Blood biomarkers
Which proteins we measure, what a result does and does not show, why it cannot stand in for a clinician's judgment, imaging or a functional test, and what the APOE result included in the panel will tell you.
The four markers
Different proteins leak from different cells on different timelines. That is why the timing of a blood draw changes what you can learn from it.
Glial fibrillary acidic protein
A structural protein of astrocytes. Its appearance in blood indicates astroglial injury, and it is one of the two markers in the FDA-cleared blood tests for suspected mild head injury — cleared as an aid, read alongside other clinical information.1
Ubiquitin C-terminal hydrolase L1
An enzyme abundant in neuronal cell bodies. Paired with GFAP because the two together are more informative than either alone; its short half-life makes early sampling essential.5
S100 calcium-binding protein B
The longest-established marker, and part of some European head-injury guidelines. Its weakness is specificity: it is also released by fat, muscle and skin, so orthopaedic injury or hard physical exertion can raise it without any brain injury at all.6,9
Neurofilament light chain
A scaffolding protein of the neuronal cytoskeleton, concentrated in subcortical white-matter axons. In one serial study plasma NF-L was roughly 483× baseline at day 2, 89× at day 9, and still 3× at day 176 — which makes it a marker of axonal damage over time rather than a snapshot.7
Three different questions
Each modality answers something the others cannot. Treating any one of them as the whole answer is how injuries get missed.
| Modality | Question it answers | Useful window | Blind spot |
|---|---|---|---|
| Blood biomarkers | Has brain tissue been damaged biologically? | Hours to days (NF-L: months) | Cannot tell you how the brain is functioning |
| CT / standard MRI | Is there bleeding or a structural lesion? | Acute | Usually normal in mild TBI |
| Symptom questionnaire | What does the patient report feeling? | Any time | Captures only what the patient can report — necessary, but not sufficient alone |
| Ocular motor testing | Is brain function measurably impaired? | Most informative early | Signal can fade as the brain compensates |
Where NeuroGlympse fits
The patient who benefits most from understanding biomarkers is the one whose results were normal. A normal CT speaks to a narrow question — whether there is acute intracranial injury a scan can see — and a normal blood result is one piece of evidence on that same question.
What those results do not measure is whether the brain is working properly. That is a functional question, which structural scans and blood tests are not designed to answer. Functional tests — functional brain imaging and ocular motor testing are two examples — look at it directly. Our assessment uses ocular motor testing, then follows autonomic recovery through continuous monitoring.
Biomarkers and functional testing are complementary, not competing. One can indicate that tissue was injured; the other shows what that injury is doing to the person.
We offer both biomarker testing and ocular motor testing, which is the point — a biomarker result read next to an ocular motor result is far more informative than either read alone.
The direction of travel
In 2025 the NIH-NINDS TBI Classification and Nomenclature Initiative proposed the CBI-M framework: characterising a traumatic brain injury across four axes — Clinical findings, Biomarkers, Imaging, and Modifiers such as prior injury and comorbidity — instead of collapsing it into a single severity label.
It is a formal acknowledgement of what clinicians treating these patients already knew: mild, moderate and severe were never adequate, and no one modality is sufficient on its own.
Genetic insight
APOE genotyping is part of the panel, but it does not come from the blood — the genetic sample is taken with a separate cheek swab. It is the one component that says nothing about your injury, and it is worth being precise about what it does and does not add.
APOE is a gene with three common forms — ε2, ε3 and ε4. Which pair you carry is fixed at conception, so unlike every other marker on this page it is tested once and never repeated, and the result is as true a year later as it is on the day it is taken.
It is also collected differently. Every other marker here comes from blood; APOE comes from a buccal swab — a swab of the inside of your cheek. Both kits come together and you do the swab and the blood collection in the same sitting — two samples, one appointment. How at-home collection works.
What it contributes is prognostic context at the group level. A meta-analysis of 2,593 patients found higher odds of a favourable outcome after TBI among people who do not carry an ε4 allele than among ε4 carriers — an odds ratio of 1.39 (95% CI 1.05–1.84).11 The authors’ own summary is that ε4 confers a small risk of poor outcome, that the data did not permit analysis by injury severity, and that the effect on neuropsychological functioning remained uncertain across conflicting studies.
Common questions
Yes. We provide biomarker testing alongside our functional ocular motor assessment, with results interpreted by our board-certified neurology team and read together rather than in isolation. A panel is ordered by a physician, not selected by the patient, and specimens are processed by a CLIA-certified laboratory identified on the result report. Contact our team and we will advise which panel is appropriate for how long ago the injury happened — or whether any of them is.
No. This is the single most misunderstood point. The FDA-cleared blood tests measuring GFAP and UCH-L1 are cleared only as an aid, used alongside other clinical information when evaluating adults within hours of a suspected mild head injury. A negative result is associated with the absence of acute intracranial lesions on a head CT. It is not a statement that your brain function is unaffected, and it neither diagnoses nor excludes a concussion. Concussion remains a clinical diagnosis.
Abbott's i-STAT TBI test measuring GFAP and UCH-L1 was cleared in 2021 for plasma, with a whole-blood cartridge that can be used at the point of care cleared in 2024; Abbott's laboratory TBI test was cleared in 2023. bioMérieux's VIDAS TBI, measuring the same two proteins, was cleared in 2024. The first such assay, from Banyan Biomarkers, was authorised through the FDA's De Novo pathway in 2018 (see references below). All are limited to adults with a suspected mild traumatic brain injury (Glasgow Coma Scale 13–15) within 12 hours of injury — 24 hours for the i-STAT whole-blood cartridge — and all are an aid used alongside other clinical information: FDA rules require their labelling to say they are not stand-alone tests (21 CFR 866.5830). None is cleared to diagnose a concussion.
In ALERT-TBI, the pivotal study, the combined test was 97.6% sensitive for intracranial injury on CT, with a 99.6% negative predictive value (American Family Physician 2022;105(3):313). Specificity was low: most people with a raised result had nothing on their CT. Taken together, in patients like those studied, a negative result makes intracranial injury on CT unlikely. It does not mean the test decides who is scanned. That is the clinician's decision, guided by validated clinical decision rules, and the American College of Emergency Physicians' 2023 clinical policy makes no recommendation for using blood tests in it (Annals of Emergency Medicine 2023;81(5):e63).
Because the test was answering a different, narrower question. A normal result does not exclude a concussion, and most mild traumatic brain injuries produce nothing a CT can see — while the functional injury — the disrupted eye movement, the disturbed sleep, the suppressed heart rate variability — can be entirely real, and measurable by functional tests.
Very early for GFAP and UCH-L1, which rise and clear within roughly the first day; the FDA-cleared indications are built around acute presentation. S100B peaks around day 1 and returns near baseline within a couple of weeks. NF-L is the exception — it climbs later and stays elevated for months, so it is the one marker still informative well after the event.
A 2025 framework from the NIH-NINDS TBI Classification and Nomenclature Initiative proposing that TBI be characterised across four axes rather than a single severity label: Clinical findings, Biomarkers, Imaging, and Modifiers such as prior injury and comorbidity. It is a formal acknowledgement that no one test defines a brain injury.
Our testing
We provide blood biomarker testing as part of the same assessment pathway as our functional testing, so a single referral produces both the biological and the functional picture rather than sending your patient to two unconnected providers.
Results are interpreted by our board-certified neurology team and read alongside ocular motor findings and monitoring data — not returned as an isolated number for someone else to make sense of.
Every figure on this page traces to a source below, so the claims can be checked rather than taken on trust.
This page is educational and is not medical advice. Prevalence figures are presented as the source reports them, including confidence intervals and ranges; a single number would imply more precision than the evidence supports.
Biomarker testing and objective functional assessment from one referral, interpreted by the same neurology team.
