ESTRO 2025 - Abstract Book

S409

Clinical - Biomarkers

ESTRO 2025

Purpose/Objective: Short-Term Androgen Deprivation Therapy (ST-ADT) has been shown to improve efficacy of definitive radiotherapy with intermediate-risk prostate cancer (IRPC) 1 . However, ST-ADT is associated with toxicities 2 and clinical parameters alone do not reliably identify those who benefit most. The ArteraAI-Prostate-Test is an artificial intelligence-based clinicopathological biomarker that gives prognostic and predictive information regarding ST-ADT benefit. While this biomarker 3 has been validated, real-world data is needed to assess its effects on shared decision-making regarding ST-ADT use. Material/Methods: This prospective registry and multicentre implementation trial collects data on the impact of the ArteraAI-Prostate Test on treatment decisions for men with IRPC undergoing curative radiotherapy. The primary endpoint is the percentage of patients for whom testing led to a change in the shared ST-ADT decision-making. Data collected included shared ST-ADT decisions pre-test and post-test. Trial registration ACTRN 12623000713695. Results: The planned interim analysis includes 200 patients enrolled across 28 centres between December 2023 and August 2024. Median age was 73 (range 56-87). Among these, 150 patients were classified as NCCN unfavourable intermediate risk (UIR) and 50 as favourable intermediate risk (FIR). The median ten-year prognostic biomarker risk for distant metastases for UIR and FIR subgroups was 2.6% (IQR 2.0 – 3.4) and 1.7% (IQR 1.4 – 2.0) respectively (figure 1), which were significantly different (p<0.001). For the predictive biomarker, 30/200 (15%) suggested a potential benefit for ST-ADT, with similar rates between FIR (7/50, 14%) or UIR (23/150, 15%) groups. Across the whole 200 patients, 74 patients had a pre-biomarker shared ST ADT decision of YES. After the biomarker result, 52 of these changed their shared ST-ADT decision to NO (decision change 70.3%; 95% CI: 59.9-80.7%). Conversely, 126 patients had a pre-biomarker shared ST-ADT decision of NO. After the biomarker result, only 3 of these changed their shared ST-ADT decision to YES (decision change 2.4%; 95% CI: 0-5.0% see figure 2). This change was statistically significant (p<0.001). Overall, 85.5% of final shared decisions aligned with the predictive biomarker. A greater magnitude of change was observed for UIR patients (35%, 53/150) compared to FIR (4%, 2/50; p<0.001).