Higher COVID-19 pneumonia risk associated with anti-IFN-α than with anti-IFN-ω auto-Abs in children

Abstract

We found that 19 (10.4%) of 183 unvaccinated children hospitalized for COVID-19 pneumonia had autoantibodies (auto-Abs) neutralizing type I IFNs (IFN-α2 in 10 patients: IFN-α2 only in three, IFN-α2 plus IFN-ω in five, and IFN-α2, IFN-ω plus IFN-β in two; IFN-ω only in nine patients). Seven children (3.8%) had Abs neutralizing at least 10 ng/ml of one IFN, whereas the other 12 (6.6%) had Abs neutralizing only 100 pg/ml. The auto-Abs neutralized both unglycosylated and glycosylated IFNs. We also detected auto-Abs neutralizing 100 pg/ml IFN-α2 in 4 of 2,267 uninfected children (0.2%) and auto-Abs neutralizing IFN-ω in 45 children (2%). The odds ratios (ORs) for life-threatening COVID-19 pneumonia were, therefore, higher for auto-Abs neutralizing IFN-α2 only (OR [95% CI] = 67.6 [5.7-9,196.6]) than for auto-Abs neutralizing IFN-ω only (OR [95% CI] = 2.6 [1.2-5.3]). ORs were also higher for auto-Abs neutralizing high concentrations (OR [95% CI] = 12.9 [4.6-35.9]) than for those neutralizing low concentrations (OR [95% CI] = 5.5 [3.1-9.6]) of IFN-ω and/or IFN-α2.

Figure S1.

Neutralizing auto-Abs against type I IFNs in children with life-threatening COVID-19. (A) Gyros (high-throughput automated ELISA) results for auto-Abs against IFN-α2 for positive controls (n = 4), pediatric patients (n = 188), and healthy controls (n = 16). (B) Plot of anti–IFN-α2 auto-Ab IgG levels, as determined by Gyros, against their neutralization capacity at 10 ng/ml in the luciferase assay. For plasma from each patient, luciferase activity was normalized against the mean induction of control plasma tested on the same day in the luciferase assay. The horizontal dotted line indicates the threshold of neutralization, defined as the level of induction below 15% of the mean value for controls tested on the same day. (C–J) Proportion by age of pediatric and adult patients from the general population positive for neutralizing auto-Abs (in plasma 1:10) against (C) IFN-α2 at 10 ng/ml, for both sexes; (D) IFN-α2, at 10 ng/ml, for men or women; (E) IFN-ω, at 10 ng/ml, for both sexes; (F) IFN-ω, at 10 ng/ml, for men or women; (G) IFN-β, at 10 ng/ml, for both sexes; (H) IFN-β, at 10 ng/ml, for men or women; (I) IFN-α2 and IFN-ω, at 10 ng/ml, for both sexes; and (J) IFN-α2 and IFN-ω, at 10 ng/ml, for men or women.

Figure 1.

Neutralizing auto-Abs against IFN-α2 and/or IFN-ω in children with life-threatening COVID-19. (A) Results for the neutralization of 10 ng/ml IFN-α2, IFN-ω, or IFN-β in the presence of plasma (1/10 dilution) from pediatric patients with critical (n = 136), severe (n = 35), or moderate (n = 12) COVID-19 pneumonia. Relative luciferase activity is shown (IFN-stimulated response element [ISRE] dual luciferase activity, with normalization against Renilla luciferase activity) after stimulation with 10 ng/ml IFN-α2, IFN-ω, or IFN-β in the presence of plasma (1/10 dilution). RLA: relative luciferase activity. All samples were tested twice independently. (B) Neutralization of 100 pg/ml IFN-α2 or IFN-ω in the presence of plasma (1/10 dilution) from pediatric patients with critical (n = 136), severe (n = 35), or moderate (n = 12) COVID-19 pneumonia. All samples were tested twice independently. (C–F) The proportion by age of pediatric and adult patients with life-threatening COVID-19 pneumonia positive for neutralizing auto-Abs (in plasma 1/10) against (C) IFN-α2 and/or IFN-ω at 10 ng/ml for both sexes, (D) IFN-α2 and/or IFN-ω at 10 ng/ml for men or women, (E) IFN-α2 and/or IFN-ω at 100 pg/ml for both sexes, and (F) IFN-α2 and/or IFN-ω at 100 pg/ml for men or women.

Figure S2.

Neutralizing auto-Abs against IFN-α2 and/or IFN-ω in children and adults with life-threatening COVID-19. (A–J) Proportion by age of pediatric and adult patients from the general population positive for neutralizing auto-Abs (in plasma 1:10) against (A) IFN-α2 and IFN-ω, at 100 pg/ml, for both sexes; (B) IFN-α2 and IFN-ω, at 100 pg/ml, for men or women; (C) IFN-α2, at 100 pg/ml, for both sexes; (D) IFN-α2, at 100 pg/ml, for men or women; (E) IFN-ω, at 100 pg/ml, for both sexes; (F) IFN-ω, at 100 pg/ml, for men or women; (G) IFN-α2 only, at 10 ng/ml, for both sexes; (H) IFN-α2 only, at 10 ng/ml, for men or women; (I) IFN-ω only, at 10 ng/ml, for both sexes; and (J) IFN-ω only, at 10 ng/ml, for men or women.

Figure 2.

Auto-Abs neutralize all 12 IFN-α subtypes and the glycosylated IFNs. (A) Results for the neutralization of 12 individual IFN-α subtypes, IFN-ω or IFN-β, from children with auto-Abs (n = 8), healthy controls (n = 2), or auto-Ab positive patients (n = 2). Relative luciferase activity is shown (ISRE dual luciferase activity, with normalization against Renilla luciferase activity, and to the non-stimulated [NS] condition) after stimulation with the various type I IFNs at a concentration of 1 ng/ml in the presence of plasma (1/10 dilution). All samples were tested once. RLA: relative luciferase activity. (B–E) Plots representing the neutralization results for glycosylated (at 1 ng/ml) or unglycosylated (at 10 ng/ml or 100 pg/ml) forms of IFN-α2 and IFN-ω. The dots in the lower right part of the plot indicate neutralization of the glycosylated form of the IFN but not of the unglycosylated form, whereas the dots in the upper right part of the plot indicate neutralization of the unglycosylated form of the IFN but not of the glycosylated form. The dots in the lower left part of the plot indicate the neutralization of either form of the IFNs, whereas those in the upper right part of the plot indicate an absence of neutralization for both forms; (B) unglycosylated form of IFN-α2 at 10 ng/ml; (C) unglycosylated form of IFN-α2 at 100 pg/ml; (D) unglycosylated form of IFN-ω at 10 ng/ml; and (E) unglycosylated form of IFN-ω at 100 pg/ml.

Figure S3.

Neutralizing auto-Abs against glycosylated type I IFNs, and proportion of children and adults from the general population with neutralizing auto-Abs against type I IFNs. (A) Results for the neutralization of various doses of the glycosylated form of IFN-α2 or IFN-ω in the presence of plasma (1/10 dilution) from children with (C−) and without (C+) auto-Abs neutralizing type I IFNs. Relative luciferase activity is shown (ISRE dual luciferase activity, with normalization against Renilla luciferase activity) after stimulation with 10 ng/ml IFN-α2 or IFN-ω in the presence of plasma (1/10 dilution). RLA: relative luciferase activity. (B–H) Proportion of children and adults from the general population with neutralizing auto-Abs against type I IFNs. (B) Prevalence of auto-Abs neutralizing type I IFNs, by type of IFN neutralized. (C–H) Proportion, by age, of pediatric and adult individuals from the general population positive for neutralizing auto-Abs (in plasma diluted 1:10) against (C) IFN-β, at 10 ng/ml, for both sexes; (D) IFN-β, at 10 ng/ml, for men or women; (E) IFN-α2 and IFN-ω, at 10 ng/ml, for both sexes; (F) IFN-α2 and IFN-ω, at 10 ng/ml, for men or women; (G) IFN-α2 and IFN-ω, at 100 pg/ml, for both sexes; and (H) IFN-α2 and IFN-ω, at 100 pg/ml, for men or women.

Figure 3.

Neutralizing auto-Abs against IFN-α2 and/or IFN-ω in children from the general population and their serological evaluation. (A) Results for the neutralization of 10 ng/ml IFN-α2, IFN-ω, or IFN-β in the presence of plasma (1/10 dilution) from children from the general population (n = 2,267). Relative luciferase activity is shown (ISRE dual luciferase activity, with normalization against Renilla luciferase activity). All samples were tested once. RLA: relative luciferase activity. (B) Neutralization of 100 pg/ml IFN-α2 or IFN-ω in the presence of plasma (1/10 dilution) from children from the general population (n = 2,267). (C) Prevalence of auto-Abs neutralizing type I IFNs, distributed by age, in individuals from the general population. (D) Serological evaluation of auto-Ab-positive children: Virscan results for children (n = 9) with auto-Abs against type I IFNs.

Figure 4.

Neutralizing auto-Abs against IFN-α2 and/or IFN-ω in the pediatric and adult general population. (A–H) Proportion by age of pediatric and adult individuals from the general population positive for neutralizing auto-Abs (in plasma 1/10) against (A) IFN-α2 and/or IFN-ω, at 10 ng/ml, for both sexes; (B) IFN-α2 and/or IFN-ω, at 10 ng/ml, for men or women; (C) IFN-α2 and/or IFN-ω, at 100 pg/ml, for both sexes; (D) IFN-α2 and/or IFN-ω, at 100 pg/ml, for men or women; (E) IFN-α2, at 100 pg/ml, for both sexes; (F) IFN-α2, at 100 pg/ml, for men or women; (G) IFN-ω, at 100 pg/ml, for both sexes; and (H) IFN-ω, at 100 pg/ml, for men or women.

Figure S4.

Neutralizing auto-Abs against IFN-α2 or IFN-ω in children from the general population. (A–H) Proportion, by age, of pediatric and adult individuals from the general population positive for neutralizing auto-Abs (in plasma diluted 1:10) against (A) IFN-α2, at 10 ng/ml, for both sexes; (B) IFN-α2, at 10 ng/ml, for men or women; (C) IFN-ω, at 10 ng/ml, for both sexes; (D) IFN-ω, at 10 ng/ml, for men or women; (E) IFN-α2 only, at 10 ng/ml, for both sexes; (F) IFN-α2 only, at 10 ng/ml, for men or women; (G) IFN-ω only, at 10 ng/ml, for both sexes; and (H) IFN-ω only, at 10 ng/ml, for men or women.

Figure S5.

Neutralizing auto-Abs against IFN-α2 and/or IFN-ω in children from the general population in Estonia and Japan. (A) Results for the neutralization of 10 ng/ml IFN-α2, IFN-ω, or IFN-β or 100 pg/ml IFN-α2 or IFN-ω in the presence of plasma (1/10 dilution) from children from Japan (n = 249). (B) Results for the neutralization of 10 ng/ml or 100 pg/ml IFN-α2, IFN-ω, or IFN-β in the presence of plasma (1/10 dilution) from children from Estonia (n = 200). (A and B) Relative luciferase activity is shown (ISRE dual luciferase activity, with normalization against Renilla luciferase activity) after stimulation with IFN-α2, IFN-ω, or IFN-β in the presence of plasma (1/10 dilution). RLA: relative luciferase activity.

Figure 5.

OR for COVID-19 pneumonia. (A) Bar plot of the calculated OR for COVID-19 pneumonia in children with auto-Abs against type I IFNs. Adjusted ORs and 95% CI were computed by penalized profile likelihood. ORs and P values were estimated by means of Firth’s bias-corrected logistic regression. (B) Bar plot of the calculated OR for life-threatening pneumonia in children with auto-Abs against type I IFNs. Adjusted ORs and 95% CI were computed by penalized profile likelihood. ORs and P values were estimated by means of Firth’s bias-corrected logistic regression.