Abstract 3864: Serial ctDNA monitoring in metastatic breast cancer using an ultrasensitive tumor informed structural variant based assay.

Abstract Background: Circulating tumor DNA (ctDNA) dynamics correlate with treatment response in metastatic breast cancer (mBC), yet evidence generated using tumor-informed, ultrasensitive assays remains limited. In addition, the landscape of serial ctDNA monitoring across different receptor subtypes and through multiple lines of therapy remains poorly characterized. Furthermore, the performance of structural variant (SV) based ctDNA detection has not been established in this setting. Validating ctDNA as a reliable biomarker of response and progression in mBC, particularly across treatment transitions, could provide clinical utility. Methods: We conducted a single-center retrospective study in patients with mBC receiving standard-of-care systemic therapy. Serial blood samples were collected prospectively at enrollment, radiographic restaging events (+/-30 days), and at disease progression. ctDNA was analyzed retrospectively using an ultrasensitive, whole genome sequencing (WGS)-based, tumor-informed SV digital PCR assay. Clinical data, real-world progression-free survival (rwPFS) across sequential lines of therapy, and overall survival (OS) were obtained from the medical record. Results: Results were available for 28/100 patients (18 ER+/HER2-, 6 TNBC, 4 HER2+), contributing 120 evaluable timepoints (median 3 per patient, range 1-18) with a median follow-up of 12.2 months (range 2.0-16.3). Patients received a median of 2 lines of metastatic therapy (range 1-8). Orthogonally validated fingerprints contained a median of 13 SVs (range 4-16). ctDNA was detected in 77% (92/120) of all sampled timepoints, with 34% (31/92) of positives falling within the ultrasensitive range (variant allele fraction ≤0.01% or ≤100 ppm). Serial ctDNA monitoring closely mirrored each patient’s treatment course. Notably, no radiologist-adjudicated progression occurred without a preceding rise in ctDNA from on-treatment nadir. Four patients (1 ER+/HER2-, 1 TNBC, 2 HER2+) had undetectable ctDNA with a median rwPFS of 27 months (range 20-40) and remain on therapy without rwPFS or OS events. Accrual is ongoing, with continued serial plasma collection and maturing rwPFS/OS data. Conclusion: These preliminary findings demonstrate the feasibility of ctDNA monitoring in a real-world, heterogeneous mBC cohort using a tumor-informed, ultrasensitive SV-based assay. ctDNA dynamics closely tracked treatment response across sequential lines of therapy, and early signals suggest that ctDNA clearance below the ultrasensitive detection threshold is associated with prolonged disease control. The high proportion of timepoints within the ultrasensitive range underscores the relevance of assay sensitivity for metastatic monitoring. Updated results from the full cohort, including correlation with all matched radiographic restaging events will be presented. Citation Format: Mitchell J. Elliott, June Roh, Kareena Thakur, Sasha Main, Karen Howarth, Sofia Birkeälv, Jennifer Yen, Völundur Hafstad, Wendy Levin, Vikash Kumar, Long Nguyen, Meredith Li, Scott Victor Bratman, Eitan Amir, Philippe L. Bedard, Lillian L. Siu, Hal Berman, David Cescon. Serial ctDNA monitoring in metastatic breast cancer using an ultrasensitive tumor informed structural variant based assay abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 3864.