Real‐World Outcomes Associated With Methotrexate, Sulfasalazine, and Hydroxychloroquine Triple Therapy Versus Tumor Necrosis Factor Inhibitor/Methotrexate Combination Therapy in Patients With Rheumatoid Arthritis

OBJECTIVE: Though randomized controlled trials have demonstrated relatively comparable clinical outcomes with triple therapy (methotrexate [MTX], sulfasalazine [SSZ], and hydroxychloroquine [HCQ]) compared to combination therapy (tumor necrosis factor inhibitor [TNFi] and MTX), real-world experiences comparing these strategies have not been well studied. METHODS: We evaluated the clinical effectiveness and effects of medication discontinuation of triple therapy with MTX/SSZ/HCQ versus combination therapy with TNFi/MTX in rheumatoid arthritis (RA) patients enrolled in the Corrona RA Drug Safety & Effectiveness Registry. Propensity score matching was used to match patients up to a ratio of 1:3 to adjust for imbalances between treatment groups, with stratification performed according to biologics-naive or biologics-exposed status of study participants. RESULTS: Patients eligible for analysis in this study included biologics-naive RA patients (3,926 who received combination therapy with TNFi/MTX and 262 who received triple therapy with MTX/SSZ/HCQ) and biologics-exposed RA patients (3,365 who received combination therapy with TNFi/MTX and 130 patients who received triple therapy with MTX/SSZ/HCQ). Before propensity score matching, numerous factors were imbalanced between the treatment groups, with triple therapy patients generally being older, having a longer disease duration of RA and lower RA disease activity, and more likely having a history of malignancy and other comorbidities. After matching, almost all (93-98%) triple therapy patients could be matched to TNFi/MTX therapy patients, and cohort characteristics were generally well balanced. Discontinuation of medication was greater in triple therapy patients referent to TNFi/MTX therapy patients (adjusted hazard ratio [HR] of 2.17 [95% confidence interval 1.63-2.88] in the biologics-naive group; adjusted HR of 1.51 [95% confidence interval 1.06-2.15] in the biologics-exposed group). At 6 months, the proportion of biologics-naive patients attaining low disease activity was significantly greater in the TNFi/MTX treatment group (49.2% in TNFi/MTX therapy patients versus 33.3% in triple therapy patients), as was the mean change in Clinical Disease Activity Index scores (-9.3 units versus -5.5 [95% confidence interval -1.5, -6.1]). Corresponding results in the biologics-exposed patients numerically favored TNFi/MTX therapy compared to triple therapy but did not reach statistical significance. CONCLUSION: Few patients receive triple therapy with MTX/SSZ/HCQ in the US. In the present study, drug persistence and clinical effectiveness outcomes were less favorable in triple therapy patients compared to TNFi/MTX therapy patients.