641 Background: We aim to determine how indices of social marginalization impact stage at diagnosis, health services use, and overall survival comparing invasive ductal (IDC), pure (pILC) and mixed lobular (mILC) breast carcinoma (ILC) between 2005 and 2025 in the province of Ontario, Canada. Methods: Using population-based administrative healthcare datasets at Institute of Clinical Evaluative Sciences (ICES) Ontario, we identified all cases of IDC, pILC and mILC in Ontario, Canada between 2005 and 2025 using ICD-O-3 codes. We compared prior healthcare utilization between groups. We further compared clinical and healthcare utilization measures between four Ontario Marginalization (ON-MARG) indices: household and dwellings (includes % living alone, housing density), material resources (includes unemployment and education level), age and labour force (includes % seniors and not participating in the work force), and racialized and newcomer populations (includes % of recent immigrants). Using a multivariable Cox proportional hazards regression analysis incorporating sociodemographic, tumour and treatment factors, we assessed the impact of social marginalization indices on survival for the overall cohort and for ILC cases. Statistical analysis was performed using SAS and P values < 0.05 were significant. Results: We identified 16,122 (9%) pILC, 7,284 (4%) mILC and 156,870 (87%) IDC cases. At diagnosis, increasing marginalization for household and dwellings, material resources, and age and labour force correlated with less likelihood to having had a mammogram within 1 and 5 years prior to diagnosis (P < .0001). Increasing marginalization in these three quintiles correlated with worse clinical stage at diagnosis for both IDC and ILC (P < .001). Patients diagnosed with pure ILC breast cancer had 1.14 times (95%CI: 1.11 - 1.18) the likelihood of mortality in the follow-up relative to females diagnosed with IDC breast cancer. Increasing marginalization in the ON-MARG indices had a worse mortality for the overall cohort and for ILC cases, except for the racialized and newcomer populations quintile (P < .0001). Conclusions: In Ontario, Canada, the indices of social marginalization found to be associated with worse survival after a breast cancer diagnosis were household and dwellings, material resources, age and labour force but not racialized and newcomer populations. These findings highlight personalized risk factors for worse stage at presentation and survival for both IDC and ILC, independent of clinical and treatment factors.
Lim et al. (2026) studied this question.