Abstract PS3-06-03: Refining adjuvant CDK4/6 inhibitor eligibility in early HR+ HER2- breast cancer with artificial intelligence

Key Points

• The study aims to refine the eligibility criteria for adjuvant CDK4/6 inhibitors in early HR+ HER2- breast cancer using an artificial intelligence approach.
• Utilized the Ataraxis Breast (ATX) AI tool to assess recurrence risk scores from H&E slides and clinical data.
• Analyzed data from 2,228 HR+ HER2- patients across five institutions.
• Defined Clinical High-Risk and Low-Risk groups based on the NATALEE trial eligibility criteria.
• Evaluated disease-free intervals using hazard ratios and concordance indices.
• 41% of patients were classified as Clinical High-Risk and eligible for CDK4/6 inhibitors.
• In the Clinical High-Risk group, ATX scores showed a hazard ratio of 1.81, indicating significant prognostic value.
• In the Clinical Low-Risk group, the ATX scores also demonstrated prognostic ability with a hazard ratio of 2.53.
• ATX identified high-risk patients who wouldn't qualify based on existing criteria, potentially improving treatment targeting.

Abstract