The review, led by Steffen Tiedt of the Institute for Stroke and Dementia Research at LMU University Hospital in Munich and Michael D. Hill of the University of Calgary, lays out a comprehensive roadmap for moving molecular biomarkers from the research bench into routine clinical practice. The authors’ central argument is deceptively simple: imaging and clinical scores show what the brain looks like, but they cannot reveal what the brain is doing at the molecular level. Blood-based biomarkers, by contrast, can resolve the activity of specific disease pathways, detect tissue stress before structural damage becomes visible on a scanner, report on processes unfolding far beyond the brain, and be measured repeatedly with minimal burden to the patient.

The strongest evidence that this vision is achievable comes from neuronal injury markers. Neurofilament light chain, a structural protein released when axons are damaged, has been shown in multiple studies to rise in blood after ischaemic stroke, track the severity of neuroaxonal injury and predict mortality and functional outcome. Even more striking is brain-derived tau, a form of the tau protein that is highly specific to the brain. Recent work, including a 2026 study in Science Translational Medicine co-authored by members of the same group, demonstrated that plasma brain-derived tau can dynamically quantify ischaemic brain injury and even capture the effects of treatment in humans. In other words, clinicians can already, in principle, watch brain damage unfold — and recede — in real time through a blood draw.

But the roadmap goes well beyond markers of dead neurons. The authors organise the biomarker landscape around four interconnected, outcome-relevant pathomechanisms: ischaemic brain injury, metabolism, systemic inflammation and gut dysbiosis. Each of these domains produces a distinct molecular signature in the circulation, and each offers a different window into the biology that determines whether a patient recovers or deteriorates. This mechanism-first framing is deliberate. The authors argue that biomarkers should be prioritised as tools for elucidating human stroke pathophysiology, not merely as therapeutic targets — a subtle but consequential shift in emphasis for a field that has struggled to translate laboratory discoveries into effective drugs.

That struggle is well documented. More than a thousand experimental treatments have shown benefit in animal models of stroke, yet virtually none have succeeded in clinical trials. Neuroprotective agents such as nerinetide, tested in the recent ESCAPE-NEXT and FRONTIER trials, and anti-inflammatory strategies including natalizumab and interleukin-1 receptor antagonists, have produced disappointing or mixed results. The review suggests a key reason: without molecular readouts of what is actually happening in each patient’s brain and body, trials enrol biologically heterogeneous populations in which a drug that helps one subgroup may harm another. Mechanism-specific biomarkers could enable biologically coherent patient stratification, allowing investigators to select patients whose disease is actually driven by the pathway a drug targets.

The metabolic dimension of stroke illustrates this potential. When blood flow is cut off, the brain’s energy economy collapses: glucose is shunted into anaerobic glycolysis, lactate accumulates, tissue becomes acidotic, and metabolites such as succinate build up before reperfusion unleashes a surge of mitochondrial reactive oxygen species. Circulating metabolites have been shown to differentiate acute ischaemic stroke from stroke mimics, and metabolic signatures measured at admission correlate with infarct growth and outcome. Lactate and ketone bodies, once viewed purely as waste products, are now understood to play signalling and even neuroprotective roles, and studies in mice suggest that microbiota-derived metabolites such as short-chain fatty acids can promote post-stroke recovery. A blood test that reports the metabolic state of the penumbra — the salvageable tissue surrounding the infarct core — could guide decisions about reperfusion and adjunctive therapy in ways no scanner currently can.

Systemic inflammation adds another layer of complexity. A stroke is not confined to the brain: within hours, the peripheral immune system is massively activated, the gut barrier is compromised and commensal bacteria can translocate into the circulation, contributing to post-stroke infection. Recent translational studies have revealed that brain-released alarmins and cell-free DNA can accelerate atherosclerosis and destabilise plaques, raising the risk of recurrent stroke, while innate immune memory triggered by brain injury can drive inflammatory cardiac dysfunction. Inflammatory markers such as interleukin-6, C-reactive protein, complement components and circulating cell-free DNA all correlate with infarct volume, severity and long-term outcome. The disappointing results of broad immunomodulation trials, including the finding that prophylactic antibiotics do not improve outcome despite reducing infections, underscore the authors’ point: post-stroke complications arise from complex immune dysregulation, and only biomarker-guided, mechanism-specific interventions are likely to succeed.

The gut-brain axis, perhaps the most surprising entry in the roadmap, is emerging as both a source of biomarkers and a therapeutic target. Stroke alters gut transit, impairs enteric neuron function and reshapes the microbiome, and microbial metabolites such as trimethylamine N-oxide have been linked to stroke severity, recurrence and outcome in human cohorts. Because the microbiome can be sampled non-invasively and modulated by diet, probiotics and antibiotics, microbiota-derived signatures offer an unusually tractable route to personalised intervention — provided that the analytical and clinical validation work is done rigorously.

Turning this promise into practice is where the six-phase roadmap earns its name. The authors propose a pragmatic progression that begins with defining the context of use — the specific clinical question a biomarker is meant to answer, whether diagnosis, prognosis, patient stratification or treatment monitoring — and then moves through assay development, analytical and biological validation, clinical validation in deeply phenotyped longitudinal cohorts, biomarker-focused experimental work, and finally biomarker-guided randomised trials in which treatment decisions are made on the basis of molecular readouts. The framework borrows deliberately from other neurological fields: Alzheimer’s disease has already redefined itself around biological markers, and the regulatory approval of tofersen for ALS, using plasma neurofilament light as a surrogate endpoint reasonably likely to predict clinical benefit, provides a precedent that biomarker-driven development can succeed with regulators. Point-of-care technology is advancing in parallel, with DNA origami-based signal amplification now demonstrating large-fold sensitivity gains in lateral flow immunoassays, pointing toward minutes-level quantification at the bedside.

The authors are candid that substantial work remains. Coordinated efforts will be needed: deeply phenotyped longitudinal cohorts with systematic biosampling in clinical trials, omics-driven discovery programmes, biomarker-focused experimental studies that validate mechanisms rather than merely correlate molecules with outcomes, and sustained collaboration between academia, industry and regulators. But the destination is compelling. If the roadmap is followed, stroke medicine could move from a discipline that treats ‘time is brain’ as its only molecular insight to one in which a blood test, drawn in the ambulance or at the bedside, tells clinicians which pathways are driving a particular patient’s injury, which treatment will help, and whether it is working — closing, at last, the translational gap that has frustrated stroke research for decades.

Subject of Research: Molecular biomarkers for clinical use in acute ischaemic stroke

Article Title: A roadmap towards clinical use of molecular biomarkers in acute ischaemic stroke

Article References: Tiedt, S., Vlegels, N., Dimitriadis, K., Li, W., Jickling, G. C., McCullough, L. D., Liesz, A., Lo, E. H., & Hill, M. D. (2026). A roadmap towards clinical use of molecular biomarkers in acute ischaemic stroke. Nature Reviews Neurology. https://doi.org/10.1038/s41582-026-01270-3

Image Credits: AI Generated

DOI: 10.1038/s41582-026-01270-3

Keywords: stroke, biomarkers, neurofilament light chain, brain-derived tau, neuroinflammation, gut-brain axis, metabolomics, thrombectomy, neuroprotection, clinical trials, precision medicine, Nature Reviews Neurology