Background: Neuroprotective drugs are needed to slow or prevent neurodegeneration and disability accrual in secondary progressive multiple sclerosis. Amiloride, fluoxetine and riluzole are repurposed drugs with potential neuroprotective effects.

Objectives: To assess whether or not amiloride, fluoxetine and riluzole can reduce the rate of brain volume loss in people with secondary progressive multiple sclerosis over 96 weeks. The secondary objectives that were assessed were feasibility of a multiarm trial design approach, evaluation of anti-inflammatory effects, clinician- and patient-reported efficacy and three mechanistic substudies.

Design: A multicentre, multiarm, randomised, double-blind, placebo-controlled, parallel-group Phase IIb trial with follow-up at 4, 8, 12, 24, 36, 48, 72 and 96 weeks. Patients, investigators (including magnetic resonance imaging analysts), and treating and independent assessing neurologists were blinded to the treatment allocation. The target sample size was 440 patients.

Setting: Thirteen UK clinical neuroscience centres.

Participants: Participants were aged 25–65 years, had secondary progressive multiple sclerosis with evidence of disease progression independent of relapses in the previous 2 years, and had an Expanded Disability Status Scale score of 4.0–6.5. Patients were ineligible if they could not have a magnetic resonance imaging scan; had a relapse or steroids in the previous 3 months; or had epilepsy, depression, bipolar disorder, glaucoma, bleeding disorders or significant organ comorbidities. Exclusion criteria were concurrent disease-modified treatments, immunosuppressants or selective serotonin reuptake inhibitors.

Interventions: Participants received amiloride (5 mg), fluoxetine (20 mg), riluzole (50 mg) or placebo (randomised 1 : 1 : 1 : 1) twice daily.

Main outcome measures: The primary end point was magnetic resonance imaging-derived percentage brain volume change at 96 weeks. Secondary end points were new/enlarging T2 lesions, pseudoatrophy, and clinician- and patient-reported measures (including the Expanded Disability Status Scale, Multiple Sclerosis Functional Composite, Symbol Digit Modalities Test, low-contrast letter visual acuity, Multiple Sclerosis Impact Scale 29 items, version 2, Multiple Sclerosis Walking Scale, version 2, and questionnaires addressing pain and fatigue). The exploratory end points included measures of persistent new T1 hypointensities and grey matter volume changes. The substudies were advanced magnetic resonance imaging, optical coherence tomography and cerebrospinal fluid analyses.

Results: Between December 2014 and June 2016, 445 patients were randomised (analysed) to amiloride [n = 111 (99)], fluoxetine [n = 111 (96)], riluzole [n = 111 (99)] or placebo [n = 112 (99)]. A total of 206 randomised patients consented to the advanced magnetic resonance imaging substudy, 260 consented to the optical coherence tomography substudy and 70 consented to the cerebrospinal fluid substudy. No significant difference was seen between the active drugs and placebo in percentage brain volume change at week 96 as follows (where negative values mean more atrophy than placebo): amiloride minus placebo 0.0% (Dunnett-adjusted 95% confidence interval –0.4% to 0.5%), fluoxetine minus placebo –0.1% (Dunnett-adjusted 95% confidence interval –0.5% to 0.3%); riluzole minus placebo –0.1% (Dunnett-adjusted 95% confidence interval –0.6% to 0.3%). There was good adherence to study drugs. The proportion of patients experiencing adverse events was similar in the treatment and placebo groups. There were no emergent safety issues.

Limitations: There was a lower than expected uptake in the cerebrospinal fluid substudy.

Conclusions: A multiarm Phase II paradigm is efficient in determining which neuroprotective agents to take through to Phase III trials. Amiloride, fluoxetine and riluzole were not effective in reducing the brain atrophy rate in people with secondary progressive multiple sclerosis. Mechanistic pathobiological insight was gained.

Future work: To use the information gained from the Multiple Sclerosis-Secondary Progressive Multi-Arm Randomisation Trial (MS-SMART) to inform future trial design as new candidate agents are identified.

Trial registration: Current Controlled Trials ISRCTN28440672, NCT01910259 and EudraCT 2012-005394-31.

Funding: This project was funded by the Efficacy and Mechanism Evaluation (EME) programme, a Medical Research Council and National Institute for Health Research (NIHR) partnership. This will be published in full in Efficacy and Mechanism Evaluation; Vol. 7, No. 3. See the NIHR Journals Library website for further project information. This trial also received funding from the UK MS Society and the US National Multiple Sclerosis Society.