Construction Of Nomogram Prediction Model For Male Breast Cancer:an Analysis Based On SEER Database

Research background:Male breast cancer(MBC)is a malignant tumor with a relatively low incidence.At present,the clinical diagnosis and treatment of male breast cancer is mainly based on the diagnosis and treatment of female breast cancer.Accurate treatment is still a big challenge for clinicians.Under normal circumstances,clinical specialists will formulate corresponding treatment plans based on the patient’s survival expectations.Therefore,accurate and scientific prediction of the prognosis for male breast cancer patients can help clinical specialists to formulate more precise,individualized treatment options.Objective:The purpose of this study is to analyze the factors affecting the prognosis of male breast cancer patients based on the National Cancer Institute(NCI)surveillance,epidemiological and follow-up results database(SEER)data,according to several important clinicopathological factors that affect male breast cancer,a comprehensive and practical Nomogram is constructed to predict breast cancer-specific survival(BCSS)and overall survival(OS)of male breast cancer patients and compared with the seventh edition of the American Joint Committee on Cancer(AJCC)TNM staging system.Methods:A total of 1752 patients diagnosed with male invasive breast cancer from 2007 to 2014 were extracted from the SEER database,and the data were randomly divided into a training set(n=1316)and a validation set(n=436).First,the Nomogram prediction model was established based on the training set:Cox proportional hazards model was used for univariate and multivariate analysis to assess the prognostic value of related variables and obtain statistically significant variables(P<0.05).Starting from the Cox proportional regression model and combining the meaningful prognostic factors obtained from multivariate analysis,a Nomogram prediction model was constructed to predict the 3-year and 5-year prognosis of male invasive breast cancer.Second,the Nomogram prediction model performance was tested in the validation set,a calibration curve was obtained by resampling the data from the validation set to assess the consistency of the model between predicted and actual survival rates.Concordance index(C-index),area under the ROC curve(AUC),net reclassification index(NRI),and decision curve analysis(DCA)were used to evaluate and compare the predictive performance of the Nomogram and TNM staging systems.Finally,risk stratification was performed according to the Nomogram prediction score,and risk stratification was performed according to the total patient prediction score in the training set.In the validation set,Kaplan-Meier was used to draw the survival curves of each risk group,and Log-rank test was used to compare between groups to evaluate the accuracy of risk stratification.Results:According to the univariate and multivariate Cox proportional hazards model analysis,eight important clinicopathological factors affecting the prognosis of male breast cancer patients were included in the Nomogram prediction model.In the validation set,the calibration curve showed a high degree of consistency between the predicted survival rate of Nomogram and the actual survival rate.In addition,the prediction performance of the Nomogram prediction model was better than that of the TNM staging system,the consistency index(C-index)of the overall survival(OS)was 0.767 to 0.671,the breast cancer-specific survival rate(BCSS)consistency index was 0.834 to 0.756,respectively.Compared with the TNM staging system,the 3-year and 5-year OS Net Reclassification Index(NRI)of the Nomogram prediction model were 0.325 and 0.562,the NRI for 3 and 5 year BCSS were 0.312 and 0.323,respectively.At the same time,the area under curve(AUC)and decision curve analysis(DCA)drawn by the ROC curve showed better prediction performance and higher clinical application value than the TNM staging system.In the training set according to the total prediction score of the patient in the Nomogram prediction model,it was divided into three layers of low,medium and high risk according to the size of the risk,In the validation group,Kaplan-Meier survival curve analysis showed the accuracy of the risk stratification of the prediction model.Conclusion:(1)For male breast cancer,age,pathological grade,tumor size,lymph node metastasis,distant metastasis,marital status,ethnicity,and ER status were independent factors affecting survival and prognosis.(2)This study showed that the Nomogram survival prediction model for male breast cancer based on the SEER database had higher accuracy,and its prediction performance was significantly better than that of the TNM staging system.It helped clinical specialists to formulate more precise and individualized treatment plans.