Comparison of cerebrospinal fluid biomarkers in patients with severe COVID-19 neurological outcomes and Alzheimer's disease

Abstract

Background: COVID-19 induces acute and long-term neurological symptoms. Links between COVID-19 neurological disturbance and Alzheimer's disease (AD) have been hypothesized because neuroinflammation plays a significant role in both diseases. However, it is unknown if COVID-19 patients with neurological disturbance present molecular alterations related to AD pathology. A better understanding of possible molecular links between COVID-19-induced neurological disease and AD would lead to improved patient follow-up and late-onset disease prevention. Here, we analyze early AD biomarkers in a Brazilian cohort of COVID-19 patients with neurological symptoms. We compared COVID-19 patients' neuroinflammatory and AD biomarker levels to controls, amnestic mild cognitive impairment (aMCI), and AD.

Methods: We analyzed cerebrospinal (CSF) biomarkers of neuroinflammation (interleukin-6 (IL6)), amyloid-beta (Aβ) proteinopathy (Aβ42/40), phosphorylated Tau (pTau181), and the neurodegeneration-associated biomarker total Tau in controls (n = 36), COVID-19 patients presenting neurological alterations (n = 35), aMCI (n = 19), and AD patients (n = 20). Comparisons were corrected by possible sex, age, and comorbidities confounding effects. CSF biomarkers were correlated with systemic and neuro-inflammation markers.

Results: We found that severe COVID-19 patients presented higher CSF Tau than controls, comparable to alterations observed in AD patients. However, we did not find changes in CSF Aβ42/40, pTau-181/Aβ42, or Tau/Aβ42 ratios. Severe COVID-19 patients presented higher Tau, Tau/Aβ42, and pTau181/Aβ42 than mild patients. In COVID-19 patients, CSF pro-inflammatory cytokine IL6 and AD biomarkers correlated with systemic inflammatory index (SII).

Conclusions: Collectively, our findings reveal that CSF tau levels are comparably elevated in COVID-19 neurological patients and AD, suggesting ongoing neurodegeneration in COVID-19 neurological disease, but no biomarker alterations related to AD pathology. Furthermore, CNS AD-related biomarker levels in COVID-19 patients change in association with disease severity and systemic inflammation. Considering that inflammation may persist post-COVID, our findings urge the assessment of possible AD-related biomarker changes in COVID-19 survivors with lingering symptoms.

Keywords: ATN; Alzheimer; Biomarkers; Coronavirus; Dementia; Long-COVID; Neuroinflammation.

Fig. 1

Study enrolment and analysis. (AD) Alzheimer’s disease; (aMCI) amnestic mild cognitive impairment; (CSF) Cerebrospinal fluid; (IL6) Interleukin-6; (pTau) phosphorylated Tau.

Fig. 2

COVID-19 patients with neurological symptoms present elevated biomarkers of systemic and neuro-inflammation. COVID-19 patients are compared to cognitively healthy controls, amnestic mild cognitive impairment (aMCI), and Alzheimer’s disease (AD) patients. Data are shown as the median, interquartile range (box), and min-max points (whiskers). Symbols indicate individual patients. Numbers over brackets indicate p-values, Multiple linear regression (MLR) of Box-Cox transformed biomarker levels and considering sex, age (months), and comorbidity (CCI) factors.

Fig. 3

Comparison of Alzheimer’s disease (AD) biomarkers in controls, mild and severe COVID-19 patients, amnestic mild cognitive impairment (aMCI), and AD patients. Data are shown as the median, interquartile range (box), and min-max points (whiskers). Symbols indicate individual patients. Numbers over brackets indicate p-values, Multiple linear regression (MLR) of Box-Cox transformed biomarker levels and considering sex, age (months), and comorbidity (CCI) factors.

Fig. 4

Systemic inflammation correlates with neuroinflammation and changes in AD biomarkers in COVID-19 patients with neurological symptoms. (a) Spearman correlation matrix (n = 31-35 patients). Spearman r’s are shown inside boxes, colors indicate direction, bold values indicate significant correlations. Multiple comparisons are corrected by the false-discovery rate (FDR) method. (b–f) Grey symbols indicate mild disease cases; black symbols represent severe cases or death, (n = 31 or 35). Spearman correlation. Abbreviations: (CSF) cerebrospinal fluid; (IL-6) interleukin-6; (Aβ) amyloid-beta; (pTau 181) tau phosphorylated at threonine 181; (RBC) red blood cells count; (SII) systemic inflammatory index; Days with symptoms are calculated considering the CSF sampling and symptoms’ start dates. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)