New Blood Test for Brain Injuries Returns Instant Results

The test measures two proteins in a drop of blood to determine whether a patient needs a CT scan

A rapid diagnostic test for mild traumatic brain injury requires only one drop of blood
Photo: UNIGE/AAAS
One drop of blood is enough for TBIcheck to diagnose a possible mild brain injury.
Advertisement

When a person hits their head in a fall, car accident, sports collision, or other mishap, he or she is often whisked to the hospital for a CT scan to identify any lesions, or bruises, on the brain.

But for 90 percent of people, the scan comes back negative, even if they are later diagnosed with a mild brain injury such as a concussion. This suggests other tools are needed to accurately diagnose mild brain injuries. Plus just one CT scan costs up to US $2,000 and delivers a large dollop of radiation exposure.

That’s why researchers have long sought ways to distinguish which patients need a CT scan and which don’t. One promising approach is to detect brain damage via bits and pieces of brain cells or blood vessels that have burst, torn, or stretched in the circulating blood.

This week, a European team announced progress toward a quick, on-the-spot test for mild traumatic brain injuries (mTBIs) that requires only a drop of blood and can tell a patient within minutes whether they should have a CT scan or feel assured that there is no injury. The test is based on the findings of a recent paper in the journal PLoS One pinpointing two proteins detected in the blood within 24 hours of a brain injury.

“The intend[ed] use is to avoid to unnecessary CT scans. But, at the same time, to be sure at 100 percent that all brain lesions are detected properly,” said study author Jean-Charles Sanchez of the University of Geneva in Switzerland.

Brain injury is complex: Many cell types are damaged and different parts of the brain are injured depending on the type and location of impact. Tests that tried to diagnose mTBI with a single blood marker—say, one protein that leaks from neurons after damage—simply haven’t been sensitive or specific enough to accurately diagnose mTBI.

The race is now on to create a panel, or set, of blood biomarkers that can accurately diagnose mTBI on the spot: whether that’s on the sports field, in an ambulance, or even on the playground.

Last year, a team at Duke University identified two proteins in the blood that, when measured together, allow doctors to differentiate between patients at low risk of a brain lesion and those at high risk.

In a different tactic, Sanchez and colleagues wondered if biomarkers for stroke, a better-studied condition in which brain cells also die, might also be good indicators of mTBI.

Across three European sites, the team enrolled 132 patients admitted to an ER with an mTBI diagnosis and at least one additional clinical symptom, such as vomiting or loss of consciousness. Blood was taken from each participant within 4 hours of their trauma and each underwent a CT scan.

The scientists then tested each participant’s blood for 13 proteins associated with stroke and two proteins previously associated with mTBI: S100B and GFAP. In earlier studies, the team tested other proposed biomarkers of brain damage including neurofilament light and tau, but these appeared in the blood only after 24 hours—too late for use in a rapid test.

Overall, four proteins—S100B, GFAP, and two stroke-associated proteins, H-FABP and IL-10—showed significantly higher levels in the blood of patients with a lesion on their CT, suggesting these proteins are good, early markers of mild brain injury.

Next, the researchers tried out different combinations of those four proteins to identify the smallest possible combination that could accurately detect brain injury. The winner was H-FABP and GFAP.

Measured together, those two markers accurately predict if a person has mTBI and needs a CT scan 100 percent of the time. It also correctly identifies those without a brain injury 46 percent of the time. That means a test based on those markers would have false positives, advising some individuals without mTBI to receive a CT scan, but it would still reduce the overall number of unnecessary CT scans. 

All four proteins together would identify those without mTBI 56 percent of the time, the researchers found. However testing four proteins in an on-the-spot test is more expensive and more difficult than testing two. 

The team confirmed their results with a second group of 109 participants.

H-FABP in particular turns out to be a great marker of mTBI, says Sanchez, because it’s a small, water-loving protein that easily slips through the blood-brain barrier, so “it leaks in the blood faster than the others, within 3 hours” after injury, he says.

In 2014, Sanchez and colleagues founded a biotech company, ABCDx, to develop biomarker tests for brain injury. Their first generation rapid mTBI blood test, called TBIcheck, tests for the presence of H-FABP. Following the new findings, the company’s next test will be a duplex, says Sanchez, to include both H-FABP and GFAP.

The company plans to initiate a clinical trial in three hospitals in Spain in September and hopes to sell the device commercially as early as the beginning of 2019.

The Human OS Newsletter

Biomedical engineering in a biweekly newsletter. Expert insights into wearable sensors, big data analytics, and implanted devices for personalized medicine.

About the Human OS blog

IEEE Spectrum’s biomedical engineering blog, featuring the wearable sensors, big data analytics, and implanted devices that enable new ventures in personalized medicine.