Primary CNS Lymphoma: Comparison of Outcomes and the Use of Radiation Therapy in Patients Treated at a Public Safety Net Hospital vs. a Tertiary Academic Center.

PURPOSE/OBJECTIVE(S) The role of whole-brain radiation therapy (WBRT) in primary CNS lymphoma (PCNSL) has changed over time and current practices vary by institution and resource availability. This study compares the treatment patterns and outcomes among PCNSL patients treated at a public safety-net hospital versus a tertiary academic institution within the same healthcare system. We hypothesize that due to limitations in resources, patients at the safety-net hospital are more likely to receive WBRT and to have inferior outcomes. MATERIALS/METHODS We reviewed records of 95 PCNSL patients treated from 2007-2020 at a public safety-net hospital (n = 33) and an academic tertiary-care center (n = 62), both serving the same metroplex. Patient, tumor, and treatment characteristics were reviewed. Categorical and continuous variables were compared using nonparametric tests. Overall survival (OS) and progression-free survival (PFS) were estimated by the Kaplan-Meier method. Multivariable Cox regression was used to assess for confounders. RESULTS Median follow-up was 38.6 months for surviving patients and 11.1 months for all patients. Compared to the tertiary academic center, patients at the safety-net hospital were significantly younger, more commonly Black or Hispanic, male, and HIV positive, and had better MSKCC prognostic class. Initial treatments used were high-dose methotrexate based systemic therapy (ST) (n = 75) or WBRT (n = 16, median dose 30.6 Gy). Four patients received steroids alone due to poor prognosis at presentation. Safety-net patients were significantly more likely to receive WBRT alone (33.3% vs 8.1%, P = 0.003) as initial treatment. OS (55% vs 60%, P = 0.599) and PFS (35.2% vs 46%, P = 0.28) were not significantly different between patients who received only WBRT vs. induction ST. Patients who had a positive response to induction ST (n = 51) were either given consolidative WBRT (n = 12), stem cell transplant (SCT) (n = 15), chemotherapy (n = 14), or no further treatment (n = 10). Safety-net hospital patients were significantly less likely to receive SCT (0% vs. 44.1%) and had higher rates of consolidative WBRT (37.5% vs 14.7%, P = 0.001). There were no significant differences when comparing consolidation strategies on OS (P = 0.763) or PFS (P = 0.908). OS (71.3% vs 49.4%, P = 0.461) and PFS (48.4% vs 39.5% P = 0.39) were not significantly different between safety-net and tertiary academic hospitals when adjusted for MSKCC prognostic class. Of patients receiving only WBRT, no neurotoxicities were documented. Three out of 12 patients (25%) receiving consolidative WBRT and 2/15 (13.3%) of SCT patients reported neurotoxicity. CONCLUSION Radiation was more often used in a safety-net hospital for both upfront and consolidative therapy, without significantly inferior outcomes. This suggests that in a resource limited setting, WBRT is an effective treatment option for PCNSL. Prospective and translational studies are required to further compare the efficacy and toxicity of WBRT to alternative treatments.