To explore the prediction value of pre-treatment systemic immune-inflammation index (SII) for pathological complete response(pCR) in breast cancer patients undergoing neoadjuvant chemotherapy (NAC), and to establish a clinical prediction model based on SII and other relevant clinicopathological characteristics.

We retrospectively collected the clinical data of 157 breast cancer patients undergoing NAC in the Department of Thyroid and Breast Surgery, Zhongnan Hospital, Wuhan University from January 2019 to April 2022. The predictive value of SII for pCR after NAC was evaluated by receiver operating characteristic (ROC) curve, and the cut-off value was determined according to the maximum Youden index. Univariate and multivariate logistic regression analyses were used to analyze the relationship between clinicopathological characteristics and pCR after NAC in breast cancer patients. Meanwhile, a clinical prediction model was established. The ROC curve was made to evaluate the model and the Bootstrap method was used for internal verification.

ROC curve analysis showed that the optimal cut-off value of pre-treatment SII was 418.92, the area under the curve (AUC) for predicting the pCR of breast cancer after NAC was 0.737(95%CI: 0.657-0.818). Multivariate logistic regression analysis showed that histological grade (OR=0.095, 95%CI: 0.024-0.292, P=0.001), tumor size(OR=0.091, 95%CI: 0.019-0.333, P=0.001), ER (OR=0.104, 95%CI: 0.026-0.348, P=0.001), HER-2 (OR=2.962, 95%CI: 1.206-7.511, P=0.019) and SII(OR=0.149, 95%CI: 0.059-0.350, P<0.001) were independent predictive factors of pCR after NAC in breast cancer patients. According to the results of multivariate logistic regression, a clinical prediction model was constructed, and the AUC of the ROC curve was 0.868 (95%CI: 0.813-0.920). The calibration plot shows that the prediction curve was close to the ideal curve, and the mean absolute error of the agreement between the predicted value and the actual value was 0.035.

Pre-treatment SII can be used as an independent factor predicting the pCR of breast cancer after NAC. Meanwhile, the clinical prediction model based on the clinicopathological characteristics (histological grade, tumor size, ER, and HER-2 and SII) can effectively predict the NAC response in breast cancer patients.