Functional Autoantibodies Targeting G-Protein-Coupled Receptors and Their Clinical Phenotype in Patients with Long-COVID

Figures

• 

  Figure A1. Variance Inflation Factor (VIF) values of GPCR-fAAb and self-reported clinical symptoms; most of the values are around 1, suggests little to no multicollinearity. Except for hair loss and AT1-fAAb and MAS-fAAb, there are no higher values (usually above 5 or 10) and this indicates the potential absence of problems with redundant information among the predictors that might affect the reliability of the regression estimates.

• 

  Figure 1. Seropositivity of GPCR-fAAb (A) and distinct patterns of combined seropositivity (B) of patients with LC. The majority demonstrated seropositivity for the following antibodies: β2-fAAb (92.8%), M2-fAAb (87.1%), AT1-fAAb (85.6%), and MAS-fAAb (85.6%). By contrast, α1-fAAb was present in 16.0%, while Nociceptin-fAAb and ET-A-fAAb were observed in 14.4% and 2.6%, respectively. The most frequent combination was M2-, MAS-, AT1- and β2-fAAb (47.1% in females and 50% in males). The combination of a seropositivity of Nociceptin-, AT1-, MAS- and M2-fAAb was observed in 13.2% of females and 9% of males LC patients. Seropositivity for a combination of β2-, M2-, AT1- and MAS-fAAb was observed in 7.5% of females and 12.5% of males; β2-M2-fAAb double seropositivity was observed in 7.5% of females and 5.6% of males. The least common combination was β2-, AT1- and MAS-fAAb (7.5% of females, 3.4% of males).

• 

  Figure 2. Self-reported clinical symptoms of the patients with LC. The most commonly self-reported symptoms were ‘fatigue/post-exertional malaise (PEM)’ (female: 88.6% and male: 88.6%) and ‘lack of concentration’ (female: 88.6% and male: 77.2%). Additional self-reported symptoms were ‘dizziness’ (female: 49% and male: 56.8%), ‘Postural Orthostatic Tachycardia Syndrome’ (POTS) (female: 51.8% and male: 55.6%), and ‘hair loss’ (female: 44.3% and male: 11.3%). A subset of participants (female: 15.1% and male: 5.5%) reported a worsening of pre-existing medical conditions, with neurological disorders being the most frequently mentioned (female: 3.7% and male: 0%).

• 

  Figure 3. Decision tree for the self-reported clinical symptoms: (A) ‘hair loss’, (B) ‘dizziness’ and (C) ‘POTS’. Each final node shows a predicted class (0 or 1), the probability of that class, and what fraction of the dataset falls into that subgroup. Each terminal node shows the predicted class (0 = symptom absent, 1 = symptom present), the probability of that class, and the percentage of the dataset in that subgroup. For each split: “yes” means the condition is true, “no” means it is false. For example, “age < 52: yes” means age is less than 52 years; “no” means age is 52 or older. Gender is coded as 0 = female, 1 = male.

• 

  Figure 4. Cluster Heatmap of the present cohort including GPCR-fAAb and the self-reported symptoms, age and gender: Nociceptin-fAAb, β2-fAAb, AT1-fAAb, α1-fAAb, MAS-fAAb, M2-fAAb, ET-A-fAAb; self-reported symptoms: lack of concentration, dizziness, hair loss, Postural Orthostatic Tachycardia Syndrome (POTS), deterioration of pre-existing disorders, deterioration of neurological disorders, fatigue/post-exertional malaise (PEM). The variables appear along both the x- and y-axes, and each cell in the central matrix indicates the strength and direction of the association between the corresponding row variable (y-axis) and column variable (x-axis). The color scale on the right ranges from negative (blue) through neutral/low (white or light shades) to positive (red) association values. Darker shades of red denote stronger positive correlations, whereas darker shades of blue denote stronger negative correlations.

• 

  Figure 5. Proposed GPCR-fAAb mediated mechanism: we assume that GPCR-fAAb act as pathogenic agonistic AAbs by inducing a permanent (i.e., chronic) activation of their targets; these GPCR-fAAb-mediated pathogenetic effects were enhanced in ischemic/inflammatory tissues, inducing a pathogenetic circulus of inflammation and ischemia.

• 

  Figure 6. Major clinical phenotype with specific GPCR-fAAb cluster of the present study cohort: a cluster of β2-fAAb, M2-fAAb, MAS-fAAb and AT1-fAAb was the most common one in LC-patients with fatigue/PEM as one major clinical symptom.