Traumatic brain injuries trigger tau pathology and raise the risk of developing dementia later in life. Could tau PET tracers help researchers pick out those on the path to neurodegeneration? Yes, according to a paper in the September 4 Science Translational Medicine. Researchers led by David Sharp at the U.K. Dementia Research Institute, Imperial College London, report that in a small cohort of people who had suffered a single severe head injury decades earlier, flortaucipir PET imaging identified widespread tangles in about one-third of the participants. Another third had localized pathology. Tau tracer uptake correlated with MRI measures of brain injury and axonal damage, as well as with tau in the cerebrospinal fluid. Sharp said the study serves as a proof of principle that flortaucipir can detect tau pathology caused by a single traumatic brain injury (TBI). “Tau PET imaging could allow us to identify patients in whom TBI has triggered a progressive degenerative process,” he told Alzforum.

  • Extensive tau pathology identified in a subset of people after a single TBI.
  • Tracer uptake appears to reflect white-matter damage.
  • It also correlates with cerebrospinal fluid measures of tau.

Gil Rabinovici at the University of California, San Francisco, said this type of study is important, but noted that questions remain about the specificity of tau-tracer binding in cases of TBI. “These are intriguing but very preliminary findings that need to be replicated in larger cohorts and with postmortem validation,” he wrote to Alzforum (full comment below).

Epidemiological studies have found that even a single mild TBI boosts the risk of dementia by 20 percent or more (Apr 2018 news; May 2018 news). Sharp and colleagues previously reported that a single severe TBI can kick-start amyloid plaque deposition, but in a pattern distinct from that seen in Alzheimer’s disease (Feb 2016 news). They wondered if tau tracers, which mostly detect the type of tangles that form in AD, would also pick up a signal after TBI. Two previous studies had found increased tau tracer uptake in football players exposed to repetitive mild TBI, but no one had looked at single head injuries with tau PET (Dickstein et al., 2016; Apr 2019 news). 

Varied Outcomes. Tau tracer uptake (yellow and red) varies from widespread (P1-P8), to patchy (P9-P15), to negligible (P16-P21) in people who had a single severe brain injury decades earlier. [Courtesy of Science Translational Medicine/AAAS.]

To investigate, Sharp and colleagues assembled a cohort of 21 patients who had been treated for TBI at either the University of Glasgow or Imperial College London. They ranged in age from 29 to 72, and two-thirds were men. All had suffered a single moderate to severe TBI, in most cases from a car accident, an average of 32 years before the study. All were given a flortaucipir PET scan and all but two donated CSF. Sharp noted that it is rare to have both CSF and PET data on TBI patients. “This is very valuable data, because there isn’t much of it available,” he said.

First author Nikos Gorgoraptis found that eight participants had widespread tracer uptake throughout their brains. Another seven had patchy uptake in cortical and subcortical regions, in both gray and white matter. The spatial pattern of tau uptake varied between individuals (see image above). “There was a very individual expression of tau pathology, depending on where the injury was,” Sharp told Alzforum. The remaining six people had tracer uptake that matched that seen in 11 healthy age-matched controls. The findings agree with a previous pathological study that found widespread tau tangles in one-third of people who had experienced a single severe TBI years earlier (Johnson et al., 2011). 

The authors correlated tracer uptake with white-matter damage. Regions with high flortaucipir binding had more axonal damage, as measured by reduced white-matter density and aberrant water diffusion in axons on tensor MRI. In addition, the worse the axon damage, the more widespread the tau signal was across the brain.

In the fluid biomarker analysis, more tau tracer uptake correlated with higher total and phosphorylated tau in CSF. It also correlated with higher CSF UCH-L1, a ubiquitin protease that is typically elevated in neurodegenerative disease and may be a marker of neuronal damage. However, the authors found no correlation with CSF neurofilament light, nor with any plasma biomarker. “The fluid biomarker data were disappointing,” noted co-author Henrik Zetterberg at the University of Gothenburg, Sweden. He oversaw the CSF analysis. “I was surprised there was no correlation with NfL, which typically shows up as a biomarker of ongoing neurodegeneration. We need to do more studies in larger cohorts.”

The authors also explored some other correlations, while acknowledging the size of the cohort precluded any solid conclusions. In the eight ApoE4 carriers, tracer uptake inversely correlated with CSF Aβ42. This may suggest some Alzheimer’s-like pathology developing in this group, Zetterberg noted. In the full cohort, tracer uptake did not correlate with any cognitive measure. Sharp noted this study was not powered to detect a cognitive effect. Twelve of the participants had notable deficits in memory, executive function, logical reasoning, and other skills as a result of their TBI, but this did not correlate with tracer binding. “It might be that the contribution of tau pathology to clinical outcome is relatively small, compared to that of the initial injury and brain cell loss,” Zetterberg speculated.

Still, Sharp believes that tau PET imaging could flag people at risk for future cognitive loss. In a previous study, his group found that people who have had a TBI lose more brain volume than healthy controls, even years after the injury, suggesting ongoing degeneration (Cole et al., 2018). Brain atrophy typically correlates with cognitive impairment as well as with tau pathology. Sharp plans to follow this cohort with MRI to find out whether areas of high tracer uptake show more volume loss over time. Zetterberg noted the importance of longitudinal tau imaging in TBI patients. “If we have larger cohorts and use these biomarker tools to follow individuals, we will learn about risk factors and resilience mechanisms,” he said.—Madolyn Bowman Rogers


  1. This study has many strengths, including the focus on long-term survivors of “civilian” TBI, a very common public health problem, and the deep phenotyping of individuals with multiple cognitive measures, MRI sequences, and genetic and fluid-based biomarkers.

    That said, significant caution is required in interpreting the biological underpinnings of flortaucipir PET signal as tau-related. The control cohort is small and there can be considerable variability in background flortaucipir signal in healthy controls (see Baker et al., 2019), such that z-scores based on such a small cohort may not be reliable or reproducible. As the authors acknowledge, there are substantial questions regarding the specificity of flortaucipir, with various sources of “off target” binding described and others still poorly understood. The anatomic patterns of binding are atypical, and don’t clearly conform in my opinion to expectations from neuropathology of CTE or acute TBI. Furthermore, though CTE and AD tangles share biochemical and micro-structural features (paired helical filaments, mix of 3R/4R tau), cryo-EM has revealed significant differences in tertiary structure that may affect PET ligand binding (Falcon et al., 2018).  Indeed autoradiography studies thus far have shown absent to low affinity of flortaucipir for CTE pathology in postmortem tissue.

    In summary, these are intriguing but very preliminary findings that need to be replicated in larger cohorts and with postmortem validation.


    . Effect of off-target binding on 18F-Flortaucipir variability in healthy controls across the lifespan. J Nucl Med. 2019 Mar 15; PubMed.

    . Structures of filaments from Pick's disease reveal a novel tau protein fold. Nature. 2018 Sep;561(7721):137-140. Epub 2018 Aug 29 PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Even Mild Traumatic Brain Injuries Raise Risk for Dementia, Parkinson’s
  2. Do Mild Traumatic Brain Injuries Double the Risk of Dementia?
  3. Traumatic Brain Injury: Aβ Ensues, but Not Quite Alzheimer’s
  4. In Football Players With Possible CTE, Tau PET Reveals Subtle Tau Deposition

Paper Citations

  1. . Cerebral [(18) F]T807/AV1451 retention pattern in clinically probable CTE resembles pathognomonic distribution of CTE tauopathy. Transl Psychiatry. 2016 Sep 27;6(9):e900. PubMed.
  2. . Widespread Tau and Amyloid-Beta Pathology Many Years After a Single Traumatic Brain Injury in Humans. Brain Pathol. 2011 Jun 29; PubMed.
  3. . Spatial patterns of progressive brain volume loss after moderate-severe traumatic brain injury. Brain. 2018 Mar 1;141(3):822-836. PubMed.

Further Reading

Primary Papers

  1. . In vivo detection of cerebral tau pathology in long-term survivors of traumatic brain injury. Sci Transl Med. 2019 Sep 4;11(508) PubMed.