Chemotherapy use and survival by demographic and socioeconomic factors in primary central nervous system lymphoma: A population-based SEER analysis.

7057 Background: Primary central nervous system lymphoma (PCNSL) is a rare non-Hodgkin lymphoma confined to the brain and spinal cord. Despite the introduction of rituximab to modern systemic therapy, previous population-based studies suggest that survival disparities by race and socioeconomic status persist. This study uses chemotherapy utilization modeling and contemporary 2000–2022 data to quantify the real-world survival impact of chemotherapy underuse across ethnic and neighborhood income strata in adults with PCNSL. Methods: Adults (≥20 years) with histologically confirmed PCNSL diagnosed from 2000–2022 were identified from SEER 21 registries. Patients with concurrent systemic lymphoma were excluded. Chemotherapy receipt was coded as yes vs no/unknown; with blank fields excluded. Binary logistic regression estimated odds of receiving chemotherapy by demographics. Multivariable Cox proportional hazards models evaluated associations of age, sex, race, neighborhood median household income (MHI), rural–urban residence, and chemotherapy with cause-specific survival. MHI was categorized into quintiles derived from the cohort distribution (approximately equal case counts per band). Kaplan–Meier methods estimated cause-specific survival and median survival times. Results: A total of 9,533 adults met the inclusion criteria. Compared with Whites, Black (OR 0.42, 95% CI 0.35–0.49) and Hispanic patients (OR 0.75, 95% CI 0.66–0.84) were less likely to receive chemotherapy; patients from neighborhoods with MHI < $65k also had reduced chemotherapy use versus ≥$100k (OR 0.53, 95% CI 0.45–0.62). Median survival was 64 months for patients receiving chemotherapy versus 3 months for those with no/unknown chemotherapy. Median survival by race was 17 months in Black patients, 25 months in Whites, and 38 months in Hispanics. Median survival was 14 months in the lowest MHI band (<$65k) and 46 months in the highest (≥$100k). In Cox models, lack of chemotherapy was associated with more than twice the hazard of lymphoma-specific death (HR 2.29, 95% CI 2.17–2.44). After adjustment for age, sex, race, MHI, residence, and chemotherapy; Black patients had higher mortality than Whites (HR 1.22, 95% CI 1.10–1.36), female gender had improved survival than males (HR 0.88, 95% CI 0.833-0.929), and MHI < $65k was associated with worse survival versus ≥$100k (HR 1.20, 95% CI 1.09–1.32). Conclusions: In this cohort of adults with PCNSL, chemotherapy use, race, neighborhood income, and age were powerful independent determinants of cause-specific survival. Chemotherapy underuse was concentrated among Black, Hispanic, and lower-income patients and was associated with a dramatic survival penalty, indicating that inequities in access to systemic therapy are likely major drivers of persistent racial and socioeconomic survival gaps in PCNSL.