S243
Clinical - Breast
ESTRO 2026
Material/Methods:
prognostic factors, including BMI, surgery type, and nodal involvement. Conclusion: By integrating an explainable AI-powered predictive tool into the clinical workflow, PRE-ACT-01 represents a significant step towards personalized AI-driven intervention in oncology. This innovative, large-scale study has the potential to establish a new standard of care, where early identification of high-risk patients enables targeted preventative strategies and empowers patients through shared decision-making, ultimately improving long-term outcomes for breast cancer survivors. References: Acknowledgements: This work was conducted in the context of the Horizon Europe project PRE-ACT (Prediction of Radiotherapy side effects using explainable AI for patient communication and treatment modification). It was supported by the European Commission through the Horizon Europe Program (Grant Agreement number 101057746), by the Swiss State Secretariat for Education, Research and Innovation (SERI) under contract number 22 00058, and by the UK government (Innovate UK application number 10061955). Keywords: breast cancer, lymphedema, AI-prediction Continuous positive airway pressure in breast cancer radiotherapy: preliminary results from a prospective clinical trial Aline Van der Vorst 1 , Jelle Verhoeven 1 , Adinda Baten 1 , Hilde Janssen 1 , Lander Van Bavel 1 , Laurence Delombaerde 1 , Truus Reynders 1 , Annouschka Laenen 2 , Maarten Lambrecht 1 , Lucas Van Aelst 3 , Caroline Weltens 1 1 Department of Radiation Oncology, UZ Leuven, Leuven, Belgium. 2 Department of Biostatistics and Statistical Bioinformatics, UZ Leuven, Leuven, Belgium. 3 Department of Cardiology, UZ Leuven, Leuven, Belgium Digital Poster Highlight 1533 Purpose/Objective: Preliminary studies suggest that continuous positive airway pressure (CPAP) can reduce organ at risk doses in left-sided breast cancer (BC) radiotherapy (RT)1–3. This prospective trial evaluates whether CPAP can reduce heart, lung and contralateral breast (CLB) doses in a larger patient cohort, including both left and right-sided cases, and identifies which patients benefit most.
Patients undergoing breast conserving surgery for invasive or in situ BC and requiring adjuvant RT are prospectively enrolled —either left-sided cases with or without regional nodal irradiation (RNI), or right-sided cases requiring RNI. Each patient undergoes an additional CT simulation with CPAP (15cm H20) —in free breathing (FB) for right-sided cases and in deep inspiration breath hold (DIBH) for left-sided cases. Left-sided patients unable to perform DIBH are evaluated in FB. Two RT plans (with and without CPAP) are dosimetrically compared. If CPAP reduces mean heart or CLB dose by ≥ 0.5 Gy, or mean lung dose by ≥ 1 Gy, treatment is delivered with CPAP. Patient comfort is assessed through surveys. Results: To date, 19 out of 53 patients planned for inclusion have been enrolled. Three patients (two right-sided and one left-sided with RNI) received CPAP-assisted RT. Across all patients, CPAP significantly reduced mean doses to the heart (1.5 ± 0.5 Gy versus 1.6 ± 0.6 Gy), lungs (5.0 ± 2.0 Gy versus 5.4 ± 2.4 Gy) , ipsilateral lung (9.8 ± 3.8 Gy versus 10.5 ± 4.6 Gy), left anterior descending artery (4.8 ± 4.7 Gy versus 5.7 ± 5.0 Gy) and left ventricle (1.2 ± 0.5 Gy versus 1.5 ± 0.8 Gy) compared to standard simulation, respectively. CPAP also significantly reduced volumetric parameters (V5Gy–V30Gy) across all these structures and lowered maximum RT dose in all these except the lungs. CLB doses remained unchanged. Lung volumes increased significantly with CPAP (+671.7cc, 95% CI: 401.73- 941.74). Furthermore, left-sided patients were able to maintain deeper (+0.3cm, 95% CI: 0.120-0.384) and longer (+12.1 seconds, 95% CI: 1.634-22.616) DIBHs when using CPAP. Moreover, CPAP was well tolerated and reported as comfortable by the majority of patients.
Made with FlippingBook - Share PDF online