Social determinants of health and survival outcomes in metastatic breast cancer (MBC): Retrospective analysis from a major Canadian cancer centre.

Abstract

e13758 Background: Social determines of health (SDOH), including geography and sociodemographic factors influence cancer outcomes, even in publicly funded healthcare systems. However, evaluation is limited by inconsistent data collection. We examined survival outcomes among patients with MBC across regional health care authorities (HCA) in British Columbia (BC), Canada, focusing on available sociodemographic data, including self-reported race and ethnicity. Methods: We conducted a retrospective cohort study of patients diagnosed with MBC between 2014-2017 across BC HCAs using a prospective provincial cancer database. Sociodemographic variables, including self-reported race and ethnicity, were manually abstracted from clinical intake and consultation records and categorized according to Canadian Cancer Clinical Trials Network (3CTN) guidelines. Endpoints were 5-year breast cancer specific (BCSS) and overall survival (OS), estimated using Kaplan-Meier methods and compared with log-rank tests. Results: Among 1917 eligible patients, 513 (26%) had missing self-reported race and ethnicity data, most commonly due to non-disclosure. Patient demographics reflected the 2021 provincial census, with White and East Asian patients comprising the largest groups (54.7% and 7.4%). No significant differences in BCSS and OS were observed across racial and ethnic groups (p = 0.2), though interpretation was limited by missing data. Survival varied by geography: patients treated in the urban HCA had higher 5-year OS relative to those treated in more remote HCAs (Table 1). No significant differences in BCSS were observed by HCA. Conclusions: Within a large, publicly funded cancer system, urban geography was associated with improved OS in MBC, highlighting geography as a key SDOH. The absence of BCSS differences may reflect standardized treatment delivery and competing non-cancer comorbidities. Although no survival differences by race or ethnicity were identified, substantial missing data limits definitive conclusions. These findings underscore the importance of standardized, prospective collection of sociodemographic and comorbidity data to inform equitable cancer care delivery. Further work is needed to clarify drivers of regional variation, including local treatment availability, clinical trial access and patient-level factors influencing treatment uptake. Five-year breast cancer mortality and overall mortality across regional HCA. 5-year Breast Cancer Mortality (95% CI) 5-year Overall Mortality (95% CI) Health authority 1 (urban) 70.0 (65.5, 74.0) 27.8 (23.9, 32.3) 2 76.0 (71.5, 79.9) 22.3 (18.6, 26.7) 3 76.6 (73.1, 79.7) 19.3 (16.4, 22.6) 4 76.3 (71.4, 80.6) 18.6 (14.8, 23.2) 5 69.8 (59.4, 78.0) 22.9 (15.9, 33.1)