At the heart of the work lies a deceptively simple observation: injured tissues are rarely static. Tendons stretch, cartilage compresses, skin flexes, and intervertebral discs cycle through millions of load-bearing events over a lifetime. Conventional regenerative biomaterials largely ignore this mechanical bustle, delivering growth factors through passive diffusion, which often leads to an initial burst release followed by a rapid decline — the opposite of the sustained, spatially patterned signaling that natural healing demands. The research team hypothesized that if a biomaterial could convert mechanical work into biochemical signaling on demand, it could transform the physical activity of a patient into a therapeutic intervention, releasing repair factors precisely when and where tissue is being loaded.

To test this idea, the investigators designed hydrogel matrices decorated with mechanically labile crosslinks — molecular tethers that remain stable under resting conditions but rupture or unfold when subjected to physiologically relevant forces. Embedded within these networks were binding domains that capture endogenous growth factors, the signaling proteins that tissues themselves produce in response to injury. Rather than flooding the wound site with exogenous factors, the material acts as a dynamic reservoir and relay station: as mechanical force passes through the matrix, the force-responsive elements transiently loosen the network, allowing locally produced growth factors to bind, concentrate, and then be presented to resident cells in a bioactive form. The result is a positive feedback loop in which tissue loading amplifies the availability of the very signals that promote repair.

The technical execution required careful tuning across multiple length scales. The researchers synthesized polymeric networks whose mesh size, degradation kinetics, and crosslink densities were calibrated so that the forces generated during normal locomotion — typically in the range of a few pascals to kilopascals of stress at the tissue interface — would activate release without triggering premature failure of the scaffold. They incorporated mechanosensitive linkers inspired by proteins such as fibronectin, which naturally unfolds under tension to expose cryptic binding sites. In the synthetic analog, these force-activated domains serve a similar purpose: they expose affinity motifs that sequester growth factors from the surrounding interstitial fluid, effectively harvesting the body’s own regenerative chemistry. Spectroscopic characterization confirmed that the binding interactions preserve the growth factors in their native conformation, a critical detail, since denatured or misfolded signaling proteins lose their biological activity.

What distinguishes this approach from earlier “mechano-activated” drug delivery systems is its reliance on endogenous rather than exogenous payloads. Recombinant growth factor therapy — whether with platelet-derived growth factor, transforming growth factor beta, or vascular endothelial growth factor — has long been hampered by prohibitive cost, short in vivo half-lives, and safety concerns stemming from supraphysiological dosing. By concentrating and presenting factors that the tissue is already producing at low, safe levels, the biomaterial sidesteps these limitations entirely. The team’s in vitro experiments showed that fibroblasts and mesenchymal stem cells cultured on force-conditioned matrices exhibited markedly enhanced proliferation, migration, and matrix deposition compared with cells grown on mechanically passive controls, even though the total growth factor concentration in the system was identical in both conditions. The difference, the authors argue, lies in presentation: spatial immobilization and force-triggered activation preserve signaling fidelity in a way that soluble delivery cannot.

The therapeutic potential became most apparent in animal models. In rodent models of skin wound healing, implants of the force-responsive material accelerated re-epithelialization and angiogenesis, producing wounds that closed significantly faster than those treated with inert scaffolds. In models of tendon and load-bearing soft tissue injury — settings where mechanical loading is unavoidable and often detrimental to passive delivery systems — the materials converted that same loading from an obstacle into an asset. Histological analysis revealed denser, more organized collagen deposition and improved mechanical integrity of the repaired tissue, suggesting that the regenerated matrix was not merely filling a defect but reconstructing functional architecture. The authors emphasize that the repaired tissue in the treated groups bore a closer resemblance to native tissue than to the disorganized scar typical of default wound healing.

Beyond the immediate clinical implications, the study contributes a conceptual advance to biomaterials science: the idea of mechanobiological feedback as a design principle. Tissue engineers have long appreciated that cells sense and respond to the stiffness and geometry of their surroundings, a field broadly known as mechanotransduction. The new work inverts the perspective, asking not how forces affect cells directly but how materials can harness force to modulate the biochemical microenvironment. This reframing opens a design space in which a patient’s own activity level, physical therapy regimen, or even rehabilitative exercise becomes part of the therapeutic dosing strategy. A clinician could, in principle, prescribe movement as a means of controlling the release of healing signals, coupling rehabilitation protocols directly to the material’s activation profile.

The safety profile of the approach also merits attention. Because the growth factors being harnessed are produced endogenously at physiological concentrations, the risk of off-target effects — aberrant vascularization, uncontrolled cell proliferation, or fibrotic scarring — is substantially lower than with bolus protein delivery. The material itself was engineered to degrade into biocompatible byproducts over a time scale matched to the healing process, ensuring that the scaffold does not persist as a foreign body once its job is done. Longitudinal studies in the animal models showed no evidence of chronic inflammation, ectopic tissue formation, or systemic signaling disturbances, findings that the authors present as an encouraging early signal for translational viability.

Challenges remain before the technology reaches the clinic. Manufacturing consistency, sterilization compatibility, and regulatory pathways for a device-drug hybrid that contains no drug in the traditional sense will all require careful navigation. The dose of “mechanical activation” will need to be standardized for different anatomical sites, since the forces experienced by a fingertip differ enormously from those in a lumbar disc. Patient populations with limited mobility — the elderly, the bedridden, or those with paralysis — may generate insufficient mechanical stimulation to fully activate the system, raising questions about whether supplemental loading devices could extend the benefit to these groups. Nonetheless, the authors argue that the platform is modular: by swapping affinity domains, the same force-responsive backbone could be adapted to concentrate different classes of endogenous signaling molecules, from cytokines that modulate inflammation to morphogens that guide stem cell differentiation.

The broader significance of the work may lie in its economy. Healthcare systems worldwide spend billions annually on recombinant biologics, and access to advanced regenerative therapies remains sharply stratified by geography and income. A material that amplifies the body’s intrinsic repair capacity — requiring no pharmaceutical ingredient, cold chain, or repeated dosing — could dramatically lower the cost barrier to regenerative medicine. In an era when the promise of tissue engineering has often been tempered by the complexity and expense of its implementations, this study offers a refreshing counterpoint: sometimes the most sophisticated therapy is the one that simply gets out of the body’s way, and then gives it a mechanical nudge in the right direction. As the field moves toward clinical translation, force-responsive biomaterials of this kind may well become a cornerstone of next-generation regenerative medicine, turning every step, stretch, and movement of the patient into a dose of self-administered healing.

The central problem is that multiple sclerosis is not a single, uniform disorder. It is an immune-mediated disease of the central nervous system in which inflammation, demyelination, axonal injury, neuronal loss and tissue repair can occur at the same time, but in different combinations and at different intensities. Two people assigned the same clinical course may therefore have very different disease mechanisms operating beneath the surface. One may have frequent focal inflammatory attacks visible on magnetic resonance imaging, while another may accumulate disability through slowly evolving neurodegeneration with few obvious relapses.

The relapsing–remitting category illustrates this limitation. It generally describes patients who experience episodes of neurological dysfunction followed by periods of partial or substantial recovery. However, clinical recovery does not necessarily mean that the disease has become biologically inactive. New lesions may develop without symptoms, and damage can accumulate silently between recognized relapses. Conversely, some symptoms may reflect residual injury rather than a new inflammatory event. The label captures the visible pattern of illness, but not always the activity taking place in the brain and spinal cord.

The transition to secondary progressive multiple sclerosis is also more gradual than the terminology suggests. In practice, worsening disability may emerge slowly, fluctuate from year to year or become apparent only after the effects of repeated inflammatory injuries have accumulated. The precise point at which a patient moves from a relapsing pattern to a progressive one can therefore be difficult to define. Primary progressive multiple sclerosis, meanwhile, identifies people whose disability increases from the outset, but it does not necessarily describe a single biological pathway. Progressive disease can involve varying contributions from inflammation, compartmentalized immune activity, mitochondrial dysfunction, synaptic damage and failure of nervous-system repair.

The roadmap proposes that future descriptions should reflect this multidimensional reality. Instead of treating disease course as a fixed sequence of named stages, clinicians could describe multiple sclerosis through continuously updated dimensions, including inflammatory activity, progression independent of relapse activity, and the degree of disability or neurological impairment. Such a framework would recognize that inflammation and progression are not mutually exclusive. A patient may have active new lesions while also experiencing gradual worsening, or may show disability progression with little evidence of conventional inflammatory activity.

This shift would require more than changing terminology. Biologically informed descriptions would depend on increasingly sensitive measurements that can detect disease processes before they become obvious in routine neurological examinations. MRI already provides information about new or enlarging lesions, contrast enhancement, brain-volume loss and damage in the spinal cord, although these measures do not capture every aspect of pathology. Fluid biomarkers, including molecules associated with axonal injury and glial activation, may offer additional insight into ongoing tissue damage. Digital assessments of walking, hand function, vision and cognition could also reveal subtle changes that are missed during intermittent clinic visits.

A dynamic system could improve both patient care and research. Clinicians would be able to distinguish ongoing inflammatory activity from disability that continues independently of relapses, potentially supporting more precise treatment decisions. Researchers could select trial participants according to the mechanism a therapy is designed to target, rather than relying primarily on broad clinical categories. This might make it easier to evaluate treatments aimed at preventing progression, protecting neurons, restoring myelin or modifying chronic inflammation—goals that are difficult to assess when all outcomes are reduced to relapse counts or a single disability scale.

The proposed transition also carries important practical challenges. New descriptors would need to be reliable across hospitals, countries and patient populations, and they would have to remain understandable to patients and healthcare professionals. Biomarkers must be validated, standardized and shown to provide information that changes clinical decisions. Any new framework should avoid creating a more complicated classification that simply replaces one set of rigid labels with another. It must also account for differences in age, disease duration, treatment exposure, comorbidities and access to care, all of which can influence how multiple sclerosis appears clinically.

The roadmap therefore presents evolution rather than an overnight replacement of the current system. Existing relapsing and progressive terms remain useful for communication and for interpreting decades of research, but they may increasingly be supplemented by biological and clinical descriptors that are updated over time. The long-term goal is a living profile of each patient’s disease, combining symptoms, disability, imaging, biomarkers and treatment response. Such a profile could bring medical language closer to the biology of multiple sclerosis—and help move care toward earlier detection, more precise intervention and a clearer understanding of why the disease follows such different paths in different people.

Subject of Research: Multiple sclerosis disease course and the development of biologically informed clinical descriptors

Article Title: Towards a biologically informed description of multiple sclerosis disease course — a roadmap for transition

Article References: Thompson, A.J., Lublin, F.D., Rechtman, L. et al. Towards a biologically informed description of multiple sclerosis disease course — a roadmap for transition. Nature Reviews Neurology (2026). https://doi.org/10.1038/s41582-026-01249-0

Image Credits: AI Generated

DOI: 10.1038/s41582-026-01249-0

Keywords: multiple sclerosis, disease progression, relapsing–remitting multiple sclerosis, secondary progressive multiple sclerosis, primary progressive multiple sclerosis, neuroinflammation, neurodegeneration, biomarkers, MRI, precision medicine