated improved PTV coverage and minor changes to OAR exposure. As a decision-support tool, CBCT-based dose prediction enables rapid clinical judgment by assessing the benefit of adaptation without requiring a full replanning process. This supports individualized adaptive workflows and promotes efficient use of clinical resources. References: [1] Sun L et al. J Appl Clin Med Phys 2019;20:115– 24. [2] De-Colle C et al. Clin Transl Radiat Oncol 2023;39:100564. [3] Brock KK. Semin Radiat Oncol 2019;29:181–4.[4] Kalinauskaite G et al. Adv Radiat
Made with FlippingBook - Share PDF online