Ultrasensitive ctDNA monitoring during CDK4/6 inhibitor therapy for metastatic breast cancer.

Abstract

1073 Background: The combination of CDK4/6 inhibitor (CDK4/6i) and endocrine therapy (ET) is the standard first-line treatment for patients with hormone receptor-positive/HER2-negative (HR+/HER2-) metastatic breast cancer (MBC). However, it exhibits highly variable efficacy, with some cancers progressing within 3–6 months while many others achieve durable and potentially indefinite complete responses (CRs). While pharmacologic strategies to escalate or deescalate this therapy exist, diagnostic tools to identify the patients who would benefit from each approach are needed. Ultrasenstive ctDNA offers the potential to assess disease burden dynamically and with more precision. In this study, we evaluate the validity of an ultrasensitive assay capable of detecting ctDNA levels in the parts per million range for monitoring patients with HR+/HER2- MBC. Methods: Patients from the MSK-LINC prospective ctDNA monitoring study, who received CDK4/6i+ET for HR+/HER2- MBC were included in the study. MRD monitoring was performed using personalized tumor-informed panels designed from whole genome sequencing (WGS) of matched tumor and normal specimens to identify up to 2,000 somatic alterations for each patient using the Precise MRD assay (Myriad Genetics). Results were reported as an overall ctDNA detection status and a quantitative tumor fraction. Results: 29 patients with HR+/HER2- MBC (8 de novo , 21 recurrent) were included in this ongoing study. The median progression-free survival (PFS) was 48.8 months (range 2.6 – 78.5) with 17/29 of patients experiencing disease progression. ctDNA panels were successfully designed for all cases, and 140/146 (95.9%) plasma samples passed QC. All pre-treatment samples had detectable ctDNA with a median tumor fraction of 1.4% (range 0.00093%, 14.0%). An early decrease in ctDNA levels, > 50% reduction from baseline or levels < 0.01% in the second sample collected within 3 months, was significantly associated with longer PFS (p < 0.001). We focused on 7 patients who achieved radiographic CR all with PFS > 3y. Notably, 3 patients had continued to have ultra low levels of ctDNA (median: 0.0086%, range 0.00032%, 0.11%), indicating stable viable micrometastatic disease below the threshold of imaging, effectively controlled by treatment. In contrast, 4 patients also achieved molecular CR (mCR) defined as sustained undetectable ctDNA suggesting that metastatic disease was either eradicated or rendered dormant without significant cell turnover. Conclusions: Ultrasensitive ctDNA monitoring is a promising tool for monitoring disease burden and treatment response. Our results highlight the ability of ctDNA to distinguish between stable molecular disease vs. mCR, highlighting the potential of ctDNA as a biomarker for tailoring treatment strategies in patients who achieve outstanding clinical responses.