1061 - Adaptive Radiotherapy in Hypersensitive and High Locoregional Risk Breast Cancer: Preliminary Results of the SAHARA04 Phase II Trial

03:10pm - 03:15pm ET

Room 156

Presenter(s)

Jessica Prunaretty, MS - Institut du Cancer de Montpellier, Montpellier, Occitanie

J. Prunaretty¹, C. Bourgier¹, A. Morel¹, L. Lopez¹, P. Hinault¹, N. Bizot¹, N. Ailleres¹, D. Azria^2,3, and P. Fenoglietto¹; ¹Institut du Cancer de Montpellier, Montpellier, France, ²Fédération Universitaire d’Oncologie Radiothérapie d’Occitanie Méditerranée, Institut du Cancer Montpellier (ICM), Univ Montpellier, INSERM U1194 ICM, Montpellier, France, ³University Federation of Radiation Oncology of Mediterranean Occitanie, Montpelier Cancer Institute (ICM), Montpellier Cancer Research Institute (IRCM), INSERM U1194, University of Montpellier, Montpellier, France

Purpose/Objective(s):

The SAHARA04 phase II trial evaluates online adaptive radiotherapy (oART) using a CBCT-guided adaptive system in breast cancer patients presenting both a high locoregional recurrence risk and a high risk of radiation-induced fibrosis. This study reports preliminary clinical experience and dosimetric results.

Materials/Methods:

Patients were treated in supine position, free-breathing, with both arms raised, using a personalized foam cushion. The prescribed doses were 52.2 Gy to the tumor bed (boost) and 42.3 Gy to the breast, internal mammary chain (IMC), and clavicular lymph nodes (Nodes), delivered in 18 fractions. CTV-to-PTV margins were 2 mm for all targets except the IMC (5 mm). Reference IMRT plans consisted of 13 fields with 6 MV FFF energy. Treatment planning followed ICRU 50/62/83 recommendations, and organ-at-risk (OAR) constraints were based on the HYPOG-01 protocol.

Each oART session began with a first CBCT (CBCT1). Artificial intelligence–based auto-segmentation generated influencer structures (both breasts, lungs, and heart), while IMC and Nodes volumes were obtained by rigid registration. The CTV_Breast was derived from the breast volume excluding the 5 mm layer beneath the skin. All contours were reviewed and edited when necessary prior to plan generation. Both adapted (reoptimized) and scheduled (recalculated) plans were then generated, followed by a second CBCT (CBCT2) to verify final positioning.

For each session, the treatment duration (from CBCT1 acquisition to the end of beam delivery) and monitor units (MUs) were recorded. Dose metrics from adapted (ADP), scheduled (SCH), and reference (PLAN) plans were compared to assess CTV coverage and OAR sparing. CBCT1–CBCT2 registrations were also analyzed.

Results:

A total of 126 oART sessions from seven patients (five right- and two left-sided breast cancers) treated between April and December 2025 were evaluated. The median treatment duration was 35 min (range: 21–61 min). MU values were comparable between reference and adaptive sessions (2474 vs. 2511). Adapted plans achieved significantly higher CTV D98% compared with scheduled plans and were similar to the reference plan metrics (Table 1). Organs-at-risk doses remained unchanged across all plan types. CTV coverage criteria (D98% > 95%) were unmet in only 7.14% of ADP sessions versus 69.84% of SCH sessions. CBCT1–CBCT2 displacements exceeding 2 mm were observed in 21.4% of fractions, mainly in the antero-posterior direction, reflecting intra-fraction positioning variability.

Conclusion:

These preliminary results demonstrate the feasibility and dosimetric benefit of oART in high-risk breast cancer patients, with improved target coverage while maintaining OAR protection. Robust patient setup and intra-fraction stability remain critical to ensure treatment accuracy.

	D_98% (%)
	CTV_Breast	CTV_IMC	CTV_Nodes
ADP	97,97 ± 1,59	99,17 ± 0,73	99,73 ± 1,31
SCH	97,00 ± 1,66	97,41 ± 5,41	91,76 ± 10,41
PLAN	98,67 ± 1,01	99,16 ± 0,43	99,83 ± 0,78