Performance of PEAR 2.0 Software in Prescribing Biotherapy in Patients With Rheumatoid Arthritis

Overview

The management of rheumatoid arthritis is based on the prescription of disease-modifying anti-rheumatic drugs (DMARDs) to induce clinical and biological remission. If the first line of treatment (methotrexate) fails, a biotherapy may be prescribed. In daily practice, the initiation of a targeted therapy must therefore be based on the prescriber's expertise or qualification in terms of his or her level of experience in the diagnosis and management of chronic inflammatory rheumatic diseases such as rheumatoid arthritis. As the therapeutic arsenal has expanded, so has the question of choosing the right treatment for the right patient at the right time. At present, in daily practice, there is no tool to help clinicians predict treatment efficacy. The choice of biotherapy based on efficacy carries relatively little weight, firstly because this choice is made in relation to other biotherapies, and secondly because there are no superiority studies that have actually demonstrated greater efficacy in favor of one of the targeted therapies. In the age of Big Data, artificial intelligence can be used to develop algorithms for predicting treatment response. mYXpression has developed medical decision support software based on the integration of transcriptomic markers to assess the probability of response and/or non-response to biotherapies for each patient. The algorithm's performance was theoretically tested by retrospectively collecting transcriptomic data and clinical responses to 6 biotherapies from 992 patients included in 17 clinical trials or cohorts. The aim of this observational study is to demonstrate the value of PEAR 2.0 medical decision support software in the management of rheumatoid arthritis patients who are candidates for biotherapy.

This study plans to include 234 patients who, like you, suffer from rheumatoid arthritis, and who will initiate biotherapy in one of the hospital departments taking part in this research. The main objective is to demonstrate the performance of the PEAR 2.0 decision support software, i.e. to show that the software's recommendations for achieving remission are accurate and reliable. The other objectives are to collect data that could help the algorithm evolve, and to provide other information such as the probability of achieving a reduction in rheumatoid arthritis activity in patients who have failed several biotherapies. Because the software must first demonstrate its reliability, the investigator will not use it and will prescribe the biotherapy he thinks is best suited to his patient. The software's results will be communicated to the investigator at the consultation scheduled 6 months after the start of the biotherapy. During two consultations (before the start of biotherapy and 6 months after the first day of biotherapy), the investigator will collect the data needed for the study and which describe the disease. These data are either present in the patient's medical record or are usually collected during consultations in the rheumatology department. For the purposes of the study, at the first visit, the investigator will prescribe a blood sample (10 ml) to be taken at the hospital laboratory before starting the biotherapy. This sample will be used to analyze the biological markers (transcriptome) that the software algorithm uses to establish its recommendations.