A Nomogram Predicts The Survival In Patients With Metastatic Breast Cancer Based On The Surveillance,Epidemiology,and End Results Database

objective: The metastasis is a widespread problem of breast cancer patients.The treatment strategy for metastatic breast cancer patients(MBC)is mostly based on the life expectancy.Therefore,predicting the survival of MBC patients allow clinicians to optimize the treatment strategy.The aim of this study is to construct a nomogram to predict the survival probability in patients with MBC.Methods: A total of 9,218 MBC patients with clinicopathological characteristics between 2010 and 2015 were obtained from the Surveillance,Epidemiology,and End Results(SEER)database.Data was divided into training set(n = 6,452)and validation set(n = 2,766).In the training set,the univariable analysis was used to analyze the association between risk factors and survival of patients with MBC.Characteristics obtained by backward stepwise analysis and achieved p < 0.05 in the univariable analysis were recruited into the Cox proportional hazards model.A statistically significant difference was identified as p < 0.05.The nomogram was established based on the Cox proportional hazards model.The performance of the nomogram was estimated by calibration and discrimination.Calibration curve subjected to 1,000 bootstraps resamples visualizes the consistency between the predicted survival and observed survival in the validation set so as to avoid over-fitting and popularization.The discrimination ability of the nomogram was quantified with the concordance index(C-index)and the area receiver operating characteristic curve(AUC)of the validation set.the Kaplan-Meier survival curve and Curves for time to mortality hazard were described by risk group stratification based on the nomogram score.Besides,The scoring mortality risk curve was used to evaluate the mortality rates in each score.On the basis of risk stratification,the distribution of MBC related factors in each risk group was statistically analyzed.Results: Ten independent clinicopathological characteristics of patients with MBC were enrolled into the nomogram.The calibration plot visualized a good coherence between predicted and actual overall survival(OS)probability.The concordance index(C-index)predicting OS probability of Nomogram was undoubtedly higher than that of the 7th AJCC TNM classification(0.72 vs.0.56).The area under the receiver operating characteristic curve(AUC)for predicting 1-year,3-year and 5-year OS based on nomogram score were 0.78(95% CI 0.76 to 0.80),0.72(95% CI 0.70 to 0.74)and 0.68(95% CI 0.67 to 0.70),respectively.Further,we stratified data by the cutoff nomogram score 170 and 300 into different groups,which demonstrated significant distinction in survival curves.The serious adverse events in low-risk groups were less common in the moderate and high-risk groups(moderate vs.low HR=2.17,95%CI: 1.82-2.59,p<0.001;high vs.low HR=7.22,95%CI:5.93-8.79,p<0.001).Conclusions Our study showed that MBC patients with older age,poor molecular typing,parenchymal organ metastasis,late local staging,high pathological grading and rejection of treatment had a worse prognosis.The long-term prognosis of patients with the nomogram score less than 170 points was relatively good,belonging to the low-risk group;Patients with a score greater than 300 have a poor long-term prognosis and are in the high-risk group.We provides evidence that the nomogram we established shows a reasonable OS probability prediction for distant MBC patients and to some degree,provides a personalized survival scale.