Am Fam Physician. 2010;82(2):196
Background: Biologic agents may help patients with rheumatoid arthritis (RA) who have an unsatisfactory response to traditional disease-modifying antirheumatic drugs (DMARDs) such as methotrexate. However, there are few studies comparing the relative effectiveness and safety of biologics against each other. Singh and colleagues compared the benefits and safety of these agents in patients with RA.
The Study: The authors performed a meta-analysis of all Cochrane reviews of biologic therapy for RA. The data were then used to calculate the relative benefits and safety of the agents compared with each other. Benefit was defined as a 50 percent improvement in symptoms of RA using the American College of Rheumatology criteria (ACR50), and safety was defined as the number of withdrawals from therapy because of adverse events.
Results: A total of six Cochrane reviews comprising 31 studies were evaluated. Most of the trials compared a biologic with placebo; both groups also received a traditional DMARD such as methotrexate. Overall, biologics were more likely than placebo to achieve an ACR50 response (odds ratio [OR] = 3.35), but were also more likely than placebo to be stopped because of adverse events (OR = 1.39). Each biologic was more likely than placebo to achieve an ACR50 (OR = 2.92 to 4.97) except for anakinra (Kineret). Patients taking adalimumab, anakinra, and infliximab were more likely than those taking placebo to discontinue treatment (OR = 1.54 to 2.21), but no difference was noted in those taking abatacept, etanercept, or rituximab (Rituxan; see accompanying table).
| Biologic agent | Likelihood of achieving ACR50 | Safety* | ||
|---|---|---|---|---|
| Odds ratio | NNT | Odds ratio | NNH | |
| Abatacept (Orencia) | 2.98 | 5 | NS | NS |
| Adalimumab (Humira) | 3.70 | 4 | 1.54 | 39 |
| Anakinra (Kineret) | NS | NS | 1.67 | 31 |
| Etanercept (Enbrel) | 4.97 | 3 | NS | NS |
| Infliximab (Remicade) | 2.92 | 5 | 2.21 | 18 |
| Rituximab (Rituxan) | 4.10 | 4 | NS | NS |
Indirect comparison of the treatments showed that anakinra was less effective than adalimumab and etanercept in achieving ACR50. Fewer patients stopped etanercept because of adverse effects than did persons taking adalimumab, anakinra, or infliximab. The treatments were similarly effective in patients regardless of duration of RA.
Conclusion: Etanercept and adalimumab are more effective than anakinra in improving the symptoms of RA, and etanercept is better tolerated than adalimumab, anakinra, and infliximab. This may help physicians make evidence-based choices about using pharma-cotherapies in the treatment of RA.
