作者
Yan Li,Kan Ni,Changjiang Gu,Yi Zhang
摘要
Aims/Background Breast signet ring cell carcinoma (SRCC) represents an uncommon tumour that has not been extensively studied. This investigation sought to assess the clinical characteristics and prognosis of breast SRCC and compare them with those of invasive ductal carcinoma (IDC). Methods We obtained clinicopathological data from the Surveillance, Epidemiology, and End Results (SEER) database, including 222 patients with breast SRCC and 492,559 patients with IDC. Clinical features, treatments, and survival outcomes were compared between the two groups. Propensity score matching (PSM) methodology was used to balance baseline characteristics when evaluating overall survival (OS) and cancer-specific survival (CSS). Sensitivity analyses employing E-values quantified the potential impact of unmeasured confounding, and multiple imputation by chained equations (MICE) was used to address missing data for molecular subtype and histological grade. Additionally, predictive models in the form of nomograms were developed to estimate OS and CSS for patients with breast SRCC. Results Compared with IDC, breast SRCC was significantly associated with older age and more advanced tumour, node, metastasis (TNM) stage (p < 0.05 for all). Kaplan-Meier analyses revealed that breast SRCC patients exhibited markedly poorer survival outcomes (OS and CSS, p < 0.05) compared to IDC patients before PSM. For breast SRCC, the median OS was 67.0 months and the median CSS was 90.0 months. The OS rates at 3, 5, and 8 years stood at 60.9%, 51.5%, and 39.6%, respectively, while the corresponding CSS rates were 67.7%, 59.7%, and 48.6%. Following PSM analysis, survival outcomes between breast SRCC and IDC patients became comparable (p > 0.05). Multivariable assessment identified age, histological grade, T stage, and surgical intervention as independent OS predictors (p < 0.05 for all) in breast SRCC, while histological grade, T stage, and surgical approach were independent CSS factors (p < 0.05 for all). The nomograms were subsequently validated using the concordance index (C-index), receiver operating characteristic (ROC), calibration curves, and decision curve analysis (DCA), demonstrating robust prognostic capability. Sensitivity analysis revealed E-values of 1.35 (OS) and 1.49 (CSS), which exceeded typical confounder effects. Multiple imputation demonstrated consistent results (OS: hazard ratio [HR] = 1.20, 95% confidence interval [CI] 0.94–1.54; CSS: HR = 1.30, 95% CI 0.86–1.97), supporting the robustness of the findings. Conclusion Breast SRCC is associated with poorer outcomes primarily due to more advanced stage at presentation rather than histological type alone. Nomograms were developed to estimate OS and CSS for patients with breast SRCC.