A Nomogram For Predicting The Level Ⅱ Axillary Lymph Node Metastasis In Level Ⅰ Node-Positive Breast Cancer After Neoadjuvant Chemotherapy

Background It is reported that breast cancer has been to be the second leading cause of cancer death and the most common malignant tumor among women in America.The breast cancer incidence rate is still increasing.It is because the utilization of neoadjuvant chemotherapy(NCT)can help to reduce tumor burden as well as the curative intervention,provide the opportunity to assess the response to treatment,and increase the breast conservation rate that more and more patients receive NCT.Many patients with clinically level Ⅰ but without level Ⅱ node-positive breast cancer after neoadjuvant chemotherapy(NCT)receive level Ⅰ and level Ⅱ axillary lymph node dissection(ALND).Several randomized clinical trials showed that the omission of levelⅡaxillary lymph node dissection could prevent the short-and long-term side effects of this procedure such as lymph edema,limited range of motion of the shoulder,and numbness of the upper arm and so provide a better quality of life compared with standard ALND.We aimed to identify predictors and construct a nomogram to predict the level Ⅱ axillary lymph node metastasis(L-Ⅱ-ALNM)in post-neoadjuvant chemotherapy breast cancer patients with level 1axillary lymph node metastasis(L-Ⅰ-ALNM)that can assist clinicians in treatment planning.Methods In total,424 patients with pathologically proven level Ⅰaxillary lymph node metastasis after NCT were from 2010 through 2015 were identified,randomized1:1 and divided into a training set and a validation set.Clinical and pathological featuresof the patients were assessed with Uni-and multivariate logistic regression.Univariate as well as multivariate logistic regression analyses were performed to screen the fit predictors in order to construct a well-calibrated nomogram for predicting the probability of L-Ⅱ-ALNM.The relation between clinicopathological variables and L-Ⅱ-ALNM of overall population,modeling group and validation group were Variables that were statistically significant(P < 0.05)in the univariate logistic analysis in the modeling set were included in the multivariate logistic regression analysis,which was used to screen independent predictors for L-Ⅱ-ALNM.Independent predictors(P <0.05 in the multivariate logistic regression analysis)were used for serve as the nomogram development.A nomogram based on independent predictors for the level Ⅱaxillary lymph node metastasis identified by multivariate logistic regression was developed.Results Among 1108 patients,612 of them were pathologically confirmed L-Ⅰ-ALNM after NCT.After excluding 188 patients with incomplete relevant information,424 eligible patients were included in the study(Fig.1).So,our study comprised 424 patients randomized 1:1 and divided into a modeling set(N = 212)and a validation set(N = 212).With the help of univariate analyses,the following statistically significant(P< 0.05)variables were entered into the multivariate model:(tumor size,response to NCT,histological grade,number of PL-Ⅰ-ALN).Univariate analysis of the modeling group and the validation group showed similar results compared with patients in the total population.Multivariate binary logistic regression analysis was used to identify independent variables for predicting the L-Ⅱ-ALNM post-NCT.The multivariate logistic regression analysis showed that tumour size,response to NCT,histological grade and number of positive level Ⅰ axillary lymph node(PL-Ⅰ-ALN)were independent predictors.The areas under the receiver operating characteristic(ROC)curve of the modeling set and the validation set were 0.895 and 0.894,respectively.the modeling group falsenegative rate of 2.6 % at cut-off points of <7%,9.7 % at cutoff points of <15%,and 11.1 % at cut-off points of <20 % for the L-Ⅱ-ALNM nomogram appeared to be acceptable.In the validation group,the falsenegative rate of 4.7 % atcut-off points of <7%,10.3 % at cutoff points of <15%,and 12 % at cut-off points of<20 %.In addition,the lowrisk subgroups of the patients relevant to a probability of L-Ⅱ-ALN involvement of <7 %,were accounting for 18.4% and 20.1% of all patients predicted by the L-Ⅱ-ALNM nomogram in the modeling and validation respectively.That is to say,about 20 % of all patients could benefit from the L-Ⅱ-ALNM nomogram.In the modeling group,the low-risk subgroup the patients could account for almost29.2 % of all the patients,if the false-negative rate of 9.7 % could be received by surgeons.Conclusion In conclusion,as the paradigm shifts to omission of level Ⅱ axillary surgery in the patients receiving NCT,using NCT to eradicate nodal disease is a clinically relevant strategy.The nomograms can be used to predict the probability of L-Ⅱ-ALNM in level 1 node-positive breast cancer after NCT with readily available clinicopathologic features.Consequently,the might have utility for making treatment decisions.Our nomogram can improve risk stratification,accurately predict the level 2axillary lymph node status and avoid unnecessary the level 2 axillary lymph node dissection.