PHREND®—A Real-World Data-Driven Tool Supporting Clinical Decisions to Optimize Treatment in Relapsing-Remitting Multiple Sclerosis

Abstract

BackgroundWith increasing availability of disease-modifying therapies (DMTs), treatment decisions in relapsing-remitting multiple sclerosis (RRMS) have become complex. Data-driven algorithms based on real-world outcomes may help clinicians optimize control of disease activity in routine praxis.ObjectivesWe previously introduced the PHREND® (Predictive-Healthcare-with-Real-World-Evidence-for-Neurological-Disorders) algorithm based on data from 2018 and now follow up on its robustness and utility to predict freedom of relapse and 3-months confirmed disability progression (3mCDP) during 1.5 years of clinical practice.MethodsThe impact of quarterly data updates on model robustness was investigated based on the model's C-index and credible intervals for coefficients. Model predictions were compared with results from randomized clinical trials (RCTs). Clinical relevance was evaluated by comparing outcomes of patients for whom model recommendations were followed with those choosing other treatments.ResultsModel robustness improved with the addition of 1.5 years of data. Comparison with RCTs revealed differences <10% of the model-based predictions in almost all trials. Treatment with the highest-ranked (by PHREND®) or the first-or-second-highest ranked DMT led to significantly fewer relapses (p < 0.001 and p < 0.001, respectively) and 3mCDP events (p = 0.007 and p = 0.035, respectively) compared to non-recommended DMTs.ConclusionThese results further support usefulness of PHREND® in a shared treatment-decision process between physicians and patients.

Framework for personalized prediction of treatment response in relapsing-remitting multiple sclerosis: a replication study in independent data

“…The NTD working group regularly improves their models by incremental learning. They used quarterly updates of the NTD database with new patients [ 24 ] and reported changes in the C-indices overtime, which we interpreted as convergence to our C-indices. Therefore, our report is not able to compare the OFSEP results to the current (based on the learning from new data) but to the starting NTD models.…”

Framework for Personalized Prediction of Treatment Response in Relapsing-Remitting Multiple Sclerosis: A Replication Study in Independent Data

“…Focusing on the model's accuracy (calibration and discrimination) and generalisability (reproducibility and transportability) is key to evaluating the performance of prediction models (22,23). In our study, we developed and validated models according to the processes and criteria described by Stühler et al (7) on an independent and representative dataset from a French MS registry and assessed their accuracy and generalizability by a temporal split of the data at hand.…”