Rheumatoid arthritis: choice of antirheumatic treatment. Methotrexate first.

Various drugs with different risk-benefit balances are claimed to reduce the intensity and frequency of exacerbations and to slow the progression of rheumatoid arthritis. What is the basis for choosing the most appropriate treatment for a given patient? Methotrexate is the best-assessed antirheumatic drug, and the one with which we have the most experience. At doses of 7.5 mg to 25 mg per week, methotrexate relieves pain, reduces the number of affected joints, and provides a functional improvement. It has the adverse effects common to all immunosuppressants, particularly gastrointestinal and haematological disorders. Treatment withdrawals due to adverse effects are infrequent at the doses used in rheumatoid arthritis. Other synthetic antirheumatic drugs such as azathioprine, chloroquine and its derivatives, ciclosporin, cyclophosphamide, D-penicillamine, leflunomide, gold salts and sulfasalazine are no more effective than methotrexate. Some are less effective, and others are more toxic. When used as monotherapy in clinical trials, TNF-alpha antagonists (etanercept and adalimumab) were no more effective than methotrexate on clinical outcome after one or two years. Etanercept and adalimumab seem to have similar risk-benefit balances. In about 10 comparative trials, combination of methotrexate and a TNF-alpha antagonist was more effective than methotrexate monotherapy on functional status and symptoms, especially in initially severe rheumatoid arthritis. Although it has not been proven, the risk of severe adverse effects is likely to be higher with such combinations, and follow-up in comparative trials does not exceed two years. The use of TNF-alpha antagonists is also less convenient, as weekly injections or monthly infusions are necessary. There is no firm evidence that other combinations of antirheumatic drugs are more effective than TNF-alpha antagonist-methotrexate combinations. In patients in whom TNF-alpha antagonists had failed, a combination of rituximab and methotrexate was more effective than methotrexate alone. There is no firm evidence that tocilizumab (an interleukin inhibitor) or abatacept (a drug acting on T lymphocytes) has a better risk-benefit balance than rituximab. The symptomatic efficacy of long-term corticosteroid therapy lasts at least three months but usually less than a year. Efficacy in patients in whom other treatments have failed is uncertain. There is no evidence that introducing methotrexate soon after diagnosis, alone or in combination with a TNF-alpha antagonist, slows the progression of rheumatoid arthritis. In practice, methotrexate is the first-line antirheumatic drug. If methotrexate monotherapy is ineffective, or when rheumatoid arthritis is initially severe, adding a TNF-alpha antagonist can be beneficial. A third-line option is to combine rituximab with methotrexate.