Comparative Efficacy of Relapsing Multiple Sclerosis Therapies: Model-Based Meta-Analyses for Confirmed Disability Accumulation and Annualized Relapse Rate

Abstract

BackgroundMultiple sclerosis (MS) is an inflammatory autoimmune disorder and the most common cause of non-traumatic disability in young adults. The Phase 3 OPTIMUM study evaluated the efficacy and safety of oral ponesimod, a selective sphingosine-1-phosphate (S1P) receptor 1 modulator, vs. teriflunomide in patients with relapsing multiple sclerosis (RMS). The aim of this analysis was to assess the effect of ponesimod and other disease modifying treatments (DMTs) compared to placebo, as measured by 12-week confirmed disability accumulation (CDA) and annualized relapse rate (ARR) in RMS patients. MethodsA database was developed by Certara Inc. (USA) based on relevant clinical trials identified from searching the following sources: PubMed, clinicaltrials.gov, FDA and EMEA documents, and conference abstracts. This database consisted of 203 unique randomized controlled trials (RCTs) with 74 MS treatments and was subsequently filtered to include RCTs with more than 25 patients receiving monotherapy to treat RMS for at least 48 weeks. A model-based meta-analysis (MBMA) was performed on the filtered database to assess treatment effects measured by CDA and ARR. Analyzed data for CDA were digitized from published Kaplan-Meier plots. A Weibull distribution was assumed to adequately capture the relationship of CDA probability over time, and hazard ratios (HRs) between treatments were assumed constant over time (proportional hazards). HRs were estimated for 12-week CDA for 17 DMTs vs. placebo. Additionally, mean ARR for each treatment arm was modelled, where relative effect versus placebo was estimated as a fixed effect parameter for each unique drug. A dose-response relationship was included if data for multiple doses were available. Relative treatment effect covariates explored for CDA and ARR included: percent of patients with relapsing-remitting MS (RRMS), trial start year, mean duration of disease, percent of patients who received DMTs within the prior 2 years (pDMT), mean number of relapses in the prior year, mean age, mean baseline EDSS score, and mean treatment duration (for ARR). ResultsThe 12-week CDA model utilized longitudinal data from 26 RCTs (18 unique treatments [including placebo]), 69 treatment arms, 31,160 patients). The ARR model utilized data from 40 RCTs (18 unique treatments [including placebo], 100 treatment arms, 33,686 patients). Compared to placebo, ponesimod significantly reduced 12-week CDA by 39% (HR: 0.61; 95% CI: 0.45–0.82) and reduced ARR by 53% (rate ratio [RR]: 0.47; 95% CI: 0.39–0.58). Except for three DMTs (interferon β-1b, glatiramer acetate, ozanimod), HR of 12-week CDA vs. placebo was significantly lower for the DMTs included in this analysis (HR range: 0.41 to 0.79). The ARR was significantly reduced for all DMTs compared to placebo (RR range: 0.29 to 0.82). A dose-response relationship indicated a potential dose-dependent effect (12-week CDA: 6 treatments; ARR: 8 treatments). Relative treatment effect was found to be significantly smaller in trials including more patients with prior DMT usage. Cross-trial heterogeneity in relative effects was assessed and found to be negligible; however, there is a possibility that confounders remain which may impact estimated relative treatment effects. ConclusionsCompared to placebo, ponesimod 20 mg significantly reduced both the risk of 12-week CDA and mean ARR, suggesting it has robust efficacy in the treatment of RMS.The study was funded by Janssen Research & Development, LLC.