Real-world data on breast pathologic complete response and disease-free survival after neoadjuvant chemotherapy for hormone receptor-positive, human epidermal growth factor receptor-2-negative breast cancer: a multicenter, retrospective study in China

Abstract

Background: The data in the real-world setting on breast pathologic complete response (pCR) after neoadjuvant chemotherapy (NAC) for hormone receptor–positive, human epidermal growth factor receptor-2-negative (HR+, HER2−) breast cancer (BC) is limited. The present study aims to screen for some predictors and investigate the prognostic significance of breast pCR after NAC in HR+, HER2− BC in China. Methods: This was a multicenter, retrospective study. In this study, three hundred eighty-four HR+, HER2− BC patients who received NAC were enrolled between 2010 and 2016 from Shanghai Jiaotong University Breast Cancer Database (SJTU-BCDB). These patients were dichotomized according to the presence of breast pCR after NAC. Logistic analysis was used to screen for predictors associated with breast pCR. Kaplan-Meier (K-M) curve and a propensity score matching (PSM) analysis were performed to compare the disease-free survival (DFS) between the two groups. Cox regression was used to analyze the prognostic significance of breast pCR on DFS in HR+, HER2− BC. A nomogram model was established to predict the probability of DFS at 1, 3, and 5 years after NAC. Results: Fifty-seven patients (14.8%) achieved breast pCR. Univariate analysis showed that tumor size, estrogen receptor (ER), progesterone receptor (PR), and Ki67 were associated with breast pCR. Further, multivariate analysis showed that tumor size, PR, and Ki67 remained statistically significant. K-M curves showed a statistical difference between the breast pCR and non-pCR groups before PSM (p = 0.047), and a more significant difference was shown after PSM (p = 0.033). Cox regression after PSM suggested that breast pCR, adjuvant ET, clinical T stage, and Ki67 status were the significant predictive factors for DFS in HR+, HER2− BC patients. The adjusted hazards ratio (aHR) for breast pCR was 0.228 (95% CI, 0.070~0.739; p = 0.014), for adjuvant endocrine therapy was 0.217 (95% CI, 0.059~0.801; p = 0.022), for Ki67 was 1.027 (95% CI, 1.003~1.052; p = 0.027), for cT stages 2 and 3 compared with 1, the values were 1.331 (95% CI, 0.170~10.389), and 4.699 (95% CI, 0.537~41.142), respectively (p = 0.043). A nomogram was built based on these significant predictors, providing an integrated probability of DFS at 1, 3, and 5 years. The values of area under the receiver operating characteristic (ROC) curve (AUC) were 0.967, 0.991, and 0.787, at 1 year, 3 years, and 5 years, respectively, demonstrating the ability of the nomogram to predict the DFS. Conclusions: This real-world study demonstrates that tumor size, PR, and Ki67 were independent predictive factors for breast pCR in HR+, HER2− BC. Breast pCR after NAC was an independent predictor for DFS in HR+, HER2− patients, regardless of a change in nodes. Furthermore, the nomogram built in our study could predict the probability of individualized DFS in HR+, HER2− BC patients.