Validation of a multi-omic biosignature for locoregional recurrence and radiation therapy benefit in breast cancer

To inform RT shared decision-making, a multi-omic test was developed [PreludeDx, AidaBREAST] to assess locoregional recurrences (LRR risk) and differential radiotherapy (RT) effect in HR+/HER2- invasive breast cancer (IBC). Herein, the test was validated in an independent cohort from Royal Melbourne Hospital, Australia. Blinded biomarker testing was performed on the tissue samples from evaluable patients with HR + HER2- T1T2NX/0M0 IBC treated with BCS and prescribed endocrine therapy between 2006 and 2021 ( n = 426) with or without RT, and/or chemotherapy. The test generated a continuous Decision Score (DS) for LRR risk and a continuous Radiation Resistance Index (RRI) for differential RT effect. Associations with LRR and differential RT effect were assessed using multivariable Cox regression, and categorical analyses used pre-specified thresholds (DS = 5, RRI = 5). Among 426 women, 319 received RT. Increasing continuous DS was prognostic for LRR risk (LRR HR = 2.4, p < 0.001). Increasing continuous RRI was associated with reduced RT effect (LRR HR = 6.0, p interaction < 0.001). The biosignature provided significant independent information beyond clinicopathological characteristics alone ( p < 0.001). In the Low Risk group (DS ≤ 5), RT provided no significant reduction in LRR risk (10-year LRR 7% with and 8% without RT). In DS Elevated Risk group, 10-year LRR risk without RT was 26%, and within this group, patients with RRI ≤ 5 had a significant reduction in LRR risk (HR = 0.2, p < 0.001) with RT, whereas those with RRI > 5, RT derived no significant reduction in LRR risk ( p = 0.8). This independent validation confirms that the continuous biosignature provided information beyond standard clinicopathological characteristics for assessing 10-year LRR risk and differential RT effect.