Multiple sclerosis (MS), notably its progressive forms, remains a major therapeutic challenge with unmet needs for neuroprotection and repair. This systematic review examined the therapeutic potential of mesenchymal stem cells (MSCs) in MS, focusing on both clinical outcomes and responses measured by biomarkers, to clarify whether MSC interventions produce measurable neuroprotective or immunomodulatory effects.
The review was conducted according to PRISMA 2020 guidelines. Searches were performed in PubMed, Scopus, Web of Science, and Embase through January 3, 2026. Seventeen publications fulfilling inclusion criteria were identified, representing 16 distinct clinical trials. Data extracted included MSC source, dosing, administration route, and outcomes across clinical, imaging, immunological, and objective biomarker domains. Study quality was appraised using Joanna Briggs Institute tools.
The evidence base comprised five unique randomized controlled trials (reported in six publications) and 11 non-randomized studies. MSCs originated from different sources and were delivered by various routes, including intravenous and intrathecal administration, and included bone marrow-derived and umbilical cord-derived MSCs. Trial designs, sample sizes, dosing regimens, follow-up durations, and chosen outcome measures varied substantially across studies, producing significant methodological heterogeneity.
Randomized trials displayed divergent clinical outcomes. A large trial of intravenous bone marrow-derived MSCs did not demonstrate significant improvement in the Expanded Disability Status Scale (EDSS). By contrast, trials that used intrathecal administration or umbilical cord-derived MSCs reported potential improvements in disability and functional measures such as the timed 25-foot walk (T25-FW) and the 9-hole peg test (9-HPT). Overall, clinical efficacy signals were inconsistent across the trial set, and no definitive, reproducible clinical benefit across MS populations was established.
Biomarker analyses across studies were heterogeneous and often inconsistent in direction and magnitude. Immunological markers such as tumor necrosis factor-alpha (TNF-α) and interleukin-17 (IL-17) were reported in several reports, but these findings were primarily derived from studies judged to have high risk of bias and did not form a consistent pattern supporting a clear immunomodulatory mechanism.
Notably, one randomized controlled trial using intrathecal MSC administration demonstrated a significant reduction in cerebrospinal fluid neurofilament light (NfL) levels. This finding suggests a possible route-dependent neuroprotective effect measurable by an objective neuronal injury biomarker. However, this result stands in the context of broader biomarker inconsistency and cannot on its own establish predictive validity or clinical translation without replication and longer-term follow-up.
The review highlights that most trials reported only expression-level changes in biomarkers without sufficient validation of their predictive accuracy for clinical outcomes. Heterogeneity in biomarker selection, timing of sampling, and laboratory methods further complicated interpretation.
Across all included studies, MSC therapy was associated with an excellent safety profile. The review states that no serious treatment-related adverse events were reported in the analyzed trials. This consistent safety signal applies across different MSC sources and administration routes included in the review.
The authors emphasize several important limitations that constrain interpretation: methodological heterogeneity across trials (including variable MSC sources, doses, and routes), short follow-up periods in many studies, small sample sizes, and inconsistent biomarker selection and reporting. Immunological marker findings came largely from studies with high risk of bias. These factors limit the ability to draw definitive conclusions about efficacy, mechanism, and biomarker predictiveness.
The review concludes that while MSC therapy appears safe in clinical studies to date, evidence for consistent clinical efficacy and biomarker improvement is tentative. Intrathecal delivery and umbilical cord-derived MSCs provided the most promising but still preliminary signals, including a reduction in CSF NfL in one RCT suggesting potential neuroprotection. The authors call for standardized, adequately powered, large randomized controlled trials to determine optimal MSC source and route of administration. They also stress the urgent need to validate biomarkers for predictive accuracy and clinical relevance rather than simply reporting expression changes.
Biomarkers; Immunological markers; Mesenchymal stem cells; Multiple sclerosis; Neurofilament light; Progressive MS; Regenerative medicine.