Background Low-dose methotrexate (MTX) is the first-line therapy in early rheumatoid arthritis (eRA). Up to 40% of eRA patients do not benefit from MTX therapy. MTX has been shown to inhibit one-carbon metabolism, which is involved in the donation of methyl groups. In this study, we investigate baseline global DNA methylation and changes in DNA methylation during treatment in relation to clinical non-response after 3 months of MTX treatment. Methods Two hundred ninety-four blood samples were collected from the Treatment in the Rotterdam Early Arthritis Cohort (tREACH, ISRCTN26791028), a multicenter, stratified single-blind clinical trial of eRA patients. Global DNA (hydroxy)methylation was quantified using liquid chromatography-electrospray ionization-tandem mass spectrometry (LC-ESI-MS/MS) and validated with a global DNA LINE-1 methylation technique. MTX response was determined as ΔDAS28. Additionally, patients were stratified into two response groups according to the European League Against Rheumatism (EULAR) response criteria. Associations between global DNA methylation and response were examined using univariate regression models adjusted for baseline DAS28, baseline erythrocyte folate levels, and body mass index (BMI). Results Higher baseline global DNA methylation was associated with less decrease of DAS28 ( β = 0.15, p = 0.013) and with MTX non-response (OR = 0.010, 95% CI = 0.001–0.188). This result was validated in LINE-1 elements ( β = 0.22, p = 0.026). Changes in global DNA (hydroxy)methylation were not associated with MTX response over 3 months. Conclusions These results show that higher baseline global DNA methylation in treatment naïve eRA patients is associated with decreased clinical response after 3 months of treatment of eRA patients and can be further evaluated as a predictor for MTX therapy non-response. Trial registration ISRCTN, ISRCTN26791028 , registered 23 August 2007—retrospectively registered Electronic supplementary material The online version of this article (10.1186/s13075-019-1936-5) contains supplementary material, which is available to authorized users.

The goals of this study were to examine whether machine-learning algorithms outperform multivariable logistic regression in the prediction of insufficient response to methotrexate (MTX); secondly, to examine which features are essential for correct prediction; and finally, to investigate whether the best performing model specifically identifies insufficient responders to MTX (combination) therapy. The prediction of insufficient response (3-month Disease Activity Score 28-Erythrocyte-sedimentation rate (DAS28-ESR) > 3.2) was assessed using logistic regression, least absolute shrinkage and selection operator (LASSO), random forest, and extreme gradient boosting (XGBoost). The baseline features of 355 rheumatoid arthritis (RA) patients from the “treatment in the Rotterdam Early Arthritis CoHort” (tREACH) and the U-Act-Early trial were combined for analyses. The model performances were compared using area under the curve (AUC) of receiver operating characteristic (ROC) curves, 95% confidence intervals (95% CI), and sensitivity and specificity. Finally, the best performing model following feature selection was tested on 101 RA patients starting tocilizumab (TCZ)-monotherapy. Logistic regression (AUC = 0.77 95% CI: 0.68–0.86) performed as well as LASSO (AUC = 0.76, 95% CI: 0.67–0.85), random forest (AUC = 0.71, 95% CI: 0.61 = 0.81), and XGBoost (AUC = 0.70, 95% CI: 0.61–0.81), yet logistic regression reached the highest sensitivity (81%). The most important features were baseline DAS28 (components). For all algorithms, models with six features performed similarly to those with 16. When applied to the TCZ-monotherapy group, logistic regression’s sensitivity significantly dropped from 83% to 69% (p = 0.03). In the current dataset, logistic regression performed equally well compared to machine-learning algorithms in the prediction of insufficient response to MTX. Models could be reduced to six features, which are more conducive for clinical implementation. Interestingly, the prediction model was specific to MTX (combination) therapy response.