Prior Traumatic Brain Injury and Alzheimer Disease Blood Biomarkers

Importance: Traumatic brain injury (TBI) is a risk factor for dementia and is known to impact levels of several Alzheimer disease (AD) blood biomarkers. The plasma phosphorylated tau 217 (p-tau217)/amyloid-β 42 (Aβ42) ratio has been reported to be 90% accurate for the detection of brain amyloid in civilian cohorts. Objective: To evaluate the accuracy of emerging AD blood biomarkers in veterans with and without a TBI history. Design, Setting, and Participants: This cross-sectional cohort study of a diagnostic test assessed the performance of the US Food and Drug Administration-approved plasma p-tau217/Aβ42 ratio and plasma levels of p-tau217 and Aβ42/40 ratio for detecting brain Aβ positivity (eg, amyloid-positron emission tomography [PET] consensus visual read). The accuracy of biomarkers was compared in veterans without TBI, those with TBI with loss of consciousness (LOC) for 0 to 5 minutes, and those with TBI with LOC for greater than 5 minutes. Existing data and banked plasma from the Alzheimer Disease Neuroimaging Initiative Department of Defense (ADNI-DOD) study were used. Years of enrollment were 2013 to 2020. Included in this study were Vietnam War veterans without dementia (cognitively unimpaired and those with mild cognitive impairment) who had amyloid-PET and concurrently collected banked plasma available for analysis. Plasma biomarker analysis took place from 2024 to 2025, and data analysis was performed from July 2025 to February 2026. Exposures: AD blood biomarkers and TBI history. Main Outcomes and Measures: p-Tau217/Aβ42 accuracy in the detection of amyloid-PET consensus visual read positivity across groups with/without TBI exposure. Results: In 272 participants, mean (SD) age was 70 (4.5) years, 270 were male (99.3%), and 83 (30.5%) were amyloid-PET positive. The plasma p-tau217/Aβ42 ratio was highly accurate in veterans with no TBI (90%; 95% CI, 84%-96%) but not in veterans with TBI with LOC for 0 to 5 minutes (78% accuracy; 95% CI, 69%-87%; P = .03 vs no TBI), nor in veterans with TBI with LOC greater than 5 minutes (63% accuracy; 95% CI, 53%-73%; P < .001 vs no TBI). Results were similar for plasma p-tau217 alone and plasma Aβ42/40 ratio. Results were also similar after excluding veterans with TBI within the past 10 years or when amyloid-PET positivity was defined using a quantitative threshold rather than consensus visual read. Conclusions and Relevance: Results of this cross-sectional cohort study of a diagnostic test suggest that among individuals who are cognitively unimpaired or have mild cognitive impairment and a TBI history, the p-tau217/Aβ42 ratio test may miss over half of amyloid-PET-positive cases. Caution is advised in interpreting AD blood test results in this context.