Global DNA (hydroxy)methylation is not associated to MTX response at 3 months in early RA patients

Introduction: Methotrexate (MTX) is first-line therapy in early rheumatoid arthritis (eRA). ~20-40% of patients does not benefit from therapy and switch medication. Immediate effective therapy is crucial to restrain joint... [ view full abstract ]

Introduction: Methotrexate (MTX) is first-line therapy in early rheumatoid arthritis (eRA). ~20-40% of patients does not benefit from therapy and switch medication. Immediate effective therapy is crucial to restrain joint damage and achieve best response. Therefore, early biomarkers for MTX response are required. MTX interferes with the folate cycle, and might thereby indirectly inhibit methylation. 5-methyl cytosine (5mC) acts as gene expression regulator and can further oxidate into 5-hydroxymethyl cytosine (5hmC), also known to have gene regulatory functions. We investigated the association between changes in global DNA (hydroxy)methylation (Δ%5hmC and Δ%5mC) and MTX response (ΔDisease Activity Score28) over the first 3 months in eRA patients. Method: 222 eRA patients were included from the Treatment in Rotterdam Early Arthritis Cohort (tREACH, ISRCTN26791028), a stratified, randomized clinical trial. Patients received triple (MTX + SSZ+ HCQ) or monotherapy (MTX) combined with corticosteroids and folic acid (10mg/wk). DNA was isolated from leukocytes at baseline (t0) and 3 months (t3) of MTX use and was degraded into single nucleotides using a DNA Degradase Plus enzyme. (Hydroxy)methylated cytosines (5hmdC/5mdC) were measured simultaneously using liquid chromatography- electrospray ionization-tandem mass spectrometry with multiple reaction monitoring (LC-ESI-MS/MS-MRM). Percentages of global DNA 5mdC and 5hmdC were calculated in relation to the total guanine concentration. Associations between ΔDAS28 and Δ%(hydroxy) methylation were adjusted for t0 DAS28 in a linear regression model. Paired analysis were conducted using a paired sample t test. Results: %5hmdC significantly increased during the first 3 months of MTX treatment (0.0364% at t0 vs 0.0372%, p=0.009). %Δ5hmdC was not associated to ΔDAS28 (B= -0.054, p=0.997). %5mdC did not change over the first 3 months of treatment (4.41% at t0 vs 4.40% at t3, p=0.393) and %Δ5mdC was not associated to ΔDAS28 (B= -0.570, p=0.220). Discussion: Our results suggest no role for global (hydroxy)methylation in relation to MTX response over the first 3 months in eRA patients. The role of increased %5hmdC during the first 3 months of MTX treatment needs further exploration.