Prognostic performance of an AI-based recurrence risk model in clinically low-risk HR+/HER2- early breast cancer
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Objective
Accurate prognostication of recurrence risk in early breast cancer is central for therapeutic decision-making, including identifying patients who may safely avoid adjuvant therapy. Here we evaluate an artificial intelligence (AI)-based method to improve risk stratification.
Methods
Ataraxis Breast CTX (ATX) is an AI test that integrates H&E-stained images with clinicopathologic features to predict risk of recurrence for individual patients. This study validates ATX in a dataset of 892 clinically low-risk patients from Dordrecht, Netherlands. Of the 892 patients, 299 did not receive adjuvant therapy. The discriminative performance of ATX was assessed using C-index and its stratification ability was evaluated by log-rank tests.
Results
ATX achieved a C-index of 0.71 and a 5-year AUC of 0.71, demonstrating strong discrimination. Among 299 patients who received no adjuvant therapy, ATX achieved a C-index and AUC of 0.78 and 0.81 respectively. ATX scores were used to stratify patients into risk groups. Notably, untreated and treated ATX low-risk patients had comparable 5-year recurrence-free survival (RFS) (untreated: RFS = 96%, 95% CI = 92-97%; treated: RFS = 96%, 95% CI = 93-97%) with identical 10-year RFS (86%, 95% CI = 83-92% for both), suggesting ATX low-risk status may identify a subgroup with favorable prognosis independent of treatment.
Conclusion
ATX provides robust prognostic stratification in an external cohort and identifies a subgroup of patients who did not receive systemic therapy with favorable observed outcomes. These results support prospective validation of ATX as a decision-support tool for adjuvant therapy de-escalation in HR+/HER2- early breast cancer.
