PB2117: TEN-YEAR SINGLE CENTER EXPERIENCE TREATING PRIMARY CENTRAL NERVOUS SYSTEM LYMPHOMA: CAN RADIOTHERAPY BE OMITTED?

Abstract

Background: Primary central nervous system lymphoma (PCNSL) is a rare form of lymphoma confined to the central nervous system (CNS) that is characterized by an aggressive course, high rates of treatment resistance and short overall survival. In recent decades, the treatment of PCNSL has undergone significant changes; its standard treatment consists of combined immunochemotherapy regimens that should include high-dose methotrexate (HD-MTX) and/or cytarabine, followed by consolidative autologous hematopoietic cell transplantation (AHCT). HD-MTX–based chemotherapy followed by consolidative whole brain radiotherapy (WBRT) has, historically, been the standard of care. In elderly and comorbid patients who are not candidates for AHCT, maintenance therapy with procarbazine or temozolomide is often used. Aims: We aim to describe a single center’s ten-year experience treating PCNSL. Methods: Retrospective study of patients with PCNSL treated in a single center from 2011 to 2021, describing the demographic, clinical features and outcomes. Data was collected from clinical records and analyzed with IBM SPSS 25.0. Results: Twenty patients, 11 (55%) female, with a median age of 61 years (range 44-77) were identified. 65% had an Eastern Cooperative Oncology Group performance status (ECOG) of 0, 25% of 1, 5% had an ECOG of 2 and 5% had an ECOG of 4. International Extranodal Lymphoma Study Group (IELSG) score was 0-1 in 7/35%, 2-3 in 11/55% and 4-5 in 2/10%; 17/85% presented with parenchymal disease, 2/10% with leptomeningeal disease only, and 1/5% with both; only one patient had eye involvement at presentation. Frontline therapy consisted of chemotherapy only (with or without rituximab) in 14 patients while 6 had chemotherapy plus WBRT; all patients received methotrexate (MTX)-based regimens (Ferreri, MATRix, DeAngelis, R-MP and R-MPV-A). Overall response was obtained in 12/60% patients (complete response in 10/50% and partial response in 2/10%). 8/40% patients relapsed, at a median time of 4.2 months after completion of therapy; 7/87.5% relapsed with parenchymal disease and 1/12.5% with mixed parenchymal and leptomeningeal disease. Median follow up was 12.1 months, median overall survival (OS) was 25.3 months and median progression-free survival (PFS) was 4.4 months. In univariate analysis (log rank), age, sex, ECOG, presence of deep brain lesions and IELSG score did not significantly impact OS and PFS, whereas radiotherapy use was significantly associated with improved OS (63.7 vs. 13.0 months; p-value 0.002) and PFS (36.1 vs. 3.3 months; p-value 0.016; Image). Univariate Cox regression analysis showed significant PFS improvement with radiotherapy use (HR 0.12; 95% CI 0.02-0.92, p-value 0.042), but no statistically significant OS improvement; in our cohort, no other variables significantly impacted survival. All surviving patients who underwent WBRT have shown cognitive impairment in neuropsychological tests. Image:Summary/Conclusion: Based on our analysis, age, ECOG and IELSG score did not significantly impact OS and PFS, whereas WBRT use was significantly associated with improved PFS, suggesting that the omission of WBRT may not be associated with improved survival, although its use should be weighed against potentially limiting neurotoxicity.