e12680 Background: Although adjuvant capecitabine may improve outcomes among patients with residual triple-negative breast cancer (TNBC) after neoadjuvant therapy, its real-world effectiveness and treatment selection patterns remain uncertain. Methods: Adults with TNBC (ER/PR ≤10%, HER2-) treated with ≥1 cycle of neoadjuvant therapy between 2014 and 2024 and ≥6 months follow-up after surgery were included. Patients with inflammatory disease or prior anti-HER2/endocrine therapy were excluded. Pathologic stage was defined by the AJCC 8th edition. Distant relapse-free survival (DRFS) and overall survival (OS) were estimated using the Kaplan-Meier method and compared using the log-rank test. Multivariable Cox regression was conducted in the contemporary cohort to adjust for pathologic stage and neoadjuvant regimen. Results: A total of 188 patients met eligibility criteria. Median age was 54 years (IQR 44-65), and all patients were female. Residual disease (RCB I-III) was present in 123 patients (65.4%), of whom 62 (50.4%) received adjuvant capecitabine. Compared with patients who did not receive capecitabine, treated patients had significantly greater residual disease burden (RCB II-III: 95% vs 70%, p = 0.001) and higher pathologic stage (AJCC II-III: 63% vs 38%, p = 0.013). The median time from surgery to capecitabine initiation was 90.5 days (IQR 76.0–112.3), and 37.1% initiated therapy within 12 weeks. Median cumulative dose was 882 mg/m² (IQR 761-1000) over 7 cycles (IQR 6–8); dose reductions and delays occurred in 46.8% and 29.0%, respectively, most commonly due to toxicity (61.1%). Median follow-up was 42 months (95% CI 37-47). Three-year DRFS was 56.8% in the capecitabine group vs 87.2% in the non-capecitabine group (p = 0.014), and 3-year OS was 69.4% vs 88.4%, respectively (p = 0.024). In multivariable analysis restricted to patients treated from 2017 onward (n = 112; 29 DRFS events, 25 OS events), adjuvant capecitabine was not independently associated with improved outcomes (DRFS: HR 2.48, 95% CI 0.98-6.33, p = 0.057; OS: HR 2.05, 95% CI 0.78-5.40, p = 0.145). Higher pathologic stage remained independently associated with worse DRFS (stage III vs I: HR 3.84, 95% CI 1.27-11.64, p = 0.017) and OS (stage III vs I: HR 4.17, 95% CI 1.32-13.23, p = 0.015). Conclusions: Use of adjuvant capecitabine in routine practice was strongly influenced by residual disease burden and pathologic stage, indicating that treatment selection was driven primarily by perceived recurrence risk rather than standardized criteria. The inferior outcomes observed in unadjusted analyses therefore, most likely reflect confounding by indication rather than treatment harm or lack of efficacy. Future integration of molecular phenotype data may further refine post-neoadjuvant risk stratification and help identify biologically defined subgroups most likely to benefit from adjuvant capecitabine.
Wong et al. (Thu,) studied this question.