Inadequate response to treat-to-target methotrexate therapy in patients with new-onset rheumatoid arthritis: development and validation of clinical predictors

Objective To identify and validate clinical baseline predictors associated with inadequate response (IR) to methotrexate (MTX) therapy in newly diagnosed patients with rheumatoid arthritis (RA). Methods In U-Act-Early, 108 disease-modifying antirheumatic drug (DMARD)-naive patients with RA were randomised to initiate MTX therapy and treated to target until sustained remission (disease activity score assessing 28 joints (DAS28) <2.6 with four or less swollen joints for >= 24 weeks) was achieved. If no remission, hydroxychloroquine was added to the treatment regimen (ie, 'MTX+') and replaced by tocilizumab if the target still was not reached thereafter. Regression analyses were performed to identify clinical predictors for IR, defined as needing addition of a biological DMARD, to 'MTX+'. Data from the treatment in the Rotterdam Early Arthritis Cohort were used for external validation of the prediction model. Results Within 1 year, 56/108 (52%) patients in U-Act-Early showed IR to 'MTX+'. DAS28 (adjusted OR (ORadj) 2.1, 95% CI 1.4 to 3.2), current smoking (ORadj 3.02, 95% CI 1.1 to 8.0) and alcohol consumption (ORadj 0.4, 95% CI 0.1 to 0.9) were identified as baseline predictors. The area under the receiver operator characteristic curve (AUROC) of the prediction model was 0.75 (95% CI 0.66 to 0.84); the positive (PPV) and negative predictive value (NPV) were 65% and 80%, respectively. When applying the model to the validation cohort, the AUROC slightly decreased to 0.67 (95% CI 0.55 to 0.79) and the PPV and NPV to 54% and 80%, respectively. Conclusion Higher DAS28, current smoking and no alcohol consumption are predictive factors for IR to step-up 'MTX+' in DMARD-naive patients with new-onset RA.