Real-world associations of pathological complete response and survival with neoadjuvant pembrolizumab in early TNBC: a Turkish Oncology Group study

Abstract

Triple-negative breast cancer (TNBC) is an aggressive subtype with limited targeted treatment options and a high risk of recurrence. Neoadjuvant chemotherapy plus pembrolizumab (NACP) is a standard approach to increase pathological complete response (pCR) rates and improve long-term outcomes, but real-world predictors of response and survival remain insufficiently defined. This multicenter retrospective real-world study included 164 patients with early-stage TNBC treated with NACP followed by surgery. The primary endpoint was pCR, and secondary endpoints were event-free survival (EFS) and overall survival (OS). Multivariable logistic regression was used to identify independent predictors of pCR, and Cox proportional hazards models were applied to evaluate factors associated with EFS and OS. pCR was achieved in 60.4% of patients. Over a median follow-up of 49.2 months (range, 4-73.8 months), recurrence occurred in 18.3% and mortality in 3.0%. In multivariable Cox analysis, weekly carboplatin scheduling, younger age, and achievement of pCR were associated with longer EFS. However, these associations were derived from a retrospective analysis without adjustment for center-level effects or relative dose intensity and should therefore be considered hypothesis-generating. Owing to the very low number of deaths (n=5), OS analyses were severely underpowered, and multivariable modeling was not statistically reliable. Therefore, OS findings should be interpreted as descriptive and hypothesis-generating rather than confirmatory. In multivariable logistic regression, weekly carboplatin administration was associated with higher odds of pCR; however, given the retrospective design and potential confounding factors, these findings should be interpreted as exploratory. In an exploratory subgroup analysis, no EFS events were observed among patients who achieved pCR and received adjuvant pembrolizumab; however, these findings are hypothesis-generating because of sparse events and potential confounding by indication. In this real-world multicenter cohort, NACP achieved high pCR rates, and pCR was associated with longer EFS in this cohort. Weekly carboplatin scheduling was associated with higher pCR rates and longer EFS; however, given the retrospective and non-randomized design, this observation should be considered exploratory rather than indicative of a modifiable causal effect. Prospective, adequately powered, biomarker-integrated trials are required to determine which patients derive incremental benefit from adjuvant pembrolizumab beyond the neoadjuvant component, particularly when stratified by pathological complete response (pCR) status. Future studies should clarify whether patients achieving pCR may safely omit adjuvant immunotherapy and, conversely, whether patients with residual disease require treatment intensification strategies guided by molecular and immune biomarkers. Such risk-adapted approaches are essential to optimize benefit while minimizing overtreatment in early-stage TNBC. The observational nature of the study precludes causal inference, and prospective validation is required before treatment modification strategies can be recommended.