Background/Purpose: In rheumatoid arthritis (RA), biologics are more efficient in combination with methotrexate (MTX) than in monotherapy. When MTX cannot be used, others csDMARD can be proposed in combination with biologics. However, the effectiveness of other csDMARD compared to MTX in combination with bDMARD is not clear. The aim of our study was to assess currently available literature on the association of csDMARD with biologics, compared with MTX.

Methods: We systematically searched literature (via Pubmed, Embase and abstracts from ACR and EULAR congresses) for studies that compare effectiveness, retention rate and safety of MTX versus other csDMARD (leflunomide [LEF] or all non-MTX csDMARD) as concomitant therapy with rituximab (RTX), TNF inhibitors (TNFi), abatacept, tocilizumab and JAK inhibitor (JAKi). A meta-analysis was performed with RevMan software using an inverse variance approach with fixed or random effects models, depending on the presence of heterogeneity. Data were extracted by one investigator and independently checked by another.

Results: From 3842 articles, the literature search revealed 144 articles and abstracts of potential interest, and further examination resulted in 8 studies fulfilling required criteria for RTX and 10 for TNFi. All of these studies were cohort studies. For tocilizumab, JAKi and abatacept, there were not enough studies to conduct meta-analysis. For patients receiving RTX, who had similar baseline characteristics in the two groups, those treated with LEF had a higher EULAR good response rate than those treated with MTX (n=3250, 5 studies, RR=1.46 (95% confidence interval [95% CI] 1.25;1.70), I²=0%, p<0.001). The variation of DAS28 at 6 months also tended to be higher in patients treated with LEF (-0.18 [-0.38;0.01], n=1449, 2 studies, I²=0%, p=0.07). The risk of adverse events also tended to be lower in patients treated with LEF (n=2718, 4 studies, RR=0.73 [0.52;1.03]; I²=0%, p=0.07).

Regarding patients receiving TNFi, those receiving MTX had a higher EULAR good or moderate response rate than those treated with other csDMARDs (n=1859, 3 studies, RR=0.88 [0.81;0.96]; I²=0%, p=0.004). The change in DAS28 at 6 months was also higher in patients treated by MTX than other csDMARD (-0.28 [-0.51;-0.05]; I²=0%, p=0.02). The risk ratio of discontinuing therapy due to adverse events at 6 months and the risk ratio of serious adverse events were similar between MTX and other csDMARD.

Conclusion: LEF or other csDMARD are safe and efficient for people who are intolerant to MTX to combine with biologics. Whereas MTX appears to be superior than other csDMARD in combination with TNFi, LEF might be superior than MTX in combination with RTX. Randomized controlled trials should be conducted to confirm this result.

« Back to ACR/ARHP Annual Meeting 2018

ACR Meeting Abstracts - https://acrabstracts.org/abstract/should-we-prefer-leflunomide-to-methotrexate-in-combination-with-biologics-a-systematic-review-and-a-meta-analysis/