To investigate the correlation between Ki-67 expression change after neoadjuvant chemotherapy (NAC) and the prognosis of HER-2-positive breast cancer patients.

We retrospectively analyzed the clinical data of 85 HER-2-positive breast cancer patients in the First People’s Hospital of Foshan from January 2014 to December 2017. All patients were diagnosed with HER-2-positive invasive ductal carcinoma of the breast by preoperative core-needle aspiration biopsy, and then received NAC. The relationship of Ki-67 expression before NAC with clinicopathological characteristics and the efficacy of NAC was analyzed by χ² test and Fisher’s exact test. The clinical and pathological efficacy of NAC in breast cancer patients receiving different regimens (TcbH group: docetaxel + carboplatin + trastuzumab; AC-TH group: epirubicin + cyclophosphamide followed by docetaxel + trastuzumab) was compared by χ² test and Fisher’s exact test. According to hormonal receptor (HR) status after NAC, the patients were divided into HR-negative group and HR-positive group, and the clinical parameters between two groups were compared by χ² test and Fisher’s exact test. The correlation between the expression changes of Ki-67 after NAC and the efficacy was analyzed by χ² test and Fisher’s exact test, the survival curve was drawn using the Kaplan-Meier method and the Log-rank method was used to compared RFS between groups. The subgroup analysis was performed according to the HR status. Univariate and multivariate Cox proportional hazards regression model was used to analyze the influencing factors of recurrence.

Before NAC, 22 patients had low expression of Ki-67 and 63 had high expression. There was no significant difference in age, menstrual status, histological grade, lymph node metastasis, ER status, PR status, pathological efficacy, tumor size and clinical efficacy between patients with low Ki-67 expression and patients with high Ki-67 expression (χ²=0.000, 0.296, 0.186, 0.276, 0.010, 0.021, 1.401, all P>0.050; P=0.646, 0.569). There was no significant difference in clinical efficacy and pathological efficacy between patients receiving TcbH and patients receiving AC-TH (P=0.154; χ²=0.232, P=0.630). After NAC, HR was negative in 51 cases and positive in 34 cases. There was no significant difference in surgical methods, Miller-Payne grade, radiotherapy, recurrence/metastasis, Ki-67 change and NAC regimen between HR-positive and HR-negative patients (P=0.157; χ²=2.215, 3.266, 0.095, 0.516, 0.297, all P>0.050). After NAC, the expression of Ki-67 was increased or unchanged in 21 cases, and decreased in 64 cases. The clinical efficacy and pathological efficacy of patients with decreased Ki-67 were significantly better than those of patients with increased or unchanged Ki-67 (P=0.003; χ²=8.729, P=0.003). The results of subgroup analysis showed that among 51 HR-negative patients, the pathological efficacy of patients with decreased Ki-67 was significantly better than that of patients with increased or unchanged Ki-67 (χ²=11.141, P=0.001), but clinical efficacy presented no significant difference (P=0.071); among 34 HR-positive patients, the clinical efficacy of patients with decreased Ki-67 was significantly better than that of patients with increased or unchanged Ki-67 (P=0.037), but there was no difference in pathological efficacy (P=0.672). Survival analysis showed that the RFS of patients with decreased Ki-67 after NAC was significantly higher than that of patients with increased or unchanged Ki-67 (χ²=26.275, P<0.001); in HR-negative and HR-positive subgroups, the RFS of patients with decreased Ki-67 after NAC was significantly higher than that of patients with increased or unchanged Ki-67 (χ²=11.302, 22.127, both P<0.001). The Cox hazard proportional regression model analysis showed that histological grade 3 and lymph node metastasis were independent risk factors for recurrence (HR=2.764, 3.550, 95%CI: 1.104-6.919, 1.026-12.281, P=0.030, 0.045). The decreased Ki-67 expression after NAC was an independent protective factor for recurrence (HR=0.197, 95%CI: 0.087-0.475, P<0.001).

The change of Ki-67 expression is correlated with clinical and pathological efficacy of NAC, and it may be considered as an independent factor affecting RFS of HER-2-positive breast cancer patients.