Abstract P4-09-09: Association between socioeconomic factors at diagnosis and survival in non-metastatic breast cancer: A population-based study

Abstract

Abstract Background: Breast cancer is a severe public health problem for women worldwide. Race disparities and regional disparities are documented regarding incidence, mortality, and survival of breast cancer patients. However, the associations between socioeconomic status and survival outcomes of breast cancer remain unclear and require a comprehensive large-scale investigation of specific socioeconomic factors. Furthermore, no model has included both histological and socioeconomic factors together to predict survival of breast cancer. In this study, we sought to develop nomograms to predict overall survival (OS) and breast cancer-specific survival (BCSS) with consideration of socioeconomic factors for non-metastasis breast cancer. Methods: We included a total of 274,108 female patients, diagnosed with malignant breast cancer between 2007 and 2014from the Surveillance, Epidemiology, and End Results (SEER) database. Socioeconomic factors involving marital status, insurance status, residence, median household income, poverty rate, unemployment rate and education level were included in the analysis. OS and BCSS were evaluated with log-rank tests and Kaplan-Meier estimates. We identified and integrated significant prognostic factors for OS and BCSS using univariate and multivariate Cox regression analysis to construct nomograms. Calibration plots and concordance indexes were used to evaluate the accuracy and discrimination of the models. Results: Among different age subgroups, insured patients were more likely to have better survival than uninsured patients or patients with Medicaid (P<0.001), and especially for patients who were aged 18 to 35 years old at diagnosis, uninsured patients associated with poor BCSS than Medicaid patients (P<0.05). Through multivariate analysis, we found non-Hispanic black patients experienced worst survival compared with the White and other races (P<0.001). Interestingly, married (vs. single vs. separated/divorced/widowed; P<0.001) and insured (vs. Medicaid vs. uninsured; P<0.001) patients had a better prognosis. Living in the non-metro area increased the risk of death (hazard ratio [HR], 1.084, P<0.05). Furthermore, living in counties with higher median household income (>US $72,800) had favorable impacts on OS (HR 0.843, P<0.001). Four and five socioeconomic factors were involved in constructing the nomograms for 3 years-, 5 years- and 7 years- OS and BCSS, respectively. The C-indexes of the final nomograms were higher than those of the TNM staging system for predicting OS (0.776 vs 0.678; P < 0.001) and BCSS (0.842 vs 0.776; P < 0.001), respectively. The performance of the nomograms for predicting OS was significantly lower when excluding the socioeconomic factors (P < 0.001). Conclusion: Some certain socioeconomic factors (i.e., marital status, insurance status, median household income, and residence) play essential roles in predicting survival of non-metastasis breast cancer. We constructed and validated nomograms including socioeconomic factors to provide more comprehensive and realistic survival estimation. Besides, these findings may highlight the importance of developing health-related policies and the necessity of targeted social support-based interventions for those high-risk patients. Citation Format: Ji P, Gong Y, Hu X, Hong D, Shao Z-M. Association between socioeconomic factors at diagnosis and survival in non-metastatic breast cancer: A population-based study [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P4-09-09.