Abstract Background: Pathological complete response (pCR) after neoadjuvant treatment (NAT) associates with improved outcomes in triple-negative breast cancer (TNBC), yet a subset of patients still experience relapse. We previously reported on risk and pattern of relapse in up to 733 patients with pCR after NAT (Massa D, SABCS 2024 and ESMO Breast 2025). Here, we present updated results on 1190 patients, with detailed analysis of site-specific relapse risk. Patients and Methods: 2458 patients with stage I-III TNBC (ER10%, PgR 10%, HER2-negative) treated with NAT were included in the multicentric real-world GAMBIT study across 18 European centers. Of these, 1193 obtained a pCR (48.5%, pCR cohort) and 1265 had residual disease (RD cohort). TILs were assessed on treatment-naïve biopsies for 691 patients with pCR (pCR-TILs cohort). We collected the first site of distant relapse and analyzed site relapse patterns by competing risk analysis. Primary survival endpoint was distant relapse-free survival (DRFS). Results: In the pCR cohort, clinical nodal involvement (cN+) was independently associated with worse outcomes (DRFS HR 2.38 95%CI 1.14-4.97; overall survival OS HR 3.19 95%CI 1.53-6.66). In the pCR-TILs cohort, cN and TILs were independently prognostic in multivariable models. The co-occurrence of cN+ and TILs30% (cN+/low-TILs) identified a subgroup of patients with distinctively high risk of relapse and death: 5-year DRFS 83.4% and OS 85.8% versus ≥95% and ≥97.3% in other strata (cN+/high-TILs; cN0/low-TILs; cN0/high-TILs). 5-year cumulative incidence of first distant relapse in pCR cohort, pCR groups defined by cN and TILs, and RD cohort are shown in the Table. While 5-year cumulative incidence of any CNS metastases as first site of relapse was lower in the overall pCR cohort vs the RD cohort (3.6% vs 7.0%, p=0.001), there was no difference in the incidence of CNS-only relapse without extracranial disease: 2.6% vs 2.1%; p=0.445. Moreover, within the pCR-TILs cohort, patients with cN+/low-TILs had 5-year cumulative incidence of CNS metastases that were comparable (and even worse for CNS-only) to patients with RD. Conclusions: Patients with TNBC and pCR after NAT are heterogeneous with regards to outcome, and the co-occurrence of low TILs and cN+ identify a subgroup of patients at clinically meaningful risk of relapse, and enriched in CNS metastases. These results challenge the assumption that pCR confers uniformly excellent prognosis and support risk-adapted allocation in post-pCR clinical trials, including escalation strategies and studies of CNS-active agents. Citation Format: M. Dieci, T. Foukakis, S. Giacchetti, C. Desmedt, J. Frenel, E. Gasparini, M. Kok, P. Vigneri, W. Jacot, C. Vernieri, M. Lambertini, F. Patanè, F. Piacentini, E. Bria, A. Zambelli, D. Trapani, A. Botticelli, O. Garrone, D. Massa, S. Lando, A. Matikas, L. Someil, G. Floris, M. Campone, M. Ragazzi, E. Lips, F. Martorana, G. Bonomi, A. Papakostantinou, L. Nicolé, I. Zerdes, P. Neven, M. Rotolo, H. Horlings, S. Nucera, A. Gudin de Vallerin, A. Vingiani, A. Dei Tos, V. Guarneri. Risk and Pattern of Relapse in Triple-Negative Breast Cancer with Pathological Complete Response after Neoadjuvant Treatment: Updated Results from the European GAMBIT Real-world Study abstract. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS4-12-24.
Dieci et al. (Tue,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: