Circulating tumor DNA (ctDNA) dynamics as a predictor of treatment response in metastatic breast cancer (mBC).
Abstract
1011 Background: ctDNA testing has emerged as a prognostic and predictive biomarker in the management of mBC. However, the relationship between ctDNA trends and real-world treatment outcomes has yet to be fully characterized. Here, we utilized a claims database to evaluate the association between ctDNA trends and the time to next treatment (TTNT) in patients with mBC. Methods: We utilized Natera's proprietary real-world database linked to commercially available claims data to identify patients who received treatment for mBC and had ctDNA testing performed commercially using a clinically validated, personalized, tumor-informed mPCR-NGS ctDNA assay (Signatera TM , Natera, Inc.). Insurance claim codes for treatment regimens were used to determine BC receptor subtype and therapy dates. Treatment lines were included in the analysis if a ctDNA test result was available within 4 weeks before treatment initiation (T1) and a subsequent ctDNA test result was available 2-6 weeks after treatment initiation (T2). TTNT was calculated as the time from initiation of the first treatment to the subsequent treatment. T1 to T2 ctDNA dynamics were analyzed using the Student’s t-test and were categorized as favorable (persistently negative, ctDNA-clearance, ctDNA-decrease) or unfavorable (ctDNA-negative to positive, ctDNA-increase). Results: A total of 7,222 treatment lines were assessed for duration of treatment and corresponding ctDNA dynamics, including3,117 lines (N=2,362 patients) for HR+/HER2- mBC, 3,717 lines (N=1,943 patients) for HER2+ mBC, and 888 lines (N=605 patients) for TNBC. Of these, 448 treatment lines met the inclusion criteria for ctDNA analysis. In HER2+ breast cancer, TTNT across 226 treatment lines was significantly longer in patients with favorable ctDNA dynamics (6.7 [3.18–10.3] months) relative to unfavorable dynamics (2.7 [1.4–5.1] months; p<0.0001). Among patients with HR+/HER2- mBC, TTNT across 156 treatment lines was longer in those with favorable ctDNA dynamics (median [Q1, Q3]: 7.51 [3.72–11.57] months) compared to unfavorable dynamics (5.02 [1.8–9.84] months; p=0.052). A similar trend was observed in TNBC, where TTNT across 66 treatment lines was longer with favorable ctDNA dynamics (6.03 [2.89–10.07] months) compared to unfavorable dynamics (2.7 [1.16–5.89] months; p=0.381), though this was not statistically significant. Conclusions: Early on-treatment ctDNA dynamics, assessed within the first 6 weeks of therapy, was associated with TTNT in a real-world ctDNA monitoring setting across different mBC subtypes and therapeutic regimens. An early rise in ctDNA levels correlated with the shortest TTNT, whereas ctDNA clearance was associated with the longest TTNT intervals. These findings highlight the potential of serial ctDNA testing in mBC for monitoring treatment response and informing clinical decisions.
Article Details
Journal Info
Journal of Clinical Oncology
Lippincott Williams & Wilkins
Authors (12)
Pedram Razavi
Luc Cabel
Jimmitti Teysir
Memorial Sloan Kettering Cancer Center, New York, NY
Julia Ah-Reum An
Samuel Rivero-Hinojosa
Sandro Satta
1University of Miami, Medicine, Miami, United States
Sara L. Bristow
Natera, Inc., Austin, TX
Ekaterina Kalashnikova
Angel A. Rodriguez
Natera, Inc., Austin, TX
Minetta C. Liu
Mark E. Robson
Sarat Chandarlapaty