Health UK (Commonwealth Union) – Juvenile Idiopathic Arthritis (JIA) is a chronic autoimmune disease that affects children and adolescents, causing inflammation in the joints. Formerly known as Juvenile Rheumatoid Arthritis, JIA is a heterogeneous condition with various subtypes, making it a complex challenge for both patients and healthcare professionals.
University of Manchester scientists are leveraging machine learning to potentially target the most responsive candidates among children and young people with arthritis for the primary treatment, methotrexate. While methotrexate is the initial drug prescribed for JIA, it proves effective or tolerable in only half of the recipients. Those not benefitting from the drug face prolonged waits for second-line therapies, exacerbating severe joint pain and related symptoms. Published in eBioMedicine, the study aims to enhance targeted research by identifying predictors of methotrexate response, such as biomarkers, leading to improved outcome predictions. The research reveals that one in eight children starting methotrexate experience improved inflammatory features, even with lingering symptoms. Additionally, 16% of children on methotrexate may show slower disease activity improvement compared to their peers.
Dr. Stephanie Shoop-Worrall, the lead author, emphasizes the importance of avoiding unnecessary exposure to potential side effects by administering methotrexate only to those likely to benefit.
“But now machine learning has opened the door made it possible to predicting which aspects of a child’s disease would be helped by the drug and so which children should start other therapies either alongside or instead of methotrexate straight away.
“In addition, this work shows how clinical trials are missing the mark in only looking at drug ‘response’ or ‘non-response’ for childhood-onset arthritis.
“This oversimplification could lead to a drug being labelled as ‘effective’ when key symptoms such as pain remain, or ‘ineffective’ where a significant improvement is seen in one aspect of this complex disease.”
Funding for this research has been provided by the Medical Research Council, Versus Arthritis, Great Ormond Street Hospital Children’s Charity, Olivia’s Vision, and the National Institute for Health Research, as part of the CLUSTER consortium.
The research team utilized data from four nationwide cohorts of children and young individuals who initiated their treatment prior to January 2018.
Components of the Juvenile Arthritis Disease Activity Score, such as the number of swollen joints, a doctor’s assessment of disease, a patient/parent-reported sense of well-being, and results from an inflammation blood test, were documented at the commencement of treatment and throughout the subsequent year.
Employing machine learning techniques, the researchers aimed to identify clusters of patients exhibiting distinct disease patterns following methotrexate treatment, predict these clusters, and subsequently compare them to existing measures of treatment response.
Out of a cohort of 657 children and young individuals, validated within a group of 1,241 patients, the researchers categorized them into different clusters: Fast Improvers (11%), Slow Improvers (16%), Improve-Relapse (7%), as well as the Persistent Disease (44%).
Additionally, two distinct clusters were identified as Persistent Physician Global Assessment (8%) and Persistent Parental Global Assessment (13%), both characterized by improvement in all activity score features except one.
Dr. Shoop-Worrall emphasized, that additional exploration is required to understand the prolonged impact of slower disease control. The research showcases the effectiveness of machine learning techniques in identifying clusters of children, providing a foundation for personalized treatment decisions.
She also indicated that this study extends the findings of previous research on methotrexate treatment response, affirming that the response is not a simple binary outcome but exhibits significant variability across distinct disease features within individuals.
Dr. Shoop-Worrall also indicated that currently, methotrexate trials for juvenile idiopathic arthritis (JIA) classify patients as either responders or non-responders. This misclassification poses a challenge for studies aiming to uncover response predictors, including potential biomarkers.
