← Cancer trial database
Trial result
Breast cancer

A tumor-informed ctDNA MRD assay flags post-surgery recurrence risk in early TNBC (HR 30.3)

Cancer type
Breast Cancer
Published in
New in Oncology - September 24, 2026

This NSABP B-59/GBG-96-GeparDouze prospective ctDNA substudy used whole-exome sequencing of each patient’s primary tumor to build a bespoke “tumor-informed” circulating tumor DNA (ctDNA) MRD assay, testing serial blood during/after neoadjuvant chemotherapy (± atezolizumab, an anti–PD-L1 checkpoint inhibitor). In 160 stage II/III TNBC patients, baseline ctDNA positivity was 96% (153/160) and fell to 6% postsurgery (9/147), but postsurgery ctDNA+ strongly predicted distant recurrence: 77.8% (7/9) vs 5.0% (7/138) for ctDNA− (HR 30.3, 95% CI 10.4–88.7; log-rank P<1E-16) at a median 37-month follow-up; among those with residual invasive disease, distant recurrence was 85.7% (6/7) if ctDNA+ vs 11.9% (5/42) if ctDNA−—a risk split far sharper than what clinicians typically get from pCR/residual disease alone, in a setting where current immuno-chemo standards (e.g., neoadjuvant/adjuvant pembrolizumab regimens) improve event-free outcomes but still leave a substantial non-pCR relapse tail.

Take home

Postsurgery ctDNA separated recurrence risk remarkably well despite the very small ctDNA-positive group, making this a compelling biomarker for trials testing whether MRD can identify which non-pCR patients should receive treatment escalation—or potentially de-escalation—rather than evidence yet that treatment should be changed based on ctDNA alone.

Source: Read the original publication

More in breast cancer

See all entries in breast cancer →