A predictive nomogram for cancer-specific survival in elderly patients with ductal carcinoma in situ after lumpectomy: A SEER-based retrospective study

• Current Knowledge Gap: Elderly patients (≥70 years) with ductal carcinoma in situ (DCIS) represent a distinct population, yet existing prognostic models lack specific validation, and the survival benefit of adjuvant radiotherapy remains uncertain, especially in hormone receptor-negative disease. • Novel Prognostic Tool: We developed and validated the first cancer-specific survival (CSS) nomogram and risk stratification system specifically tailored for elderly (≥70 years) DCIS patients treated with lumpectomy. • Informed Radiotherapy Decision: Adjuvant radiotherapy was associated with a significant CSS benefit specifically in hormone receptor-negative subgroups, providing crucial evidence to guide treatment selection in this demographic. • Clinical Translation: This nomogram offers clinicians a practical, individualized tool for prognosis assessment and therapeutic planning, facilitating a more personalized management strategy. • Future Implications: The findings underscore the necessity of age-specific prognostic models in DCIS and suggest that future research should focus on age-related biological and clinical differences. To develop and validate a prognostic nomogram to predict cancer-specific survival (CSS) in elderly patients with ductal carcinoma in situ (DCIS) following breast-conserving surgery (BCS). Data from the SEER database (2000–2019) were used, comprising 40,502 eligible patients who were randomly assigned to training and validation cohorts. Cox proportional hazards and Fine-Gray competing risks regression models were employed. A nomogram was constructed and evaluated using time-dependent C-index, calibration curves, and decision curve analysis. Risk stratification and subgroup analyses were conducted. Multivariable analysis identified older age (≥80 years; HR 1.85, 95% CI 1.53–2.24), hormone receptor negativity (HR 1.33, 95% CI 1.04–1.69), and omission of radiotherapy (HR 1.34, 95% CI 1.11–1.61) as independent predictors of worse CSS. The nomogram demonstrated moderate discriminative ability, with time-dependent C-indices of 0.606 (training) and 0.659 (validation). Calibration and decision curve analysis demonstrated favorable clinical utility. Subgroup analyses indicated a significant CSS benefit from radiotherapy among hormone receptor-negative patients across all age groups (70–79: HR 0.349, p < 0.001; ≥80: HR 0.375, p = 0.045). We developed and validated a nomogram that integrates age, hormone receptor status, and radiotherapy use to predict CSS in older patients with DCIS following breast-conserving surgery. This tool could aid in identifying a high-risk patients, particularly those with hormone receptor-negative disease, who may derive significant survival benefit from adjuvant radiotherapy.