Longitudinal measurement of methotrexate liver concentrations does not correlate with liver damage, clinical efficacy, or toxicity during a 3.5 year double blind study in rheumatoid arthritis.

Abstract

OBJECTIVES: In patients with rheumatoid arthritis (RA), we examined whether methotrexate (MTX) and MTX polyglutamate accumulation in the liver correlated with clinical efficacy or clinical/laboratory toxicity. We also began preliminary examination of a new histologic index of liver histology (the Iowa Score) relative to the Roenigk grading system. METHODS: Forty patients with RA participated in a prospective, double blind, 3.5 year study of MTX treatment. Liver biopsies, liver MTX and MTX polyglutamate concentrations, laboratory tests, evaluation of disease activity, and evaluation of adverse events were done prospectively at baseline and at 1, 2, and 3.5 years. Biopsies were examined using the Roenigk grading system and an additional histological scoring system. Radiochemical ligand binding assays and HPLC methods were used to measure MTX and MTX polyglutamates. Statistical analysis included ANOVA, linear regression, and logistic regression modeling. RESULTS: No significant changes in the mean values of aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase, albumin, or hemoglobin occurred. A significant percentage of patients had at least one abnormal alkaline phosphatase, AST, or ALT (25 to 52%), although most abnormalities were small and transient. Histological abnormalities did not progress using either the Roenigk or the Iowa score. The last abnormal AST, the number of abnormal AST and ALT, and female sex correlated with histological liver abnormalities (r2 = 0.41) using a new preliminary histologic scoring system (the Iowa Score). Amount of alcohol use correlated with fatty change, and the MTX dose at biopsy was associated with liver histological abnormalities (p = 0.03 and 0.049, respectively). Total liver MTX concentrations were stable from Year 1 to Year 3.5 and the percentage of higher order polyglutamates was relatively high (38 to 56%) relative to monoglutamates. No correlation of these concentrations with clinical response or toxicity, histology, or liver function tests could be documented. CONCLUSION: This analysis describes the accumulation and stabilization of MTX concentrations in the liver and examined correlations between MTX liver concentrations, patient demographics, liver histology, concomitant medications, and disease activity. No such correlations were found, decreasing the likelihood that MTX concentrations in serum would be useful measures to predict significant hepatotoxicity.