REFINE-MS: A randomised, rater-blind, phase 3 trial to determine the optimal dose frequency of ocrelizumab for patients with relapsing remitting multiple sclerosis using a novel multi-arm multi-stage frequency-optimisation design.

RESEARCH QUESTION: Can the dosing frequency of ocrelizumab for patients with stable relapsing remitting MS be reduced, without compromising efficacy? BACKGROUND: Ocrelizumab is an anti-CD20 monoclonal antibody that depletes circulating B-cells, thereby reducing disease activity in multiple sclerosis (MS). Almost 9,000 patients with MS are currently on ocrelizumab in England alone. Treatment continues with 6-monthly infusions or subcutaneous injections until secondary immunodeficiency, new disease activity, or advanced disability, but, as there are no better treatment options, patients can be treated indefinitely. This has raised concerns, as ocrelizumab is associated with risks of serious side effects, including severe and recurrent infections. Observational studies during the COVID-19 pandemic demonstrated no association between B-cell repopulation and disease activity and reducing dose frequency did not promote disease reactivation, compared with 6-monthly interval dosing. OBJECTIVES: (1) To determine the optimal ocrelizumab dosing frequency, defined as the least frequent dose that is non-inferior to the standard 6-monthly regime, in terms of the proportion of patients without new T2 lesions on brain MRI. (2) To establish whether reduced dose frequencies minimize serious side effects, mitigate healthcare costs and improve patient quality-of-life and convenience, whilst maintaining control of disease activity. TRIAL DESIGN: This is a multi-centre, randomised, rater-blinded, open label, phase 3 trial recruiting RRMS patients who have already received at least 2 years of 6-monthly ocrelizumab, had no disease activity for the last 12 months, and wish to continue treatment. We will adopt an innovative multi-arm, multi-stage response over continuous interventions (MAMS-ROCI) design: 540 patients will be randomised to 5 arms across a range of frequencies (6, 9, 12, 15 and 18 months) of ocrelizumab over 2 stages. In Stage 1, 6-monthly vs 12-monthly frequencies will be tested. If, at an interim analysis, 12-monthly is not less effective than 6-monthly, we will move to Stage 2, and additional patients will be recruited to all arms; if it is less effective, the trial will convert into a 2-arm non-inferiority trial of 9-monthly vs. 6-monthly frequencies. All participants will be followed-up for 2 years. Patients with disease activity with or without disability progression on reduced-frequency arms will be offered the 6-monthly schedule. The primary outcome is the proportion of patients without new MRI lesions at 2 years. Secondary outcomes are: relapse rate, proportions of patients without disability progression; rate of side effects; rate of brain atrophy on MRI; changes in serum neurofilament light levels, cognition, and quality-of-life; cost savings. The whole frequency-response curve will be modelled across these 5 frequencies, sharing information across them, to identify the optimal reduced frequency. IMPACT: Our results will directly translate to patient and NHS benefits, including reduced risk of side effects, increased convenience and better quality-of-life, lower drug costs and reduced pressure on resources. Our trial design can serve as a model for future trials to optimise the dose, frequency or duration of other therapies. DISSEMINATION will occur throughout the trial lifetime, through open-access publications, conference presentations, NIHR library, webinars, newsletters, and NHS/departmental social media accounts.