A systematic review of oral melatonin in adults with multiple sclerosis found no demonstrated reduction in relapse activity, disability progression or new MRI activity.
The symptom evidence was less decisive. One trial reported a small difference in total sleep time, another reported lower fatigue scores, and a separate study reported improvements in sleep, fatigue and pain, but the favorable findings lacked independent replication and came with methodological limitations.
A small and varied body of evidence
The review included 10 independent studies across 17 publications and 435 participants. Nine were randomized trials, including two crossover studies, while one was a prospective non-randomized intervention study.
Participants were adults aged 18 or older with confirmed multiple sclerosis. The studies tested exogenous oral melatonin, either alone or alongside stable therapy, in immediate-release or prolonged-release forms. Doses ranged from 0.5 to 300 milligrams a day, and treatment lasted from two weeks to 14 months.
The reviewers did not combine the results into a single pooled estimate. They said the studies were too different clinically and methodologically, and too few independent studies were available for many outcomes, so they used a structured synthesis without meta-analysis.
Sleep and fatigue results point in different directions
In one crossover trial, mean total sleep time was 7.0 hours with melatonin and 6.7 hours with placebo, a difference of about 18 minutes. The reported comparison had a p-value of 0.03, but the gap was below the prespecified 20-minute threshold for clinical significance.
The fatigue finding came from another trial. After adjusting for baseline differences, the melatonin group had an average score of 30.42, compared with 41.00 in the placebo group. The reported variation was 12.40 and 12.66 points, respectively, and the reported p-value was 0.001. But the allocation was uneven, with 47 participants in the melatonin group and 15 in placebo, and reporting of the trial methods was limited.
Jallouli reported improvements in sleep, fatigue and pain, but those results remained unreplicated. The review therefore did not establish a consistent symptom benefit.
Why the conclusions remain cautious
Risk of bias was assessed with RoB 2 and ROBINS-I. Eight randomized trials had some overall concerns about bias, one trial had high overall risk, and no randomized trial was judged to have low overall risk.
The review followed PRISMA 2020 and registered its protocol prospectively in PROSPERO. Two reviewers independently screened titles, abstracts and full texts, and two reviewers independently extracted data using a standardized pilot-tested form. Disagreements were settled through discussion or consultation with a third reviewer.
The search covered MEDLINE via PubMed, Scopus, Web of Science Core Collection and CENTRAL from database inception through August 15, 2026. The reviewers also checked reference lists and forward citations and imposed no publication-date or language restrictions.
The small number of independent studies and their substantial clinical and methodological differences were the main reason the reviewers did not give a single pooled treatment effect. They reported results by outcome instead.
What the review establishes
The bottom line is narrow but important: current evidence does not demonstrate reductions in relapses, disability progression or new MRI activity. Sleep and fatigue signals remain mixed, and the favorable findings have not been independently replicated.
Paper data and sources
Original title: Melatonin supplementation in multiple sclerosis: a systematic review of clinical, imaging, biomarker, and safety outcomes
Authors: Nabil Y, Ghazou A, Abodabesh S et al.
Journal/Repository: Frontiers in neurology
Status: Peer-reviewed
First online: 2026-10-10
DOI: Not available
Original paper · Full text