ESTRO 2025 - Abstract Book

S3968

Radiobiology - Tumour radiobiology

ESTRO 2025

995

Proffered Paper Genomic analysis of radiosensitivity in breast cancer: identifying pathological determinants and assessing GARD for personalized dose escalation Pierre Loap 1,2 , Irène Buvat 2 , Alain Fourquet 1 , Youlia Kirova 1 , Gilles Crehange 1,2 1 Department of radiation oncology, Institut Curie, Paris, France. 2 Laboratoire d'imagerie translationnelle en oncologie, Institut Curie, Orsay, France Purpose/Objective: Adjuvant radiotherapy improves recurrence-free survival in breast cancer patients. However, tumor radiosensitivity varies significantly, even within the same histological subtype. The Radiosensitivity Index (RSI), derived from the expression of 10 genes, and the Genomic-Adjusted Radiation Dose (GARD) model enable individualized radiotherapy dosing. This study evaluates the impact of histopathological factors on RSI and quantifies the biological effects of different boost doses across standard and hypofractionated radiotherapy regimens in a large cohort of breast cancer patients. Material/Methods: Transcriptomic RNA-seq data from 1,284 breast cancer patients in the TCGA-BRCA cohort were analyzed to calculate RSI using a rank-based regression model. GARD was calculated for multiple fractionation schemes and boost dose, including conventional regimens (50 Gy/25 fractions, 40.05 Gy/15 fractions, 26 Gy/5 fractions) and regimens with simultaneous integrated boost (64.4 Gy/28 fractions, 48 Gy/15 fractions, 30 Gy/5 fractions). Statistical analyses identified histopathological factors associated with RSI. Results: The median RSI across the TCGA cohort was 0.471. Lower RSI values, indicating higher radiosensitivity, were significantly associated with basal (p < 0.001) and luminal B (p < 0.001) subtypes. Other histopathological features such as necrosis, inflammation, and high mitotic index were also linked to greater radiosensitivity.

Without a boost, 70% of patients achieved a GARD > 21 (associated with improved tumor control in retrospective series) with the 50 Gy/25 fractions regimen, compared to 49.5% and 15.6% for 40.05 Gy/15 fractions and 26 Gy/5 fractions, respectively. The addition of an integrated boost significantly increased GARD values: 85.9% of patients receiving 64.4 Gy/28 fractions and 67.3% receiving 48 Gy/15 fractions achieved a GARD > 21. In contrast, only 26.0% of patients reached GARD > 21 with the 30 Gy/5 fractions regimen, indicating that extreme hypofractionation would require substantial dose escalation to achieve similar biological effectiveness. When stratified by molecular subtype,