Overview
CDK4/6 inhibitors in combination with endocrine therapy are a preferred first-line treatment for patients with hormone-receptor positive (HR+), HER2 negative (HER2-) metastatic breast cancer. However, some patients experience rapid disease progression within the first year despite treatment, representing a major unmet clinical need. Identifying which patients will progress rapidly before treatment starts could enable more personalized treatment strategies and clinical trial designs.
Researchers used Flatiron Health’s US Breast Cancer Panoramic Database, inclusive of more than 985,000 patients with breast cancer, to analyze data from over 10,800 patients with HR+/HER2- metastatic breast cancer who received first-line CDK4/6 inhibitors. They developed three machine learning models using routinely collected electronic health record data (patient demographics, clinical characteristics, laboratory values) to predict which patients would progress within 12 months. The models performed similarly well, with the ability to identify high-risk and low-risk groups with clinically meaningful and significantly different progression-free survival.
Key predictors of rapid progression included advanced age, specific genetic mutations (BRCA, ESR1, AKT), newly diagnosed stage IV disease, shorter time from metastatic diagnosis, higher disease burden, abnormal liver enzymes, and bone-only metastases.
Why this matters
These findings establish that rapid progressors in metastatic breast cancer can, in many cases, be predicted using data readily available from electronic health records. This suggests an approach to helping oncologists identify high-risk patients upfront, enabling trial enrichment strategies and informing individualized treatment decisions. Further development could ultimately improve outcomes for breast cancer patients at risk for aggressive disease progression.