Sleep Disturbance Compromises Adaptive and Innate Immune Indices in Long COVID: T-cell Responses to SARS-CoV-2 Variants, Activation of Pro-inflammatory and Pro-resolving Pathways, and Glucocorticoid Sensitivity

Introduction: Pre-existing and/or new onset sleep disturbance is common in about 30% of individuals with Long COVID (LC). Sleep disturbance compromises optimal functioning of the adaptive and innate immune system, thereby worsening infection outcomes and hindering full recovery from infections.

Objective: To investigate whether sleep disturbance prior to SARS-CoV-2 infection in LC was associated with (a) compromised T cell responses to SARS-CoV-2 variants, (b) activation of inflammatory pathways, (c) compromised glucocorticoids (GC) sensitivity, and (d) dysregulated inflammatory resolution pathways.

Methods: Seventy-four participants, enrolled into the acutely infected or uninfected cohort of the RECOVER clinical protocol, were selected for this study. Based on the Long COVID Research Index, participants were categorized into groups of Likely LC (N=19), Possible LC (N=24), and No LC (N=31), matched by age and sex. Reports on the presence of sleep problems (i.e., stopping breathing, trouble falling asleep, nighttime awakenings, trouble staying awake during the day) in the year prior to infection were utilized for this study. Biospecimens (PBMC, plasma) were analyzed from the 6- or 12-month collection timepoint following infection. Analysis included (a) IFN-gamma expression in CD4 and CD8 T cells in response to stimulation with Omicron spike overlapping peptide variants BA.5 and XBB, (b) IL-6, TNF, and COX-2 expression in LPS-stimulated and unstimulated monocytes, (c) sensitivity of monocytes to the synthetic GC dexamethasone, and (d) plasma levels of pro-inflammatory and specialized pro-resolving mediators (SPMs), including the precursors of lipoxins and D-/E-series resolvins. For comparisons of the effect of pre-existing sleep disturbance on each assay in the Likely LC group compared to the Possible LC and No LC groups, a log-transformed linear regression was used with a covariate for LC status and pre-existing sleep disturbance and an interaction between the two covariates. Estimates were generated to compare the Likely LC group to the No LC or combined Possible/No LC group.

Results: Likely LC participants with sleep disturbance showed nominally lower IFN-gamma-positive BA.5- and XBB-stimulated CD4 T cells compared to those without sleep disturbance (p=0.068 and 0.054, resp.), though this effect did not differ from Possible/No LC. Likely LC participants with sleep disturbance further showed nominally less suppression of IL-6 by GC compared to those without sleep disturbance (p=0.056), with the same directional trend for TNF and COX-2. This effect was larger in Likely LC compared to the combined group of Possible/No LC (p=0.010, 0.034, 0.099 for IL-6, TNF, COX-2, resp.). Likely LC participants with sleep disturbance further had nominally higher levels of prostaglandin F2alpha and resolvin D2 than those without sleep disturbance (p=0.036 and 0.062, resp.), and this sleep effect was nominally larger compared to the No LC group (p=0.084 and 0.032, resp.).

Conclusion/Discussion: Findings suggest that the presence of pre-existing sleep disturbance compromises certain immune indices in LC. In particular, sleep disturbance impairs the ability of GCs to suppress inflammation to a greater extent in Likely LC compared to Possible or No LC, indicating reduced inflammatory control by GC. Research on the type of sleep disturbance (e.g., insomnia, hypersomnia, breathing-related problems) driving GC sensitivity and other changes will help to identify subtype-specific intervention targets.

Key Topics:

• Assay and in vitro studies to gain mechanistic insights