World’s First Clinical Test Predicts Best Rheumatoid Arthritis Treatment

Rheumatoid arthritis (RA) is a chronic condition affecting 1 in 100 people in the UK today, causing the immune system to attack its joints. Unlike osteoarthritis, which is caused by wear and tear, RA can strike suddenly and is most common in people aged 40-60. The disease is often difficult to treat due to the variation in patients' responses to therapies, with biological therapies offering the best treatment options. However, there is no diagnostic test that can predict which therapy will work for a specific patient. As a result, many patients endure multiple rounds of failed treatments, each carrying the risk of severe side effects and infections due to the suppression of the immune system. New research has identified a way to predict which therapies are most likely to succeed for individual patients, thereby minimizing ineffective treatments and improving patient outcomes.

This breakthrough solution was developed by scientists at Queen Mary University of London (London, UK) by combining deep molecular phenotyping and machine learning to predict patient responses to biological therapies for RA. By extracting RNA from a biopsy taken from an affected joint, they measure the activity of 524 specific genes. This data is then analyzed using machine learning models designed to predict the likelihood of success for three main biological therapies: etanercept, tocilizumab, and rituximab. The approach aims to enable clinicians to select the most effective treatment from the start, reducing the chances of adverse effects and unnecessary treatments.

The research was tested in clinical trials, with results showing that the predictive models accurately forecasted patient responses for 79%-85% of patients. Published in Nature Communications, the findings suggest that this method could significantly improve patient well-being by guiding clinicians toward the most suitable therapy faster. The researchers are in discussions with commercial partners to help bring this predictive tool to clinical practice and are conducting large, randomized controlled trials to further refine the technology. With a clinical trial underway, the team is on course to be the first to introduce a predictive test into a real-world medical setting. Their new approach could revolutionize the way RA treatments are personalized, providing patients with better outcomes and a more efficient healthcare system.  

“This innovation could have major benefits for patients and healthcare providers alike,” said Prof Costantino Pitzalis, Professor of Rheumatology at Queen Mary University of London. “Prescribing the right treatment first time would reduce patient suffering and make our healthcare system more efficient.”

Related Links:
Queen Mary University