Brain-injury and Alzheimer’s disease biomarkers are elevated in patients with suspected infection and physiological derangement: importance for context-specific interpretation of Alzheimer’s biomarkers

Abstract Following acute infective illness, patients frequently exhibit neurological symptoms, with persistent neurological symptoms commonly observed following severe infection. However, the association between systemic infection and the concentration of blood based-biomarkers of brain injury and the relationship between these and markers of the host response to infection, is poorly characterised in the literature. Further, the association between acute illness and the Alzheimer’s disease associated biomarker phosphorylated-tau-217 is unknown. In acute samples from 26 patients attending the emergency department with suspected sepsis (clinically suspected or proven infection, a National Early Warning Score or National Early Warning Score 2 score ≥ 5), the levels of serum biomarkers of brain injury neurofilament light, glial fibrillary acidic protein, total tau and ubiquitin C-terminal hydrolase L1 and phosphorylated-tau-217 were compared to age and sex matched non-infected controls, with further assessment of the correlation between the biomarker levels and cytokine profiles. Phosphorylated-tau-217 levels were additionally compared to a positive control group of patients with diagnosed Alzheimer’s disease. Both total tau (p=0.006, Wilcoxon rank-sum test) and neurofilament light (p=0.044, Wilcoxon rank-sum test) levels were significantly higher in patients with suspected sepsis as compared to controls, with no significant differences in levels of glial fibrillary acidic protein or ubiquitin C-terminal hydrolase L1. Within suspected sepsis patients, serum total tau levels were associated with multiple cytokines and a summary cytokine score (Spearman’s rank correlation coefficient ρ = 0.65, p0.001). There were significantly higher phosphorylated-tau-217 levels in suspected sepsis patients as compared to non-infected controls (p0.001, Dunn-Kruskal-Wallis test), with no significant difference compared to Alzheimer’s disease controls (p=0.118, Dunn-Kruskal-Wallis test). Of patients with suspected sepsis, 29% had a p-tau-217 level classified as high (0.63 pg/ml) with a further 17% classified as intermediate (0.4-0.63 pg/ml). In conclusion, we have identified elevated levels of total tau and neurofilament light compared to age and sex matched controls, along with significant correlations between these tau levels and cytokine levels. Additionally, we observed elevated levels of phosphorylated-tau-217 in the patient cohort, with levels comparable to those seen in Alzheimer's disease patients. Further analysis is required to replicate the findings of this study in larger cohorts. However, the results suggest a potentially extracranial source of tau expression in the context of infection or physiological stress. Given the potential for acute illness to influence phosphorylated-tau-217 levels, our results raise important considerations regarding the interpretation of phosphorylated-tau-217 as a diagnostic marker for Alzheimer’s disease in patients with active infection.