Postoperative radiotherapy for ductal carcinoma in situ: survival prediction and clinical decision support using a...

Whether postoperative radiotherapy (RT) is required for ductal carcinoma in situ (DCIS) after breast-conserving surgery (BCS) remains controversial. In this study, we aimed to analyze the association between postoperative RT and survival outcomes in DCIS patients and develop nomograms to predict these outcomes. Using data on 50,580 DCIS patients obtained from the surveillance, epidemiology, and end results (SEER) database, the Chi-squared tests revealed that DCIS patients with younger age, partial mastectomy, larger tumor size, negative estrogen receptor (ER) status, and higher nuclear grade, were more likely to receive postoperative RT. Additionally, Postoperative RT could improve the overall survival (OS), disease-specific survival (DSS), and disease-free survival (DFS) across most patient subgroups, although no significant association was observed in DSS among older patients or those with smaller tumor size. Univariate and multivariate Cox regression analyses identified that age, tumor size, ER status, and nuclear grade were independent predictors for DSS and DFS. Based on these findings, we further constructed a nomogram, which demonstrated strong discriminative ability and good calibration as validated by C-index and calibration curve. A predictive online tool was created to visualize personalized DSS and DFS prediction for DCIS patients with different treatment regimens ( https://nordaraail.github.io/breast-calculator/ ). Our study suggests that postoperative RT is associated with improved survival in most DCIS patients. The nomograms showed good performance in predicting the DSS and DFS of DCIS patients, and our online tool well visualized the DSS and DFS prediction to support clinical decision-making.