Treatment with methotrexate plus oral prednisolone versus triple therapy (methotrexate/sulfasalazine/hydroxychloroquine) plus intra-articular glucocorticoids in early rheumatoid arthritis: a prespecified nonrandomised subgroup analysis of clinical and radiographic data at 48 weeks from the NORD-STAR trial's conventional treatment arm

OBJECTIVES: In the NOrdic Rheumatic Diseases Strategy Trials And Registries (NORD-STAR) trial, the active conventional arm had 2 nonrandomised regimens: arm 1A (oral group; Sweden, Norway, Netherlands, and Iceland) and arm 1B (injection group; Denmark and Finland). We report clinical, patient-reported, safety, and radiographic outcomes after 48 weeks.

METHODS: Oral group received methotrexate plus oral prednisolone (20.0 mg/d, tapered rapidly, discontinued week 36); Injection group received triple therapy (methotrexate, sulphasalazine, hydroxychloroquine) and mandatory intra-articular glucocorticoid injections. The primary end point was analysed by logistic regression with several approaches for handling missing outcomes.

RESULTS: In total, 137 and 80 patients were included in the oral group and injection group; 78% vs. 89% completed, respectively. At 48 weeks, adjusted clinical disease activity index remission ≤2.8 rates (95% CI) were 36% (28-44) and 55% (42-68), respectively; the risk difference (primary outcome) was 19% (2-35). Similarly, key secondary clinical, patient-reported and safety outcomes showed numerically better results in the injection group vs oral group, for example, infections occurred in 53% vs 30%, respectively. Radiographic progression (Δtotal van der Heijde-modified Sharp Score) was low: oral group: adjusted mean, 0.26 (95% CI, 0.08-0.43); injection group: adjustedd mean, 0.80 (95% CI, 0.55-1.05). Cumulative dose of oral/intra-articular glucocorticoids (median) was 1905 mg prednisolone for the oral group and 165 mg for the injection group.

CONCLUSIONS: In treatment-naïve patients with early rheumatoid arthritis, triple therapy and mandatory glucocorticoid joint injections had numerically better clinical outcomes, fewer withdrawals, fewer adverse events, and lower cumulative dose of glucocorticoids, but slightly worse radiographic outcomes than treatment with methotrexate and oral prednisolone. These findings, although nonrandomised, suggest a potential for optimising treatment strategy with conventional therapies in early rheumatoid arthritis.