Background:  Cognitive impairment is a prominent symptom of Long COVID (LC), yet the underlying mechanism remains obscure.  We evaluated phosphorylated-Tau-217 (pTau217), a robust marker of Alzheimer’s Disease (AD) type pathology, in LC subjects and controls.

Methods:  pTau217 levels were measured in blood samples collected in a prospective observational study (excluding patients with pre-COVID dementia or cognitive impairment) in LC subjects (persistent or new symptoms lasting ≥3-months post-SARS-CoV-2 infection), Recovered-COVID subjects (no history of any LC symptom), and COVID-negative controls (no clinical or laboratory SARS-CoV-2 infection and negative nucleocapsid antibodies).  Multivariable linear regression models were constructed to evaluate the relationship of log₁₀-transformed pTau217 levels with COVID status adjusting for relevant confounders.  A subset of patients underwent formal neuropsychological testing and clinician interview following Alzheimer’s Disease Research Center protocols.

Results:  We enrolled 209 subjects (N=120 LC, N=59 Recovered-COVID, N=30 COVID-negative controls).   LC subjects were significantly younger (median 60 versus 72 years among Recovered-COVID and 75 years among COVID-negative,  P<0.001), more often hospitalized for acute COVID (18% versus 2% of Recovered-COVID, P<0.001), and had a longer duration of time from index SARS-CoV-2 to pTau217 measurement (2.0 versus 1.5 years among Recovered-COVID, P<0.001).  Overall, pTau217 levels were significantly higher in LC subjects (0.210 pg/mL versus 0.129 pg/mL in Recovered-COVID and 0.189 pg/mL in COVID-negative, P=0.024), and particularly in those with brain fog (median 0.229 pg/mL).  In multivariable linear regression adjusting for age, race, BMI, time from index SARS-CoV-2, and prior hospitalization for acute COVID, LC was associated with 94% (95%CI 32%-185%, P<0.001) higher geometric mean pTau217 levels than those measured in Recovered-COVID and COVID-negative controls.  In a subset of COVID patients that underwent formal neuropsychological testing and expert clinician interview (N=130), patients diagnosed with Alzheimer’s type mild cognitive impairment (MCI-AD) had significantly higher pTau217 levels (0.233 pg/mL) compared to thosed without MCI-AD (0.130 pg/mL, P=0.006).

Discussion: PTau217 levels were significantly higher in LC subjects, particularly among those with brain fog, compared to Recovered-COVID and COVID-negative controls, after accounting for age, race, BMI, time from index infection, and hospitalization for acute COVID.  Significantly higher pTau217 levels were detected in COVID patients clinically diagnosed with MCI-AD.