Treg and Oxidative Stress Biomarker Profiles in Long COVID and Myalgic Encephalomyelitis/Chronic Fatigue Syndrome

Chronic immune dysregulation is the possible pathophysiological mechanism in myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) and long coronavirus infectious disease (long COVID, LC). However, a detailed characterization of regulatory and effector T-cell subsets in these conditions remains limited. To assess and compare T-cell subsets in relation to serum biomarkers of chronic activation and oxidative/nitrosative stress in an attempt to differentiate between LC and ME/CFS patients. Peripheral blood mononuclear cells (PMNC) from LC patients (n = 89), ME/CFS patients (n = 125), and healthy controls (HC, n = 80) were analyzed using multiparameter flow cytometry. The frequencies of naïve (CD45RA+), effector (CD45RA-), CD31+, CD39+, CD73+, GARP+, PI16+, CD15s/CD161+, and CD147+ Treg subsets were determined in relation to CD4+ and CD8+ T-cell differentiation, activation, and effector subsets. Catalase (CAT), superoxide dismutase 1 (SOD1), glutathione peroxidase 1 (GPX1), malondialdehyde (MDA), nitric oxide synthase 1 and 3 (NOS1, NOS3), zonulin, lipopolysaccharide (LPS), immunoglobulin G (IgG) against lipopolysaccharide (IgG-LPS), Toll-like receptor 4 (TLR4), indoleamine 2,3-dioxygenase (IDO), and Dickkopf-related protein 1 (DKK1) were measured in serum by ELISA. Clinical symptom scores (HAMA, HAMD, FF, dePol) were additionally assessed. The immune-serum biomarkers distinguishing LC, ME/CFS, and HC comprised Th17, CD15s+CD161−eTreg, and CD161+ Treg. The five-marker signature retained significance after adjustment for age, BMI, and sex. Altered T-cell subsets were strongly associated with markers of oxidative (GPX1) and nitrosative (NOS3) stress but not directly with symptom severity. ME/CFS showed a strong separation from HC across Random Forest (AUC = 0.907) and logistic regression (AUC = 0.924). The modest distinction of LC from HC (AUC = 0.727 and 0.577 respectively) possibly reflected important endotypic diversity, though differences in disease duration, vaccination history, biomarker selection, and other cohort characteristics might add to it. MDA was an independent predictor for three-class classification, ME/CFS vs. HC, and LC vs. HC pairwise comparisons, while GPX1 discriminated between ME/CFS and LC. In spite of shared biomarkers of oxidative stress and T-cell response dysregulation, our results support the hypothesis that LC and ME/CFS differ in the extent of cellular damage and the exhaustion of compensatory mechanisms.