Elisa Agostinetto MD; 2026 EBCC: Circulating Tumor DNA Trumps Clinical Prognostic Markers After Neoadjuvant Therapy for Patients with Early Breast Cancer

Circulating Tumor DNA Trumps Clinical Prognostic Markers After Neoadjuvant Therapy for Patients with Early Breast Cancer

An interview with:

Elisa Agostinetto MD, Institut Jules Bordet, Brussels, Belgium

BARCELONA, Spain—Among patients being treated with neoadjuvant therapy for their early breast cancers, circulating tumor DNA performed better as an independent prognostic marker for predicting relapse and disease progression than clinical markers such as pathological complete remission. This was according to prospective study data from Belgium and Italy reported at the 2026 European Breast Cancer Conference by Elisa Agostinetto MD, a medical oncologist from the Institut Jules Bordet in Brussels. She discussed her findings with Audio Journal of Oncology reporter, Peter Goodwin:

AUDIO J0URNAL OF ONCOLOGY: Elisa Agostinetto MD

IN: [GOODWIN] “I’m here at the ……

OUT:  ……of Oncology, I’m Peter Goodwin  10:04secs

EBCC 2026:

Abstract no: 12

Circulating tumor DNA at completion of neoadjuvant therapy is an independent prognostic marker: an individual patient-level pooled analysis of two prospective studies

Agostinetto1, V. Appierto2, P. Minicozzi3, C. Sotiriou1, F. Rothé1, E. Tamborini2, F. Lebrun4,
Belfiore2, D. t’Kint4, L. De Cecco5, L. Buisseret4, M.C. De Santis6, A. Gombos4, G. Bianchi7,
Aftimos4, P. Verderio3, D. Vincent4, G. Pruneri2, M. Ignatiadis4, S. Di Cosimo MD2

1Institut Jules Bordet- Université libre de Bruxelles ULB- Hôpital Universitaire de Bruxelles H.U.B, Breast Cancer Translational Research Laboratory, Brussels, Belgium

2Fondazione IRCCS Istituto Nazionale dei Tumori, Advanced Diagnostics, Milan, Italy

3Fondazione IRCCS Istituto Nazionale dei Tumori, Epidemiology and Data Science, Milan, Italy

4Institut Jules Bordet- Université libre de Bruxelles ULB- Hôpital Universitaire de Bruxelles H.U.B, Medical Oncology, Brussels, Belgium

5Fondazione IRCCS Istituto Nazionale dei Tumori, Experimental Oncology, Mian, Italy

6Fondazione IRCCS Istituto Nazionale dei Tumori, Radiation Oncology, Milan, Italy

7Fondazione IRCCS Istituto Nazionale dei Tumori, Medical Oncology, Milan, Italy

Background:

Circulating tumor DNA (ctDNA) is a promising biomarker in early breast cancer. However, limited sample sizes in available clinical studies weaken neoadjuvant evidence.

Materials and methods: Data from two independent prospective observational studies, one at the Institut Jules Bordet (Brussels), and one at the Istituto Nazionale dei Tumori (Milan), were pooled at the individual patient level for joint analysis. In both studies, women with early breast cancer received neoadjuvant therapy and underwent primary tumor-informed ctDNA assays at predefined time points: baseline (before initiation of neoadjuvant therapy), end of neoadjuvant therapy before surgery (EoT), and during follow-up. Associations between ctDNA detection and clinico- pathological variables (age, tumor size [e.g. T ≤5 cm vs. >5 cm], hormone receptor status, HER2 status, and pathological complete response (pCR) were evaluated using χ2 or Wilcoxon tests, as appropriate. The effect of ctDNA on event-free survival (EFS) was analyzed by univariable and multivariable Cox proportional hazards models adjusted for relevant covariates. A two-sided p value< 0.05 was considered statistically significant.

Results:

A total of 81 patients were analyzed. Median age was 48 (range 27-75) years at diagnosis; most had T≤5 cm (72%), node positive (68%) and triple negative (60%) disease. Breast cancer events were 26 throughout a median follow-up of 7 years (IQR 5.3-8.8). ctDNA was detected in 31/54 (57.4%) plasma samples at baseline, and in 11/64 (17%) plasma samples at the EoT. The detection of ctDNA was significantly associated with hormone receptor-negative status both at baseline and at the EoT (p<0.05). No association with pCR was observed at any time point. Patients with baseline ctDNA detection showed a non significant trend toward worse EFS (HR 1.56, 95% CI 0.58-4.22). Notably, detection of ctDNA at the EoT predicted breast cancer events and remained independently associated with decreased EFS even after adjustment for all other clinicopathological variables including pCR (HR 3.58, 95% CI 1.33-9.60).

Conclusions:

This individual patient-level pooled analysis includes one of the largest numbers of events reported to date in the ctDNA literature. ctDNA predicted breast cancer relapse, particularly when detected at the end of treatment, supporting the use of ctDNA for post-neoadjuvant risk stratification and subsequent therapeutic strategies.