Twitter Facebook LinkedIn
Center for Global Research Data

Individualized Prediction of Treatment Response to Methotrexate in Patients with Rheumatoid Arthritis: A Machine Learning Approach

Lead Investigator: Elena Myasoedova, Mayo Clinic
Title of Proposal Research: Individualized Prediction of Treatment Response to Methotrexate in Patients with Rheumatoid Arthritis: A Machine Learning Approach
Vivli Data Request: 5352
Funding Source: None.
Potential Conflicts of Interest: None

Summary of the Proposed Research:

Timely and effective treatment for rheumatoid arthritis (RA), a chronic disabling autoimmune disease, is essential to avoid irreversible joint and organ damage. Methotrexate (MTX) is the most common anchor drug for RA, but the risk of missing the opportunity for early effective treatment with alternative medications is substantial given the delayed onset of MTX action and 30-40% non-response rate. American College of Rheumatology recommendations for escalation of RA therapy are based on a trial-and-error approach with a moderate/low level of evidence, resulting in a substantial knowledge gap regarding early identification of treatment responders. Clinical and genetic markers have been suggested as potential predictors of non- response to MTX, however a predictive algorithm efficiently and effectively identifying patients at RA onset who are likely to respond to MTX is lacking. We aim to apply methods of artificial intelligence (AI) for integration and processing of sociodemographic, clinical, and serological data obtained from patients with early RA to develop and validate a clinically meaningful predictive algorithm of individual’s response to MTX.

Specifically, we aim to: (1) Identify clusters of patients based on RA disease activity scores. We hypothesize that: (H1) Certain RA disease characteristics (e.g., serologic status) and sociodemographic factors can be linked to a particular RA disease activity cluster. (2) Develop and validate an individualized predictive algorithm for response to MTX in RA patients. We hypothesize that: (H2) Remission at 3 months of MTX use can be robustly predicted using a combination of clinical, serological and sociodemographic data; (H3) The predictive algorithm for MTX response differs depending on patients’ serological status. (H4) Our predictive algorithm is valid in discriminating MTX responders and non-responders at 3 months in our validation cohort; (H5) Our predictive algorithm is valid in discriminating MTX responders and non-responders at 6 and 12 months, using clinical trial data.

This proposal will take advantage of a large pre-identified development cohort (n=1424) and an independent validation cohort (n=429) of patients with RA on MTX with available clinical data as a part of an established collaboration between the Mayo Clinic, National Institute of Health (NIH)-Pharmacogenetics Research Network (PGRN), the PhArmacogenetics of Methotrexate in Rheumatoid Arthritis (PAMERA) consortium and the UK Maximizing Therapeutic Utility in RA (MATURA) consortium. Innovative findings from this study will provide pivotal information for individualized prediction of MTX response by applying a machine learning (ML) approach to sociodemographics, clinical, and serological data in patients with early RA. This information is instrumental for overcoming critical barriers in early effective management of RA and will seed ground for the development of predictive models for other antirheumatic medications (i.e., biologics), reshaping the current rheumatology care and improving RA outcomes.

Requested Studies:

Efficacy and Safety of CDP870 400 mg in Combination With Methotrexate Compared to Methotrexate Alone in the Treatment of the Signs and Symptoms of Patients With Rheumatoid Arthritis Who Are Partial Responders to Methotrexate
Sponsor: UCB
Study ID: NCT00544154
Sponsor ID: C87014

A Phase III Multi-center, Double-blind, Placebo-controlled, Parallel Group 24-Week Study to Assess the Efficacy and Safety of Two Dose Regimens of Liquid Certolizumab Pegol as Additional Medication to Methotrexate in the Treatment of Signs and Symptoms of Rheumatoid Arthritis and in Prevention of Joint Damage in Patients With Active Rheumatoid Arthritis Who Have an Incomplete Response to Methotrexate.
Sponsor: UCB
Study ID: NCT00160602
Sponsor ID: C87050

A Phase IIIb Multicenter Study With a 12-week Double-blind, Placebo-controlled, Randomized Period Followed by an Open-label, Extension Phase Evaluating Safety/Efficacy of Certolizumab Pegol Given to Patients With Active Rheumatoid Arthritis.
Sponsor: UCB
Study ID: NCT00717236
Sponsor ID: C87094

A Multi-center, Randomized, Double-blind, Placebo-controlled Study to Evaluate the Efficacy and Safety of Certolizumab Pegol in Combination With Methotrexate for Inducing and Sustaining Clinical Response in the Treatment of DMARDNaïve Adults With Early Active Rheumatoid Arthritis
Sponsor: UCB
Study ID: NCT01521923
Sponsor ID: RA0055 Period 2

A Multi-center, Randomized, Double-blind, Placebo-controlled Study to Evaluate the Efficacy and Safety of Certolizumab Pegol in Combination With Methotrexate for Inducing and Sustaining Clinical Response in the Treatment of DMARDNaïve Adults With Early Active Rheumatoid Arthritis
Sponsor: UCB
Study ID: NCT01519791
Sponsor ID: RA0055 Period 1