Skip to main content

27-07-2018 | Rheumatology | News | Article

Model developed to predict methotrexate nonresponse in RA patients

medwireNews: Researchers have a developed a model including clinical and psychosocial factors to predict which patients with rheumatoid arthritis (RA) are unlikely to respond to methotrexate treatment.

Suzanne Verstappen (University of Manchester, UK) and colleagues used data from the prospective observational UK RAMS study to identify predictors of methotrexate nonresponse in 1050 patients with RA or undifferentiated polyarthritis who were starting methotrexate treatment for the first time, either as monotherapy or in combination with other conventional DMARDs or oral steroids.

After 6 months of treatment, 43% of patients were classified as methotrexate nonresponders, defined as “no response” according to the EULAR response criteria (Disease Activity Score at 28 joints [DAS28] improvement ≤0.6, or DAS28 improvement >0.6 but ≤1.2 and 6-month DAS28 score >5.1), treatment discontinuation due to inefficacy, or commencing biologic DMARD treatment. DAS28 scores were calculated using C-reactive protein, 28-joint counts and visual analogue scales for general well-being.

On multivariable analysis, patients who were rheumatoid factor (RF)-positive were a significant 38% more likely to respond to methotrexate than RF-negative patients, whereas the likelihood of nonresponse was significantly higher among patients with higher Health Assessment Questionnaire (HAQ) scores, tender joint counts and Hospital Anxiety and Depression Scale (HADS) anxiety scores, at 64%, 6%, and 7% per unit increase, respectively.

Conversely, the researchers found that higher baseline disease activity was linked to greater chance of methotrexate response, with each unit increase in DAS28 score conferring a 71% higher likelihood of response. They suggest that this “surprising result” may have been “an inevitable consequence of the method of calculating non-response”, because patients with lower DAS28 scores at baseline have less potential to achieve at least a 0.6-point improvement in this score.

In sensitivity analyses excluding patients in remission or low disease activity at baseline, however, lower DAS28 score remained a significant predictor of methotrexate nonresponse.

When the independent predictors of methotrexate nonresponse (RF negativity, higher HAQ score, higher tender joint count, higher HADS anxiety score and lower DAS28 score) were combined, the model correctly distinguished between methotrexate responders and nonresponders on 74% of occasions after correcting for optimism.

“[T]his is the first model to be developed using a large cohort of patients with RA starting [methotrexate] for the first time and recruited from routine clinical care”, write Verstappen and colleagues in Arthritis Research & Therapy.

Such a model “could enable earlier access to alternative medications such as biologic therapy and the avoidance of disease progression for some patients”, the authors add. They caution, however, that their model now requires independent validation in an external dataset.

By Claire Barnard

medwireNews is an independent medical news service provided by Springer Healthcare. © 2018 Springer Healthcare part of the Springer Nature group