Long COVID: plasma levels of neurofilament light chain in mild COVID-19 patients with neurocognitive symptoms

Abstract

It is well known the potential of severe acute respiratory coronavirus type 2 (SARS-CoV-2) infection to induce post-acute sequelae, a condition called Long COVID. This syndrome includes several symptoms, but the central nervous system (CNS) main one is neurocognitive dysfunction. Recently it has been demonstrated the relevance of plasma levels of neurofilament light chain (pNfL), as a biomarker of early involvement of the CNS in COVID-19. The aim of this study was to investigate the relationship between pNfL in patients with post-acute neurocognitive symptoms and the potential of NfL as a prognostic biomarker in these cases. A group of 63 long COVID patients ranging from 18 to 59 years-old were evaluated, submitted to a neurocognitive battery assessment, and subdivided in different groups, according to results. Plasma samples were collected during the long COVID assessment and used for measurement of pNfL with the Single molecule array (SIMOA) assays. Levels of pNfL were significantly higher in long COVID patients with neurocognitive symptoms when compared to HC (p = 0.0031). Long COVID patients with cognitive impairment and fatigue symptoms presented higher pNfL levels when compared to long COVID patients without these symptoms, individually and combined (p = 0.0263, p = 0.0480, and 0.0142, respectively). Correlation analysis showed that levels of cognitive lost and exacerbation of fatigue in the neurocognitive evaluation had a significative correlation with higher pNfL levels (p = 0.0219 and 0.0255, respectively). Previous reports suggested that pNfL levels are related with higher risk of severity and predict lethality of COVID-19. Our findings demonstrate that SARS-CoV-2 infection seems to have a long-term impact on the brain, even in patients who presented mild acute disease. NfL measurements might be useful to identify CNS involvement in long COVID associated with neurocognitive symptoms and to identify who will need continuous monitoring and treatment support.

Fig. 1. Comparison of pNfL levels between HC samples collected before the pandemic and long COVID patients with alteration of at least one neurocognitive test.

The long COVID patients had significantly higher levels of pNfL (p = 0.0031).

Fig. 2. Comparison of pNfL levels (pg/mL) between neurocognitive assessment subgroups.

A Long COVID patients with cognitive impairment and long COVID patients without cognitive impairment. B Long COVID patients with fatigue and long COVID patients without fatigue. C Long COVID patients with anxiety and long COVID patients without anxiety. D Long COVID patients with depression and long COVID patients without depression.

Fig. 3. Comparison of pNfL levels (pg/mL) between long COVID patients with cognitive impairment and fatigue, and long COVID patients without cognitive impairment and fatigue.

The first group had significantly higher levels of pNfL (p = 0.0031).

Fig. 4. Correlation between pNfL levels (pg/mL) among neuropsychological exams.

A Correlation between pNfL levels (pg/mL) and SDMT Score T results. B Correlation between pNfL levels (pg/mL) and FSS results. C Correlation between pNfL levels (pg/mL) and HADS-A results. D Correlation between pNfL levels (pg/mL) and HADS-D results.