R. F. van Vollenhoven*¹, S. Ernestam¹, P. Geborek², I. F. Petersson², L. Cöster³, E. Waltbrand⁴, A. Zickert⁵, J. Theander⁶, Å. Thörner⁷, H. Hellström⁸, A. Teleman⁹, C. Dackhammar¹⁰, F. Akre¹¹, K. Forslind¹², L. Ljung¹³, R. Oding¹⁴, M. Wörnert¹, J. Bratt^1
1Rheumatology, Karolinska University Hospital, Stockholm, ²Rheumatology, University Hospital, Lund, ³Rheumatology, University Hospital, Linköping, ⁴Rheumatology, Lasarettet, Borås, ⁵Rheumatology, Danderyds Hospital, Stockholm, ⁶Rheumatology, Centralsjukhuset, Kristianstad, ⁷Rheumatology, Mälarsjukhuset, Eskilstuna, ⁸Rheumatology, Lasarettet, Falun, ⁹Rheumatology, Spenshult Hospital, Halmstad, ¹⁰Rheumatology, University Hospital, Mölndal, ¹¹Rheumatology, University Hospital, Örebro, ¹²Rheumatology, Lasarettet, Helsingborg, ¹³Rheumatology, University Hospital, Umeå, ¹⁴Rheumatology, Centrallasarettet, Västerås, Sweden

Background: New treatment strategies for early rheumatoid arthritis (RA) are evolving rapidly. The SWEFOT trial was designed to compare two strategies in patients who had failed an initial 3-4 months trial period with methotrexate (MTX) monotherapy.

Objectives: To compare, in an open, randomized, controlled trial, the addition of conventional DMARDs (sulfasalazine + hydroxychloroquine, with the option of switching to cyclosporin A) versus the addition of anti-TNF (infliximab, with the option of switching to etanercept).

Methods: A total of 487 patients with early RA (symptom duration <1 year) were started on MTX at a rapidly escalated dosage up to 20 mg/week. After 3-4 months, the 258 patients who had not achieved DAS28<3.2 (but who could tolerate MTX) were randomized. A single switch within the same strategy was allowed for intolerance only. At 12 months, the continuation rate on therapy according to protocol was assessed.

Results: The two arms were well-matched in terms of demographics, disease activity at baseline, and disease activity at randomization. In arm A, conventional DMARD combination therapy, 88 out of 130 patients were receiving treatment according to protocol at the 12 months visit. Of these, 4 had switched to cyclosporin A. In arm B, anti-TNF therapy, 105 out of 128 patients were on treatment according to protocol (5 of whom had switched to etanercept). The difference between the groups was statistically significant (p<0.01, Fisher’s exact test), and was explained by a higher number of drop-outs due to lack of efficacy (group A: 18; group B: 3), whereas the numbers of drop-outs due to side-effects and to other causes were similar between the groups. Final database validation is progressing in stages and the primary outcome in this trial (Eular response at 12 months) will be available in April 2008.