ABSTRACT
Objective: To develop a predictive model for pathologic complete response (pCR) of ipsilateral supraclavicular lymph nodes (ISLN) after neoadjuvant chemotherapy for breast cancer and guide the local treatment. Methods: Two hundred and eleven consecutive breast cancer patients with first diagnosis of ipsilateral supraclavicular lymph node metastasis who underwent ipsilateral supraclavicular lymph node dissection and treated in the Breast Department of Henan Cancer Hospital from September 2012 to May 2019 were included. One hundred and forty two cases were divided into the training set while other 69 cases into the validation set. The factors affecting ipsilateral supraclavicular lymph node pCR (ispCR)of breast cancer after neoadjuvant chemotherapy were analyzed by univariate and multivariate logistic regression analyses, and a nomogram prediction model of ispCR was established. Internal and external validation evaluation of the nomogram prediction model were conducted by receiver operating characteristic (ROC) curve analysis and plotting calibration curves. Results: Univariate logistic regression analysis showed that Ki-67 index, number of axillary lymph node metastases, breast pCR, axillary pCR, and ISLN size after neoadjuvant chemotherapy were associated with ispCR of breast cancerafter neoadjuvant chemotherapy (P<0.05). Multivariate logistic regression analysis showed that the number of axillary lymph node metastases (OR=5.035, 95%CI: 1.722-14.721, P=0.003), breast pCR (OR=4.662, 95%CI: 1.456-14.922, P=0.010) and ISLN size after neoadjuvant chemotherapy (OR=4.231, 95%CI: 1.194-14.985, P=0.025) were independent predictors of ispCR of breast cancer after neoadjuvant chemotherapy. A nomogram prediction model of ispCR of breast cancer after neoadjuvant chemotherapy was constructed using five factors: number of axillary lymph node metastases, Ki-67 index, breast pCR, axillary pCR and size of ISLN after neoadjuvant chemotherapy. The areas under the ROC curve for the nomogram prediction model in the training and validation sets were 0.855 and 0.838, respectively, and the difference was not statistically significant (P=0.755). The 3-year disease-free survival rates of patients in the ispCR and non-ispCR groups after neoadjuvant chemotherapy were 64.3% and 54.8%, respectively, with statistically significant differences (P=0.024), the 3-year overall survival rates were 83.8% and 70.2%, respectively, without statistically significant difference (P=0.087). Conclusions: Disease free survival is significantly improved in breast cancer patients with ispCR after neoadjuvant chemotherapy. The constructed nomogram prediction model of ispCR of breast cancer patients after neoadjuvant chemotherapy is well fitted. Application of this prediction model can assist the development of local management strategies for the ipsilateral supraclavicular region after neoadjuvant chemotherapy and predict the long-term prognosis of breast cancer patients.