Development and validation of a SEER-based prognostic nomogram for overall survival in elderly Asian American women with breast cancer.

BACKGROUND: Breast cancer is the most common malignancy among Asian American women. Although Asian American patients as a group often have favorable survival outcomes, elderly Asian American women may face distinct prognostic challenges. General prognostic models for breast cancer may have reduced accuracy in elderly Asian American women, who have distinct tumor biology, treatment patterns, and social contexts. No population-specific prognostic nomogram exists for this understudied intersectional group. The present study aimed to establish and validate a prognostic nomogram to predict overall survival (OS) in elderly Asian American breast cancer patients. METHODS: Cases were identified via the International Classification of Diseases for Oncology, 3rd edition (ICD-O-3) morphology and topography codes in Surveillance, Epidemiology, and End Results (SEER) [2000-2021], and Asian ethnicity was defined per SEER race/ethnicity classification. The median follow-up time was 50 months. Patients were randomly split into training (70%) and validation (30%) cohorts. Of 109,317 Asian female patients initially identified from the SEER database, 11,968 eligible patients aged ≥65 years with stage I-III primary breast cancer were included after applying exclusion criteria for secondary tumors, missing data, and other ineligible conditions. Univariable and multivariable Cox regression identified independent prognostic factors, which were incorporated into a nomogram predicting 1-, 3-, and 5-year OS probabilities. The nomogram's performance was evaluated using calibration, time-dependent receiver operating characteristic (tROC) curves, concordance index (C-index), and decision curve analysis (DCA). RESULTS: Most patients had stage I disease (69.0%), hormone receptor (HR)+/human epidermal growth factor receptor 2 (HER2)- subtype (81.4%), and were aged 65-74 years (68.0%). A total of 11,968 individuals were enrolled, with 8,380 patients in the training cohort and 3,588 in the validation cohort. The nomogram incorporated age, tumor grade, marital status, tumor-node-metastasis (TNM) stage, molecular subtype, surgery, and radiotherapy. It demonstrated moderate-to-good discriminative ability, with C-indices of 0.779 and 0.763 in the training and validation cohorts, respectively, outperforming the traditional TNM staging system. Calibration plots showed strong agreement between predicted and observed survival. tROC curves and DCA confirmed the nomogram's superior predictive accuracy and clinical utility over TNM staging. CONCLUSIONS: This nomogram provides a tailored risk stratification tool to support clinical decision-making for elderly Asian American breast cancer patients. External validation is required to confirm its generalizability before clinical implementation.