A Population-Based Analysis of a Risk Stratification System for Predicting Radiotherapy Benefits in Medullary Breast Carcinoma

Background Medullary breast carcinoma (MBC), a rare histological subtype representing 3-5% of breast malignancies, presents unique therapeutic challenges due to its distinct clinicopathological characteristics and uncertain radiotherapy (RT) benefit profile. While current guidelines extrapolate treatment protocols from invasive ductal carcinoma, the prognostic heterogeneity among MBC patients and lack of validated biomarkers necessitate precision stratification tools to optimize RT decision-making. Methods Based on the data of patients from the SEER database between 2010 and 2018, we used univariate and multivariate Cox to develop a prognostic stratification model, and stratified the whole cohort into different risk groups to determine the optimal candidates to benefit from radiotherapy. The accuracy of the nomogram was evaluated by discrimination and calibration evaluation. Results A total of 677 patients were randomly divided into training set (n = 535) and verification set (n = 132) at 8:2. Then we identified five independent prognostic factors for MBC patients. Together, the 3 - and 5-year nomograms were made up of these 5 variables and patients were divided into two prognostic cohorts based on optimal cutoff value. The results showed that radiotherapy improved the prognosis of low-risk MBC patients compared to their non-radiotherapy-receiving counterparts (P = 0.017), while radiotherapy could not beneficial for patients with high-risk cohort (P = 0.47). The prognostic model predicts OS with excellent performance, the 3- and 5-year AUC of the training group were 0.777 and 0.775, the 3- and 5-year AUC of the validation set were 0.747 and 0.712, respectively. And the 3-year and 5-year calibration diagrams showed good consistency between the predicted results and the actual results. Conclusion The current study developed a prognostic stratification nomogram of patients with MBC and found that patients in the low-risk group were more likely to benefit from radiotherapy.