Adult and pediatric relapsing multiple sclerosis phase II and phase III trial design and their primary end points: A systematic review

Abstract

No systematic review of trial designs in patients with relapsing multiple sclerosis (RMS) was reported. This systematic review was conducted on the trial designs and primary end points (PEs) of phase II and III trials intended to modify the natural course of the disease in patients with RMS. The purpose of the study is to explore trends/topics and discussion points in clinical trial design and PE, comparing them to regulatory guidelines and expert recommendations. Three trial registration systems, ClinicalTrials.gov, the EU Clinical Trials Register, and the Japan Registry of Clinical Trials, were used and 60 trials were evaluated. The dominant clinical trial design was a randomized controlled parallel‐arms trial and other details were as follows: in adult phase III confirmatory trials (n = 32), active‐controlled double‐blind trial (DBT) (53%) and active‐controlled open‐label assessor‐masking trial (16%); in adult phase II dose‐finding trials (n = 9), placebo‐ and active‐controlled DBT (44%), placebo‐controlled DBT (22%), and placebo‐controlled add‐on DBT (22%); and in pediatric phase III confirmatory trials (n = 8), active‐controlled DBT (38%) and active‐controlled open‐label non‐masking trial (25%). The most common PEs were as follows: in adult confirmatory trials, annual relapse rate (ARR) (56%) and no evidence of disease activity‐3 (NEDA‐3) (13%); in adult dose‐finding trials, the cumulative number of T1 gadolinium‐enhancing lesions (56%), combined unique active lesions (22%), and overall disability response score (22%); and in pediatric confirmatory trials, ARR (38%) and time to first relapse (25%). It was suggested that some parts of the regulatory guidelines and expert recommendations need to be revised.

Study Highlights.

• WHAT IS THE CURRENT KNOWLEDGE ON THIS TOPIC?

Regarding multiple sclerosis (MS) trial designs and endpoints, the EU regulatory guidelines and pediatric MS expert recommendations have been published, respectively. Mostly, the confirmatory trial design for relapsing MS (RMS) was a randomized active‐controlled double‐blind parallel‐arms trial and the primary endpoint (PE) was annual relapse rate (ARR).

• WHAT QUESTION DID THIS STUDY ADDRESS?

This study addressed what were the current trends/topics and points to be discussed regarding RMS trial designs and their PEs.

• WHAT DOES THIS STUDY ADD TO OUR KNOWLEDGE?

ARR was the most common AE in pediatric confirmatory trials, which differed from the expert recommendations of time to relapse. As more potent DMTs are developed, PE closer to complete remission has become necessary, like NEDA‐3. A novel trial design incorporating historical data in a pediatric non‐inferiority trial has been agreed with regulatory agencies.

• HOW MIGHT THIS CHANGE CLINICAL PHARMACOLOGY OR TRANSLATIONAL SCIENCE?

For incurable progressive diseases and/or rare diseases, a new approach incorporating historical data could become a mainstream option for confirmatory trials, avoiding a placebo arm and optimizing sample sizes.

INTRODUCTION

Multiple sclerosis (MS) is a chronic, immune‐mediated, and neurodegenerative disease of the central nervous system and is a multifactorial disease involving genetic, environmental, and immunological factors. ¹ , ² Since there are no specific markers for MS, the diagnosis of MS must be done as a differential diagnosis after ruling out other conditions that might produce similar symptoms and signs. The prevalence of MS varies from region to region, and although it is not a rare disease in North America or Europe, it is a rare disease in Eastern Asia including Japan and sub‐Saharan Africa. ¹ Disease‐modifying therapies (DMTs) have been developed as the mainstay of treatment to modify the natural course of MS, such as relapses and disability progression. Clinical trials of DMTs have been conducted for relapsing–remitting MS (RRMS) or relapsing MS (RMS), and the number of approved DMTs has become quite substantial. Historically, the development of treatments for progressive MS (PMS) has lagged but in recent years many researchers have focused on its clinical development.

Regarding clinical trial design and end points, “Guideline on clinical investigation of medical products for the treatment of Multiple Sclerosis – Revision 2” (The GL ³ ) was issued by the European Medicines Agency (EMA) and Committee for Medical Products for Human Use (CHMP) in 2015 ³ and “Clinical trials of disease‐modifying agents in pediatric MS: Opportunities, challenges, and recommendations from the IPMSSG” (The Recommedations ⁴ ) were published in 2019 by the International Pediatric Multiple Sclerosis Study Group (IPMSSG). ⁴ For PMS clinical trials, a very detailed study design and outcome measurement lessons learned and an end‐point systematic review have been reported. ⁵ , ⁶ Although there is no systematic review of RMS clinical trial design and primary end points (PEs), it is expected that further trials for RMS will be conducted, suggesting that the current RMS trial designs and PEs are still controversial, as they vary from trial to trial.

The purpose of the study is to explore trends/topics and discussion points in current clinical trial design and PE, comparing them to regulatory guidelines and expert recommendations. We chose to review information from clinical trial registers rather than literature searches because it could provide comprehensive and up‐to‐date insights from ongoing trials and terminated/unpublished trials, possibly limiting publication bias. This systematic review was conducted on the trial designs and PEs of phase II and III clinical trials intended to modify the natural course of the disease in patients with RMS from information on clinical trials registered in the clinical trial registers. As for regulatory guidelines for MS clinical trials, the only one available in the EU, the USA, and Japan is the GL, ³ which has come into usage since October 1, 2015. Given the time lag until the GL ³ is fully disseminated and the trial design and PE are set following the GL, ³ the inclusion of this study was limited to trials planned to be initiated on or after January 1, 2017, and by December 31, 2022, the last day of the previous year in which this study was planned.

METHODS

Data sources and search strategy

A systematic review was prepared according to the latest Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) guidelines. ⁷ This systematic review of trial registration information regarding phase II and III clinical trials in patients with MS was performed using three clinical trial registries: ClinicalTrials.gov, European Union Clinical Trials Register, and Japan Registry of Clinical Trials. The search strategy used is presented in Figure 1. The last search was performed on July 7, 2023. This search yielded 185, 127, and 8 results, respectively.

FIGURE 1.

The search strategy.

A search of chemical names of marketed drugs for the prevention of relapse and/or progression of patients with MS as of March 24, 2023 was also done using ProQuest Dialog/AdisInsight and the National Multiple Sclerosis Society's site. The search strategy used is presented in Figure 1 as well. The last search was performed on March 24, 2023, and June 13, 2023, respectively. This search yielded 38 and 25 results, in order.

Selection criteria

The selection criteria for trial information were that the adult phase III confirmatory, adult phase III extension, adult phase II dose finding, adult phase II extension, pediatric phase III confirmatory, and pediatric phase II trial information intended to modify the natural course of the disease in RMS (including RRMS only) and that the trials were scheduled to begin between 2017 and 2022. Each clinical trial registry has different search functions, and as shown in Figure 1, some could not be searched using the same criteria for selection. In addition, because each registry is independent and not mutually compatible, duplicate or triplicate information cannot be excluded by their search functions. Therefore, eligible records were screened first by title, estimated enrollment, and other study ID numbers, and those that did not meet search criteria, such as duplicate or triplicate information, were visually excluded. The remaining records were visually excluded if they did not meet the selection criteria. An overview of the included and excluded records for the systematic search is summarized in Figure 2. Adult trials were defined as trials that included eligible patients between the ages of 18 and 64, and pediatric trials were defined as trials that included eligible patients from birth to 17 years of age.

FIGURE 2.

PRISMA flow diagram demonstrating included and excluded trials, and the reasons for exclusion in this systematic review. *Including duplicates. EU, Europe Union; MS, multiple sclerosis; PRISMA, Preferred Reporting Items for Systematic Reviews and Meta‐Analyses; RMS, relapsing MS.

The selection criterion for chemical names of marketed drugs was chemical names of drugs approved and available for the prevention of relapse and/or progression of patients with MS as of March 24, 2023, excluding the bioequivalent alternatives, generics, biosimilars, and bioequivalent non‐biological complex drugs. ⁸ , ⁹ After the criterion was applied, 43 drugs were all excluded manually (Figure S1).

Referenced guidelines and expert recommendations

In conducting this review, the following were referenced and compared: The GL, ³ the Recommendations, ⁴ and CPMP/ICH/378/95, as well as dose–response information to support drug registration—ICH Harmonised Tripartite Guideline ¹⁰ (the ICH‐GL ¹⁰ ). Table 1. summarizes some of the GLs ³ , ¹⁰ and expert recommendations ⁴ , ¹¹ for clinical trials that intend to modify the natural course of RMS.

TABLE 1.

Summary of regulatory guidelines and expert recommendations on trials that intend to modify the natural course of RMS.

Statements/recommendations	Reference #
1. Trial design
1‐1. The EMA/CHMP guideline on clinical investigation of medical products for the treatment of multiple sclerosis–revision 2
1. Confirmatory trial design and general remarks for a clinical trial for patients with RMS For new drugs intended for the treatment of RMS, efficacy should be established using randomized double‐blind (double dummy if needed) controlled parallel group trials. The preferred approach would be a development showing superiority versus placebo or an active comparator (i.e., first‐line DMTs like β‐interferons, glatiramer acetate). A non‐inferiority approach would also be acceptable as long as assay sensitivity and a reasonable non‐inferiority margin can be determined and properly justified, for example, in cases when an active comparator with consistent pronounced efficacy versus placebo or first‐line DMTs. However, non‐inferiority trials versus first‐line DMT products like interferons or glatiramer acetate (or other products with a similar efficacy range), in the absence of a placebo arm, would not be sufficient, as in such a scenario an apparent efficacy could be explained by the regression to the mean, a placebo effect, as well as by the natural course of the disease. As several subjective decisions and assessments will have to be performed, with a considerable risk of bias, all possible efforts should be made to keep the design double blind. In cases where a double‐blind evaluation is not possible, a blind observer design with a blinded examining physician different than the treating physician may be used. All measures to ensure reliable single‐blind evaluation should be guaranteed (i.e., patches that cover injection sites to hide reddening or swellings, education of examining physicians, etc.). Criteria to refer the patient to evaluation of relapse should be established a priori in the protocol to avoid selective referral. The superiority trial is preferable showing superiority versus placebo or an active comparator (i.e., first‐line DMTs). Add‐on study designs may be considered as an alternative as long as an additive immunosuppressant effect is considered unlikely. In add‐on trials, one might include a third monotherapy arm with the new product to establish if the superiority of the combination arm is due only to the new product or to the combination. A useful design is a three‐arm trial seeking superiority of the combination versus both products in monotherapy.	3
2. Long‐term trial design	3
An extended follow‐up either blinded or open‐label should be performed to address the maintenance of the effect and to gather information on the long‐term course of patients under treatment. For a distinct claim on disability, large‐scale long‐term parallel group trials will be required to establish clinically relevant treatment effects on disease progression. Study duration will depend on the population studied, and should be sufficient to show a reliable and relevant effect on disability. Such a study may need to last ~3 years. For chronic treatment, it is expected that at the time of marketing authorization, safety data of at least 2 years are available for a meaningful number of patients.
3. Clinical trial design in pediatric MS The clinical manifestations of pediatric‐onset MS resemble that of adult‐onset RRMS. However, compared to adult‐onset RRMS, younger children appear to have more frequent relapses, earlier cognitive deficits, recover better from relapses, and have a slower disease progression. Depending on the mechanism of action, the expected safety profile and the targeted age group generation of specific data could be done by performing clinical trials tailored to children, by incorporating adolescent MS patients into the adult trials, and/or by partial extrapolation of efficacy from adult trial(s). The approach chosen should be justified and depends among others on the mechanism of action, on the availability of suitable biomarkers for efficacy as confirmed by the adult trials, and on the existence of a valid pharmacokinetic target exposure. Trials in children should only be initiated when there is sufficient evidence that the product may meet a pediatric medical need with a suitable safety profile. Data on safety in children should be generated. Considering the life‐long treatment, this should include long‐term safety data concerning immune response to infectious disease—vaccination when appropriate, mental, cognitive function, growth, and sexual development. All patients should preferably be included in registries to monitor long‐term safety and efficacy.	3
4. PK study Pharmacokinetics of the drug should be thoroughly investigated in accordance with relevant guidelines.	3
1‐2. The CPMP/ICH dose–response information to support drug registration—ICH Harmonised Tripartite Guideline
1. Dose‐finding trial design Trials intended to evaluate dose or concentration response should be well controlled, using randomization and blinding (unless blinding is unnecessary or impossible) to assure comparability of treatment groups and to minimize potential patient, investigator, and analyst bias, and should be of adequate size. Randomization to several fixed‐dose arms (the randomized, parallel, dose–response study) is a design that has had extensive use and considerable success. No placebo arm is acceptable if a positive slope, even without a placebo arm, provides evidence of a drug effect. However, including a placebo arm is desirable to measure the absolute size of the drug effect and can salvage, in part, a study that used doses that were all too high and therefore showed no dose–response slope, by showing that all doses were superior to placebo. The inclusion of both placebo and active control arms allows for the evaluation of measurement sensitivity, permitting a distinction between an ineffective drug and an ineffective trial. A randomized multiple crossover study of different doses can be successful if the drug effect develops rapidly and patients return to baseline conditions quickly after cessation of therapy.	10
1‐3. Clinical trials of disease‐modifying agents in pediatric MS—opportunities, challenges, and recommendations from the International Pediatric Multiple Sclerosis Study Group
1. Phase III trial design in pediatric MS Placebo‐controlled trials of immunomodulatory agents proven effective in adult MS are inappropriate in pediatric MS. For randomized controlled trials of immunomodulating agents, the control drug could be fingolimod or other drugs used in pediatric MS. For add‐on trials, the new agent (such as a therapy with potential remyelinating or neuroprotective properties, nutraceuticals, or others) could be compared to placebo, provided that immunomodulatory therapies are maintained. No more than one phase III clinical trial should be performed for the same pharmacologic agent. 2. Phase II/III trial design in pediatric MS For drugs that have been well studied in adults, and have demonstrated, through pharmacodynamic and statistical modeling, extrapolated predictions for the pediatric age range that would support efficacy, then open‐label study (including PK/PD and safety end points) should be considered sufficient for registration in pediatric MS. In the case that an open‐label study (including PK/PD and safety end points) is deemed insufficient for registration in pediatric MS, a short controlled trial with an MRI primary end point is recommended, provided that the pivotal study of the same agent in adult‐onset MS demonstrated a robust relationship between clinical efficacy and MRI end points. 3. Safety monitoring trial design For safety monitoring, open‐label extension studies should be mandated for all clinical trials in pediatric MS populations. Open‐label studies or registries should be designed to monitor safety in patients aged 12 and under.	4
1‐4. The challenge of comorbidity in clinical trials for multiple sclerosis and expert recommendations from members of the International Workshop on Comorbidity in Multiple Sclerosis
1. Phase III trials of new therapies in MS Shift phase III clinical trials of new therapies in MS along the continuum from explanatory to more pragmatic trials. Specifically, we propose changes to eligibility criteria for clinical trials of pharmacologic and non‐pharmacologic therapies. First, relax age restrictions. Second, consider relaxing restrictions on the inclusion of individuals with comorbidity. This has been feasible in other fields. Such decisions are a tradeoff between homogeneity of the population and the strength of the signal being sought, safety, and generalizability of the findings to clinical practice. We suggest that, in the absence of strong a priori safety concerns, individuals with the most common comorbidities in the MS population, including depression, anxiety, hypertension, hyperlipidemia, and chronic lung disease, be considered for inclusion in trials. Furthermore, individuals with a history of cancer should also be considered as potential participants in some settings. These changes will be particularly important in trials of progressive MS, as these individuals will be older and will have a greater burden of comorbidity. These changes will require larger trials.	11, 59
2. Primary end point
2‐1. The EMA/CHMP Guideline on clinical investigation of medical products for the treatment of multiple sclerosis–revision 2
1. Primary efficacy end points in confirmatory trials	3
In patients with RMS, both the ARR and the time to relapse are considered acceptable as primary end points. A relapse‐based primary end point though cannot be taken as a surrogate for disability progression. If the primary end point is based on relapse assessment, progression of disability should be evaluated as key secondary end point. Time to relapse would limit the exposure to placebos or the time during which the patient experiences suboptimal active control of his symptoms. However, even if an effect on the time to relapse is shown, maintenance of the effect on relapses should be demonstrated. Thus, time to relapse is acceptable as a primary end point provided that data are generated to show maintenance of effect. Time to second or third relapse may be useful for this. MRI measurements have not been demonstrated to be a reasonably validated surrogate end point for the clinical outcome and are, therefore, not acceptable as a primary end point in pivotal studies evaluating new agents.
2. Primary efficacy end points in dose‐finding trials and proof‐of‐concept trials	3
In exploratory trials in RMS, the use of MRI‐derived parameters as the main end point for assessing dose‐finding or preliminary efficacy is acceptable where applicable. MRI findings are useful as a first indication of dealing with a potentially clinically effective product.
2–2. The CPMP/ICH dose–response information to support drug registration—ICH Harmonised Tripartite Guideline
1. Guidance and advice It is prudent to carry out dose‐ranging or concentration–response studies early in development as well as in later stages in order to avoid failed phase III studies or accumulation of a database that consists largely of exposures at ineffective or excessive doses. The end points of studies may vary at different stages of drug development. For example, in studying a drug for heart failure, a pharmacodynamic end point might be used early (e.g., cardiac output and wedge pressure), an intermediate end point might be used later (e.g., exercise tolerance, symptoms), and a mortality or irreversible morbidity end point might be the final assessment (survival and new infarction). It should be anticipated that the dose–response for these end points may be different. Of course, the choice of end points that must be studied for marketing approval will depend on the specific situation.	10
2‐3. Clinical trials of disease‐modifying agents in pediatric MS—opportunities, challenges, and recommendations from the International Pediatric Multiple Sclerosis Study Group
1. Primary efficacy end point in pediatric MS phase III trial For agents for which a phase III trial with a clinical primary end point is mandated by regulatory agencies, time‐to‐event (relapse) analyses (rather than annualized relapse rate) should be favored as this design allows a prompter switch to more effective therapies. 2. Primary efficacy end point in pediatric MS phase II/III trial In the case that an open‐label study (including PK/PD and safety end points) is deemed insufficient for registration in pediatric MS, a short controlled trial with an MRI primary end point is recommended rather than a clinical end point, provided that the pivotal study of the same agent in adult‐onset MS demonstrated a robust relationship between clinical efficacy and MRI end points. 3. Safety monitoring For safety monitoring, open‐label extension studies should be mandated for all clinical trials in pediatric MS populations. Open‐label studies or registries should be designed to monitor safety in patients aged 12 and under.	4

Abbreviations: CHMP, Committee for Medical Products for Human Use; CPMP, Committee for Proprietary Medical Products; DMT, disease‐modifying treatment; EMA, European Medicines Agency; ICH, International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use; MRI, magnetic resonance imaging; MS, multiple sclerosis; RMS, relapsing multiple sclerosis; RRMS, relapsing–remitting multiple sclerosis.

Analysis

In this study, DMTs were classified into three categories: first‐line drugs (Interferon β‐1a, interferon β‐1b, and glatiramer acetate), middle‐class drugs (dimethyl fumarate, fingolimod, and teriflunomide), strongest‐class drugs (alemtuzumab, cladribine, natalizumab, ocrelizumab, ofatumumab, and rituximab). All data analyses were performed using EZR Version 1.61 software.

RESULTS

Eighty‐eight clinical trials were selected through screening and categorized for review as follows: adult RMS phase III confirmatory trials (n = 32), adult RMS phase II dose‐finding trials (n = 9), adult RMS phase III extension trials (n = 10), adult RMS phase II extension trials (n = 9), and pediatric RMS trials (n = 10) (each, in part, includes duplicates) (Figure 2).

Trial designs in adult RMS

All of the trials reviewed are listed in Table 2, including the registration number, trial design, year of trial initiation, age limits of inclusion, interventions, main inclusion criteria related to MS severity, PE and its timeframe, estimated enrollment number of patients, follow‐up period, consistency/concerns with the GL, ³ the ICH‐GL, ¹⁰  and the Recommendations ⁴ in trial design and PE, and tentative evaluations of consistency. A summary of the trial design breakdown of the trials evaluated is shown in Figure S2. The summary statistics of the number of arms, the estimated enrollment number, and the follow‐up period of all trials reviewed are shown in Table S2.

TABLE 2.

Registered phase II and phase III trials intended to modify the natural course of the disease in adult patients with RMS (planned start years 2017–2022).

Registration #	Trial design	Arms and interventions	Main inclusion criteria related to MS severity	Primary/key end point and its timeframe	Consistency/concerns with GLs and recommendations in trial designs/PEs
Target MS type	Trial status at the moment of the systematic review			Main objective
Phase type of trial	Year of trial initiation
Estimated enrollment #
Trial acronym	Age limits of inclusion			Follow‐up period
1. Adult RMS, phase III confirmatory trials for new drugs/biologicals/treatments (n = 32)
NCT05385744 Adult RMS Phase III Confirmatory trial (Superiority trial) MIRANTIBUS	Randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial Recruiting 2021 18–60	Arm 1: BCD‐132 + placebo to match teriflunomide Arm 2: Teriflunomide + placebo to match BCD‐132 Active comparator: Teriflunomide	1. (a) ≥1 relapse within the past year (12 months), or (b) 2 relapses within the past 2 years (24 months), or (c) ≥1 T1 Gd‐enhanced brain lesion and 1 relapse within the 2 past years 2. EDSS score 0–5.5 (inclusive)	ARR [timeframe: 48 weeks] To evaluate the efficacy and safety of BCD‐132 in the treatment of patients with RMS 336 pts 100 weeks	1. Trial design GLs: Consistent 2. PE GLs: Consistent
NCT03277261/2017–000638‐75 Adult RMS Phase III Confirmatory trial (Superiority trial) ULTIMATE I	Randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial Completed 2017 18–55	Arm 1: Ublituximab + placebo to match teriflunomide Arm 2: Teriflunomide, orally + placebo to match ublituximab Active comparator: Teriflunomide	1. (a) ≥2 relapses in the past 2 years, or (b) 1 relapse in the past year, and/or (c) ≥1 Gd‐enhancing lesion 2. Active disease 3. EDSS score 0–5.5 (inclusive)	ARR [timeframe: 96 weeks] To determine the ARR in subjects with RMS after 96‐week treatment with i.v. infusion of ublituximab/oral placebo compared to 14 mg oral teriflunomide/i.v. placebo 440 pts 116 weeks	1. Trial design GLs: Consistent 2. PE GLs: Consistent
NCT03277248/2017–000639‐15 Adult RMS Phase III Confirmatory trial (Superiority trial) ULTIMATE II	Randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial Completed 2017 18–55	Arm 1: Ublituximab + placebo to match teriflunomide Arm 2: Teriflunomide, orally + placebo to match ublituximab Active comparator: Teriflunomide	1. (a) ≥2 relapses in the past 2 years, or (b) 1 relapse in the past year, and/or (c) ≥1 Gd‐enhancing lesion 2. Active disease 3. EDSS score 0–5.5 (inclusive)	ARR [timeframe: 96 weeks] To determine the ARR in subjects with RMS after 96‐week treatment with i.v. infusion of ublituximab/oral placebo compared to 14 mg teriflunomide/i.v. placebo 440 pts 116 weeks	1. Trial design GLs: Consistent 2. PE GLs: Consistent
NCT04410978/2020–000637‐41/jRCT2021200035 Adult RMS Phase III Confirmatory trial GEMINI 1	Randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial Active, not recruiting 2020 18–55	Arm 1: Tolebrutinib, orally + placebo to match teriflunomide Arm 2: Teriflunomide, orally + placebo to match tolebrutinib Active comparator: Teriflunomide	1. (a) ≥1 relapse within the past year, or (b) ≥2 relapses within the past 2 years, or (c) ≥1 Gd‐enhancing lesion within the past year 2. EDSS score 0–5.5 (inclusive)	ARR [timeframe: approximately 36 months] To assess the efficacy of tolebrutinib compared to teriflunomide measured by ARR in participants with RMS 900 pts 3 years	1. Trial design GLs: Consistent 2. PE GLs: Consistent
NCT04410991/2020–000644‐55 Adult RMS Phase III Confirmatory trial GEMINI 2	Randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial Active, not recruiting 2020 18–55	Arm 1: Tolebrutinib, orally + placebo to match teriflunomide Arm 2: Teriflunomide, orally + placebo to match tolebrutinib Active comparator: Teriflunomide	1. (a) ≥1 relapse within the past year, or (b) ≥2 relapses within the past 2 years, or (c) ≥1 Gd‐enhancing lesion within the past year 2. EDSS score 0–5.5 (inclusive)	ARR [Timeframe: approximately 36 months] To assess the efficacy of tolebrutinib compared to teriflunomide measured by ARR in participants with RMS 900 pts 3 years	1. Trial design GLs: Consistent (because teriflunomide is not a first‐line DMT based on the definition of the EMA/CHMP GL, a placebo arm is not mandatory even if it is a non‐inferiority trial. No information on whether it is a non‐inferiority or superiority trial.) 2. PE GLs: Consistent
NCT04338022/2019–004972‐20 Adult RMS Phase III Confirmatory trial (Superiority trial) Evolution RMS 1	Randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial (double‐blind treatment period: DBTP) + double‐blind extension (DBE) + OLE Active, not recruiting 2020 18–55	For DBTP and DBE Arm 1: Evobrutinib + placebo to match teriflunomide Arm 2: Teriflunomide + placebo to match evobrutinib Active comparator: Teriflunomide	For DBTP ≥1 relapse within the past 2 years with either: (a) 1 relapse within the past years, or (b) ≥1 Gd‐enhancing T1 lesion within the past 6 months EDSS score 0–5.5 (inclusive)	For DBTP ARR [timeframe: 156 weeks] For DBE ARR [timeframe: 96 weeks in DBE period] For the DBTP: to demonstrate the superior efficacy of evobrutinib compared to teriflunomide in terms of ARR For the DBE: To further evaluate the efficacy of evobrutinib compared to teriflunomide in terms of ARR 930 pts DBTP: 156 weeks; DBE: 96 weeks; OLE: 96 weeks; Overall: 348 weeks	1. Trial design for DBTP and DBE GLs: Consistent 2. PE for DBTP and DBE GLs: Consistent
NCT04338061/2019–004980‐36 Adult RMS Phase III Confirmatory trial (superiority trial) Evolution RMS 2	Randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial (double‐blind treatment period: DBTP) + double‐blind extension (DBE) + OLE Active, not recruiting 2020 18–55	For DBTP and DBE Arm 1: Evobrutinib + placebo to match teriflunomide Arm 2: Teriflunomide + placebo to match evobrutinib Active comparator: Teriflunomide	For DBTP ≥1 relapse within the past 2 years with either: (a) 1 relapse within the past years, or (b) ≥1 Gd‐enhancing T1 lesion within the past 6 months EDSS score 0–5.5 (inclusive)	For DBTP ARR [timeframe: 156 weeks] For DBE ARR [timeframe: 96 weeks in DBE period] For the DBTP: To demonstrate the superior efficacy of evobrutinib compared to teriflunomide in terms of ARR For the DBE: To further evaluate the efficacy of evobrutinib compared to teriflunomide in terms of ARR 930 pts DBTP: 156 weeks; DBE: 96 weeks; OLE: 96 weeks; Overall: 348 weeks	For DBTP and DBE: 1. Trial design GLs: Consistent 2. PE GLs: Consistent
NCT04032158/2018–004701‐11 Adult RRMS Phase III Confirmatory trial (Superiority trial) EVOLUTION MS1	Randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial Terminated: Following analysis of OLE data from the RMS phase II study (MS200527–0086), it was determined that a change in active comparator warranted in phase III RMS comprised of this trial. Consequently, this trial terminated early. 2019 18–55	Arm 1: Evobrutinib + placebo to match interferon β‐1a Arm 2: Interferon β‐1a + placebo to match evobrutinib Active comparator: Interferon β‐1a	≥1 relapse within the past 2 years with either: (a) 1 relapse within the past year, or (b) ≥1 Gd‐enhancing lesion within the past 6 months EDSS score 0–5.5 (inclusive)	ARR [timeframe: 96 weeks] To demonstrate superior efficacy with evobrutinib compared to interferon β‐1a (Avonex®) in terms of ARR 950 pts 96 weeks	1. Trial design GLs: Consistent 2. PE GLs: Consistent
NCT04032171/2018–004700‐19 Adult RMS Phase III Confirmatory trial (Superiority trial) EVOLUTION MS2	Randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial Terminated: Following analysis of OLE data from the RMS phase II study (MS200527–0086), it was determined that a change in active comparator warranted in phase III RMS comprised of this trial. Consequently, this trial terminated early. 2019 18–55	Arm 1: Evobrutinib + placebo to match interferon β‐1a Arm 2: Interferon β‐1a + placebo to match evobrutinib Active comparator: Interferon β‐1a (Avonex®)	≥1 relapse within the past 2 years with either: (a) 1 relapse within the past year, or (b) ≥1 Gd‐enhancing lesion within the past 6 months EDSS score 0–5.5 (inclusive)	ARR [timeframe: 96 weeks] To demonstrate superior efficacy with evobrutinib compared to interferon β‐1a (Avonex®) in terms of ARR 950 pts 96 weeks	1. Trial design GLs: Consistent 2. PE GLs: Consistent
NCT04586010/2019–004857‐10 Adult RMS Phase III Confirmatory trial FENhance	Randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial Recruiting 2021 18–55	Arm 1: Fenebrutinib + placebo to match teriflunomide Arm 2: Teriflunomide + placebo to match fenebrutinib Active comparator: Teriflunomide	EDSS score 0–5.5 (inclusive) Ability to complete the 9HPT for each hand in <240 s Ability to perform the T25FWT in <150 s	ARR [timeframe: 96 weeks] To evaluate the efficacy and safety of fenebrutinib on disability progression and relapse rate in adult participants with RMS 736 pts 96 weeks	1. Trial design GLs: Consistent (because teriflunomide is not a first‐line DMT based on the definition of the EMA/CHMP GL, a placebo arm is not mandatory even if it is a non‐inferiority trial. No information on whether it is a non‐inferiority or superiority trial.) 2. PE GLs: Consistent
NCT04586023/2020–001168‐28 Adult RMS Phase III Confirmatory trial FENhance 2	Randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial Recruiting 2021 18–55	Arm 1: Fenebrutinib + placebo to match teriflunomide Arm 2: Teriflunomide + placebo to match fenebrutinib Active comparator: Teriflunomide	EDSS score 0–5.5 (inclusive) Ability to complete the 9HPT for each hand in <240 s Ability to perform the T25FWT in <150 s	ARR [timeframe: 96 weeks] To evaluate the efficacy of fenebrutinib compared with teriflunomide based on ARR 734 pts 96 weeks	1. Trial design GLs: Consistent (because teriflunomide is not a first‐line DMT based on the definition of the EMA/CHMP GL, a placebo arm is not mandatory even if it is a non‐inferiority trial. No information on whether it is a non‐inferiority or superiority trial.) 2. PE GLs: Consistent
NCT05147220/2020–005899‐36 Adult RMS Phase III Confirmatory trial (Superiority trial) NA	Randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial + OLE Recruiting 2021 18–55	Double‐blind part Arm 1: Remibrutinib + placebo to match teriflunomide Arm 2: Teriflunomide + placebo to match remibrutinib Active comparator: Teriflunomide	Double‐blind part 1. (a) ≥1 relapse within the past year, or (b) ≥2 relapses within the past 2 years, or (c) ≥1 active Gd‐enhancing lesion in the 12 months 2. EDSS score 0–5.5 (inclusive)	Double‐blind part ARR [timeframe: 30 months] To assess that remibrutinib is superior to teriflunomide in reducing the frequency of confirmed relapses. 800 pts Double‐blind part: 30 months; overall: 5 years	Double‐blind part 1. Trial design GLs: Consistent 2. PE GLs: Consistent
NCT05156281/2020–005929‐89 Adult RMS Phase III Confirmatory trial (Superiority trial) REMODEL‐2	Randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial + OLE Recruiting 2021 18–55	Double‐blind part Arm 1: Remibrutinib + placebo to match teriflunomide Arm 2: Teriflunomide + placebo to match remibrutinib Active comparator: Teriflunomide	Double‐blind part 1. (a) ≥1 relapse within the past year, or (b) ≥2 relapses within the past 2 years, or (c) ≥1 active Gd‐enhancing lesion in the 12 months 2. EDSS score 0–5.5 (inclusive)	Double‐blind part ARR [timeframe: 30 months] To assess that remibrutinib is superior to teriflunomide in reducing the frequency of confirmed relapses. 800 pts Double‐blind part: 30 months; overall: 5 years	Double‐blind part 1. Trial design GLs: Consistent 2. PE GLs: Consistent
NCT04578639/2020–001205‐23 Adult RRMS Phase III Confirmatory trial (non‐inferiority trial) OVERLORD‐MS	Randomized, active‐controlled, double‐blind, parallel‐arms trial Recruiting 2020 18–60	Arm 1: Rituximab Arm 2: Ocrelizumab Active comparator: Ocrelizumab	Treatment‐naive ≥1 relapse or ≥1 new MRI lesion within the past 12 months EDSS score 0–4.0 (inclusive)	The proportion of patients with no new or enlarging T2‐weighted brain MRI lesions [timeframe: from month 6 (re‐baseline) to month 24] To assess if rituximab is non‐inferior to ocrelizumab with regards to efficacy and safety in the treatment naive RRMS patients, diagnosed within the last 12 months. 211 pts 24 months	1. Trial design GLs: Consistent (because ocrelizumab is not a first‐line DMT, a placebo arm is not mandatory even in a non‐inferiority trial.) 2. PE GLs: Inconsistent (MRI findings are not acceptable)
NCT03477500/2017–001362‐25 Adult RRMS Phase III Confirmatory trial RAM‐MS	Randomized, active‐controlled, open‐label, assessor‐masking, parallel‐arms trial Recruiting 2018 18–50	Arm 1: aHSCT Arm 2: DMTs (alemtuzumab, cladribine or ocrelizumab) Active comparator: above DMTs	Significant inflammatory disease activity in the last year despite treatment with standard DMTs (interferon β, glatiramer acetate, dimethyl fumarate, teriflunomide, fingolimod, and natalizumab) (a) ≥1 relapse within the past year, and (b) ≥1 Gd‐enhancing lesion within the past year, or (c) ≥3 new or enlarging T2 lesions within the past year 2. EDSS score 0–5.5 (inclusive)	NEDA‐3 [timeframe: 96 weeks] To investigate the efficacy and the safety of aHSCT compared to alemtuzumab, cladribine, or ocrelizumab in patients with aggressive RRMS 100 pts 96 weeks	1. Trial design GLs: Consistent (because alemtuzumab, cladribine, and ocrelizumab are not first‐line DMTs, a placebo arm is not mandatory even if it is a non‐inferiority trial. No information on whether it is a non‐inferiority or superiority trial.) 2. PE GLs: Inconsistent (NEDA‐3 is not recommended)
NCT04047628 Adult RMS Phase III Confirmatory trial BEAT‐MS	Randomized, active‐controlled, open‐label, assessor‐masking, parallel‐arms trial Recruiting 2019 18–55	Arm 1: aHSCT Arm 2: DMTs (cladribine, natalizumab, alemtuzumab, ocrelizumab, rituximab, or ofatumumab after approval by the FDA) Active comparator: above DMTs	Highly active treatment‐resistant RMS, defined as ≥2 episodes of disease activity (relapse or MRI evidence) and must meet all the following criteria: (a) ≥1 episode must occur after ≥1 month of treatment of DMTs approved by the FDA for the treatment of RMS, and (b) ≥1 episode within the past 12 months, and (c) ≥1 relapse 2. EDSS score 0–6.0 (inclusive)	MS relapse‐free survival [timeframe: 3 years] To compare the treatment strategy of aHSCT to the treatment strategy of the best available therapy for treatment‐resistant RMS 156 pts 6 years	1. Trial design GLs: Consistent (because cladribine, natalizumab, alemtuzumab, ocrelizumab, rituximab, and ofatumumab are not first‐line DMTs, a placebo arm is not mandatory even if it is a non‐inferiority trial. No information on whether it is a non‐inferiority or superiority trial.) 2. PE GLs: Inconsistent (MS relapse‐free survival is not recommended)
NCT03283397 Adult RRMS Phase III Confirmatory trial (Superiority trial) NA	Randomized, active‐controlled, open‐label, assessor‐masking, parallel‐arms trial Recruiting 2019 18–55	Arm 1: EK‐12 (metenkefalin + tridecactide) Arm 2: Interferon β‐1a Active comparator: Interferon β‐1a	1. (a) ≥1 relapse within the past year, or (b) ≥1 relapse within the past 2 years and ≥1 Gd‐enhancing lesion within the past year 2. EDSS score 0–4.5 (inclusive)	ARR [timeframe: 144 weeks] To prove the superiority of efficacy of EK‐12 compared to interferon β‐1a in patients with RRMS based on ARR by 144 weeks. 400 pts 144 weeks	1. Trial design GLs: Possibly consistent (this assessor‐masking design may be acceptable) 2. PE GLs: Consistent
NCT04688788/2020–002981‐15 Adult RRMS Phase III Confirmatory trial (non‐inferiority trial) DanNORMS	Randomized, active‐controlled, open‐label, assessor‐masking, parallel‐arms trial Recruiting 2021 18–65	Arm 1: Rituximab Arm 2: Ocrelizumab Active comparator: Ocrelizumab	1. Active MS Treatment naive RRMS patients (never treated, or no DMT within the past 2 years) (a) 1 relapse within the past year with severe residual symptoms and EDSS ≥3.0, or (b) ≥2 relapses within the past 2 years, or (c) 1 relapse past year and ≥9 T2 lesions on brain and/or spinal cord MRI, and 1 contrast‐enhancing lesion or ≥1 new or enlarging T2 lesion on brain and/or spinal cord MRI within the past year Previously treated RRMS patients: (a) ≥1 relapse within the past year, or (b) ≥1 contrast‐enhancing lesion or ≥2 new/enlarging T2 lesions on brain and/or spinal cord MRI within the past year Progressive MS patients: (a) ≥1 relapse within the past year, or (b) ≥1 contrast‐enhancing lesion within the past year or ≥1 new/enlarging T2 lesions on brain and/or spinal cord MRI within the past year or ≥2 new/enlarging T2 lesions on brain and/or spinal cord MRI within the past 2 years, or (c) Increased levels of neurofilament light chain (NFL) in serum or cerebrospinal fluid (CSF) in samples collected in the past year (details omitted.) 2. EDSS score 0–6.5 (inclusive)	Percentage of patients without new or enlarging T2 white matter lesions on brain MRI scans [timeframe: 24 months] To assess whether treatment of active multiple sclerosis with rituximab is non‐inferior to ocrelizumab regarding efficacy and safety. 594 pts 24 months	1. Trial design GLs: Consistent (because ocrelizumab is not a first‐line DMT, a placebo arm is not mandatory even in a non‐inferiority trial.) 2. PE GLs: Inconsistent (MRI findings are not acceptable)
NCT04121403 Adult RMS Phase III Confirmatory trial (non‐inferiority trial) NOR‐MS	PROBE, active‐controlled, assessor‐masking, parallel‐arms trial Recruiting 2019 18–65	Arm 1: Rituximab Arm 2: Cladribine Active comparator: Cladribine	1. Disease activity was seen as either a relapse or MRI activity during the past year 2. EDSS score 0–5.5 (inclusive)	Number of new/enlarging cerebral MRI T2 lesions [timeframe: 96 weeks] To assess whether rituximab is non‐inferior to cladribine for the treatment of RMS 264 pts 96 weeks	1. Trial design GLs: Consistent (In this trial, blinded radiologists assess the primary end points and do not know the treatment allocation. This assessor‐masking design would be acceptable. However, since the end points are public, the PROBE design may not work.) 2. PE GLs: Possibly inconsistent (MRI findings are not acceptable in principle. However, this trial is a comparative trial of already approved two DMTs and the PE may be acceptable depending on the intention of use of the results.)
NCT05242133 Adult RRMS Phase III Confirmatory trial (non‐inferiority trial) NA	Randomized, active‐controlled, open‐label, non‐masking, parallel‐arms trial Completed 2017 18–50	Arm 1: Pegylated interferon β‐1a (CinnaGen, Iran) Arm 2: Interferon β‐1a (CinnoVex®, CinnaGen, Iran) Active comparator: CinnoVex®	≥1 relapse within the past year EDSS score 0–5.0 (inclusive)	ARR [timeframe: 96 weeks] To verify the non‐inferiority of peginterferon β‐1a versus CinnoVex® in reducing the ARR in participants with RRMS at 2 years 168 pts 96 weeks	1. Trial design GLs: Inconsistent (at least an assessor‐masking design is required) 2. PE GLs: Consistent
2019–001549‐42 Adult RRMS (including pediatric RRMS) Phase III Confirmatory trial StarMS	Randomized, active‐controlled, open‐label, non‐masking, parallel‐arms trial No update information due to a trial in Great Britain—no longer in the EU/European Economic Area 2020 16–55	Arm 1: aHSCT Arm 2: DMTs (alemtuzumab or ocrelizumab) Active comparator: alemtuzumab or ocrelizumab	1. Severe inflammatory disease defined as RRMS course with ≥2 relapses, or 1 such relapse and evidence of MRI disease activity >3 months before or after its onset, in the past 12 months despite being on a DMT* 2. EDSS score 0–6.0 (inclusive)* If the EDSS score is 6.0, this must be due to confirmed relapse rather than progressive disease. *Patients with EDSS scores of 0–1.5 or those who failed only first‐line treatments must also fulfill the following criteria: short illness duration (<5 years), active disease clinically and radiologically (i.e., ≥2 relapses in the last 12 months, evidence of multiple Gd‐enhancing MRI lesion), high brain lesion load, and brain or spinal cord atrophy.	The proportion of patients who have maintained NEDA‐3 status [timeframe: 2 years] To assess whether aHSCT is more effective at maintaining clinical stability in patients with highly active RRMS than treatment with a DMT (alemtuzumab or ocrelizumab). 198 pts 2 years	1. Trial design GLs: Inconsistent (at least an assessor‐masking design is required) 2. PE GLs: Inconsistent (NEDA‐3 is not recommended)
NCT04121221/2018–000284‐93 Adult RMS Phase III Confirmatory trial NA	Randomized, placebo‐controlled, double‐blind, parallel‐arms trial Active, not recruiting 2019 18–55	Arm 1: GA Depot (long‐acting intramuscular injection of glatiramer acetate) Arm 2: Placebo	1. (a) 1 relapse within the past year, or (b) 2 relapses within the past 2 years, or (c) 1 relapse between 12 and 24 months, with ≥1 T1 Gd‐enhancing lesion within the past year 2. EDSS score 0–5.5 (inclusive)	ARR [timeframe: 52 weeks] To assess the efficacy, safety, and tolerability of GA Depot as compared to placebo in a study design of 52 weeks' duration. 960 pts 52 weeks	1. Trial design GLs: Consistent 2. PE GLs: Consistent
NCT05134441/2021–000028‐36 Adult RMS Phase III Confirmatory trial ENSURE‐1	Randomized, placebo‐controlled, double‐blind, parallel‐arms trial + OLE Recruiting 2022 18–55	Double‐blind part Arm 1: IMU‐838 (vidofludimus calcium) Arm 2: Placebo	Double‐blind part 1. (a) ≥1 relapse within the past year, or (b) ≥2 relapses within the past 2 years, or (c) A positive Gd‐enhancing lesion on an MRI scan (brain and/or spine) within the past year 2. EDSS score 0–5.5 (inclusive)	Double‐blind part Time to first relapse [timeframe: 72 weeks] To demonstrate the efficacy of IMU‐838 versus placebo in adult patients with active RMS in delaying the occurrences of relapses based on time to first relapse 1050 pts Double‐blind part: 72 weeks; overall: 8 years	Double‐blind part 1. Trial design GLs: Consistent 2. PE GLs: Consistent
NCT05201638/2021–000029‐28 Adult RMS Phase III Confirmatory trial ENSUER‐2	Randomized, placebo‐controlled, double‐blind, parallel‐arms trial + OLE Recruiting 2022 18–55	Double‐blind part Arm 1: IMU‐838 (vidofludimus calcium) Arm 2: Placebo	Double‐blind part 1. (a) ≥1 relapse within the past year, or (b) ≥2 relapses within the past 2 years, or (c) A positive Gd‐enhancing lesion on an MRI scan (brain and/or spine) within the past year 2. EDSS score 0–5.5 (inclusive)	Double‐blind part Time to first relapse [timeframe: 72 weeks] To demonstrate the efficacy of IMU‐838 versus placebo in adult patients with active RMS in delaying the occurrences of relapses based on time to first relapse 1050 pts Double‐blind part: 72 weeks; overall: 8 years	Double‐blind part 1. Trial design GLs: Consistent 2. PE GLs: Consistent
NCT02907177/2012–000541‐12 Adult RMS Phase III Confirmatory trial (Superiority trial) POINT	Randomized, placebo‐controlled, double‐blind, add‐on, parallel‐arms trial Terminated due to low recruitment 2017 18–55	Arm 1: Ponesimod + dimethyl fumarate Arm 2: Placebo + dimethyl fumarate	1. Ongoing treatment with dimethyl fumarate for ≥6 months 2. Active disease after at least 3 months of dimethyl fumarate treatment 3. ≥1 relapse within the past year or ≥2 relapses within the past 2 years, or ≥1 Gd‐enhancing lesion on an MRI scan within the past year 4. EDSS score 0–6.0 (inclusive)	ARR [timeframe: 68 weeks] To determine whether add‐on therapy with ponesimod reduces relapse frequency as compared to placebo in subjects with active RMS who are treated with dimethyl fumarate 600 pts 167 weeks	1. Trial design GLs: Possibly inconsistent (for add‐on therapy, the GLs state a useful design is a three‐arm trial seeking superiority of the combination versus both products in monotherapy.) 2. PE GLs: Consistent
NCT03342638 Adult RRMS Phase III Confirmatory trial MOST	Randomized, controlled, open‐label, non‐masking, add‐on, parallel‐arms trial Terminated 2107 18–58	Arm 1 (Control arm): aHSCT will be performed as follows: Autologous stem cells will be infused after conditioning with cyclophosphamide, mesna, rabbit anti‐thymoglobulin (rATG), and methylprednisolone. Granulocyte‐colony stimulating factor (G‐CSF) will be administered post‐transplant. Arm 2: aHSCT will be performed as follows: Autologous stem cells will be infused after conditioning with cyclophosphamide, mesna, rATG, and methylprednisolone. intravenous immunoglobulin (IVIg). G‐CSF will be administered post‐transplant.	1. ≥1 relapse within the past year or ≥2 relapses within the past 2 years, or ≥1 Gd‐enhancing lesion on an MRI scan within the previous year 2. EDSS score 0–6.0 (inclusive) 3. An EDSS >6.0 may be included if still relapsing–remitting disease and at least two enhancing lesions on MRI within the last 3 months 4. Inflammatory disease despite treatment with standard DMT including at least 6 months of interferon or copaxone. Inflammatory disease is defined based on either MRI (Gd‐enhancing lesion, new T2 lesion) or *steroid‐treated relapses (prescribed by a neurologist). 5. Minimum disease activity required: (a) Failed a first‐line DMT (Copaxone or interferon), defined as two or more *steroid‐treated relapses within the last 12 months. A relapse may also be evidence of active inflammation on MRI (Gd‐enhancing lesion or new T2 lesion) in the last 12 months on two MRIs at least 3 months apart (b) Failed a second‐ or third‐line MS drug: Zinbryta (daclizumab), Aubagio (teriflunomide), Gilenya (fingolimod), Tecifidera (dimethyl fumarate), Lemtrada (alemtuzumab), Ocrevus (ocrelizumab), Tysabri (natalizumab), and Rituxan (rituximab) or IVIg, defined as one *steroid‐treated relapse within the last 12 months or evidence of active inflammation on MRI (Gd‐enhancing lesion or new T2 lesion) in the last 12 months. (c) Cognitive dysfunction that prevents gainful employment, but competent to comply with treatment and informed consent A steroid‐treated relapse will include a relapse that was severe enough to justify treatment but due to patient intolerance of steroids, they were offered but not used.	NEDA‐3 [timeframe: 5 years] To compare two regimens: (1) cyclophosphamide and rATG versus (2) cyclophosphamide, rATG, and IVIg 200 pts 5 years	1. Trial design GLs: Inconsistent (at least an assessor‐masking design is required) 2. PE GLs: Inconsistent (NEDA‐3 is not recommended)
NCT02744222 Adult RRMS Phase III/II Confirmatory trial NA	Randomized, placebo‐ and active‐controlled, double‐blind, parallel‐arms trial + OLE Completed 2017 18–60	Double‐blind part Arm 1: Pegylated interferon β‐1a, dose 1 (180 μg) Arm 2: Pegylated interferon β‐1a, dose 2 (240 μg) Arm 3: Interferon β‐1a (Avonex®) Arm 4: Placebo Patients will receive blinded placebos for the first 20 weeks. Active comparator: Avonex®	Double‐blind part 1. ≥1 relapse within the previous year or ≥2 relapses within the previous 2 years, or ≥1 Gd‐enhancing lesion on an MRI scan within the previous year 2. EDSS score 0–5.5 (inclusive)	Double‐blind part Time to first relapse [timeframe: 52 weeks] To compare the efficacy, safety, and tolerability of pegylated interferon β‐1a, 180 μg, and 240 μg, versus Avonex® in patients with RRMS 165 pts Double‐blind part: 52 weeks; OLE: 52 weeks; Overall: 104 weeks	Double‐blind part 1. Trial design GLs: Possibly consistent (the shorter 20‐week placebo administration period will be a point of contention) 2. PE GLs: Consistent
NCT04966338 Adult RRMS Phase III Confirmatory trial (Assess the equivalency) NA	Randomized, active‐controlled, double‐blind, parallel‐arms trial Completed 2019 18–55	Arm 1: Ocrelizumab (Xacrel®, CinnaGen, Iran) Arm 2: Ocrelizumab (Ocrevus®, Roche, Switzerland) Active comparator: Ocrevus®	≥2 relapses within the past 2 years or 1 relapse within the past year EDSS score 0–5.5 (inclusive)	ARR [timeframe: 48 weeks] To verify the equivalency of Ocrelizumab (CinnaGen) versus Ocrevus® (Roche) in reducing the ARR in participants with RRMS 170 pts 96 weeks	1. Trial design GLs: NA (the GLs do not mention the trial design required for trials to assess the equivalence between the biosimilar and the original product) 2. PE GLs: NA (the GLs do not mention the primary end points required for trials to assess the equivalence between the biosimilar and the original product)
NCT04115488/2018–004751‐20 Adult RMS Phase III Confirmatory trial (Assess the equivalency) Antelope	Randomized, active‐controlled, double‐blind, parallel‐arms trial Completed 2019 18–60	Arm 1: Natalizumab (PB006, Polpharma Biologics S.A., Poland) Arm 2: Natalizumab (Tysabri®, Biogen, U.S.A.) Active comparator: Tysabri®	1. ≥1 relapse within the past year, and either (a) ≥1 Gd‐enhancing T1‐weighted brain lesions, or (b) ≥9 T2‐weighted brain lesions 2. EDSS score 0–5.0 (inclusive)	Cumulative number of new active lesions [timeframe: 48 weeks] To assess the equivalence in efficacy and similarity in the safety of biosimilar PB006 compared to Tysabri® in patients with RRMS 260 pts 48 weeks	1. Trial design GLs: NA (the GLs do not mention the trial design required for trials to assess the equivalence between the biosimilar and the original product) 2. PE GLs: NA (the GLs do not mention the primary end points required for trials to assess the equivalence between the biosimilar and the original product)
NCT04544436/2020–000893‐69 Adult RMS Phase III Confirmatory trial (Superiority trial) NA	Randomized, active‐controlled, double‐blind, parallel‐arms trial + OLE Active, not recruiting 2020 18–55	Double‐blind part Arm 1: Ocrelizumab, higher dose Arm 2: Ocrelizumab, approved dose Active comparator: Ocrelizumab, approved dose	Double‐blind part ≥2 relapses within the past 2 years or ≥1 relapse within the past year EDSS score 0–5.5 (inclusive) Average T25FWT score over two trials at screening and two trials at baseline, respectively, up to 150 (inclusive) seconds Average 9HPT score over four trials at screening and four trials at baseline, respectively, up to 250 (inclusive) seconds	Double‐blind part Reduction in cCDP12 [timeframe: approximately 4.3 years] To demonstrate the superiority of a higher dose of ocrelizumab over the approved dose of ocrelizumab as assessed by risk reduction in cCDP12 865 pts Double‐blind part: A minimum of 120 weeks; OLE: 96 weeks; overall: 4.3 years	Double‐blind part 1. Trial design GLs: Consistent 2. PE GLs: Possibly consistent (cCDP12 is not recommended. However, this PE may be acceptable given that the next therapeutic target for ocrelizumab, which has already proven effective against relapse, is to evaluate its efficacy against disability progression.)
NCT03979456/2018–000721‐31 Adult RRMS Phase III Confirmatory trial (non‐inferiority trial) RIDOSE‐MS	Randomized, active‐controlled, single‐blind (investigator‐blind), assessor‐masking, parallel‐arms trial Note: Each patient will have one treating physician responsible for all ongoing medical questions and decisions regarding the continuation of the study and one examining physician performing the blinded EDSS examination and assessments of exacerbations. The coordinating nurse will administer the study‐related tests and administer the rituximab infusions. MRI investigations will be performed blinded for the dosing arm allocation. Note: On the EC Clinical Trials Register website, this study is classified as a phase II dose comparison trial. Active, not recruiting 2018 20–52	Arm 1: Rituximab 500 mg every 6 months Arm 2: Rituximab 500 mg every 12 months Active comparator: Rituximab 500 mg every 6 months	1. The patient has completed the RIFUND‐MS trial and is treated with either of the study medications rituximab or dimethyl fumarate at the last visit of the RIFUND trial or has been treated with rituximab with a dose regimen of 500–1000 mg followed by 500 mg every 6 months for up to 2 years as part of clinical practice 2. EDSS score 0–5.5 (inclusive)	NEDA‐3 [timeframe: 4 years] The proportion of patients maintaining NEDA‐3 during Years 2–4 of the trial To compare two dosing regimens of rituximab in the long‐term treatment of MS in a non‐inferiority analysis between a 12‐month dosing interval of 500 mg rituximab with a 6‐month dosing interval 200 pts 4 years	1. Trial design GLs: Consistent (this assessor‐masking design would be acceptable) 2. PE GLs: Inconsistent (NEDA‐3 is not recommended)
NCT03689972/2018–002145‐11 Adult RRMS Phase III Confirmatory trial NOVA	Randomized, active‐controlled, open‐label, assessor‐masking, parallel‐arms trial + Open‐label, 2*2 crossover extension Note to double‐blind part: The centralized MRI reader, independent neurology evaluation committee, site examining neurologists, site backup examining neurologists, and site examining technicians were masked to study group assignments. (DOI: 10.1016/S1474‐4422(22)00143‐0) Active, not recruiting 2018 18–60	Double‐blind part Arm 1: Natalizumab, i.v. 4‐week standard interval dosing up to Week 72 Arm 2: Natalizumab, i.v. 6‐week extended interval dosing up to Week 72 Active comparator: Natalizumab, i.v. 4‐week standard interval dosing	Double‐blind part 1. ≥1 relapse within the previous year or ≥2 relapses within the previous 2 years, or ≥1 Gd‐enhancing lesion on an MRI scan within the previous year 2. EDSS score 0–5.5 (inclusive)	Double‐blind part Number of new or newly enlarging T2 hyperintense lesions [timeframe: 72 weeks] To evaluate the efficacy of natalizumab extended interval dosing (EID) in participants who have previously been treated with natalizumab standard interval dosing (SID) for at least 12 months, in relation to continued SID treatment 500 pts Double‐blind part: up to Week 96; OLE: Week 108 to Week 156; overall; 156 weeks	Double‐blind part 1. Trial design: GLs: Consistent (this assessor‐masking design would be acceptable) 2. PE GLs: Inconsistent (MRI findings are not recommended)
2. Adult RMS, phase II dose‐finding trials (n = 9)
NCT04056897 Adult RRMS Phase II Dose‐finding trial NA	Randomized, placebo‐ and active‐controlled, double‐blind, parallel‐arms trial Active, not recruiting 2019 18–60	Arm 1: BCD‐132, dose 1 Arm 2: BCD‐132, dose 2 Arm 3: Teriflunomide Arm 4: Placebo Active comparator: Teriflunomide	1. (a) ≥1 relapse within the past 12 months, or (b) ≥2 relapses within the past 24 months, or (c) ≥1 Gd‐enhancing T1 lesion on an MRI and 1 relapse over the past 24 months 2. EDSS score 0–5.5 (inclusive)	The cumulative number of T1 Gd‐enhancing lesions [timeframe: 24 weeks] To assess the efficacy and safety of BCD‐132 in RRMS 270 pts 100 weeks	1. Trial design GLs: Consistent 2. PE GLs: Consistent
NCT02975349/2016–001448‐21 Adult RMS Phase II Dose‐finding trial NA	Randomized, placebo‐ and active‐controlled, double‐blind, parallel‐arms trial + DBE + OLE Note: The active control group is a parallel open‐label group, not a double‐blinded group Note: The only available information on OLE is that it is ongoing and that the follow‐up period is 96 weeks. Because the trial design and primary end points are unknown, this OLE is not included in the list of phase II extension trials in this table. Active, not recruiting 2017 18–65	Double‐blind part Arm 1: Placebo Arm 2: Evobrutinib, dose 1 Arm 3: Evobrutinib, dose 2 Arm 4: Evobrutinib, dose 3 Arm 5: Dimethyl fumarate (open‐label) Active comparator: Dimethyl fumarate	Double‐blind part 1. ≥1 relapse within the past 2 years with either (a) ≥1 relapse within the past year, or (b) ≥1 Gd‐positive T1 lesion within past 6 months 2. EDSS score 0–6.0 (inclusive)	Double‐blind part The cumulative number of Gd‐enhancing T1 Lesions [timeframe: 24 weeks] To evaluate the efficacy and dose–response of M2951 on the number of Gd‐positive T1 MRI lesions versus placebo after 24 weeks of treatment 250 pts Double‐blind part: 24 weeks; DBE: 24 weeks; OLE: 96 weeks; overall: 144 weeks	Double‐blind part 1. Trial design GLs: Consistent 2. PE GLs: Consistent
2019–001967‐58 Adult RMS Phase II Dose‐finding trial NA	Randomized, placebo‐ and active‐controlled, double‐blind, parallel‐arms trial Prematurely ended 2020 No detailed information on the target age range, but at least within the range of 18 to 64 years old	Arm 1: Dapirolizumab pegol, dose 1 Arm 2: Dapirolizumab pegol, dose 2 Arm 3: Dapirolizumab pegol, dose 3 Arm 4: Ocrelizumab Arm 5: Placebo Active comparator: Ocrelizumab	1. (a) ≥2 relapses in the past 24 months (but not within 30 days before Day 1/baseline) and ≥1 relapse in the past 12 months, or (b) ≥1 relapse in the past 24 months (but not within 30 days before Day 1/baseline) and ≥1 new brain MRI lesion (Gd‐positive and/or new or enlarging T2 hyperintense lesion) within the last 12 months, or (c) ≥1 Gd‐positive lesion on brain MRI within the past 6 months, or (d) ≥1 relapse and ≥1 new brain MRI lesion (Gd‐positive and/or new or enlarging T2 hyperintense lesion) in the past 12 months 2. EDSS score 0–5.0 (inclusive)	The cumulative number of Gd‐enhancing T1 lesions over two brain MRI scans [timeframe: 12 weeks] To evaluate the effect of dapirolizumab pegol versus placebo on key brain MRI outcomes. 220 pts 48 weeks	1. Trial design GLs: Consistent 2. PE GLs: Consistent
NCT05417269/2021–004974‐67 Adult RRMS Phase II/I Dose‐finding trial ISEMIS Study	Phase IIb part randomized, placebo‐ and active‐controlled, double‐blind, parallel‐arms trial Note: The active control group is a parallel open‐label group, not a double‐blinded group Note: The study will be conducted under a Bayesian adaptive design approach comprising two phases: Phase I where three IMCY‐0141 doses will be administered following a dose escalation approach. An IDMC safety assessment will allow the escalation from lower to upper dose. Phase IIa leads to a preliminary estimate of the efficacy of each dose and determines the promising doses. The two doses that emerge as most promising will be recommended to advance to phase IIb, where final efficacy will be judged, again relative to placebo. The least‐promising dose may be dropped, at the discretion of the trial's IDMC. Recruiting 2022 18–45	Phase IIa Part Arm 1: IMCY‐0141, dose 1 Arm 2: IMCY‐0141, dose 2 Arm 3: IMCY‐0141, dose 3 Arm 4: Placebo Arm 5: Dimethyl fumarate (open label) Active comparator: Dimethyl fumarate Phase IIb part The number of IMCY‐0141 arms could be two by dropping the least‐promising dose arm.	Phases I and II Patients should be newly diagnosed or have a disease duration ≤3 years If not newly diagnosed, patients should have at least one relapse in the past 12 months. ≥1 Gd‐enhancing T1 lesion or ≥2 new or enlarging T2 lesions in the last 6 months EDSS score 0–5.0 (inclusive)	Phase II part The cumulative number of CUAL [timeframe: 36 weeks] To determine if any IMCY‐0141 dose(s) offer superior efficacy relative to placebo and to assess immune responses and biomarker data as potential early predictors of efficacy of IMCY‐0141 in adults presenting with RRMS Phase IIa part (the adaptive design phase): a minimum of 40 patients Phase IIb part: up to a maximum of 150 patients The sample size estimation is based on the total cumulative number of CUAL observed on brain MRI scans from Week 12 till Week 36. The study sample size has been estimated as adequate to determine if any IMCY‐0141 doses offer superior efficacy (as measured by CUAL) relative to placebo. Using the negative binomial model for CUAL, a maximum total of 150 patients are planned to be enrolled (including the 12 patients enrolled in phase I), with 30 patients randomized to each of the five groups: 36 weeks	Phase II part 1. Trial design GLs: Possibly consistent (regulatory agencies may have specific requirements for adaptive trials) 2. PE GLs: Consistent
NCT03846219/2018–001896‐19 Adult RRMS Phase II Dose‐finding trial EMPhASIS	Randomized, placebo‐controlled, double‐blind, parallel‐arms trial + OLE Active, not recruiting 2019 18–55	Double‐blind part Arm 1: IMU‐838 (vidofludimus calcium), 30 mg/day Arm 2: IMU‐838 (vidofludimus calcium), 45 mg/day Arm 3: Placebo Arm 4: IMU‐838 (vidofludimus calcium), 10 mg/day (for substudy)	Double‐blind part 1. (a) ≥1 relapse within the past 12 months and ≥2 relapses within the past 24 months, and (b) ≥1 Gd‐enhancing brain lesion within the last 6 months 2. EDSS score 0–5.5 (inclusive)	Double‐blind part The cumulative number of CUAL [timeframe: 24 weeks] To assess the efficacy and safety of 2 doses of IMU‐838, 30 mg/day and 45 mg/day in the main study, cohort 1 (and 10 mg/day for the patients in the cohort 2 substudy), in patients with RRMS 195 pts Double‐blind part: 24 weeks; overall: 10 years	Double‐blind part 1. Trial design GLs: Consistent 2. PE GLs: Consistent
NCT04711148/2020–005117‐41 Adult RRMS Phase II Dose‐finding trial NA	Randomized, placebo‐controlled, double‐blind, parallel‐arms trial + OLE Active, not recruiting 2021 18–55	Double‐blind part Arm 1: Orelabrutinib, dose 1 Arm 2: Orelabrutinib, dose 2 Arm 3: Orelabrutinib, dose 3 Arm 4: Placebo	Double‐blind part ≥1 relapse within the past 2 years with either: (a) 1 relapse within the past year, or (b) ≥1 Gd‐enhancing T1 lesion on an MRI within the last 6 months EDSS score 0–5.5 (inclusive)	Double‐blind part The cumulative number of new Gd‐enhancing T1 MRI brain lesions [timeframe: 12 weeks] To evaluate the efficacy of orelabrutinib on the cumulative number of new Gd‐enhancing T1 MRI brain lesions versus placebo over 12 weeks of treatment. 160 pts 120 weeks Double‐blind part: 24 weeks; overall: 120 weeks	Double‐blind part 1. Trial design GLs: Consistent 2. PE GLs: Consistent
NCT03737851 Adult RMS Phase II Dose‐finding trial NA	Randomized, placebo‐controlled, double‐blind, add‐on, parallel‐arms trial Completed 2018 18–65	Arm 1: Elezanumab, dose 1 + Standard care Arm 2: Elezanumab, dose 2 + Standard care Arm 3: Placebo + Standard care	1. Participant has evidence of physical disability according to EDSS or T25FW or 9HPT	Mean ODRS [timeframe: 52 weeks] The ODRS is a composite score derived from four components: EDSS, T25FW, and 9HPT in the dominant hand (9HPT‐D), and 9HPT in the non‐dominant hand (9HPT‐ND) To evaluate the safety and efficacy of elezanumab when added to the standard of care in RMS 165 pts 52 weeks	1. Trial design GLs: Possibly consistent (for add‐on therapy, the GLs state a useful design is a three‐arm trial seeking superiority of the combination versus both products in monotherapy. However, no clear statement as to at which stage of the clinical trials the three‐arm design should be applied. The 3‐arm trial design may not be mandatory for a phase II dose‐finding trial.) 2. PE GLs: Inconsistent (mean ODRS is not recommended)
NCT04079088/2019–001847‐28 Adult RMS Phase II Dose‐finding trial NA	Randomized, placebo‐controlled, double‐blind, add‐on, parallel‐arms trial Withdrawn: due to lack of stronger preclinical effects of BIIB061 on remyelination relative to opicinumab 2021 18–55	Arm 1: BIIB061 dose 1 + (Interferon β1 or Glatiramer acetate) Arm 2: BIIB061 dose 2 + (Interferon β1 or Glatiramer acetate) Arm 3: BIIB061 dose 3 + (Interferon β1 or Glatiramer acetate) Arm 4: Placebo + (Interferon β1 or Glatiramer acetate)	Must have at least one of the following occurring within past 12 months (a). ≥1 relapse or objective disability worsening, (b). ≥1 Gd‐enhancing T1 lesion on brain/spinal cord MRI, or (c). ≥1 new T2 lesion on brain/spinal cord MRI EDSS score 2.0–6.0 (inclusive)	1. The number of participants with treatment‐emergent AEs and SAEs [timeframe: 84 weeks] 2. ODRS [timeframe: 48 weeks] To evaluate the safety of BIIB061 versus placebo in RMS, and to evaluate the efficacy of BIIB061 to improve disability outcome versus placebo in RMS. 300 pts 84 weeks	1. Trial design GLs: Possibly consistent (for add‐on therapy, the GLs state a useful design is a three‐arm trial seeking superiority of the combination versus both products in monotherapy. However, no clear statement as to at which stage of the clinical trials the three‐arm design should be applied. The three‐arm trial design may not be mandatory for a phase II dose‐finding trial.) 2. PE GLs: Inconsistent (AE and ODRS are not recommended)
NCT03889639/2018–003927‐12 Adult RMS Phase II Dose‐finding trial NA	Randomized, placebo‐controlled, double‐blind, 2*8 crossover trial Completed 2019 18–55	Arm1: Tolebrutinib, dose 1. then placebo Arm 2: Tolebrutinib, dose 2, then placebo Arm 3: Tolebrutinib, dose 3, then placebo Arm 4: Tolebrutinib, dose 4, then placebo Arm 5–8: in reverse order	1. (a) ≥1 relapse within the past year, or (b) ≥2 relapses within the past 2 years, or (c) ≥1 Gd‐enhancing brain lesion on an MRI scan within the past 6 months	The cumulative number of new Gd‐enhancing T1‐hyperintense lesions [timeframe: 12 weeks] To determine the dose–response relationship for tolebrutinib to reduce the number of new active brain lesions 160 pts 24 weeks	1. Trial design GLs: Possibly inconsistent (if the drug effect develops rapidly and patients return to baseline conditions quickly after cessation of therapy, a randomized multiple crossover trial is acceptable. A crossover design with placebo may be a point that could be considered for acceptance) 2. PE GLs: Consistent
3. Adult RMS, phase III long‐term extension trials (n = 10)
NCT04338022/2019–004972‐20 Adult RMS Phase III Extension trial Evolution RMS 1	Open‐label, single‐arm trial Note: This extension trial is one of the parts of the following trial; randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial + DBE + OLE Active, not recruiting 2020 No age limit (subjects must have completed the double‐blind part of this trial)	OLE Part Arm 1: Evobrutinib	OLE part 1. Subjects who have completed the double‐blind part of the trial	OLE part Primary end point: Number of participants with AEs and SAEs [timeframe: 96 weeks in OLE period] To evaluate the long‐term safety and tolerability of evobrutinib 45 mg twice daily (BID) in participants with RMS over time 930 pts (this is the estimated number of patients who will be enrolled in the double‐blind part of the trial. Of these, subjects who have completed the double‐blind part of the study will participate in the OLE part.) DBTP: 156 weeks; DBE: 96 weeks, OLE: 96 weeks; overall: 348 weeks	OLE part 1. Trial design GLs: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points)
NCT04338061/2019–004980‐36 Adult RMS Phase III Extension trial Evolution RMS 2	Open‐label, single‐arm trial Note: This extension trial is one of the parts of the following trial; randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial + DBE + OLE Active, not recruiting 2020 No age limit (subjects must have completed the double‐blind part of this trial)	OLE Part Arm 1: Evobrutinib	OLE part 1. Subjects who have completed the double‐blind part of the trial	OLE part Primary end point: Number of participants with AEs and SAEs [timeframe: 96 weeks in OLE period] To evaluate the long‐term safety and tolerability of evobrutinib 45 mg twice daily (BID) in participants with RMS over time 930 pts (this is the estimated number of patients who will be enrolled in the double‐blind part of the trial. Of these, subjects who have completed the double‐blind part of the study will participate in the OLE part.) DBTP: 156 weeks; DBE: 96 weeks, OLE: 96 weeks; Overall: 348 weeks	OLE part 1. Trial design GLs: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points)
NCT05134441/2021–000028‐36 Adult RMS Phase III Extension trial ENSURE‐1	Open‐label, single‐arm trial Note: This extension trial is one of the parts of the following trial: randomized, placebo‐controlled, double‐blind, parallel‐arms trial + OLE Recruiting 2022 No age limit (subjects must have completed the double‐blind part of this trial)	OLE Part Arm 1: IMU‐838 (vidofludimus calcium)	OLE part 1. Subjects who have completed the double‐blind part of the trial	OLE part No information on end points To evaluate the efficacy, safety, and tolerability of IMU‐838 in adult patients with RMS 1050 pts (this is the estimated number of patients who will be enrolled in the double‐blind part of the trial. Of these, subjects who have completed the double‐blind part of the study will participate in the OLE part.) Double‐blind part: 72 weeks; overall: 8 years	OLE part 1. Trial design GLs: Consistent 2. PE GLs: NA
NCT05201638/2021–000029‐28 Adult RMS Phase III Extension trial ENSUER‐2	Open‐label, single‐arm trial Note: This extension trial is one of the parts of the following trial; randomized, placebo‐controlled, double‐blind, parallel‐arms trial + OLE Recruiting 2022 No age limit (subjects must have completed the double‐blind part of this trial)	OLE Part Arm 1: IMU‐838 (vidofludimus calcium)	OLE Part 1. Subjects who have completed the double‐blind part of the trial	OLE Part No information on end points To evaluate the efficacy, safety, and tolerability of IMU‐838 in adult patients with RMS 1050 pts (this is the estimated number of patients who will be enrolled in the double‐blind part of the trial. Of these, subjects who have completed the double‐blind part of the study will participate in the OLE part.) Double‐blind part: 72 weeks; overall: 8 years	OLE Part 1. Trial design GLs: Consistent 2. PE GLs: NA
NCT03232073/2016–004719‐10 Adult RMS Phase III Extension trial OPTIMUM LT	Open‐label, single‐arm trial Note: This study is the long‐term extension for the phase III trial (OPTIMUM; NCT02425644). The trial design of OPTIMUM was as follows: randomized, active‐controlled, double‐blind, parallel‐arms trial Active, not recruiting 2017 No age limit (subjects must have completed OPTIMUM)	Arm 1: Ponesimod	1. Completers of the OPTIMUM are eligible for participation in this extension trial	Primary end points: ARR; Time to first confirmed relapse; Time to first 12‐week CDA; Time to first 24‐week CDA; Patients with absence of relapses; Change from baseline in EDSS; NEDA‐3; NEDA‐4; Percent change from baseline in brain volume (PCBV); Cumulative number of CUA; Determination of the number of Gd + T1 lesions; Cumulative number of new or enlarging T2 lesions; Assessment of the volume of brain lesions; Absence of MRI lesions; Determination of the proportion of Gd + lesions at baseline evolving to persistent black holes (PBHs); Estimation of incidence rates of AEs; Estimation of incidence rates of treatment‐emergent morphological ECG abnormalities; Assessment of cardiac rhythms measured by ECG parameters; Change from baseline values by visit for cardiac rhythms; Change in ECG parameters from pre‐dose to selected post‐dose assessments; Absolute values and percent change from baseline in forced expiratory volume and forced vital capacity; Assessment of treatment‐emergent decrease from baseline in forced expiratory volume and forced vital capacity; Absolute change from baseline to end‐of‐study (EOS) versus change from baseline to end‐of‐treatment (EOT) in forced expiratory volume and forced vital capacity. The timeframe for change in ECG parameters from pre‐dose to selected post‐dose assessments: From day 1 in the extension study to EOT in the extension study, i.e., for up to 240 weeks The timeframe for others: from day 1 in OPTIMUM to EOT in the extension trial, i.e., for up to 354 weeks Safety objectives: To describe the long‐term safety and tolerability of ponesimod 20 mg in subjects with RMS. To describe the effects of re‐initiation of ponesimod treatment after interruption in subjects with RMS. Efficacy objectives: To describe the long‐term disease control in subjects with RMS receiving ponesimod 20 mg. To describe the effect of a switch from teriflunomide to ponesimod 20 mg on disease control in subjects with RMS 800 pts OPTIMUM: 108 weeks; OPTIMUM LT: 246 weeks; Overall: 354 weeks	1. Trial design GLs: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points)
NCT04130997/2019–003625‐16 Adult RMS Phase III Extension trial NA	Open‐label, single‐arm trial Note: This study is the long‐term extension for two phase III trials, ULTIMATE 1 (NCT03277261) or ULTIMATE 2 (NCT03277248). The two trials' design was as follows: Randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial. This study is also the long‐term extension for a phase II OLE trial (NCT03381170). Enrolling by invitation 2019 No age limit (subjects must have completed the above three trials)	Arm 1: Ublituximab	1. Completers of the ULTIMATE 1, ULTIMATE 2, or phase II OLE trial (NCT03381170) are eligible for participation in this extension trial.	Primary end point: ARR [timeframe: up to Week 336] To evaluate the long‐term safety and efficacy of ublituximab in participants with RMS 1100 pts ULTIMATE 1: 96 weeks; UMITIMATE 2: 96 weeks; NCT03381170: 96 weeks on therapy; overall: 336 weeks	1. Trial design GLs: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points)
NCT05156281/2020–005929‐89 Adult RMS Phase III Extension trial REMODEL‐2	Open‐label, single‐arm trial Note: This extension trial is one of the parts of the following trial: randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial + OLE Recruiting 2021 No age limit (subjects must have completed the double‐blind part of this trial)	OLE Part Arm 1: Remibrutinib	OLE part 1. Subjects who have completed the double‐blind part of the trial	OLE part Secondary end points (no primary end point): Number of participants with AEs and SAEs; ARR; Annualized rate of new or enlarging T2 lesions; Time to 6‐month confirmed disability progression (6mCDP) on EDSS; Change from baseline in the Symbol Digit Modalities Test (SDMT); Neurofilament light chain (NfL); Change from baseline in Multiple Sclerosis Impact Scale (MSIS‐29). Timeframe: Day 1 extension up to 5 years To assess long‐term safety, tolerability, and efficacy parameters in participants treated with remibrutinib 800 pts (this is the estimated number of patients who will be enrolled in the double‐blind part of the trial. Of these, subjects who have completed the double‐blind part of the study will participate in the OLE part.) Double‐blind part: 30 months; OLE part: 30 months; overall: 5 years	OLE part 1. Trial design GLs: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points)
NCT05147220/2020–005899‐36 Adult RMS Phase III Extension trial NA	Open‐label, single‐arm trial Note: This extension trial is one of the parts of the following trial: randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial + OLE Recruiting 2021 No age limit (subjects must have completed the double‐blind part of this trial)	OLE part Arm 1: Remibrutinib	OLE part 1. Subjects who have completed the double‐blind part of the trial	OLE part Secondary end points (no primary end point): Number of participants with AEs and SAEs; ARR; Annualized rate of new or enlarging T2 lesions; Time to 6‐month confirmed disability progression (6mCDP) on EDSS; Change from baseline in the SDMT; Neurofilament light chain (NfL); Change from baseline in Fatigue Symptoms and Impacts Questionnaire–Relapsing Multiple Sclerosis (FSIQ‐RMS); Change from baseline in Generalized Anxiety Disorder Scale (GAD‐7); Change from baseline in Patient Health Questionnaire‐9 (PHQ‐9); Change from baseline in Brief Pain Inventory–short form (BPI‐SF); Change from baseline in Health Utilities Index (HUI‐Ill); Change from baseline in Multiple Sclerosis Impact Scale (MSIS‐29) timeframe: Day 1 Extension up to 5 years To assess long‐term safety, tolerability, and efficacy parameters in participants treated with remibrutinib 800 pts (this is the estimated number of patients who will be enrolled in the double‐blind part of the trial. Of these, subjects who have completed the double‐blind part of the study will participate in the OLE part.) Double‐blind part: 30 months; OLE part: 30 months; overall: 5 years	OLE part 1. Trial design GLs: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points)
NCT02744222 Adult RRMS Phase III/II Extension trial NA	Open‐label, parallel‐arms trial Note: This extension trial is one of the parts of the following trial: randomized, placebo‐ and active‐controlled, double‐blind, parallel‐arms trial + OLE Completed 2017 18–60	OLE part Arm 1: Pegylated interferon β‐1a 180 μg Arm 2: Pegylated interferon β‐1a 240 μg From Week 52 until Week 100, patients will receive open‐label test drugs 180 μg or 240 μg	OLE part 1. Subjects who have completed the double‐blind part of this trial	OLE part Secondary end points (no primary end point): The cumulative number of new accumulating contrast lesions on MRI in T1 mode, and new lesions in T2‐weighted or any increase in foci in T2 mode; Proportion of patients without contrast‐enhancing lesions; Number of new or enlarging T2‐weighted lesions Proportion of patients without new or enlarging T2‐weighted lesions; Changes in T2‐weighted lesion volume; Changes in hypointense T1‐weighted lesion volume; Annual average frequency of relapses; Proportion of relapse‐free patients Proportion of patients with sustained disability progression; EDSS; T25FW; 9HPT; SDMT; The proportion of patients who developed AEs/SAEs; The proportion of patients, in each group, who developed СТСАЕ v. 4.03 Grade 3–4 AEs The proportion of patients, in each group, who discontinued the study due to AEs/SAEs; The proportion of BAb‐ and NAb‐positive patients Timeframe: 104 weeks No information on the main objective 165 pts (this is the estimated number of patients who will be enrolled in the first part of the trial. Subjects who have completed the prior part of the trial will participate in the OLE part.) Double‐blind part: 52 weeks; OLE: 52 weeks; overall: 104 weeks	OLE part 1. Trial design GLs: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points)
NCT03689972/2018–002145‐11 Adult RRMS Phase III Extension trial NOVA	Open‐label, 2*2 crossover trial Note: This extension trial is one of the parts of the following trial: randomized, controlled, open‐label, assessor‐masking, parallel‐arms trial + open‐label, 2*2 crossover extension Active, not recruiting 2018 No age limit (subjects must have completed the double‐blind part of this trial)	OLE Part Arm 1: Natalizumab 300 mg s.c. injection every 6 weeks from Week 108 through Week 126, followed by natalizumab 300 mg IV. infusion every 6 weeks from Week 132 through Week 150 Arm 2: Natalizumab 300 mg i.v. infusion every 6 weeks from Week 108 through Week 126 followed by natalizumab 300 mg s.c. injection every 6 weeks from Week 132 through Week 150	OLE part 1. Subjects who have completed the double‐blind part of NOVA	OLE part Primary end point: Percentage of participants indicating a preference for natalizumab SC. administration at the end of Part 2 [timeframe: 156 weeks] To evaluate participant preference for s.c. versus i.v. route of natalizumab administration. 500 pts (this is the estimated number of patients who will be enrolled in the first part of the trial. Subjects who have completed the prior part of the trial will participate in the OLE part.) Double‐blind part: up to Week 96; OLE part: Week 108 to Week 156; overall; 156 weeks	OLE part 1. Trial design: GLs: Possibly consistent (if the drug effect develops rapidly and patients return to baseline conditions quickly after cessation of therapy, a randomized multiple crossover trial is acceptable. It could be acceptable given the main objective and the primary end point of this OLE) 2. PE GLs: NA (the GLs do not recommend specific end points)
4. Adult RMS, phase II long‐term extension trials (n = 9)
NCT03249714 Adult RMS Phase II Extension trial NA	Open‐label, single‐arm trial Note: This extension trial is one of the parts of the following trial: randomized, placebo‐controlled, double‐blind, parallel‐arms trial + OLE 2018 Completed No age limit (age between 18 and 55 years old at the time of the trial participation. Subjects must have completed the double‐blind part.)	OLE Part Arm 1: Ofatumumab	OLE part 1. Subjects who have completed the double‐blind part of the trial (NCT03249714)	OLE part Secondary end points (no primary end point): Total number of Gd‐enhancing T1 lesions [Timeframe: Week 24 up to Week 48]; Number of new or enlarging T2 Lesions (annualized T2 lesion rate) [timeframe: Week 24 up to Week 48]; ARR [timeframe: Week 24 up to Week 48]; Pharmacokinetic concentrations of ofatumumab [timeframe: Weeks 24, 28, 36, and 48]; The CD19+ B‐cell counts [timeframe: Weeks 24, 36, and 48]; Participants with confirmed relapse [timeframe: baseline up to Week 48] To access efficacy, safety, and pharmacokinetics data of ofatumumab for patients with RMS 60 pts (this is the estimated number of patients who will be enrolled in the double‐blind part of the trial. Of these, subjects who have completed the double‐blind part of the study will participate in the OLE part.) Double‐blind part: 24 weeks; OLE part: 24 weeks; overall: 48 weeks	OLE part 1. Trial design GLs: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points)
NCT03222973/2017–001224‐22 Adult RMS Phase II Extension trial AFFINITY	Open‐label, single‐arm trial Note: This extension trial is one of the parts of the following trial: randomized, placebo‐controlled, double‐blind, parallel‐arms trial + OLE Terminated: The top‐line results from part 1 did not meet the pre‐specified primary end point or the key secondary end points. 2017 No age limit (age between 18 and 58 years old at the time of the trial participation. Subjects must have completed the double‐blind part).	OLE Part Arm 1: Opicinumab	OLE part 1. Subjects who have completed the double‐blind part of the AFFINITY	OLE part Primary end point: Number of participants experiencing AEs and SAEs [timeframe: Week 73 to Week 169] To investigate long‐term efficacy (disability improvement) and additional safety measures of opicinumab as an add‐on therapy to DMT in participants with RMS. 240 pts (this is the estimated number of patients who will be enrolled in the double‐blind part of the trial. Of these, subjects who have completed the double‐blind part of the study will participate in the OLE part.) Double‐blind part: 72 weeks; OLE: 96 weeks; overall: 169 weeks	OLE part 1. Trial design GLs: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points)
NCT03381170 Adult RMS Phase II Extension trial NA	Open‐label, single‐arm trial Note: This study is the long‐term extension for the phase II dose‐finding trial (TG1101‐RMS201 trial, NCT02738775). The trial design of TG1101‐RMS201 was as follows: Randomized, placebo‐controlled, double‐blind, 2*6 crossover trial 2017 Completed No age limit (age between 18 and 55 years old at the time of the TG1101‐RMS201, NCT02738775, participation. Subjects must have completed the TG1101‐RMS201 trial).	Arm 1: Ublituximab	1. Completers of the TG1101‐RMS201 (NCT02738775) trial are eligible for this extension trial	Primary end point: Number of participants with treatment‐related events as assessed by CTCAE, Version 4.0 [timeframe: 96 weeks] To evaluate the long‐term use of ublituximab in patients with RMS 48 pts TG1101‐RMS201 trial: 48 weeks; this extension trial: 96 weeks; overall: 144 weeks	1. Trial design GLs: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points)
NCT03996291/2018–004731‐76 Adult RMS Phase II Extension trial NA	Open‐label, single‐arm trial Randomized, placebo‐controlled, double‐blind, 2*8 crossover trial Active, not recruiting 2019 No age limit (age between 18 and 55 years old at the time of the DRI15928 trial participation. Subjects must have completed the DRI15928 trial).	Arm 1: Tolebrutinib	1. Completers of the DRI15928 (NCT03889639) trial are eligible for this extension trial	Primary end points: Number of participants with AEs and SAEs [timeframe: baseline to final follow‐up visit (Month 60 plus 8 weeks)]; Number of participants with potentially clinically significant abnormalities [timeframe: baseline to final follow‐up visit (Month 60 plus 8 weeks)]. To determine the long‐term safety and tolerability of tolebrutinib in RMS participants 105 pts DRI15928: 16 weeks; overall 60 months and 8 weeks.	1. Trial design GLs: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points)
NCT03846219/2018–001896‐19 Adult RRMS Phase II Extension trial EMPhASIS	Open‐label, parallel‐arms trial Note: This extension trial is one of the parts of the following trial: randomized, placebo‐controlled, double‐blind, parallel‐arms trial + OLE Active, not recruiting 2019 No age limit (subjects must have completed the double‐blind part of this trial)	OLE Part Arm 1: IMU‐838 (vidofludimus calcium), 30 mg/day Arm 2: IMU‐838 (vidofludimus calcium), 45 mg/day	OLE part 1. Subjects who have completed the double‐blind part of the EMPhASIS	OLE part No information on end points No information on the main objective 195 pts (this is the estimated number of patients who will be enrolled in the double‐blind part of the EMPhASIS. Of these, subjects who have completed the double‐blind part of the study will participate in the OLE part.) Double‐blind part: 24 weeks; OLE part: 9.5 years; overall: 10 years	OLE part 1. Trial design GLs: Consistent 2. PE GLs: NA
NCT04711148/2020–005117‐41 Adult RRMS Phase II Extension trial NA	Open‐label, parallel‐arms trial Note: This extension trial is one of the parts of the following trial: randomized, placebo‐controlled, double‐blind, parallel‐arms trial + OLE Active, not recruiting 2021 No age limit (subjects must have completed the double‐blind part of this trial)	OLE Part Arm 1: Orelabrutinib, dose 1 Arm 2: Orelabrutinib, dose 2 Arm 3: Orelabrutinib, dose 3 Note: The number of arms will be determined by the results of a double‐blind part of this trial	OLE part 1. Subjects who have completed the double‐blind part of the trial (NCT04711148)	OLE part Primary end point: the cumulative number of new Gd‐enhancing T1 MRI brain lesions [timeframe: 120 weeks] To collect additional data on the long‐term safety and efficacy of orelabrutinib at doses suggested by the double‐blind results of this trial 160 pts (this is the estimated number of patients who will be enrolled in the double‐blind part of the trial. Of these, subjects who have completed the double‐blind part of the study will participate in the OLE part.) Double‐blind part: 24 weeks; overall: 120 weeks	OLE part 1. Trial design GLs: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points)
NCT04879628/2020–004785‐19 Adult RMS Phase II Extension trial NA	Open‐label, parallel‐arms trial Note: This extension trial is one of the parts of the following trial: randomized, placebo‐controlled, double‐blind, parallel‐arms trial + OLE Active, not recruiting 2021 No age limit (subjects must have completed the double‐blind part for 18–55 years old)	OLE Part Arm 1: Frexalimab i.v. Arm 2: Frexalimab s.c.	OLE part 1. Subjects who have completed the double‐blind part of the trial (NCT04879628)	OLE Part Secondary end points (no primary end point) Number of participants with AEs and SAEs; Number of participants with antidrug antibodies (ADA); Maximum concentration; Time to Cmax; Area under the curve over the dosing interval; Elimination half‐life. Timeframe: up to 172 weeks To evaluate the safety and tolerability of frexalimab and the pharmacokinetics of frexalimab 129 pts (this is the estimated number of patients who will be enrolled in the double‐blind part of the trial. Of these, subjects who have completed the double‐blind part of the study will participate in the OLE part.) Double‐blind part: 12 weeks; overall: no longer than 176 weeks	OLE Part 1. Trial design GLs: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points)
NCT03239860/2016–004935‐18 Adult RMS Phase II Extension trial ANGEL‐MS	Randomized, controlled, double‐blind, parallel‐arms trial Note: This study is the long‐term extension for the phase II trial (CHANGE‐MS, NCT02782858). The trial design of CHANGE‐MS was as follows: randomized, placebo‐controlled, double‐blind, parallel‐arms trial Terminated: Based on R&D strategic reasons 2017 No age limit (must have completed the CHANGE‐MS for 18–55 years old)	Arm 1: Temelimab, dose 1 Arm 2: Temelimab, dose 2 Arm 3: Temelimab, dose 3	1. Completers of the CHANGE‐MS are eligible for this extension trial	Primary end points: AEs and SAEs; Clinical safety laboratory (hematology, chemistry, and coagulation); IgG4 dosing; Physical examination and vital signs: blood pressure, heart rate, temperature, and body weight; 12‐lead ECGs; Human anti‐GNbAC1 antibodies in serum; Suicidality assessed with the Columbia Suicide Severity Rating Scale (C‐SSRS). Timeframe: 96 weeks To assess the long‐term safety of temelimab in patients with RRMS and the long‐term efficacy of temelimab in terms of MRI outcomes, relapse rate, disability, and disease progression. 200 pts CHANGE‐MS: 48 weeks; ANGEL‐MS: 48 weeks; Overall: 96 weeks	1. Trial design GLs: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points)
NCT05049161/2021–001973‐21 Adult RMS Phase II Extension trial NA	Randomized, controlled, double‐blind, parallel‐arms trial Note: This study is the long‐term extension for the phase II trial (ProTEct‐MS, NCT04480307). The trial design of ProTEct‐MS was as follows: Randomized, placebo‐controlled, double‐blind, parallel‐arms trial. In ProTEct‐MS, temelimab was administered subsequent to rituximab, i.e., no co‐administration of rituximab and temelimab was done in the study. Terminated: Drug product unavailability 2021 No age limit (subjects must have completed the ProTEct‐MS for 18–55 years old)	Arm 1: Temelimab, dose 1 Arm 2: Temelimab, dose 2 Arm 3: Temelimab, dose 3	1. Completers of the ProTEct‐MS are eligible for this extension trial	Primary end point: Number of patients with treatment‐related AEs [timeframe: 48 weeks] To assess the long‐term safety and tolerability of temelimab in patients with RMS who are treated with rituximab 40 pts ProTEct‐MS: 48 weeks; This extension trial: 48 weeks; Overall: 96 weeks	1. Trial design GLs: Consistent 2. PE GLs: NA (The GLs do not recommend specific end points)
5. Pediatric RMS, phase III trials, and phase II trials (n = 10)
NCT04926818/2020–002700‐39 Pediatric RRMS Phase III Confirmatory trial (non‐inferiority trial) NEOS	Randomized, active‐controlled, double‐blind, triple‐dummy, parallel‐arms trial + OLE Recruiting 2021 10–17	Double‐blind part Arm 1: Ofatumumab + placebo to match fingolimod + placebo to match siponimod Arm 2: Siponimod + placebo to match fingolimod + placebo to match ofatumumab Arm 3: Fingolimod + placebo to match ofatumumab + placebo to match siponimod Active comparator: Fingolimod OLE Part No detailed information Note: The trial design elements for this study have been discussed and agreed with the FDA and the CHMP. (https://doi.org/10.1177/17562864211070449)	Double‐blind part 1. (a) ≥1 MS relapse/attack within the past year, or (b) 2 MS relapse within the past 2 years, or (c) ≥1 T2 lesion within the past year 2. EDSS score: 0–5.5 (inclusive) OLE part No detailed information	Double‐blind part ARR [timeframe: 24 months] <Reference information on one of the secondary end points > ARR as compared to historical interferon β‐1a data [timeframe: baseline up to 24 months] The historical data for interferon β‐1a will be derived from prior phase III trials. To demonstrate the non‐inferiority of ofatumumab and/or siponimod as compared to fingolimod 180 pts Double‐blind part: 2 years; OLE part: 5 years (except for the first 12 weeks transition which will remain double blind); overall: 7 years OLE part Number of AEs and SAEs [timeframe: baseline up approximately 66 months] No other detailed information	Double‐blind part 1. Trial design GLs: Possibly inconsistent, but agreed by the regulatory agencies (as detailed in the left column, it was agreed to avoid setting up a placebo arm.) Recommendations: Consistent 2. PE GLs: Consistent Recommendations: Inconsistent (time to the first relapse is recommended) OLE part 1. Trial design: OLE GLs: Consistent Recommendations: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points) Recommendations: NA (no specific primary end points are recommended, but safety monitoring by OLE is recommended)
NCT05123703/2020–004128‐41 Pediatric RRMS Phase III Confirmatory trial (non‐inferiority trial) Operatta 2	Randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial + OLE, single‐arm trial Recruiting 2022 10–17	Double‐blind part Arm 1: Ocrelizumab + placebo to match fingolimod Arm 2: Fingolimod + placebo to match ocrelizumab Active comparator: Fingolimod OLE part Arm 1: Ocrelizumab	Double‐blind part 1. (a) ≥1 MS relapse/attack within the past year, or (b) 2 MS relapse within the past 2 years, or (c) ≥1 Gd‐enhancing lesion within the past 6 months 2. EDSS score: 0–5.5 (inclusive) OLE part No detailed information	Double‐blind part ARR [timeframe: approximately 4 years] To demonstrate the non‐inferiority of ocrelizumab compared with fingolimod based on ARR 223 pts Double‐blind part: 96 weeks; OLE part: at least 144 weeks; overall: 8 years OLE part Incidence and SAEs, with severity determined according to CTCAE, Version 5.0 [timeframe: baseline up approximately 8 years] Prevalence of anti‐drug antibodies against ocrelizumab [timeframe: baseline up approximately 8 years] No other detailed information	Double‐blind part 1. Trial design GLs: Possibly inconsistent (for pediatric MS patients, it is unclear whether the GLs interpret fingolimod as a first‐line DMT. Given that there are only three DMTs approved and fingolimod was first approved in the EU, it would probably be considered the first‐line DMT in pediatric MS patients.) Recommendations: Consistent 2. PE GLs: Consistent Recommendations: Inconsistent (time to the first relapse is recommended) OLE part 1. Trial design: OLE GLs: Consistent Recommendations: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points) Recommendations: NA (no specific primary end points are recommended, but safety monitoring by OLE is recommended)
2020–004431‐24 Pediatric RRMS Phase III Confirmatory trial (Non‐inferiority trial) PIONEER	Randomized, active‐controlled, double‐blind, double‐dummy, parallel‐arms trial Ongoing 2019 10–17	Arm 1: Ponesimod + placebo to match fingolimod Arm 2: Fingolimod + placebo to match ponesimod Active comparator: Fingolimod	1. (a) ≥1 MS relapse during the past year, or (b) 2 MS relapses in the past 2 years, or (c) ≥1 Gd‐enhancing lesions on MRI within the past 6 months 2. EDSS score: 0–5.5 (inclusive)	Time to first relapse [timeframe: 108 weeks] To demonstrate that ponesimod is not less effective than fingolimod in reducing relapses. 212 pts 108 weeks	1. Trial design GLs: Possibly inconsistent (for pediatric MS patients, it is unclear whether the GLs interpret fingolimod as a first‐line DMT. Given that there are only three DMTs approved and fingolimod was first approved in the EU, it would probably be considered the first‐line DMT in pediatric MS patients.) Recommendations: Consistent 2. PE GLs: Consistent Recommendations: Consistent
NCT03958877/2018–003008‐38 Pediatric RRMS Phase III Confirmatory trial NA	Randomized, active‐controlled, open‐label, non‐masking, parallel‐arms trial + OLE Recruiting 2019 10–17	Double‐blind part Arm 1: Peginterferon β‐1a Arm 2: Interferon β‐1a Active comparator: Interferon β‐1a OLE part No detailed information	Double‐blind part 1. (a) ≥1 MS relapse within the past year, or (b) 2 MS relapse within the past 2 years, or (c) ≥1 Gd‐enhancing lesion within the past 6 months 2. EDSS score: 0–5.5 (inclusive) OLE part 1. Subjects who completed the study treatment in the double‐blind part (Week 96 visit), as per protocol.	Double‐blind part ARR [timeframe: 96 weeks] To evaluate the safety, tolerability, and descriptive efficacy of peginterferon β‐1a 142 pts Double‐blind part: 96 weeks; OLE part: 100 weeks; overall: 196 weeks OLE part: Percentage of participants with AEs, SAEs, and AEs leading to study treatment discontinuation [timeframe: from Week 96 to Week 196] To evaluate the long‐term safety of peginterferon β‐1a No other detailed information	Double‐blind part 1. Trial design GLs: Inconsistent (at least an assessor‐masking design is required) Recommendations: Consistent 2. PE GLs: Consistent Recommendations: Inconsistent (time to the first relapse recommended) OLE part 1. Trial design: OLE GLs: Consistent Recommendations: Consistent 2. PE GLs: NA (the GLs do not recommend specific end points) Recommendations: NA (no specific primary end points are recommended, but safety monitoring by OLE is recommended)
2019–001549‐42 Pediatric RRMS (including adult RRMS) Phase III Confirmatory trial StarMS	Randomized, active‐controlled, open‐label, non‐masking, parallel‐arms trial No update information due to a trial in Great Britain—no longer in the EU/European Economic Area 2020 16–55	Arm 1: aHSCT Arm 2: DMTs (alemtuzumab or ocrelizumab) Active comparator: Alemtuzumab or ocrelizumab	1. Severe inflammatory disease defined as RRMS course with ≥2 relapses, or 1 such relapse and evidence of MRI disease activity >3 months before or after its onset, in the past 12 months despite being on a DMT* 2. EDSS score 0–6.0 (inclusive)* If the EDSS score is 6.0, this must be due to confirmed relapse rather than progressive disease. *Patients with EDSS scores of 0–1.5 or those who failed only first‐line treatments must also fulfill the following criteria: short illness duration (<5 years), active disease clinically and radiologically (i.e., at least 2 relapses in the last 12 months and evidence of multiple Gd‐enhancing MRI lesion), high brain lesion load, and brain or spinal cord atrophy.	The proportion of patients who have maintained NEDA‐3 status [timeframe: 2 years] To assess whether aHSCT is more effective at maintaining clinical stability in patients with highly active RRMS than treatment with a DMT (alemtuzumab or ocrelizumab) 198 pts (20 pts, under 18) 2 years	1. Trial design GLs: Inconsistent (at least an assessor‐masking design is required) Recommendations: Consistent 2. PE GLs: Inconsistent (NEDA‐3 is not recommended) Recommendations: Inconsistent (NEDA‐3 is not recommended)
NCT03368664/2016–003100‐30 Pediatric RRMS Phase III Confirmatory trial LemKids	Open‐label, non‐masking, single‐arm trial Recruiting 2017 10–17	Arm 1: Alemtuzumab Pre‐treatment: DMT (compared with the effect of DMT before alemtuzumab administration) (switch trial, not add‐on trial)	≥2 MS attacks and ≥1 MS attack (relapse) within the past year during treatment with a β interferon therapy (IFNβ) or glatiramer acetate (GA) after being on that therapy for at least 6 months, and is currently still taking the same therapy At least 1 of the following: (a) ≥1 new or enlarging T2 hyperintense lesion or Gd‐enhancing lesion while on that same prior therapy (IFNβ or GA), or (b) ≥2 relapses in the past year, or (c) tried at least 2 MS DMTs. EDSS score: 0–5.5 (inclusive)	The number of new or enlarged T2 lesions [timeframe: Period 1: Month 4 up to Month 0, Period 2: Month 4 to Month 8] To evaluate the efficacy, safety, and tolerability of alemtuzumab in pediatric patients who have disease activity on prior DMT. 65 pts 5 years	1. Trial design GLs: Inconsistent (at least an assessor‐masking design is required) Recommendations: Consistent (if an open‐label trial is insufficient, conduct a controlled trial) 2. PE GLs: Inconsistent (MRI findings are unacceptable) Recommendations: Possibly consistent (a robust relationship between clinical efficacy and MRI end points must have been demonstrated in the pivotal study of the same drug in adult‐onset MS)
2017–005129‐18 Pediatric RRMS Phase III Confirmatory trial Assess the equivalency or superiority) NA	Randomized, controlled, open‐label, pragmatic, parallel‐arms trial Note: No information on whether the outcome assessor was blinded or not. Ongoing 2019 10–17	Arm 1: Glatiramer acetate Arm 2: Interferon β‐1a Active comparator: None	1. Treatment naive 2. ≥1 MS relapse during the past year 3. EDSS score: 0–5.5 (inclusive)	The proportion of subjects free of new or newly enlarging T2 hyperintense on brain MRI scans [timeframe: 96 weeks] To demonstrate the equivalence of two injectable drugs or the superiority of one of them on MRI and clinical effectiveness outcomes. 142 pts 96 weeks	1. Trial design GLs: Consistent, if this is a reliable assessor‐masking design Inconsistent, if this is not a reliable assessor‐masking design Recommendations: Consistent 2. PE GLs: Inconsistent (MRI findings are unacceptable) Recommendations: Possibly consistent (a robust relationship between clinical efficacy and MRI end points must have been demonstrated in the pivotal study of the same drug in adult‐onset MS)
NCT03870763/2018–000516‐22 Pediatric RRMS Phase III Confirmatory trial NA	Randomized, placebo‐controlled, double‐blind, double‐dummy, parallel‐arms trial Terminated: The decision to stop the trial was based on long‐term difficulties in fulfilling our enrolment commitments and changes in the pediatric MS landscape which no longer support placebo‐controlled trials. 2019 10–17	Arm 1: Dimethyl fumarate + placebo to match peginterferon β‐1a Arm 2: Peginterferon β‐1a + placebo to match dimethyl fumarate Arm 3: Placebo Active comparator: None	1. At least 1 relapse during the past year, or must have evidence of asymptomatic disease activity seen on MRI in the past 6 months, or ≥2 relapses in the past 2 years 2. EDSS score: 0–5.0 (inclusive)	Time to first relapse [timeframe: 96 weeks] To evaluate the efficacy of dimethyl fumarate and peginterferon β‐1a, both compared with placebo. 340 pts (11, actual) 100 weeks	1. Trial design GLs: Consistent Recommendations: Inconsistent (the trial design with a placebo arm is not recommended) 2. PE GLs: Consistent Recommendations: Consistent
NCT04075266/2016–002667‐34 Pediatric RRMS Phase II PK/PD trial Immunogenicity and exploratory trial NA	Open‐label, non‐masking, parallel‐arms trial Active, not recruiting 2020 10–17	Arm 1: <40 kg, ocrelizumab 300 mg Arm 2: 40 kg≤, ocrelizumab 600 mg Arm 3: <40 kg, ocrelizumab another dose Arm 4: 40 kg≤, ocrelizumab another dose	1. At least 6 contiguous months of DMT within the past 1 year must have evidence of disease activity occurring after the full 6‐month course of treatment, i.e., at least one relapse or ≥1 Gd‐enhancing lesion(s) on a T1‐weighted brain MRI 2. EDSS score: 0–5.0 (inclusive)	1. Serum concentration of ocrelizumab [timeframe: up to 6 months, up to 5 years] 2. Levels of CD19+ B‐cell count in blood [timeframe: up to 6 months, up to 5 years] To characterize the ocrelizumab pharmacokinetic profile To evaluate the relationship between drug exposure and pharmacodynamics (CD19+ B‐cell count) 36 pts 5 years	1. Trial design GLs: NA Recommendations: Consistent 2. PE GLs: NA Recommendations: NA (no specific primary end points are recommended, but PK/PD and safety end points are recommended)
NCT04062331 Pediatric/adult RRMS Phase II/I Exploratory trial NA	Randomized, placebo‐controlled, patient‐blind, assessor‐masking, add‐on, parallel‐arms trial Unknown status 2019 14≤	Arm 1: TMS 1 Hertz + Natalizumab Arm 2: TMS 5 Herts + Natalizumab Arm 3: Placebo 0 Herts (Sham) + Natalizumab	1. Patients diagnosed with RRMS in their inflammatory forms who have completed a 14‐dose treatment with natalizumab 2. EDSS score: 3.0–6.5	1. EDSS [timeframe: 1 year] 2. Comprehensive clinical assessment of the disease [timeframe: 1 year] It consists of three tests: T25FW, 9HPT, and PASAT 3. Cognitive function [timeframe: 1 year] Brief Repeatable Battery of Neuropsychological Test To demonstrate the therapeutic effect of TMS through measurement of clinical changes in EDSS 90 pts 1 year	1. Trial design GLs: Possibly inconsistent (for add‐on therapy, the GLs state a useful design is a three‐arm trial seeking superiority of the combination versus both products in monotherapy. However, no clear statement as to at which stage of the clinical trials the three‐arm design should be applied. The three‐arm trial design may not be required for clinical trials in the early stages. This assessor‐masking design may be acceptable.) Recommendations: Consistent 2. PE GLs: NA Recommendations: NA

Abbreviations: AE, adverse event; aHSCT, autologous hematopoietic stem cell transplantation; ARR, annualized relapse rate; CDA, confirmed disability accumulation; cCDP12, composite 12‐week confirmed disability progression; CHMP, the European Committee for Medicinal Products for Human Use; CTCAE, common terminology criteria for adverse events; CUAL, combined unique active lesions; DBE, double‐blind extension trial; DMT, disease‐modifying therapy; ECG, electrocardiogram; EDSS, Expanded Disability Status Scale of Kurtzke; EMA, the European Medicines Agency; EMA/CHMP GL, “the Guideline on clinical investigation of medical products for the treatment of Multiple Sclerosis (issued by the EMA and CHMP in 2015)”; FDA, U.S. Food & Drug Administration; Gd, gadolinium; GLs, EMA/CHMP GL and “DOSE‐RESPONSE INFORMATION TO SUPPORT DRUG REGISTRATION‐ICH Harmonised Tripartite Guideline‐(issued by CPMP/ICH/378/95 in 1994)”; IDMC, Independent Data Monitoring Committee; i.v., intravenous infusion; MRI, magnetic resonance imaging; MS, multiple sclerosis; NA, not available/not applicable; NEDA, no evidence of disease activity; NEDA‐3, no evidence of disease activity‐3 (MS relapse, new MRI lesions, and increase in EDSS); NEDA‐4, no evidence of disease activity‐4 (MS relapse, new MRI lesions, increase in EDSS, and brain volume loss); OLE, open‐label extension trial; ODRS, overall disability response score; PK/PD, pharmacokinetics/pharmacodynamics; PROBE, prospective randomized open‐label blinded end point; PASAT, paced auditory serial addition task; Recommendations, “Clinical trials of disease‐modifying agents in pediatric MS: Opportunities, challenges, and recommendations from the IPMSSG (Neurology. 92, e2538‐e2548, 2019)”; RMS, relapsing‐multiple sclerosis; RRMS, relapsing–remitting multiple sclerosis, SAE, severe adverse event; s.c., subcutaneous administration; SDMT, Symbol Digit Modalities Test; TMS, transcranial magnetic stimulation; T25FW, timed 25‐foot walk; 9HPT, 9‐hole peg test.

In adult RMS phase III confirmatory trials (n = 32), the most common clinical trial design of the core part was randomized, active‐controlled, double‐blind, parallel‐arms trial (17/32, 53%) which includes double‐dummy trial (13/32, 41%), following randomized, active‐controlled, open‐label, assessor‐masking, parallel‐arms trial (5/32, 16%). Three trials had designs that did not meet the GL ³ : open‐label, non‐masking trial (NCT05242133, 2019–001549‐42, and NCT03342638). Another trial had a design that did not possibly meet the GL ³ : add‐on trial without a test product monotherapy arm (NCT02907177). One trial used the prospective randomized open‐label blinded end‐point (PROBE) design (NCT04121403), which is not specifically recommended by the GL, ³ but this trial met the GL ³ because of an assessor‐masking design. The breakdown of active comparators of phase III confirmatory trials is as follows: first‐line drugs (interferon β‐1a, 5/32, 16%), middle‐class drugs (teriflunomide, 11/32, 34%), strongest‐class drugs (alemtuzumab, cladribine, natalizumab, ocrelizumab, ofatumumab, or rituximab, 11/32, 34%), and none (5/32, 16%).

In adult RMS phase II dose‐finding trials (n = 9), the most common clinical trial design of the core part was randomized, placebo‐ and active‐controlled, double‐blind, parallel‐arms trial (4/9, 44%), following randomized, placebo‐controlled, double‐blind, parallel‐arms trial (2/9, 22%) and randomized, placebo‐controlled, double‐blind, add‐on, parallel‐arms trial (2/9, 22%). One trial had a design that did not possibly meet the GL ³ and the ICH GL ¹⁰ : randomized, placebo‐controlled, double‐blind, 2*8 crossover trial between four dose and placebo arms (NCT03889639). Three trials had designs that possibly met the GL ³ and the ICH GL ¹⁰ : randomized, placebo‐ and active‐controlled, double‐blind, parallel‐arms trial with Bayesian adaptive design (NCT05417269) and add‐on trial without a test product monotherapy arm (NCT03737851 and NCT04079088). The breakdown of active comparators of phase II dose‐finding trials is as follows: first‐line drugs (0%), middle‐class drugs (dimethyl fumarate or teriflunomide, 3/9, 33%), strongest‐class drugs (ocrelizumab, 1/9, 11%), and none (5/9, 56%).

In adult RMS phase III extension trials (n = 10), the most common clinical trial design of the core part was open‐label, single‐arm trial (8/10, 60%), following open‐label, parallel‐arms trial (1/10, 10%) and open‐label, 2*2 crossover trial (1/10, 10%). One trial had a design that possibly met the GL ³ : open‐label, 2*2 crossover trial between natalizumab subcutaneous administration and natalizumab intravenous infusion to evaluate participant preference for subcutaneous versus intravenous route of natalizumab administration (NCT03689972).

In adult RMS phase II extension trials (n = 9), the most common clinical trial design of the core part was open‐label, single‐arm trial (4/9, 44%), following open‐label, parallel‐arms trial (3/9, 33%) and randomized, controlled, double‐blind, parallel‐arms trial (2/9, 22%). There were no trials with designs that were inconsistent with the GL. ³

PEs in adult RMS

A summary of the PEs breakdown of the trials evaluated is shown in Figure S3.

In adult RMS phase III confirmatory trials, the most common PE of the core part was annual relapse rate (ARR) (18/32, 56%), following no evidence of disease activity‐3 (NEDA‐3) (4/32, 13%) and time to first relapse (3/32, 9%). Ten trials had PEs that did not meet the GL ³ : NEDA‐3 (NCT03477500, 2019‐001549‐42, NCT03342638, and NCT03979456), number of new or enlarging T2 lesions (NCT04121403 and NCT03689972), percentage of patients without new or enlarging T2 lesions (NCT04578639 and NCT04688788), MS relapse‐free survival (NCT04047628), and composite 12‐week confirmed disability progression (NCT04544436).

In adult RMS phase II dose‐finding trials, the most common PE of the core part was cumulative number of T1 gadolinium‐enhancing lesions (5/9, 56%), following number of combined unique active lesions (2/9, 22%) and mean overall disability response score (2/9, 22%). Two trials had PEs that did not meet the GL ³ : mean overall disability response score (NCT03737851), number of participants with treatment‐emergent adverse events (AEs), severe AEs (SAEs), and mean overall disability response score (NCT04079088).

In adult RMS phase III extension trials, the most common PE of the core part was number of participants with AEs and SAEs (2/10, 20%) and ARR (2/10, 20%). In phase II extension trials, the most common PE of the core part was number of participants with treatment‐related AEs and SAEs (2/9, 22%) and number of participants with AEs and SAEs (2/9, 22%). The GL ³ does not specifically mention the PE of the extension trial, so no comparison with the GL ³ was made. All extension trial subjects were completers of prior phase III or phase II trials.

Trial designs and PEs in pediatric RMS

Eight pediatric phase III trials and two pediatric phase II trials were identified, all of which were in patients with RRMS. In pediatric phase III confirmatory trials, the most common clinical trial design of the core part was randomized, active‐controlled, double‐blind, parallel‐arms trial (3/8, 38%) which includes two double‐dummy trials (2/8, 25%) and one triple‐dummy trial (1/8, 13%) and randomized, active‐controlled, open‐label, non‐masking, parallel‐arms trial (2/8, 25%). Three trials had designs that did not meet the GL ³ : open‐label, non‐masking design (NCT03958877, 2019–001549‐42, and NCT03368664). Another three trials had designs that did not possibly meet the GL ³ : non‐inferiority trial without a placebo arm with an active comparator, fingolimod (NCT04926818, NCT05123703, and 2020‐004431‐24). The GL ³ calls for the addition of a placebo arm in non‐inferiority trials with first‐line DMTs or other products with a similar efficacy range. One trial (NCT03870763) had designs that did not meet the recommendations ⁴ : a placebo‐controlled trial.

In pediatric phase II trials, one of the two trial designs was an open‐label, non‐masking, parallel‐arms trial as a pharmacokinetics and pharmacodynamics trial. Another design was a randomized, placebo‐controlled, patient‐blind, assessor‐masking, add‐on, parallel‐arms trial. No designs were inconsistent with the GL ³ or the Recommendations. ⁴ The latter trial (NCT04062331) was an add‐on trial with natalizumab maintenance therapy and primary not only for adults but also for children 14 years and older.

In phase III confirmatory trials, the most common PE of the core part was ARR (3/8, 38%), following time to first relapse (2/8, 25%). Three trials had a PE that did not meet the GL ³ : NEDA‐3 (2019–001549‐42), number of new or enlarged T2 lesions (NCT03368664), and proportion of subjects free of new or newly enlarging T2 lesions (2017‐005129‐18). Four trials had a PE that did not meet the Recommendations ⁴ : ARR (NCT04926818, NCT05123703, and NCT03958877) and NEDA‐3 (2019‐001549‐42). Two trials had a PE that did not possibly meet the Recommendations ⁴ : number of new or enlarged T2 lesions (NCT03368664) and proportion of subjects free of new or newly enlarging T2 lesions (2017‐005129‐18).

Chemical names of marketed drugs for the prevention of relapse and/or disability progression in patients with MS

The following 19 chemical names were identified for marketed drugs somewhere in the world for the prevention of relapse and/or disability progression in patients with RMS: metenkefalin/tridecactide, alemtuzumab, cladribine, dimethyl fumarate, diroximel fumarate, monomethyl fumarate, glatiramer acetate, interferon β‐1a, interferon β‐1b, natalizumab, ofatumumab, ozanimod, peginterferon β‐1a, ponesimod, teriflunomide, ublituximab‐xiiy, fingolimod, and ocrelizumab (Table S1). Of these, fingolimod (e.g., in the USA and the EU), dimethyl fumarate (e.g., in the EU), and teriflunomide (e.g., in the EU) also have an approved indication for pediatric MS use, and ocrelizumab also has an approved indication for primary progressive MS (PPMS) as of March 24, 2023 (Table S1). Apart from the 19 drugs listed above, one drug, mitoxantrone, was identified, and the type of MS for which it was indicated was secondary progressive MS (SPMS), progressive RMS, and worsening RRMS (Table S1).

DISCUSSION

This is the first systematic review of trial designs of phase II and phase III trials intended to modify the natural course of the disease in patients with RMS. One of the critical elements in the thrust of RMS clinical trials is the identification of trial design and end points. Regulatory guidelines and expert recommendations are very helpful tools for guiding such identification and should be updated timely based on the latest findings as appropriate to maintain their usefulness. The authors' recommendations on the trial design and PEs are shown in Table 3.

TABLE 3.

The authors' recommendations on clinical trials that intend to modify the natural course of RMS.

Recommendations
1. Phase III confirmatory trial 1‐1. Adult phase III confirmatory trial for a new agent/treatment as a monotherapy in patients with RMS Trial design Randomized, active‐controlled, double‐blind, parallel‐arms trial (double dummy if necessary) is the first option. Randomized, active‐controlled, open‐label, assessor‐masking, parallel‐arms trial is the second option. If a double‐blinding trial design is not feasible, an outcome assessor‐masking trial design is acceptable. The term single blind is commonly used to mean that either the investigator or the patient is blinded; in the area of MS treatment, MRI findings and disease progression are often evaluated by non‐investigator assessors. Active comparator One of the strongest‐class DMTs if a new agent/treatment is expected to be innovative and potent. Primary end point NEDA‐3 if a new agent/treatment is expected to be innovative and potent, otherwise the primary end point is ARR. Time to first relapse is not recommended because the durability of the effect cannot be evaluated. 1‐2. Adult phase III confirmatory trial for a new agent/treatment as an add‐on therapy in patients with RMS Trial design For trials involving patients who do not respond adequately to existing therapy but are at risk of rapid deterioration if the existing therapy is discontinued. Randomized, placebo‐controlled, double‐blind, parallel‐arms trial without a test agent/treatment monotherapy arm (double‐dummy if needed) is the first option. Arm 1: Test agent/treatment + approved specific potent care Arm 2: Placebo to match test agent/treatment + approved specific potent care (placebo arm) If a double‐blinding trial design is not feasible, an outcome assessor‐masking trial design is acceptable. Primary end point NEDA‐3 if a new agent/treatment is expected to be innovative and potent, otherwise the primary end point is ARR. Time to first relapse is not recommended. 1‐3. Pediatric RMS phase III confirmatory trial of a DMT approved in adults with RMS Trial design Randomized, active‐controlled, double‐blind, parallel‐arms trial (double dummy if necessary) is the first option. If a double‐blinding trial design is not feasible, an outcome assessor‐masking trial design is acceptable. Active comparator Fingolimod in a non‐inferiority trial. In a non‐inferiority trial, if the addition of a placebo arm or interferon β‐1a arm as a placebo‐like arm is required by the regulatory authorities, the trial design should demonstrate indirect superiority to interferon β‐1a by comparison with historical interferon β‐1a data from adult and pediatric prior phase III trials, as in NEOS (NCT04926818), to avoid adding a placebo arm or interferon β‐1a arm. Primary end point ARR. Time to first relapse is not recommended. Statistical analysis/approaches It is crucial that the appropriate statistical analysis/approaches should be selected which can detect even a smaller number of cases. For example, the Bayesian approach may integrate adult phase III trial data to interpret pediatric results.
2. Phase II dose‐finding trial 2‐1. Adult phase II dose‐finding trial for a new agent/treatment as a monotherapy in patients with RMS Trial design Randomized, placebo‐controlled, double‐blind, parallel‐arms trial is the first option. If necessary, consider adding a reference arm of a middle‐class or strongest‐class DMT If a double‐blinding trial design is not feasible, an outcome assessor‐masking trial design is acceptable. Primary end point. MRI findings. Specific examples include the following: ‐ The cumulative number of new gadolinium‐enhancing T1 lesions ‐ The number of combined unique active lesions ‐ The number of new or newly enlarging T2 lesions 2‐2. Adult phase II dose‐finding trial for a new agent/treatment an add‐on therapy in patients with RMS Trial design For trials involving patients who do not respond adequately to existing therapy but are at risk of rapid deterioration if the existing therapy is discontinued. Randomized, placebo‐controlled, double‐blind, parallel‐arms trial without a test agent/treatment monotherapy arm (double dummy if needed) as the first option. If a double‐blinding trial design is not feasible, an outcome assessor‐masking trial design is acceptable. Primary end point NEDA‐3 if a new agent/treatment is expected to be innovative and potent, otherwise the primary end point is ARR. Time to first relapse is not recommended.
3. Phase II/III long‐term extension trial design 3‐1. Adult phase III long‐term extension for a new agent/treatment in patients with RMS Trial design Open‐label, non‐masking, pragmatic, multiple‐cohorts trial is the first option if there are no major safety concerns with the test agents/treatments. This trial will be a descriptive, not a comparative one because comparisons of real‐world data may introduce various biases. If a comparative analysis is needed, it should be devised to prevent the analysis from becoming arbitrary. Specifically, the protocol should declare the approach to the analysis in advance. The purpose of the trial To confirm the long‐term usefulness of the new agents/treatment. Subjects Completers of phase III confirmatory trials. New subjects will receive treatment with the test agent/treatment from the beginning of the extension trial, loosening the age limit and allowing patients with typical complications of MS to participate. Reference cohort shall be those receiving an alternative treatment to the test agents/treatments at the same time as the target subjects are receiving the test agents/treatments. Primary end point No specific end points are recommended. Select a primary end point to be evaluated in this trial based on the clinical trial results gathered to date. In principle, one primary end point. Sensitivity analysis Participants from phase III confirmatory trials and those from the extension trial differ in terms of patient background, whether or not they have already been treated with the test agents/treatments, and are not a homogeneous population. To confirm the impact of these differences on the analysis, a stratified analysis should be performed for the group that participated in phase III confirmatory and the group that participated in the extension trial only, respectively, and discuss the differences and impact of these differences on the results of the overall analysis. Patient participation in protocol development Patients are encouraged to participate in the protocol development because the trial is designed to generate evidence of the benefits of new drugs/treatments for a wider range of patients. Their input will enhance the significance of the trial. Second option The trial design is an open‐label, non‐masking, single‐arm, extension trial as usual. No specific end points are recommended. Select primary end points of more interest in this trial based on the clinical trial results gathered to date. In principle, one primary end point. In this case, we recommend that consideration be given to conducting one of the following studies after launch to assess the real‐world effectiveness and safety of the test products/treatments in a broad patient population: observational studies with a reference cohort, post‐authorization safety study (PASS) encompassing at least one effectiveness outcome (e.g., NEDA‐3 or ARR), or prospective non‐interventional study (e.g., NCT03593590). 3‐2. Adult phase II long‐term extension trial for a new agent/treatment in patients with RMS Trial design Open‐label, non‐masking, parallel‐arms, extension trial or open‐label, non‐masking, single‐arm, extension trial. Note: The number of arms will be determined by the results of prior clinical trials. Primary end point No specific end points are recommended. Select primary end points of more interest in this trial based on the clinical trial results gathered to date. In principle, one primary end point.

Abbreviations: ARR, annual relapse rate; DMT, disease‐modifying therapy; EDSS, Expanded Disability Status Scale of Kurtzke; MRI, magnetic resonance imaging; MS, multiple sclerosis; NEDA‐3, no evidence of disease activity‐3 (MS relapse, new MRI lesions, and increase in EDSS); RMS, relapsing multiple sclerosis.

Adult RMS phase III confirmatory trial design

MS is an immune‐mediated disease, and the products and other interventions under development for MS are predominantly those that modulate immune abnormalities, such as DMT, and several other novel approaches, such as hematopoietic stem cell transplantation (HSCT). We believe that RMS trial design should consider the following: (1) Mechanism of action of test agent/treatment, (2) trial objectives, (3) the need for a certain period to demonstrate efficacy and sustainability of effects for a while after discontinuation, (4) safety concerns such as carcinogenesis with long‐term treatment, and (5) the necessity and feasibility of double‐blind trial (DBT). There is no doubt that the first option for the confirmatory trial core design would be a randomized, controlled, double‐blind, parallel‐arm trial recommended by the GL, ³ given that a new agent/treatment directed against immune‐related targets will take time to show efficacy and that the PEs, magnetic resonance imaging (MRI) findings, ARR, and expanded disability status scale of Kurtzke (EDSS) have raised concerns about regression to the mean and placebo effect (Table 1.). ³ , ¹² , ¹³ , ¹⁴ , ¹⁵ The results of this study showed that about 10% of the trial designs were inconsistent with the GL. ³ Below, we discuss examples that may not fit the GL ³ found and the components of the trial design.

Non‐masking design

In the field of MS, it is often the case that the investigator is not the same person who evaluates the MRI findings or the progression of the disease. If it is difficult to conduct a double‐blind comparative study, then a common single‐blind design (blinded to the investigator or patient) is not acceptable, and an open‐label, assessor‐masking design or single‐blind (investigator‐ or patient‐blind), assessor‐masking design is acceptable. As with the GL, ³ we recommend that non‐masking designs for confirmatory trials be not acceptable because they introduce assessment bias. Specifically, the two cases are as follows; one trial (NCT05242133) was an open‐label, non‐masking trial comparing pegylated interferon β‐1a with interferon β‐1a. Another trial (2019‐001549‐42) was an open‐label, non‐masking trial comparing HSCT to DMTs (alemtuzumab or ocrelizumab), which is understandably difficult to double blind because of the operation involved. However, both examples seem to require an assessor‐masking design to prevent evaluation bias.

PROBE design

One trial (NCT04121403) was a PROBE, assessor‐masking trial of rituximab versus cladribine. The PROBE design was often used in the field of the cardiovascular area because there are many candidate outcomes to be evaluated and it was thought difficult to predict which one would be selected. ¹⁶ , ¹⁷ In several cases, it was reported as if the PROBE design had not worked well, and the scandal of paper withdrawal continued in the area of hypertension such as the Jikei Heart study and the Kyoto Heart study. ¹⁸ , ¹⁹ However, we believe that the failures of the scientific evaluation were not due to the PROBE design, but to the non‐masking design. In the case of NCT04121403, the scientific evaluation would be possible because blinded radiologists assess the primary end points and do not know the treatment allocation. One concern is that the PROBE design masks the end points, but not completely, since the end points of the trial are disclosed on ClinicalTrials.gov.

Add‐on trial design

One trial (NCT02907177) was a placebo‐controlled, double‐blind, add‐on, parallel‐arms trial comparing ponesimod plus dimethyl fumarate to placebo plus dimethyl fumarate. For add‐on therapy, the GL ³ states a useful design is a three‐arm trial‐seeking superiority of the combination versus both products in monotherapy. Although the rationale for the recommendation is reasonable, there may be few examples that can be adapted in practice. Most of the existing therapies for add‐on therapy would be DMT, and rapid deterioration is assumed if DMT is discontinued. Therefore, the said three‐arm trial design would not be applicable in most cases of add‐on trials. We recommend that add‐on trials for patients with inadequate response to existing therapy but who are at risk of rapid deterioration if the existing therapy is discontinued should be designed as a two‐arm comparison with a placebo arm that is a maintained existing therapy. We do not recommend a 3‐arm trial seeking superiority of the combination versus both products in monotherapy.

Placebo arm

Overall, there were five placebo‐controlled trials. The first (NCT02744222) was a mixed trial with a dose‐finding trial and a confirmatory trial, so the inclusion of a placebo arm was unavoidable. The second (NCT02907177) was an add‐on trial and the placebo arm could continue the existing treatment, so it was acceptable as discussed above. The third (NCT04121221) may be understandable because it was a long‐acting product of a first‐line DMT and there was no appropriate DMT to be compared. The remaining two trials (NCT05134441 and NCT05201638) were unclear as to why trials were designed as placebo comparisons. The GL ³ recommends a superiority trial versus placebo or an active comparator (i.e., first‐line DMTs like β‐interferons, glatiramer acetate) as the first option following a non‐inferiority trial as the second option. However, given the current availability of nearly 20 DMTs (Table S1), with established long‐term efficacy and safety, comparisons to placebo should not be made in principle from an ethical perspective. We recommend that no comparison with a placebo arm be made in confirmatory trials in principle, except for add‐on trials and mixed trials consisting of both dose‐finding trial and confirmatory trial.

Active comparator

The GL recommends first‐line DMTs as an active comparator. However, as noted earlier, given the number of available DMTs, new agents/treatments should not in principle be compared to first‐line DMTs. Acceptable comparisons with first‐line drugs should be limited to biosimilars of first‐line DMTs, long‐acting formulations of first‐line DMTs, and products with slightly modified formulas of first‐line DMTs aimed at improving safety. We recommend that for a potent new agent/treatment, confirmatory trials should compare them with one of the strongest class DMTs in principle.

Based on the above, we recommend the following trial designs for adult RMS phase III confirmatory trials of potent new agents/treatment as a monotherapy: randomized, active‐controlled, double‐blind, parallel‐arms trial (double dummy if needed) as the first option. We recommend the following trial designs for adult RMS phase III confirmatory trials of potent new agents/treatment as a combination therapy: randomized, placebo‐controlled, double‐blind, parallel‐arms trial without a test agent/treatment monotherapy arm (double dummy if needed) as the first option.

Adult RMS phase II dose‐finding trial design

The ICH GL ¹⁰ recommends the randomized blinding parallel‐arms trial design (Table 1), which we believe is an acceptable core design part without debate. The results of this study showed that two‐third of the designs met the GL ³ and the ICH‐GL. ¹⁰ Below, we discuss examples that may not fit the GL ³ and the ICH‐GL ¹⁰ found and the components of the trial design.

Adoptive design

One (NCT05417269) was a randomized, placebo‐ and active‐controlled, double‐blind, parallel‐arms trial that had a unique trial design that consisted of a sequential three‐part, phase I trial with dose escalation, phase IIa trial with Bayesian adaptive design which may identify the least‐promising dose and drop it in phase IIb, and phase IIb trial. Regulatory agencies may have specific requirements for adaptive trials. Adaptive design has benefited from application in rare diseases because of its ability to optimize the number of cases. ²⁰ , ²¹ On the other hand, there is a challenge from a practical standpoint because it requires a system capable of rapid data collection and analysis for interim analysis, and a timely evaluation by the trial Independent Data Monitoring Committee. At this time, we do not actively recommend the adoption of adaptive design, but it is worth considering its application when optimization of the number of cases is necessary.

Add‐on trial design

Two trials (NCT03737851 and NCT04079088) were add‐on trials without a test agent/treatment monotherapy arm. For add‐on therapy, our recommendations are the same as those described in the section on adult RMS III confirmatory trial design in adult RMS.

Crossover trial design

One trial (NCT03889639) was a randomized, placebo‐controlled, double‐blind, 2*8 crossover trial between four dose and placebo arms. If the test agent/treatment effect develops rapidly and patients return to baseline conditions quickly after cessation of therapy, a randomized multiple crossover trial would be acceptable. If there is concern about some delay in the onset of effect or carry‐over effect, no matter how much crossover with placebo, a double‐blinded parallel‐arms design would be better to assess the dose–response.

Placebo arm

As the ICH‐GL ¹⁰ states that no placebo arm is acceptable if the effect slope is positive, but that the inclusion of a placebo arm is desirable because it can measure the absolute size of the drug effect and can partially salvage a trial that showed no dose‐response slope if all dose were too high (Table 1). The T1 gadolinium‐enhancing lesions most used as PE, which has a reported regression to the mean, ¹² and if it is used as PE, inclusion of a placebo arm would help achieve the trial objective. In this study, all dose‐finding trials included a placebo arm. We recommend including a placebo arm in dose‐finding trials.

Active comparator

The ICH‐GL ¹⁰ also discusses the benefits of including an active comparator arm in dose‐finding trials (Table 1). Although the inclusion of an active comparator arm is not mandatory, the results of an active comparator arm would be helpful when considering development strategies. In this study, approximately 30% of the active comparators were in the middle class and 10% were in the strongest class. Uniquely, two of the trials used dimethyl fumarate in an open‐label setting, which could avoid the implementation burden of double dummy, and would be easy to apply to patients because it is an oral drug with a relatively safe profile. It would be a good option to obtain reference data from an active comparator arm to estimate the magnitude of the effect compared to an existing DMT at the dose‐finding trial stage.

Based on the above, we recommend the following trial designs for adult RMS phase II dose‐finding trials of potent new agents/treatment as a monotherapy: randomized, placebo‐controlled, double‐blind, parallel‐arms trial as the first option. If necessary, consider adding a reference arm of a middle‐class or strongest‐class DMT. We recommend the following trial designs for adult RMS phase II dose‐finding trials of potent new agents/treatment as a combination therapy: randomized, placebo‐controlled, double‐blind, parallel‐arms trial without a test agent/treatment monotherapy arm (double dummy if needed) as the first option.

Adult RMS phase III extension trial design

The GL ³ states an extended follow‐up, either blinded or open label, should be performed to address the maintenance of the effect and to gather information on the long‐term course of patients under treatment (Table 1). This trial objective is reasonable because the majority of products tested in RMS are considered to act as immunomodulators and are used for long periods and there are potential concerns about safety risks such as malignancies and opportunistic infections. ³ The results of this study showed that 90% of the designs met the GL. ³ The GL ³ states that for the treatment of chronic diseases, the availability of safety data for a minimum of 2 years in a sufficiently meaningful number of patients at the time of launch is expected. All of the phase III extension trials met this requirement by evaluating safety for 2 years and over. Below, we discuss examples that may not fit the GL ³ found, components of the trial design, another perspective, and our recommendation.

Crossover trial design

One trial (NCT03889639) was an open‐label, 2*2 crossover trial between natalizumab subcutaneous administration and natalizumab intravenous infusion in which PE was the percentage of participants indicating a preference for natalizumab subcutaneous administration. Given the PE, it seems acceptable.

Masking

In most cases, the trial will be a single‐arm one, so there is very little need for blinding and masking. Therefore, an open‐labeled, non‐masking single‐arm extension trial is applicable and would be a standard design.

Another perspective

Experts recommend shifting phase III clinical trials of new therapies in MS along the continuum from explanatory to more pragmatic trials relaxing age restrictions and comorbidity ¹¹ (Table 1). They also state such decisions are a tradeoff between the homogeneity of the population and the strength of the signal being sought for the safety and generalizability of the findings to clinical practice ¹¹ (Table 1). We agree with the experts' suggestion. Clinical evidence on patients who do not meet phase III criteria will be needed in clinical settings after launch, but it is not clear who will generate the evidence for such patients and when. Current phase III extension trials cannot enroll such patients. In addition, there is no comparison arm or reference cohort except test agents/treatments. When an event occurs, it is impossible to evaluate whether it is due to test agents/treatments or not.

Our recommendations

If there are no major safety concerns with the test agents/treatments, the phase III extension trial should be an open‐label, non‐masking, pragmatic, multiple‐cohorts trial. The target subjects shall be those who have completed phase III confirmatory trials and new subjects receiving treatment with the test agents/treatments, loosening the age limit and allowing patients with typical complications of MS to participate. Also, a reference cohort shall be those receiving an alternative treatment to the test agents/treatments at the same time as the target subjects are receiving the test agents/treatments. The purpose of the trial is to confirm the long‐term usefulness of the new agents/treatment, and the alternative treatment cohort should serve as reference data. This trial will be descriptive, not a comparative one, because comparisons of real‐world data may introduce various biases. If a comparative analysis is needed, it should be devised to prevent the analysis from becoming arbitrary. Specifically, the protocol should declare the approach to the analysis in advance. As an example, a procedure for checking the distribution of collected data, confirming comparability, predeclaring candidate analysis methods, and, if comparability is confirmed, selecting the analysis method to be adopted (e.g., inverse probability of treatment‐weighting analysis to effectively utilize a small number of cases) from among the analysis candidates. No specific end points are recommended as PE. Select PEs of more interest in this trial based on the clinical trial results gathered to date. Patients are encouraged to participate in the protocol development because the trial is designed to generate evidence of the benefits of new drugs/treatments for a wider range of patients. Their input will enhance the significance of the trial.

Regarding the limits of this approach, participants from phase III confirmatory trials and those from extension trials differ in terms of patient background, whether or not they have already been treated with the test agents/treatments, and are not a homogeneous population. To confirm the impact of these differences on the analysis, a stratified analysis should be performed for the group that participated in phase III confirmatory and the group that participated in the extension trial only, respectively, and discuss the differences and impact of these differences on the results of the overall analysis. The extent to which selection criteria should be loosened should also be carefully discussed and decided based on the results of previous clinical trials.

If this recommendation is not feasible or acceptable, we recommend the following trial designs for adult RMS phase III extension trials of potent new agents/treatment as a monotherapy or a combination therapy: open‐label, non‐masking, single‐arm, extension trial as the first option. No specific end points are recommended as PE. Select a PE to be evaluated in this trial based on the clinical trial results gathered to date.

In this case, we recommend that consideration be given to conducting one of the following studies after launch to assess the real‐world effectiveness and safety of the test products/treatments in a broad patient population: observational studies with a reference cohort, post‐authorization safety study (PASS) encompassing at least one effectiveness outcome (e.g., NEDA‐3 or ARR), or prospective non‐interventional study (e.g., NCT03593590).

Adult RMS phase II extension trial design

The results of this study showed that 100% of the designs met the GL. ³ All trials were extensions from randomized, placebo‐controlled, double‐blind trials. Below, we discuss the components of the trial design.

Arms

In the present study, 44% of the trials were in a single‐arm trial and 55% in multiple‐arm trials. Differences in the number of dose arms were likely determined based on the results of the dose‐finding trial and likely dropped the least effective dose arm.

Double‐blind

In general, open‐label, non‐masking trial designs can achieve the objective of confirming the maintenance of efficacy and safety. In this study, there were two double‐blind, parallel, three‐arm trial designs (about 20%). It is unknown how such a design was chosen. The dose‐setting study may not have been able to determine the optimal dose, and such cases would be rare.

Based on the above, we recommend the following trial designs for adult RMS phase II extension trials of potent new agents/treatment as a monotherapy or a combination therapy: open‐label, non‐masking, parallel‐arms, extension trial or open‐label, non‐masking, single‐arm, extension trial.

Note: The number of arms will be determined by the results of prior clinical trials.

PEs of adult RMS phase III confirmatory trials

PEs in confirmatory trials should be true end points rather than surrogate end points if they are feasible; in the case of RMS, true end points would be related to relapses and/or disabilities, but the latter would require a longer evaluation period because of the slower progression and this may have the disadvantage of delaying deliver of the new drug/treatment to the patients. The GL ³ states that ARR or the time to relapse (means time to first relapse) is considered acceptable as PE and MRI measurements are not acceptable because it is not a validated surrogate end point (Table 1.). The results of this study showed that about one‐third of the PEs were inconsistent with the GL. ³ The following is a sequential discussion of the PEs recommended by the GL ³ and PEs that are not consistent with the GL. ³

ARR

This is the most common PE in this study and has been used in many previous confirmatory trials. ²² , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ³² , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ The pros and cons of ARR and NEDA‐3 are summarized in Table 4. ARR is the most‐established PE. However, as more potent therapies are developed, PE closer to complete remission or free of disease is needed instead of ARR. We recommend that ARR be PE of adult RMS phase III confirmatory trials for new agents/treatments not expected to have a potent effect, generics/biosimilars, long‐acting products of approved drugs, and agents that involve a partial structural change in an approved drug to improve the safety in clinical settings.

TABLE 4.

Pros and cons of annual relapse rate and no evidence of disease activity‐3 as a primary end point.

End point	Pros and cons with specific explanation	Reference #
ARR	Pros 1. One of the most established and accepted clinical end points with extensive historical data Relapses are one of the clinically meaningful events. ARR is one of the established measures of relapse. ARR can better evaluate the drug's efficacy because it also evaluates the second and subsequent relapses, as compared to the evaluation of time to first relapse. In other words, ARR can assess the maintenance of the effect of relapse control, unlike time to first relapse. Most of the marketed drugs for RMS have made the ARR a PE in phase III confirmatory trials. These rich historical data could potentially be available in the planning and design of both adult and pediatric clinical trials. From the results of this study, ARR was also the most common PE in both adult and pediatric phase III confirmatory trials. There is an innovative pediatric MS phase III trial (NCT04926818/2020–002700‐39) that has been agreed upon with both the FDA and the CHMP with a study design that uses historical data on ARR with appropriate analysis methods, avoids setting up a placebo arm in a non‐inferiority trial, optimizes the number of patients needed, and uses ARR as the primary end point. The confirmatory trial of fingolimod, which has been approved for pediatric MS patients in both the USA and Europe, had ARR as its primary end point. 2. Easy to measure: ARR is an easily measurable end point. It involves counting the number of relapses a patient experiences over the course of a year. Cons 1. Not a highly reproducible outcome measure Systematic reviews suggest ARRs decrease in placebo groups in RCTs in RMS over calendar time as well as throughout the study. An apparent efficacy could be explained by the regression to the mean, a placebo effect, as well as by the natural course of the disease. 2. Not a holistic outcome measure As ARR does not correlate perfectly with MRI findings and the progression of disability, it is not a substitute for those assessments. A multiple sclerosis‐related disability could occur not only as relapse‐associating worsening but also as a progression independent of relapse activity.	22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32 49 51 13, 14, 60, 61 3, 13, 14 3, 42, 43 62
NEDA‐3	Pros Holistic and stringent outcome measure NEDA‐3 incorporates three key end points such as the absence of relapses, no sustained EDSS score progression, and no new or enlarging T2 or T1 gadolinium‐enhancing lesions, providing a more holistic assessment of disease activity rather than assessing each in isolation. Each of the three components is a well‐established multiple sclerosis treatment assessment end point. NEDA‐3 is a kind of treatment goal, not a total score like the usual composite index. Although regression to the mean and placebo effects have been reported for ARR, EDSS, and MRI findings, respectively, the NEDA‐3 evaluates the absence of each and thus may be an ideal PE with fewer of those effects. NEDA‐3 has greater clinical relevance than the ARR because it includes an assessment of pre‐ and post‐relapse stages. NEDA‐3 could be one of the most stringent outcome measures and suggests complete remission of the disease. Patient‐centric goal NEDA‐3 aligns with the treatment goal of not just reducing relapses but also preventing new lesions on MRI and further disability progression, which could be an ideal goal for patients with RMS. Not too complicated end point Various factors are involved in the progress of disease disability. If eight factors (clinical relapse, EDSS, new MRI lesions, brain atrophy, cognitive impairment, cerebral spinal fluid neurofilament level, patient‐related outcome, and oligoclonal bands) are taken into account, NEDA‐8 can be developed, but it would be too complicated to implement it as an index. NEDA3 is the ideal index at this stage. Cons Risk of overemphasis on inflammatory aspects of multiple sclerosis Controversial whether the detection of a single new magnetic resonance imaging lesion is considered as significant as the clinical relapse. NEDA‐3 may be weighted toward inflammatory activity and heavily rely on MRI outcomes. A recent systematic review and meta‐analysis reporting the difference in the association of NEDA‐3 and NEDA‐4 with no disability progression observed no differences between NEDA‐3 and NEDA‐4. Neurodegenerative aspects such as brain volume loss are not evaluated. Brain volume loss has been shown to independently predict long‐term disability progression in RRM. Emerging evidence suggests progression independent of relapse activity may be the most important contributor to disability progression, more so than relapse‐associated worsening and the presence of new focal inflammatory lesions. Unknown predictability of longer‐term clinical stability Due to the difficulty of NEDA predicting sustained remission or complete recovery of disease, it has been proposed to assess the biological activity of the disease by measuring the levels of biomarkers in the CSF. The burden of implementation Measurement of NEDA‐3 is feasible in clinical trials, but often difficult to perform in routine clinical practice.	12, 13, 14, 15 63, 64 65 65, 66, 67, 68, 69 3, 70 71

Abbreviations: ARR, annual relapse rate; CHMP, the European Committee for Medicinal Products for Human Use; EDSS, Expanded Disability Status Scale of Kurtzke; FDA, U.S. Food & Drug Administration; MRI, magnetic resonance imaging; NEDA‐3, no evidence of disease activity‐3 (MS relapse, new MRI lesions, and increase in EDSS); NEDA‐4, no evidence of disease activity‐4 (MS relapse, new MRI lesions, increase in EDSS, and brain volume loss); RCT, randomized controlled trial; RMS, relapsing multiple sclerosis; RRMS, relapsing–remitting multiple sclerosis.

Time to first relapse

Although it is reported to be more sensitive than ARR and to reduce the number of cases to be evaluated, ⁴¹ only 9% of trials in this study had it as PE. The reason for the low utilization may be that time to first relapse evaluates the first relapse only, whereas ARR evaluates the first relapse and the second or more relapses and is superior because it can assess the durability of the effect. We recommend that time to first relapse not be PE.

NEDA‐3

Four trials had NEDA‐3 as PE in this study. Three of them were trials of HSCT. Another was a trial of rituximab 500 mg every 6 months versus rituximab 500 mg every 6 months. At this time, NEDA‐3 is being used, for the most part, in trials of a breakthrough treatment, HSCT, and not even for other test agents. NEDA‐3 incorporates three key end points such as the absence of relapses, no sustained EDSS score progression, and no new or enlarging T2 or T1 gadolinium‐enhancing lesions, providing a more holistic assessment of disease activity rather than assessing each in isolation. If MRI findings are viewed broadly as a reserve for relapse ⁴² , ⁴³ and physical disability as an accumulation due to relapse and other factors, NEDA‐3 has greater clinical relevance than the ARR because it includes an assessment of pre‐ and post‐relapse stages. Although regression to the mean and placebo effects have been reported for MRI findings, ARR, and EDSS, respectively, ¹² , ¹³ , ¹⁴ , ¹⁵ NEDA‐3 evaluates the absence of each and thus may be an ideal PE with fewer of those effects. We recommend that NEDA‐3 be PE of adult RMS phase III confirmatory trials for new agents/treatments expected to be innovative and potent.

MRI findings

MRI findings were PE in four trials in this study: ocrelizumab versus rituximab (two non‐inferiority trials), cladribine versus rituximab (non‐inferiority trial), and natalizumab, i.v. 4‐week standard interval dosing versus natalizumab, and i.v. 6‐week extended interval dosing. It is unknown why MRI findings in those trials were PE and their timeframes were longer than 1 year (72 weeks to 2 years). We have no objection to the GL ³ that MRI findings are not acceptable as PE of confirmatory trials.

MS relapse‐free survival

MS relapse‐free survival was PE in one trial of HSCT in this study. NEDA‐3 is a more comprehensive end point because it also includes no relapse as a component of the evaluation. We do not recommend MS relapse‐free survival as PE.

Composite 12‐week confirmed disability progression

It was defined by an increase in 1 or more measures (EDSS, timed 25‐foot walk, and 9‐hole peg test) and can capture overall (EDSS), upper (9‐hole peg test), and lower (timed 25‐ft walk) extremity function. Although it is an end point to evaluate disability progression from multiple angles, we do not recommend this as a PE for a new agent/treatment because the ideal PE can evaluate the stage before disability progression at the same time. However, it would be useful as a PE when exploring the efficacy against disability progression of a drug that has already been proven on relapses and approved.

PEs of adult RMS phase II dose‐finding trials

Since phase II dose‐finding trials should ideally be conducted with a relatively small number of subjects and over a relatively short period, which requires PEs with a sensitive surrogate end point; in the case of RMS, MRI lesions are PE candidates because they have been reported to correlate with relapse. ⁴² , ⁴³ The GL ³ states that MRI‐derived parameters as the main end point for assessing dose finding are acceptable. We have no objections to this and agree with it. Specifically, the following representative end points are available: cumulative number of T1 gadolinium‐enhancing lesions, number of combined unique active lesions, and number of new or newly enlarging T2 lesions. The results of this study showed that three‐fourth of PEs met the GL. ³ PEs found from the results of this study are discussed in turn.

Cumulative number of T1 gadolinium‐enhancing lesions

This was the most common PE that met the GL ³ in this study. The timeframe for evaluation was 12 or 24 weeks, which was relatively shorter than the evaluation period for ARR.

Number of combined unique active lesions

This was the second common PE that met the GL ³ in this study. The timeframe for evaluation was 24 or 36 weeks, which was relatively shorter than the evaluation period for ARR.

Mean overall disability response score

This end point was not consistent with the GL. ³ It was a composite score derived from four components (EDSS, timed 25‐foot walk, 9‐hole peg test in the dominant hand, and 9‐hole peg test in the non‐dominant hand). It was PE of an add‐on trial (added to the standard care). Elezanumab, the test agent, may have a strategy to find the optimal dose by its effect on disability progression, even though it takes more time than MRI findings (in this case, the timeframe was 52 weeks).

PEs of adult RMS phase II and phase III extension trials

The GL ³ presents the purpose of the extension trials but makes no specific mention of PE. The most common PEs were safety‐related PEs in both phase II and phase III extension trials. The number of PEs in one trial varied from 0 to 27. We do not suggest that any particular end point be PE but recommend that the end points of most interest during long‐term use be PE based on the information obtained up to that point.

Pediatric RMS phase III confirmatory trial design

We believe that pediatric RMS trial design should consider the following: (1) Very low prevalence ranging from 0.69 to 26.92 per 100,000 people, ⁴⁴ (2) high relapse rate, ⁴⁵ , ⁴⁶ (3) cognitive impairment at younger age, ⁴⁷ , ⁴⁸ (4) only one or three DMTs approved for pediatric applications (Table S1), and (5) various considerations for being non‐adult patients.

The GL ³ recommends randomized double‐blind (double dummy if needed) controlled parallel group trials and the Recommendations ⁴ state placebo‐controlled trials of immunomodulatory agents proven in adult MS are inappropriate in pediatric MS. We have no objections to both recommendations and agree with them. Trial designs that did not or may not meet the GL ³ and/or the Recommendation ⁴ are discussed below:

Non‐masking

There were three applicable trials, two of which were active‐controlled ones. The remaining one was designed to evaluate efficacy in subjects who switched from another DMT treatment. As there is a risk of bias in the evaluation or it would be unclear whether the effect observed in the trial was due to the test agent or to the natural course of the disease, the randomized, controlled, double‐blind, parallel‐arms trial design or at least the randomized controlled assessor‐masking parallel‐arms trial is required.

Placebo

In this study, one trial (NCT03870763) was placebo controlled but terminated due to long‐term difficulties in fulfilling the enrolment commitments and changes in the pediatric MS landscape which no longer support placebo‐controlled trials by the people involved. This case illustrates the reality that placebo‐controlled phase III trials in pediatric MS are not acceptable. Considering the early disease progression specific to pediatric MS (e.g., high relapse frequency ⁴⁵ , ⁴⁶ and early cognitive impairment ⁴⁷ , ⁴⁸ ), a phase III confirmatory trial design should be without a placebo arm because of the potential risk of placebo treatment outcome on long‐term prognosis in pediatric MS patients.

Active comparator

The Recommendations ⁴ recommend fingolimod or other drugs used in pediatric MS. We recommend fingolimod as the active comparator for pediatric RMS phase III confirmatory trials of DMTs approved in adults. The reason is that it is the only DMT approved for pediatric indications in both the USA and the EU, and it is easier for pediatric patients to take because it is an oral drug.

Superiority trial/non‐inferiorly trial

Given the small number of DMTs that are indicated for pediatric MS, the priority now is to increase treatment options (i.e., other DMTs). Therefore, we recommend that until there are enough DMTs available for pediatric MS patients, a non‐inferiority trial design can be chosen for phase III confirmatory trials in pediatric RMS of DMTs approved in adults. The GL ³ calls for the addition of a placebo arm in non‐inferiority trials with first‐line DMTs or other products with a similar efficacy range. It is unclear whether this applies to fingolimod in the case of trials in pediatric MS patients, but given a few DMTs approved for pediatric MS use, a placebo arm would be requested by regulatory agencies. As a way to avoid setting up a placebo arm, one of the three non‐inferiority trials found in this study provided helpful information. That is NEOS (NCT04926818), Graves JS, et al. reported that the NEOS trial design had been discussed and agreed upon with the U.S. Food & Drug Administration (FDA) and the CHMP. ⁴⁹ It is designed to demonstrate non‐inferiority to fingolimod and, instead of not having a placebo arm, to indirectly demonstrate superiority over interferon β‐1a by comparing historical interferon β‐1a data from adult and pediatric prior phase III trials. This would be a practical solution to avoid setting up a placebo arm or interferon β‐1a arm.

Statistical analysis/approaches

When examining trial designs for pediatric patients (e.g., comparison with historical cohorts' data), it is not possible to reach as large a sample size as is possible in adult trials. It is crucial that the appropriate statistical analysis/approaches can detect even if a smaller number of cases should be selected, which would be different for adult trials. Whenever trials of the same agent have been conducted in both adults and children, respectively, a Bayesian approach may allow us to integrate results from trials in adults in the interpretation of pediatric trials. For example, in the TERIKIDS trial, the initial standard frequentist analysis showed no significant difference, but when the Bayesian approach was applied, a significant difference was detected. ⁵⁰

Based on the above, we recommend the following trial designs for pediatric RMS phase III confirmatory trials of DMTs approved in adults: randomized, active‐controlled, double‐blind, parallel‐arms trial (non‐inferiority trial, double dummy if needed) as the first option. If a placebo arm is required by regulatory authorities, we recommend a design that does not include a placebo arm but instead indirectly compares interferon β‐1a with historical data.

PEs of pediatric RMS phase III confirmatory trials

The GL ³ recommends either ARR or time to first relapse, and the Recommendations ⁴ recommend time to first relapse to switch to more effective therapies after the first relapse. In this study, ARR was the most common PE (38%) and time to first relapse was a minor PE (13%). Time to first relapse cannot assess the durability of the efficacy, but ARR can. In the first place, if we are comparing two therapies, we will not know which therapy is better during the trial until the results are available, so switching patients who experience a relapse to a better treatment will not be possible while a randomized controlled double‐blind trial is being conducted. We recommend that the ARR be PE when conducting pediatric RMS phase III confirmatory trials of DMT approved in adult MS. The reasons are as follows: ARR can assess the durability of the efficacy, many DMTs have been evaluated and approved with ARR data as PE in adults, ²² , ²³ , ²⁴ , ²⁵ , ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ³² there is abundant historical adult ARR data, fingolimod has been approved for pediatric MS use with ARR data as PE, ⁵¹ NEOS is underway with ARR as PE, and NEOS trial design was approved by the FDA and the CHMP. ⁴⁹

Trial designs and PEs of phase II trials in pediatric RRMS

Two trials were found in this study: one is an open‐label, non‐masking, parallel‐arms trial as a pharmacokinetics and pharmacodynamics trial and another is a placebo‐controlled multi‐dose arm trial. Both trials are exploratory and it seemed the trial designs and PEs were selected for their intended purposes. The sample size in this study was small and we do not discuss more.

Regulatory differences in the EU, USA, and Japan

In the EU, they have the GL ³ for MS clinical trials, which no doubt is a useful guideline for sponsors in setting trial designs and PEs, not only for new drug applications to the EU but also for applications in other countries. In the USA, the FDA has published various guidance related to clinical trials, but there is none on MS clinical trials. Differences are recognized between drugs approved for pediatric indications in the EU and the USA, but clarifying on what basis the differences are occurring would allow for better development methodology discussions.

In Japan, MS is a rare disease and under the regulatory support system for designated new developing drugs for rare diseases, sponsors can consult with the regulatory authorities in advance on individual trial design and key end points. In principle, new drugs must be validated in the Japanese population by two or more randomized controlled trials (RCTs), ⁵² but the Pharmaceutical and Medical Devices Agency (PMDA) is flexible on trial designs and PEs, depending on scientific considerations, feasibility, availability of treatment options, and other ethical considerations. As specific examples, seven available DMTs for RRMS or RMS patients in Japan, except interferon β‐1b, which was approved first, the confirmatory trials in which the Japanese participated were mainly randomized placebo‐controlled DBTs with a PE of MRI findings and with around 24‐week double‐blind term, not global pivotal trials with a PE of clinical end points. ⁵³ , ⁵⁴ , ⁵⁵ , ⁵⁶ , ⁵⁷ , ⁵⁸ These Japanese results, along with results from larger, longer‐term, international phase III confirmatory studies with clinical end points, were reviewed and they were approved. The primary reasons for the regulatory exemption from conducting long‐term confirmatory trials with clinical PEs for Japanese patients were as follows; considering Japan's universal health insurance system and the availability of existing treatments, long‐term administration of a placebo for around 2 years was ethically unacceptable and extremely difficult to obtain consent from patients.

If more international phase III confirmatory trials comparing an available active comparator without a placebo arm are conducted in the future, Japanese patients will have more opportunities to participate in them, which is expected to help eliminate drug loss and drug lag in Japan.

CONCLUSIONS

The results of this system review confirmed consistency with the GL, ³ the Recommendations, ⁴ and the ICH‐GL ¹⁰ in many cases. We made several recommendations based on the insights obtained. Some of them differ from the GL ³ and the Recommendations. ⁴ Here are some highlights from our recommendations:

1. Adult RMS phase III confirmatory trials for a new agent/treatment

Randomized, active‐controlled, double‐blind, parallel‐arms trial (double dummy if necessary) is the first option for the trial design. An active comparator is one of the strongest‐class DMTs if a new agent/treatment is expected to be innovative and potent. PE is NEDA‐3 if a new agent/treatment is expected to be innovative and potent, otherwise, PE is ARR.

2. Pediatric RMS phase III confirmatory trials of a DMT approved in adults MS

Randomized, active‐controlled, double‐blind, parallel‐arms trial (double dummy if necessary) is the first option for the trial design. An active comparator is fingolimod in a non‐inferiority trial. In a non‐inferiority trial, if the addition of a placebo arm or interferon β‐1a arm as a placebo‐like arm is required by the regulatory authorities, the trial design should demonstrate indirect superiority to interferon β‐1a by comparison with historical interferon β‐1a data from adult and pediatric prior phase III trials, as in NEOS (NCT04926818), to avoid adding a placebo arm or interferon β‐1a arm. PE is ARR. It is key that the appropriate statistical analysis/approaches should be selected which can detect even a smaller number of cases. For example, the Bayesian approach may integrate adult phase III trial data to interpret pediatric results.

STUDY LIMITATION

This study was conducted using clinical trial registry information and is therefore informationally limited. Information that is not registered may contain information necessary for consideration and may hinder sufficient discussion. This study did not take into account clinical pharmacology factors specific to the investigational drugs/treatments. Even if their study designs are not aligned with what the guidelines recommended, some of them may guarantee a reliable method of evaluation. However, we believe that such cases are exceptional and to be discussed one by one and that this study covers most of the main things to be discussed.

AUTHOR CONTRIBUTIONS

K.H. wrote the manuscript; K.H. and H.M. designed the research; K.H. and H.M. performed the research; and KH analyzed the data.

FUNDING INFORMATION

No funding was received for this work.

CONFLICT OF INTEREST STATEMENT

KH declared that he was an employee of Amgen and owned stock in Amgen at the time this study was planned and prepared. HM declared no competing interests in this work.

ETHICS STATEMENT

Our study did not require an Institutional Review Board approval or an Institutional Animal Care and Use Committee approval because it is a systematic review of trial registration information identified through a clinical trial registry search.

Supporting information

Figure S1.

Figure S2.

Figure S3.

Table S1.

Table S2.

Hiramatsu K, Maeda H. Adult and pediatric relapsing multiple sclerosis phase II and phase III trial design and their primary end points: A systematic review. Clin Transl Sci. 2024;17:e13794. doi: 10.1111/cts.13794