Whereas the optimum therapy for early rheumatoid arthritis (RA) has not been established, there is still interest in combination therapy with existing disease-modifying antirheumatic drugs, due to the cost and toxicity of biological agents. We previously reported a 24-month open-label, randomised study comparing a methotrexate and ciclosporin A combination with sulfasalazine monotherapy in 82 early, poor-prognosis RA patients (n  =  40 combination, n  =  42 sulfasalazine).1 In the combination arm, first ciclosporin A then methotrexate doses were optimised. After 24 months all patients were treated according to standard clinical care. The aim of this report is to present long-term efficacy and toxicity data.

At 12 months there was no difference in the mean disease activity score based on 28 joints (DAS28; 3.2 for both groups). Mean serum creatinine had increased in both groups (combination 93.6 μmol/l, sulfasalazine 90.6 μmol/l, p = 0.05).1 24-Month data demonstrated no differences between groups for American College of Rheumatology (ACR) response rates (⩾ACR50: combination 23 (58%), sulfasalazine 22 (52%), p = 0.644). However, there were fewer withdrawals due to lack of efficacy in the combination arm (combination 3 (<1%), sulfasalazine 11 (26%), p = 0.025).

At 24 months ciclosporin A was stopped and patients were evaluated over 6 months for short-term effects (n  =  19). The mean doses of methotrexate and ciclosporin A were 16 mg/week and 159 mg/day. Those requiring increase/additional disease-modifying antirheumatic drugs (n  =  9) had higher swollen joint counts at 24 months and had needed drug changes during the past 6 months of combination therapy. Renal function improved (creatinine clearance diagnosis 97.0 ml/min, 24 months 81.3, 27 months 89.4), although creatinine remained unchanged (30 months 93.1 μmol/l). Three out of six patients who started antihypertensives during ciclosporin A treatment were able to withdraw them.

At 7.2 years (range 6.0–10.4), data were available for 26 patients in each group. Reasons for incomplete follow-up included death (two patients/group) and relocation. Table 1 demonstrates no significant differences in clinical outcomes. One patient in the combination group compared with four in the sulfasalazine group required biological therapy for disease control. Fourteen in the combination group remained on methotrexate monotherapy after ceasing ciclosporin A, whereas only four in the sulfasalazine group continued on sulfasalazine monotherapy (p = 0.003) and a further six switched to methotrexate monotherapy. The antihypertensive requirement was similar (combination eight patients, sulfasalazine five patients, p = 0.346). There was no difference in creatinine between groups. Creatinine only were available for a further 11 patients (n  =  63) and again, there was no difference between groups (combination 93.4 μmol/l, sulfasalazine 96.1 μmol/l, p = 0.945).

This study reported patients treated according to standard clinical care after 24 months of protocol-driven therapy. More patients in the sulfasalazine group withdrew due to lack of efficacy from the initial 24-month study and subsequently more required further therapy, including biological agents. In addition, more patients in the combination group were maintained long term on monotherapy. Despite the initial study open-label design, drop-out rate and lack of power to detect significant differences between the groups for therapy, these findings give further insight into initial non-biological combination therapy in early, poor-prognosis RA. Creatinine had not deteriorated, although creatinine clearance values were not collected. There was no excess of antihypertensive use in the combination group.

In conclusion, long-term follow-up of early, poor-prognosis RA patients treated with an initial methotrexate–ciclosporin A combination may confer better outcomes compared with sulfasalazine monotherapy.