From insights to innovations: evaluating preclinical paradigms in demyelinating disease therapeutics

lab animal Review article https://doi.org/10.1038/s41684-026-01725-6 From insights to innovations: evaluating preclinical paradigms in demyelinating disease therapeutics Check for updates Melika Karbalaee    1, Ally Lin1, Luca Peruzzotti-Jametti Sabah Mozafari    1    1,2, Stefano Pluchino    1 & Demyelinating disorders such as multiple sclerosis and leukodystrophies are on the rise, posing substantial challenges due to their progressive nature and the current limitations of therapies that effectively restore lost myelin. Over the past decade, advancements in regenerative neuroscience, including cutting-edge stem cell therapies, advanced biomaterials and groundbreaking gene-editing technologies, offer promising avenues for remyelination, immunomodulation and neural repair. Yet, to successfully transition these innovations into clinical therapies, we need robust preclinical models that accurately reflect disease pathology and predict treatment efficacy. This Review offers a thorough overview of the preclinical models utilized in regenerative neurology for demyelinating diseases, highlighting the rapid progress in biomaterial and gene-editing research, which requires meticulous testing and validation in both in vitro and in vivo environments. We begin by explaining the pathophysiology of demyelination, then provide an exhaustive discussion on various preclinical models, including toxin-induced, autoimmune, genetic, viral-induced and large animal models. This is followed by an exploration of emerging regenerative strategies, from cell-based and pharmacological approaches to bioengineered techniques, and we conclude with an analysis of current challenges, translational barriers and future directions in the field. By synthesizing insights from multiple disciplines, this Review strives to engage a diverse audience eager to connect laboratory discoveries with clinical applications in regenerative neuroscience. Demyelination disrupts the intricate wiring of the central nervous system (CNS), stripping neurons of their protective myelin sheath and driving progressive axonal loss. Multiple sclerosis (MS)—the most common acquired demyelinating disease—is marked by neuroinflammation, axonal degeneration and accumulating neurological disability1. Other acquired disorders, including neuromyelitis optica spectrum disorder and acute disseminated encephalomyelitis, present distinct but equally destructive inflammatory profiles. By contrast, leukodystrophies—genetic disorders of myelin formation and maintenance—such as adrenoleukodystrophy, metachromatic leukodystrophy, Krabbe disease and Pelizaeus–Merzbacher disease (PMD), cause relentless neurodegeneration from early life2. These largely human-specific diseases share a unifying hallmark: the loss of myelin integrity, which disrupts neural communication and amplifies maladaptive neuroimmune interactions. Deciphering the mechanisms governing myelin homeostasis, plasticity and intercellular regulation 1 across development, maintenance, adaptation and repair is essential to identify therapeutic targets and halt neurodegeneration. As myelin provides trophic, structural and metabolic support to axons, prolonged demyelination renders axons highly vulnerable to degeneration. Although remyelination can promote functional recovery and protect axons from degeneration, this process is often inadequate in diseases like MS with frequent episodes of demyelination3. Current first-line therapies for demyelinating diseases, including MS, primarily focus on symptom management, mainly through anti-inflammatory and immunomodulatory effects, with their efficacy largely limited to the early stages of the disease4. However, these treatments do not address the underlying causes of myelin loss or promote true repair5. Consequently, there is an urgent need for regenerative approaches that not only halt disease progression but also promote myelin restoration and functional repair. Department of Clinical Neurosciences, National Institute for Health Research Biomedical Research Centre, University of Cambridge, Cambridge, UK. Department of Metabolism, Digestion and Reproduction, Imperial College London, London, UK. e-mail: ; 2 Lab Animal https://doi.org/10.1038/s41684-026-01725-6 Neuroregenerative strategies are advancing to address the root causes of myelin diseases, aiming to restore myelin, suppress inflammation and protect neurons and glia6. Preclinical efforts now target the limited repair capacity of the CNS through diverse approaches. Pharmacological agents, such as clemastine fumarate, promote oligodendrocyte precursor cell (OPC) differentiation and enhance endogenous remyelination7. Cell-based therapies, using neural stem cells (NSCs), mesenchymal stem cells (MSCs) or induced pluripotent stem (iPS) cell-derived neuroglia and their products, combine immune modulation with tissue repair8–10. Tissue-engineering platforms such as hydrogels and scaffolds are being developed to improve cell delivery, reduce glial scarring and create a pro-regenerative niche11–13. Neurostimulation techniques, including transcranial magnetic stimulation (TMS) and electrical stimulation, have also shown potential to boost myelin repair and functional recovery in preclinical models14,15. However, translating these advances into clinical therapies requires robust preclinical models that closely recapitulate human demyelination for mechanistic studies, efficacy testing and safety validation. In vitro and ex vivo models, including neuroglia cultures, organoids, organotypic and microfluidic systems, provide controlled platforms to investigate myelination, remyelination and immunomodulation16,17. While invaluable for dissecting mechanisms, they cannot fully replicate the complexity of the human brain, particularly immune interactions and long-range connectivity, and they also face limitations in cost, scalability and their ability to model chronic disease progression18. In vivo models capture demyelination and repair in a whole-organism context19. These include toxin-induced paradigms (for example, cuprizone (CPZ), lysolecithin)20, autoimmune models such as experimental autoimmune encephalomyelitis (EAE) that closely mimic MS pathology21, genetic models such as Shiverer mice and PMD22,23, viral models24 and combined approaches25,26. Large animal models, such as nonhuman primates and canines, offer greater anatomical and immunological similarity to humans9, but face major drawbacks, including high costs, ethical constraints, long lifespans and disease courses, limited genetic tools and increased variability. Moreover, the adult human CNS has more restricted regenerative capacity than other species, a limitation that worsens with age, underscoring the need for therapies tailored to human-specific biology27, particularly in chronic and progressive demyelinating diseases. This Review critically examines the evolving spectrum of preclinical models in regenerative neuroscience (...truncated)