Tau Positron-Emission Tomography in Former National Football League Players

Abstract

Background: Chronic traumatic encephalopathy (CTE) is a neurodegenerative disease that has been associated with a history of repetitive head impacts. The neuropathological diagnosis is based on a specific pattern of tau deposition with minimal amyloid-beta deposition that differs from other disorders, including Alzheimer's disease. The feasibility of detecting tau and amyloid deposition in the brains of living persons at risk for CTE has not been well studied.

Methods: We used flortaucipir positron-emission tomography (PET) and florbetapir PET to measure deposition of tau and amyloid-beta, respectively, in the brains of former National Football League (NFL) players with cognitive and neuropsychiatric symptoms and in asymptomatic men with no history of traumatic brain injury. Automated image-analysis algorithms were used to compare the regional tau standardized uptake value ratio (SUVR, the ratio of radioactivity in a cerebral region to that in the cerebellum as a reference) between the two groups and to explore the associations of SUVR with symptom severity and with years of football play in the former-player group.

Results: A total of 26 former players and 31 controls were included in the analysis. The mean flortaucipir SUVR was higher among former players than among controls in three regions of the brain: bilateral superior frontal (1.09 vs. 0.98; adjusted mean difference, 0.13; 95% confidence interval [CI], 0.06 to 0.20; P<0.001), bilateral medial temporal (1.23 vs. 1.12; adjusted mean difference, 0.13; 95% CI, 0.05 to 0.21; P<0.001), and left parietal (1.12 vs. 1.01; adjusted mean difference, 0.12; 95% CI, 0.05 to 0.20; P = 0.002). In exploratory analyses, the correlation coefficients in these three regions between the SUVRs and years of play were 0.58 (95% CI, 0.25 to 0.79), 0.45 (95% CI, 0.07 to 0.71), and 0.50 (95% CI, 0.14 to 0.74), respectively. There was no association between tau deposition and scores on cognitive and neuropsychiatric tests. Only one former player had levels of amyloid-beta deposition similar to those in persons with Alzheimer's disease.

Conclusions: A group of living former NFL players with cognitive and neuropsychiatric symptoms had higher tau levels measured by PET than controls in brain regions that are affected by CTE and did not have elevated amyloid-beta levels. Further studies are needed to determine whether elevated CTE-associated tau can be detected in individual persons. (Funded by Avid Radiopharmaceuticals and others.).

Figure 1. Statistical Parametric Maps of Flortaucipir Positron-Emission Tomography (PET).

The maps show voxels with higher regional:cerebellar gray-matter flortaucipir standard uptake value ratios (SUVRs) among former National Football League (NFL) players than among controls. The arrow indicates the point along the color gradient above which the differences between the groups, indicated by the t value, are significant (P<0.005, uncorrected for multiple regional comparisons). The numbers below each spatially standardized horizontal brain section correspond to the distance in millimeters above (positive numbers) or below (negative numbers) a plane through the anterior and posterior commissures. The left hemisphere is on the right.

Figure 2. Three-Dimensional Stereotactic Surface Projection Maps of Flortaucipir PET.

Higher flortaucipir SUVRs in the former-player group than in the control group were found in the bilateral superior frontal (Panel A), bilateral medial temporal (Panel B), and left parietal (Panel C) regions of the brain. The regions shown in red in these surface projection images correspond to the statistical parametric maps after restriction of the map to those clusters of at least 100 contiguous voxels associated with higher regional:cerebellar gray-matter flortaucipir SUVRs in the former-player group than in the control group (P<0.005, uncorrected for multiple regional comparisons).

Figure 3. Statistical Evaluation of Regional Flortaucipir SUVRs.

The box plots in the top row show the distributions of statistically characterized regional SUVRs in 26 former NFL players and 31 controls in the bilateral superior frontal, bilateral medial temporal, and left parietal regions of the brain. The horizontal line in each box represents the median, the lower and upper boundaries of the boxes the interquartile range, and the ends of the I bars 1.5 times the interquartile range. The scatterplots in the bottom row show the relationship between regional flortaucipir SUVRs and total years playing tackle football. The gray-shaded areas in the bottom row represent the standard error of the regression line (blue). The r values are based on Spearman correlations of the residuals from age-adjusted partial correlations.