Background/Purpose: Recently, there has been a trend to start methotrexate (MTX) in higher doses, either as monotherapy or in combination with other drugs in rheumatoid arthritis (RA) trials –and likely also in daily practice. It is unclear whether higher initial doses are associated with better short term clinical responses, especially in combination therapy with other effective disease modifying anti-rheumatic drugs (DMARDs). We investigated the short term relationship with early clinical response of various doses of MTX in monotherapy and combination therapy in DMARD naive early RA patients.

Methods: A systematic literature search was performed, including early, DMARD naive RA patients, treated with MTX, showing disease activity results within 3 to 6 months follow-up. Cohen’s effect sizes (ratio of mean change in score and baseline SD, with negative scores indicating improvement) were calculated for the health assessment questionnaire (HAQ), erythrocyte sedimentation rate/c-reactive protein (ESR/CRP) and/or DAS/DAS28 in 4 treatment groups: MTX monotherapy, and MTX in combination with synthetic (cs)DMARDs, or with biologic (b)DMARDs or with oral glucocorticoids. Multivariate random-effects meta-regression analyses were performed for each outcome, with treatment group as predictor corrected for standardized baseline disease activity and month of assessment.

Results: Out of 2567 articles and 417 meeting abstracts, 31 studies including 5589 patients were extracted. The meta-regression (table 1) did not show any indication for  higher effectiveness of increasing MTX doses in monotherapy. In combination with glucorticoids a higher MTX dose was statistically significantly associated with higher HAQ (β=0.012, 95% CI 0.00070;0.023), but not with DAS/DAS28 or ESR/CRP. In combination with bDMARDs a higher MTX dose was statistically significantly associated with higher HAQ (β=0.042, 95% CI 0.012;0.073) and DAS/DAS28 (β=0.033, 95% CI 0.0070;0.059). These effect sizes were too small to indicate clinical relevant effects on disease activity/functional ability. There were too few treatment groups using MTX in combination with csDMARDs to evaluate in the meta-regression.

Conclusion: In DMARD naive early RA patients, a higher initial dose of MTX either as monotherapy or in combination with a bDMARD or glucocorticoid was not associated with better clinical responses within 3 to 6 months of treatment start. Initial combination therapies with a bDMARD or glucocorticoid show higher effect sizes than initial monotherapy. The results suggest that such initial combination therapies give better short term clinical outcomes, but that the benefits to be expected from higher in stead of lower doses of MTX is negligible.