Abstract Background: Neoadjuvant chemotherapy (NACT) has been more frequently used in breast cancer, but it’s controversial whether NACT or adjuvant chemotherapy (ACT) is more effective in improving survival outcomes for triple-negative breast cancer (TNBC) patients. Recent studies also proposed pathologic complete response (pCR) could not be a robust surrogate for survival outcomes in certain types of breast cancer. To address these questions, a general meta-analysis and a network meta-analysis (NMA) were conducted to compare the survival outcomes of patients with pCR, non-pCR after NACT and ACT. The cut-off pCR rate was also declared to indicate when NACT produces equivalent survival outcomes to ACT (Registration: PROSPERO CRD42022336732). Methods: Databases including PubMed, Embase, Web of science and Cochrane library were searched up to June 2023 to investigate studies comparing NACT and ACT, as well as randomized controlled trials (RCTs) comparing different regimens in NACT or ACT settings in early operable TNBC patients. Heterogeneity was assessed using χ² based Q-test and I², combined with hazard ratios (HRs) with 95% confidence intervals (CI) computed for overall survival (OS), and disease-free survival (DFS, or event-free survival). The NMA with a Bayesian framework was conducted using both random and fixed effect model. The Weighted Least Square method and the Least Absolutely Deviation method were used to determine the cut-off pCR rates of OS and DFS. Results: A total of 35 studies involving 21 RCTs and 34143 TNBC patients were included. The general meta-analysis comprised 14 cohort studies. Eight high-quality cohort studies with propensity score matching and 21 RCTs were included in NMA. The pooled pCR rate for NACT was 42.2% (95% CI 37.8%-46.8%). Overall, the NACT cohort showed significantly worse OS and DFS than ACT cohort (HR=1.67, 95% CI 1.22-2.31; HR=1.37, 95% CI 1.14-1.64). However, patients with pCR after NACT showed improved OS than ACT (HR=0.58, 95% CI 0.50-0.66), while the patients without pCR after NACT had pooper OS than ACT (HR=2.03, 95% CI 1.88-2.19). As for DFS, there was no statistically difference between pCR after NACT and ACT groups (HR=1.04, 95% CI 0.63-1.73), while patients without pCR after NACT had significantly worse DFS than ACT (HR=2.65, 95% CI 1.98-3.55). From the NMA, the OS and DFS were better when combining modern target therapy such as Bevacizumab, PARPi, and PD-1/PD-L1 inhibitor (Figure 1), but not achieving pCR completely negated the benefits of newer drugs. Adding adjuvant capecitabine to patients not achieving pCR seemed to favor the OS. It was predicted that when the pCR rates were 64.7% and 46.9% (Figure 2), respectively, the OS and DFS of patients treated with NACT would be the same as those of patients treated with ACT. Conclusions: Higher pCR rate after NACT was associated with improved OS compared with ACT in TNBC patients. However, failure to achieve pCR after NACT resulted in worse survival outcomes than ACT. NACT had similar survival outcomes with ACT only when the pCR rate was at least 46.9%. The development of effective NACT regimens is beneficial for breast cancer patients. Figure 1. Forest plot of the Hazard Ratio for OS and DFS in TNBC patients pCR: pathological complete response; Bev: Bevacizumab; Pt: platinum; NACT: neoadjuvant chemotherapy; ACT: adjuvant chemotherapy; vel: veliparib; PDi:PD1/PDL1 inhibitor; X: capecitabine Figure 2. The cut-off value of pCR when NACT and ACT have the same Hazard Ratio based on OS and DFS. pCR: pathological complete response; Bev: Bevacizumab; Pt: platinum; NACT: neoadjuvant chemotherapy; ACT: adjuvant chemotherapy Citation Format: Xiaoying Qiao, Shu Wang, Taobo Hu, Baosheng Liang. Rethinking the value of pathologic complete response rate for the survival outcomes of early triple-negative breast cancer treated with neoadjuvant chemotherapy: a systematic review and network meta-analysis [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO1-28-09.
Read full abstract