Amy Carpenter Aquino
April 2023—Traumatic brain injury triage in the emergency department is badly in need of biomarkers—and ones that can change practice.
“If biomarkers don’t change practice, they’re a waste of time,” said W. Frank Peacock IV, MD, professor of emergency medicine, vice chair of research, and research director, Department of Emergency Medicine, Baylor College of Medicine.
“We have no biomarkers now that tell me who has damage to their brain,” he said in an AACC session last year. “We have a couple that say, ‘I don’t need a scan,’ but there are a lot of people with negative CT scans who have subsequent brain injuries. We’re only sorting that out now.”
Dr. Wu
Dr. Peacock painted a bleak picture of the biomarker practice gap in traumatic brain injury. His co-speakers, Alan Wu, PhD, D(ABCC), of Zuckerberg San Francisco General Hospital, and Pradip Datta, PhD, D(ABCC), of Siemens Healthineers, talked about short- and long-term laboratory diagnostics for TBI today and to come. (For Dr. Datta’s rundown, see “Traumatic brain injury biomarkers”. )
If the patient’s TBI is not a clear emergency, and most are not, “I have to decide what to do,” Dr. Peacock said. The American College of Emergency Physicians advises getting a CT scan “in almost every case,” he said. A neurology consult follows a positive scan. If the CT scan is negative, “you go into the black box,” which Dr. Peacock describes as “I don’t have a clue, but I know what to do: I’m going to send you home.” The worry, he said, is post-concussive syndrome. The ACEP’s level A guidelines include a long list of symptoms including headache. “I’ve never seen anybody with a TBI who didn’t have a headache,” Dr. Peacock said, so patients with mild TBI automatically get a head CT scan, “and 95 percent of them are negative.”
ACEP level B guidelines go further, he said. If a patient falls more than three feet, for example—“and most people are taller than three feet”—they get a CT scan.
“You can see where I’m going with this: Everybody’s getting a CT scan.”
The ALERT-TBI study of almost 2,000 patients found a 98 percent sensitivity and high negative predictive value (0.996) for the ubiquitin C-terminal hydrolase-L1 (UCH-L1) and glial fibrillary acidic protein (GFAP) tests in ruling out the need for a CT scan (Bazarian JJ, et al. Lancet Neurol. 2018;17[9]:782–789). “If it’s negative, I am done,” Dr. Peacock said of the utility of the assay results. The specificity is low: 0.364 for patients with a Glasgow Coma Scale score of nine to 15; 0.367 for GCS 14–15; 0.344 for patients with neurologically manageable lesions. Even so, “I can get rid of a third of CT scans,” he said. If a patient with a TBI has assay results above the cutoff values, “that’s a positive, and you’re off to the CT scanner.” If the results are below the cutoff values, the patient can go home.
“What I’m always interested in are the people whose Glasgow Coma scores are 14 or 15. I could even go down to 13,” Dr. Peacock said. “But below that, you’re getting a CT scan.”
Time from injury to blood draw is important, he said, and in the ALERT-TBI study, it’s 3.2 hours. “So if you show up one hour after your injury, what does it mean? It means I don’t have a clue. You can’t trust these labs, and one of the problems is the literature is sloppy about this,” with some studies saying within one hour of injury “this works,” he said. However, “if you really look into the data, it’s always three hours,” because there is delay in when patients arrive in the ED for most major events, and getting a patient involved in a clinical trial adds another hour. “This is just the realities of studies.”
For patients with a negative CT scan, there are no guidelines for what to tell them to do, Dr. Peacock said. “We’re winging it.” But post-concussive syndrome and its symptoms and aftermath—in some cases job loss—are the problem.
Dr. Peacock said the sensitivity of an ED physician knowing who has a TBI and who does not is eight percent (Korley FK, et al. Acad Emerg Med. 2019;26[12]:1384–1387). “It’s the worst number I’ve ever seen published. . . . It implies emergency docs don’t have a clue, and the reality is we don’t. I cannot tell who has a TBI. In fact, the patients can’t tell—that’s the other reality.” Symptoms sometimes don’t show up for a couple of weeks.
“What we’re really talking about is acute traumatic encephalopathy,” Dr. Peacock said, known more commonly as chronic traumatic encephalopathy. “It’s not a chronic problem with no beginning,” he said of CTE. “The beginning is acute traumatic encephalopathy,” and there is no ability to diagnose it at this time. Before markers, there was no way to detect non-ST-elevation myocardial infarction, but there was subclinical damage, leading to heart failure five years later. “It’s the same thing here,” he said. “Five years later after multiple hits, you’re going to have chronic traumatic encephalopathy.” These are patients who may have post-concussive syndrome symptoms. They’ll have post-traumatic biomarker abnormalities and may have abnormal functional testing.
Dr. Peacock and colleagues evaluated the use of three biomarker proteins in adult patients examined for head injury at two emergency departments within the Johns Hopkins Hospital system and enrolled in the prospective observational HeadSMART study. All had a CT scan and a GCS score of 13 to 15. Two control cohorts were used. They found that neurogranin (NRGN) and neuron-specific enolase (NSE) were elevated and that metallothionein 3 (MT3) decreased in mild TBI patients compared with controls (Peacock WF IV, et al. Front Neurol. Published online Nov. 30, 2017. doi:10.3389/fneur.2017.00641).
“These are three markers that have the characteristics of something that would be useful for ATE,” he said.
TBI biomarkers for short-term injury focus on proteins, and autoantibody production is the focus of long-term injury, said Dr. Wu, co-core lab director at Zuckerberg San Francisco General Hospital and professor of laboratory medicine, UCSF School of Medicine.
Autoantibodies are released at the same time as other TBI biomarkers, he said. “We have these biomarkers that are normally sequestered in the central nervous system, that don’t circulate in the blood. And when they show up in blood because we have a disruption of the blood-brain barrier, the theory is that our body recognizes it as foreign and it tends to try to protect itself by raising antibodies.”
Some antibodies are protective and others are not. Those that are not should be the focus, he said, “because pathology results from these autoantibody responses,” not unlike what is seen in type one diabetes, rheumatoid arthritis, or other autoimmune-mediated diseases.
TBI risk stratification is an unmet need, Dr. Wu said. Even among the people who have mild symptoms, “there’s still a need to look for prognosis. And predicting short- and long-term consequences, even with people who have had imaging studies, is the key to moving forward.”
As of now, he said, “neurologists have no tools” to advise whether it is safe to return to sports or work, or to determine whether the patient will have long-term difficulties. “So a lot of gaps in diagnosis and risk stratification and prediction and pathophysiology” of TBIs. “And most importantly, the therapeutics are lagging,” he said. “But because we are biomarker people, it is my hypothesis that you first have to have tools, and then you can start evaluating next-generation pharmacology.”
The pathophysiology of TBIs is complex, Dr. Wu said. “First there’s the acute injury, and the mechanical damage is what we see in the breakdown of the blood barrier and circulating biomarkers.” He calls it the tip of the iceberg. The chronic phase is the bottom of the iceberg, he said, and “we need to know, Is there anything below that iceberg? This is where we have to look at the immune response.”
The central nervous system has what is termed “immune privilege,” he said, drawing from a chapter in a book he and Dr. Peacock co-edited (Kobeissy FH, et al. Autoantibodies in central nervous system trauma: new frontiers for diagnosis and prognosis biomarkers. In: Wu AHB, Peacock WF, eds. Biomarkers for Traumatic Brain Injury. Elsevier Science; 2020:431–445). It means “we have a separate layer of cells, of membranes, that protect us from ourselves. Circulating antibodies don’t enter the central nervous system because the CNS is privileged. And rightfully so,” he said. “We want to protect brain function, and nature and evolution have allowed us to have extra layers of protection.” He describes injury as immune privilege being revoked, “because now you see the breakdown of the blood-brain barrier and the release of macrophages and inflammatory agents that release harmful cytokines and stimulate antibody protection.”
Much research is underway to understand long COVID-19, and the question, Dr. Wu said, is what can be taken from it for understanding TBI. Interleukin-6, which is important in diagnosing concussions (Edwards KA, et al. BMC Neurol. 2020;20[1]:209), is a “harbinger of the kind of cytokine release storm and damage to the neuronal system.”
He and UCSF colleagues have studied IL-6 after COVID. “At least in the reasonable short term, levels do correlate with the disease severity and recovery. But when you have sequelae that come about six months or nine months later, can we use this model to also predict for people who have this TBI sentinel event? I don’t know the answer to that question, but it’s something to consider,” Dr. Wu said (Fitch BA, Lynch KL. Clinical utility of an automated IL-6 immunoassay for COVID19 prognosis and follow-up. Abstract presented at: 2021 AACC Annual Meeting; Sept. 26-30, 2021; Atlanta. Session B-120).
Controlled animal and human studies have demonstrated the detrimental effect of autoantibodies to neuroproteins. One study in which mice were subjected to controlled spinal cord injury demonstrated that autoantibodies potentiate neuronal injury, Dr. Wu said. Wild-type mice were compared with C3-/- knockout mice that don’t have functioning B cells. The data indicated that B cells, through the production of antibodies, affect pathology in such injuries. “If you can’t produce an autoimmune response, does that protect you against injury? This morphologic study seems to suggest the answer is yes,” Dr. Wu said (Ankeny DP, et al. J Clin Invest. 2009;119[10]:2990–2999).
Hypopituitarism is common after TBI and persists for long periods, he said. In a prospective follow-up study of anterior pituitary function in patients with mild, moderate, and severe TBI, Tanriverdi, et al., reported finding elevated pituitary hormone levels up to five years later. Head injury is expected to affect hormonal function, Dr. Wu said. “What I didn’t realize was that it also elicits an autoimmune response” (Tanriverdi F, et al. J Neurotrauma. 2013;30[16]:1426–1433).
In a study of GFAP antibodies, GFAP levels were found to be higher in individuals who have TBI. More interesting, he said, is the higher incidence of GFAP antibodies in those who have poor Glasgow Outcome Scale-Extended scores at six months, relative to those who have higher GOS-E scores and more favorable outcomes. “So there are links to antibody productions” (Zhang Z, et al. PLoS One. 2014;9[3]:e92698).
A study of the consequences of repeated blood-brain barrier disruptions in college football players found changes in S100B autoantibodies. “This was interesting because they had a baseline value—a preseason and postseason value,” Dr. Wu said. Some of the 67 athletes studied had no increase in S100B levels relative to preseason and lower levels of cognitive defects, but a handful had an increase in S100B and neurocognitive changes. “So we have markers that could perhaps predict whether someone is going to have long-term sequelae” (Marchi N, et al. PLoS One. 2013;8[3]:e56805).
“How do we reduce long-term risk?” Dr. Wu asked. One study of mice reported treatment of spinal cord injury with a glycoengineered anti-muCD20 antibody that reduces the development of inflammation and tissue injury by altering the immune system associated with the injury (Casili G, et al. Neurotherapeutics. 2016;13[4]:880–894). Another investigated the role of complement in triggering neuroinflammation after TBI (Alawieh A, et al. J Neurosci. 2018;38[10]:2519–2532).
Intravenous immunoglobulin therapy blocks cytokine release and complement-mediated damage. For patients with TBI, Dr. Wu said, “maybe this should be added to the armamentarium, but you don’t necessarily want to give this to everybody.”
Amy Carpenter Aquino is CAP TODAY senior editor.