Comparison of differing dose levels of methotrexate for patients with primary central nervous system lymphoma.

INTRODUCTION: It has long been established that high-dose methotrexate is an essential part of therapy for primary central nervous system lymphoma. When regimens utilizing high-dose methotrexate were first studied, a dose of 8 g/m2 was used. More recently, reduced dosing strategies have been studied and adopted in attempts to reduce rates of adverse events. Studies utilizing 3.5 g/m2 of methotrexate have shown promising outcomes and improved rates of adverse events but there have never been any randomized head-to-head studies of differing dose levels of high-dose methotrexate. The purpose of this study was to compare efficacy and safety of different dosing strategies of high-dose methotrexate (HD-MTX) for primary central nervous system lymphoma (PCNSL). METHODS: This single center retrospective review was conducted between 07/01/2013 to 6/3/2020. The patient population was separated into two arms based upon dose of methotrexate. The high intensity (HiHD) arm was defined as patients who received doses > 3.5 g/m2, while the low intensity (LiHD) arm received ≤ 3.5 g/m2. The primary endpoint was overall response rate (ORR) and secondary endpoints include efficacy via 2-year overall survival (OS), progression to transplant, and utilization of consolidation or salvage therapy. Safety was assessed through monitoring of relevant laboratory studies. RESULTS: A total of 92 patients were included in this analysis. Baseline demographics were similar between groups, with the LiHD group trending toward older age. There were 78 patients eligible for assessment for ORR; there was no significant difference between the two groups (42.0% LiHD vs. 44.4% HiHD; p = 1.0). Rates of OS, progression to transplant and progression to consolidation chemotherapy were not different between groups. There were statistically significantly higher rates of renal and/or hepatic dysfunction with the first dose in the HiHD group compared with the LiHD group (11.5% LiHD vs. 64.3% HiHD; p ≤ 0.01). CONCLUSIONS: In this PCNSL patient cohort, there is no difference in terms of efficacy between HiHD LiHD methotrexate, but patients in the HiHD group had higher rates of renal and hepatic dysfunction. Limitations include small sample size and disparity between group sizes.