MRI biomarkers of post-stroke stem cell implantation and motor recovery: systematic review of clinical trials

Abstract

Background: Stroke causes long-term disability, and stem cell implantation may augment neural repair. Advanced MRI may provide objective biomarkers of tissue remodeling associated with motor recovery. Methods: Five databases were searched per PRISMA 2020 (PROSPERO CRD420251085067). Clinical trials enrolling stroke patients treated with mesenchymal stem cells (MSCs), mononuclear cells (MNCs), or neural stem cells (NSCs) were included if they reported standardized motor scales and advanced MRI (DTI, fMRI/rs-fMRI, or MRS). Risk of bias was assessed with RoB 2 or ROBINS-I. Results: Eleven trials (5 randomized, 6 non-randomized) comprising 513 participants were included. DTI was used in 9 studies, fMRI in 4, rs-fMRI in 3, and MRS in 1. MSC trials demonstrated the most consistent associations with motor improvement, though many lacked robust controls and were accompanied by preserved or increased corticospinal tract fractional anisotropy and greater task-related BOLD activity within motor regions (BA4/BA6). MNC trials reported NIHSS reductions and later increases in relative FA; the single MRS study suggested rising Nacetylaspartate with decreasing choline. Evidence for NSCs was limited to one small trial reporting motor gains with increased fractional anisotropy. Conclusions: Multimodal MRI, especially DTI and fMRI, provides tractable biomarkers for monitoring post-stroke stem cell interventions; current clinical evidence shows associations with motor improvement and MRI biomarker changes most frequently in MSC trials.