P14.01 MTHFR genotype may affect methotrexate toxicity, survival and disease occurrence in patients with central nervous system lymphoma

Abstract

INTRODUCTION: Two common methylenetetrahydrofolate reductase (MTHFR) single nucleotide polymorphisms (SNPs)‐C677T and A1298C‐have been implicated in MTX toxicity in patients receiving high dose methotrexate (HD‐MTX) therapy. Our objective was to explore whether MTHFR polymorphisms are associated with treatment‐related toxicity and with tumor‐related outcomes. METHODS: A retrospective cohort study was conducted which included all patients who received HD‐MTX for a central nervous system lymphoma between June 2010 and November 2016. MTHFR genotyping was performed on all patients prior to their first course of HD‐MTX as part of our standard management protocol. Patient demographic data, hematologic, hepatic, cutaneous and renal toxicity data, daily MTX levels, and overall survival (OS) and PFS were all recorded. RESULTS: Fifty‐one patients were included, and 86.3% [74.3‐93.2] had at least one MTHFR SNp. 52.9% [39.5‐66.6] of patients were either heterozygous or homozygous forC677T, 58.8% [45.2‐71.3] were either heterozygous or homozygous for A1298C, and 25.5% [15.5‐38.9] were compound heterozygotes. Peak methotrexate levels were 656.0 [579.6‐730.4] in patients with a mutation and 513.43 [376.3‐605.5] in non‐mutated patients (p=0.11). Serum methotrexate levels >0.05 on day 6 only occurred in patients with a mutation (34.1%, p=0.06). Any grade 2 or greater toxicity occurred in 42.9% of non‐mutated and 84.1% of mutated patients (p=0.031). Grade 3 toxicities were only observed in patients with at least one mutation: 14.8% [5.92‐32.5] and 23.3% [11.8‐40.9] in C677T and A1298C mutations respectively. There was no statistically significant difference between the two mutations (p=0.51). We hypothesized that overall survival might be longer in patients with MTHFR SNPs, but in our cohort that difference was not statistically significant: 730 days in the mutated vs. 284 days in the non‐mutated patients (p=0.4). The frequency of MTHFR SNPs was dramatically greater in our lymphoma cohort than in the general population ( 33% for C766T and 24‐40% for A1298C mutations) raising the possibility that these mutations are themselves associated with CNS lymphomas. CONCLUSIONS: MTHFR mutations are associated with an increased frequency of toxicity and prolonged need for leucovorin rescue. Larger patient series may disclose a survival advantage for patients with MTHFR mutations. These mutations may be associated with the occurrence of CNS lymphoma.