Article In Brief
TBI subjects with higher tau deposition on PET showed more severe neurodegeneration, and tau deposition was also linked to damage to the brain's white matter years after experiencing a moderate to severe injury, researchers reported.
A PET scan that measures abnormal tau deposition was sensitive enough to pick up tangle pathology in people who had suffered a moderate to severe head injury decades earlier, suggesting that the initial insult is not a static event but continues to cause damage, months, years, and even decades later.
The findings, reported September 4 in Science Translational Medicine, suggests that PET could be used to study the neurodegenerative process in TBI, identify tangle pathology in patients who suffered a moderate to severe head injury, and ultimately lead to the development of new treatment strategies.
The patients with TBI who continued to have a poor outcome have tangles in white matter and cortical regions, which could explain some of their ongoing cognitive and behavioral problems, said the senior author of the study, David J. Sharp, MD, professor and consultant neurologist at the computational, cognitive and clinical neuroimaging laboratory at Imperial College London.
“We study the long-term effects of traumatic brain injury and one of the frustrations is that it is hard to disentangle the ongoing and progressive problems from the TBI,” said Dr. Sharp. “We just don't know what links back to the original injury. We need objective measures to identify the underlying disease process. Having a way to measure abnormal tau deposition gives us a marker of neurodegeneration and should help us better manage our patients.”
Study Methods, Findings
To explore that further, the researchers aimed to compare the PET tau ligand flortaucipir F 18 to structural MRI, markers of neurodegeneration in cerebrospinal fluid (CSF), and neuropsychological measures of cognition (the Mini-Mental State Examination, or MMSE) in patients who had experienced a single moderate to severe TBI years, or decades, earlier. They also recruited controls without a history of TBI.
The study included 21 people (seven women and 14 men, 29 to 72 years old) who had a history of a single and serious TBI. Nineteen had been seen at the Institute of Health and Wellbeing at the University of Glasgow and were part of a longitudinal study in the head injury research group there. The other two were recruited from the TBI clinic at Imperial College Healthcare NHS Trust in London.
The majority of the study participants (18) were injured in a traffic accident; two had been assaulted; and one fell from a high spot. The examinations for this study took place 18 to 51 years after the initial injury. The median Glasgow Outcome Scale-Extended score was 6. (Three is the worst and 15 is the best outcome.) They also had other clinical measures, including the MMSE that were done approximately 16 years after their injury and an equal amount of time before the tests obtained for this study. And they had genetic information, including their APOE status.
Most of the study participants (19) had focal lesions that were found on an MRI. One had a cerebellar lesion, also seen on structural MRI.
They divided the group into a “good recovery group” and a disabled group. Twelve participants were in the disabled group and nine were in the good recovery group. The scientists had access to longitudinal clinical data on nine patients in the disabled group and six in the good recovery group.
The 11 healthy controls were part of the Institute of Health and Wellbeing's head injury research group or from the National Institute for Health Research program. They were similar in age and education as the TBI participants, although they reported that the disabled TBI group had a lower estimated premorbid intelligence score on the Wechsler Test of Adult Reading compared to controls and those who made a good recovery.
Not surprisingly, the group identified as disabled had a long post-injury history of cognitive problems, including executive function, processing speed, inhibition, and verbal and visual memory. The only clinical difference between the good recovery group and the healthy controls was on measures of executive function and apathy. The disabled group seemed to suffer progressive cognitive decline on the MMSE, losing about a point a year, compared with the good recovery group that showed improvements in the MMSE scores over time.
All study participants underwent a tau PET scan with the flortaucipir ligand. This ligand binds phosphorylated tau. Some TBI participants had increased binding in both gray and white matter while others were indistinguishable from the healthy controls.
The researchers measured the volume of the binding in voxels that range from zero to almost 16,000. About two-thirds of the TBI participants had an average of 841 voxels above zero compared with the controls where the median number of voxels was four (a range of zero to 253.) Eight of the TBI participants (38 percent) had over 2000 voxels and another seven (33 percent) had anywhere from 254 to 1999 voxels.
Flortaucipir binding was significantly higher in TBI participants (in both subgroups) than in healthy controls in the right lateral occipital cortex, said Dr. Sharp. There was no significant difference in flortaucipir binding in the disabled and good outcome TBI subgroups, and why that is remains a question for future studies.
Dr. Sharp and his colleagues also looked at brain regions affected in Alzheimer's to see if there was a signal that puts them at increased risk for dementia. They looked for differences in tau PET binding in the entorhinal, perirhinal, and parahippocampal cortices and the hippocampus, and found nothing that would distinguish the TBI group from the healthy controls.
The team also had access to CSF and plasma biomarkers of neurodegeneration, including total tau, phosphorylated tau, ubiquitin, amyloid-beta 42, neurofilament light, glial fibrillary acidic protein (GFAP), and protein S100.
In TBI, the total tau levels in CSF were significantly correlated with flortaucipir PET binding in the cortical gray matter but not in the white matter. But phosphorylated tau was correlated with flortaucipir binding in the cerebral white matter but not in the gray matter.
Plasma T-tau concentration did not correlate with flortaucipir binding in either TBI or healthy controls.
They did find abundant axonal injury in the white matter, which Dr. Sharp said was probably the catalyst that triggers inflammation and neurodegeneration.
“We know from MRI that the brain is shrinking, and post-mortem we see a lot of inflammation and damage in white matter as well as amyloid and tau pathology,” Dr. Sharp explained. “This is our first attempt to use the ligand to look at areas with abnormal tau over time.”
Scientists want to understand why the clinical outcomes are so variable. All of the TBI patients in this study had significant brain injuries and some improved and others did not.
“We did not find a clear correlation between the increased tau signal and the ongoing cognitive problems,” added Dr. Sharp. He said the study was not powered to address this question. The numbers were too small.
Still, he said, “the findings unlock the door to a new diagnostic pathway. More work needs to be done to understand the natural history of traumatic brain injuries.”
Tau PET would be a useful technique to look at early pathological signs and understand how this might play out in the clinical realm, he added.
Commenting on the study, Elisa R. Zanier, MD, director of the laboratory of acute brain injury and therapeutic strategies at the Mario Negri Institute for Pharmacological Research in Milano, Italy, said: “We need to change the way we look at traumatic brain injuries. These types of neuroimaging studies combining tau-PET and MRI are fundamental in shedding light on the progressive nature and long-term pathological effects of a single moderate to severe TBI.”
“Until recently, TBI was regarded as a disruptive but static event,” she said. “Scientists focused on the acute stages where all the gains were made and little attention was paid to the progressive nature of this disease. The field will advance now that we understand that there are long-term pathological and clinical effects. We now have a much wider window of time to help people but we need to understand the drivers that lead to neurodegeneration so we can develop new treatment strategies.”
Dr. Zanier published a study in Brain last year showing that TBI in mice generates a neurotoxic form of tau that propagates throughout the brain like a prion. “This new study in Science Translational Medicine also suggests that TBI-tau pathology might spread from local sites to distant regions along neural networks,” she said.
As the scientists noted in the paper, “the study was not adequately powered to assess the relationship between flortaucipir load and clinical outcome, thus leaving the question open as to whether tau pathology is pathogenetic in driving cognitive decline or is an epiphenomenon of late neurodegeneration.”
“Late and very late functional deterioration occurs in some patients after single moderate-severe TBI,” added Richard Greenwood, MD, consultant member and lead clinician of the novel Acute Brain Injury Service at The National Hospital for Neurology and Neurosurgery and University College London Hospitals, “This is often likely to be at least in part the result of a failure to optimize disability management, but whether that is the explanation for the increased long-term risk of dementia in later life after TBI seems unlikely.”
“If confirmed, this finding is likely to encourage clinical investigation and monitoring of late deterioration after single moderate-severe TBI, whether due to degenerative, microvascular, or other mechanisms,” Dr. Greenwood continued. “This in turn will help lead to and a wider understanding and implementation of a variety of other important interventions that optimize disability management of TBI long-term rescue many survivors from a variety of backwaters in which they find themselves.”
“This study is the first to show that the levels of tau—measured by a radioisotope binding to this protein—are elevated decades after a single TBI, compared to age-matched controls,” said Olli Tenovuo, MD, professor of neurotraumatology and head of the Turku Brain Injury Centre at Turku University Hospital and University of Turku in Finland.
“Moreover, this elevation was connected with other signs of brain damage: Structure of the white matter and levels of brain injury biomarkers in the cerebrospinal fluid. The study also raises some major questions. Why was the only region where a significant result was seen compared to controls on the right occipital lobe, a region that is usually relatively spared in TBI? Why was there no significant difference in binding between those with a good and bad outcome after a TBI?”
The study “still leaves open how the accumulation of this protein affects the long-term outcome and associates with the clinical symptoms. Importantly, it still leaves unanswered if there are people who have a specific risk for adverse long-term outcome, and how could we recognize these persons, and maybe find preventive interventions.”
Drs. Sharp, Zanier, Tenovuo, and Greenwood had no disclosures.