Response to methotrexate predicts long-term mortality of patients with rheumatoid arthritis independent of the degree of response: results of a re-evaluation 30 years after baseline.

Objective To assess if there is a correlation between the degree of response to treatment with methotrexate (MTX) and long-term mortality in a cohort of patients with rheumatoid arthritis (RA) established in Germany in the early eighties. Methods RA patients who had started MTX treatment between 1980 and 1987 were included. One year after baseline, the treatment response was evaluated. Responders were defined as patients with at least 20% decline in the swollen joint count (out of 32 joints) and the ESR with a prednisone dosage <5 mg/day. Thereafter, assessments were performed at 10, 18, and 30 years after baseline. Standardised mortality ratios (SMR) were calculated, Cox regression and logistic regression were performed. Results The cohort comprised 271 patients. In 2015, about 30 years after the initiation of MTX therapy, 185 patients (68%) were deceased, 52 (19%) lost to follow-up and 34 alive. The response after the first year of MTX treatment was the strongest predictor of survival with a hazard ratio of 0.44 (95% confidence interval [CI]: 0.30-0.65). However, even responders still had an SMR of 1.37 (95% CI 1.31-1.65), but this was much worse for non-responders who had an SMR of 4.22 (95% CI 3.13-5.56). Using Cox regression analysis no difference was detected between responders with more than 50% improvement (38% of all patients) and those with 20-50% improvement (28%). Conclusion The predictive value of a response to one year of MTX therapy for long-term mortality of RA patients is independent of the degree of response.