Risk Factors Analysis And Prediction Model For Brain Metastasis Of HER2 Positive Breast Cancer

Objective:This study analyzed and determined the risk factors of invasive breast cancer with HER-2 positive brain metastasis at initial diagnosis,developed and verified the risk prediction model of brain metastasis,and provided a basis for improving the clinical ability to predict the risk of HER-2 positive breast cancer with brain metastasis and the efficacy of breast cancer with brain metastasis.Methods:The data were downloaded from SEER database,and the qualified patients were selected through strict inclusion and exclusion criteria.The clinical characteristics of the patients were collected and the influencing factors of HER2-positive breast cancer brain metastasis were screened by univariate and multivariate Logistic regression analysis Based on the above factors,the prediction model of HER-2 positive BMS in invasive ductal carcinoma of breast was established With the Bootstrap method to internal restructuring of data validation,consistency index index(C)used to assess the accuracy of the models,drawing the calibration curve to further evaluate the clinical value of the model,the sensitivity and specificity of ROC curve analysis is applied to forecast model,decision analysis curves were used to verify whether invasive ductal breast cancer her-2 positive benefited from this modelResults:A total of 32,483 HER-2 positive patients with invasive ductal breast cancer were screened from the SEER database for inclusion in this study,including 197 patients with BMS,accounting for 0.61% of the entire cohort by Logistic single factor Multivariate regression equation analysis showed that women aged 61-70 years,T3 stage,T4 stage,histological grade IV,bone metastases,liver metastases,lung metastases,and hormone receptor(HR)negative HER-2 positive breast cancer were more likely to be diagnosed with brain metastases at initial diagnosis Histologically graded bone metastases,lung metastases,liver metastases,AND HR status in this predictive model,the C-index was 0.936(95%CI:0.913-0.955;P<0.01),with good discriminant and correction ability,internal validation can still achieve a high C index 0.927 decision analysis curve validation,suggesting that the model can bring clinical benefits.Conclusion:Patients with high T stage,61--70 years old,histological grade IV,bone metastasis,lung metastasis,liver metastasis,and hormone receptor-negative HER-2positive breast cancer were more likely to develop brain metastasis at initial diagnosis.This rosette,which integrates age,T stage,histological grade,bone metastasis,lung metastasis,liver metastasis and HR status,can better predict the risk of BMS in breast cancer patients with HER-2 positive and may have potential clinical benefits.