The Frequency of aAbs to Type I IFNs Was High in Patients with Critical COVID-19
We first measured aAbs to type I IFNs by ELISA in 622 Japanese COVID-19 patients aged 0–104 years, including 170 critical, 235 severe, 112 moderate, and 105 mild cases. We detected aAbs to IFN-α2 or IFN-ω at the following frequencies: 5.9% critical cases, 1.7% severe cases, 0.9% moderate cases, 3.8% mild case (Fig. 1C, Table 2). In detail, 4.7% (95% CI: 2.4–9.0) of patients with critical disease had aAbs to IFN-α2, 3.5% (95% CI: 1.6–7.5) to IFN-ω, and 2.4% (95% CI: 0.9–5.9) to both IFN-α2 and IFN-ω (Table 2). Among patients who had IFN-α2 or IFN-ω aAbs, there were several patients who had isolated aAb solely to IFN-α2 or IFN-ω (Table S2). The aAbs to IFN-α2 or IFN-ω were also detected in 3.8% of patients with mild disease and 0.9% of those with moderate disease. Unlike patients with critical COVID-19, none of the patients with mild to severe disease had aAbs to both interferon subtypes (Table 2). Among patients over 50 years old, 3.6% (95% CI: 2.2–5.7) had aAbs to IFN-α2 or IFN-ω, while 1.7% (95% CI: 0.6–4.9) of patients younger than 50 years had these aAbs (Table 2). Overall, these aAbs to type I IFNs were detected more frequently in patients with critical disease and patients over 50 years old. However, isolated aAbs to IFN-α2 or IFN-ω was also detected in some of the patients with mild or moderate disease in the current study.
naAbs to Type I IFNs Were Frequently Detected in Patients with Critical COVID-19
aAbs which react with type I IFNs were detected by ELISA; however, their neutralizing activity could not be assessed by ELISA. We thus measured neutralizing activity against type I IFNs using the ISRE reporter assay in sera from 622 patients with COVID-19 . Sera were considered to have neutralizing activity if the induction of ISRE activity, which was normalized to Renilla luciferase activity, was less than 15% of the median values of healthy controls . These data are summarized in Table 3 and Table S3. Strongly neutralizing naAbs, capable of neutralizing 10 ng/mL of IFN-α2 or IFN-ω, were found in 5.9% of critical cases, 2.1% of severe cases, 0.9% of moderate cases, and 0% of mild cases (Fig. 2A, Table 3). In patients with critical disease, antibody prevalence was as follows: 5.9% (95% CI: 3.2–10.5) had naAbs to IFN-α2, 4.1% (95% CI: 2.0–8.3) to IFN-ω, and 4.1% (95% CI: 2.0–8.3) to both IFN-α2 and IFN-ω (Table 3). On the other hand, less than 1% of patients with mild to moderate disease had naAbs to type I IFNs (Table 3). Among patients over 50 years old, 3.6% (95% CI: 2.2–5.7) had naAbs to IFN-α2, 2.2% (95% CI: 1.2–4.1) to IFN-ω, 2.2% (95% CI: 1.2–4.1) to both IFN-α2 and IFN-ω, and 3.6% (95% CI: 2.2–5.7) to IFN-α2 or IFN-ω (Table 3). By contrast, none of the patients younger than 50 years old had naAbs to type I IFNs (Table 3). These results are summarized according to disease severity in Fig. 2B. All patients having neutralizing activity against IFN-ω had neutralizing activity against IFN-α2. Of note, in contrast to the prevalence of aAb (Table S2), no patients had isolated naAbs to IFN-ω (Table S3, Fig. S1).
Next, we analyzed serum neutralizing activity under more sensitive conditions by stimulating cells at lower concentrations (100 pg/mL) of IFN-α2 or IFN-ω. Under this condition, consistent with previous reports , the prevalence of naAbs was observed in 10.6% of critical cases, 2.6% of severe cases, 0.9% of moderate cases, and 1.0% of mild cases (Fig. 3A, Table 3, Table S3). In detail, 7.1% (95% CI: 4.1–11.9) of critical cases had naAbs to IFN-α2, 10.0% (95% CI: 6.3–15.4) to IFN-ω, and 6.5% (95% CI: 3.7–11.2) to both IFN-α2 and IFN-ω (Table 3). Only 1% or less of the patients with mild to moderate disease had these naAbs to IFN-α2 or IFN-ω (Table 3). Among patients over 50 years old, 4.5% (95% CI: 2.9–6.8) had naAbs to IFN-α2, 4.7% (95% CI: 3.1–7.1) of them to IFN-ω, 3.4% (95% CI: 2.0–5.5) to both IFN-α2 and IFN-ω, and 5.8% (95% CI: 4.0–8.4%) to IFN-α2 or IFN-ω. By contrast, none of the patients younger than 50 years old had naAbs to IFN-α2 or IFN-ω (Table 3).
Using this more sensitive condition, the percentage of the patients with naAbs to IFNs was higher than in the condition with 10 ng/mL (Table 3). We detected naAbs against IFN-α2 in an additional 4 patients at the 100 pg/mL condition compared to the 10 ng/ml condition. Among these 4 patients, 3 had critical/severe disease and 1 patient had mild disease (Fig. 4A, Fig. S3). Regarding naAbs to IFN-ω, an additional 11 patients showed neutralizing activity only against 100 pg/mL. All 11 patients had critical/severe disease (Fig. 4B, Fig. S4). It is known that the concentration of type I IFNs in the blood of patients with acute and benign SARS-CoV-2 infections ranges from 1 to 100 pg/mL [13, 27]. Moreover, it has been experimentally proven that 100 pg/mL of type I IFNs can impair SARS-CoV-2 replication in epithelial cells . Therefore, a neutralization assay using 100 pg/mL of type I IFNs, which reflects physiological conditions, detected naAbs more precisely than the assay using 10 ng/mL, especially naAbs to IFN-ω.
The prevalence of naAbs by sex was 5.5% at 100 pg/mL and 3.4% at 10 ng/mL for males and 1.1% at 100 pg/mL and 0.5% at 10 ng/mL for females (Table S4, Fig. S5). NaAbs to IFNs were significantly associated with critical disease (P = 0.0152 at 10 ng/mL, P = 0.0012 at 100 pg/mL) compared to mild disease, age over 50 (P = 0.0085, P = 0.0002), and male sex (P = 0.0488, P = 0.137) (Table 4). COVID-19 aggravation was strongly associated with naAbs among critical patients using both assay conditions (at 10 ng/mL, IFN-α2 and IFN-ω odds ratio (OR) = 9.3, IFN-α2 or IFN-ω OR = 13.5; at 100 pg/mL, IFN-α2 and IFN-ω OR = 14.9, IFN-α2 or IFN-ω OR = 12.7) (Figs. 2C and 3C). These data are consistent with previous reports that identified a high prevalence, 10.2–18% in patients with critical disease, of naAbs to type I IFNs (Table S5) [18,19,20,21,22,23,24,25,26].
Comparison of the Results of the Neutralization Assay and ELISA
While the IFN neutralization assay is the gold standard in assessing the biological effect of aAbs and the ISRE reporter assay is a sensitive method, it is time-consuming. On the other hand, ELISA is more high-throughput with faster turnaround times. We thus compared the results of neutralizing activity against type I IFNs measured by the ISRE reporter assay with the results of aAbs to type I IFNs measured by ELISA. When the presence of naAbs to IFN-α2 was predicted by the results of aAbs to IFN-α2, the sensitivity was 50%, the specificity was 99.3%, the positive predictive value (PPV) was 66.7%, the negative predictive value (NPV) was 98.7% at 10 ng/mL (Fig. 4C), and these two detection methods had a weak negative correlation (a correlation coefficient − 0.307 (95% CI: − 0.376 to − 0.234, P value < 0.0001)). For the 100 pg/mL condition, the sensitivity was 40%, the specificity was 99.3% (PPV of 66.7% and NPV of 98.0%), and these two detection methods had a weak negative correlation (a correlation coefficient − 0.199 (95% CI: − 0.273 to − 0.123, P value < 0.0001)) (Fig. 4E). We thus realized that ELISA-based detection of aAbs to IFN-α2 can be an alternative method to enable testing of multiple samples, e.g., screening tests for the general population, and to evaluate antibodies to type I IFNs in sera. In contrast, for IFN-ω, ELISA failed to adequately detect the presence of naAbs to IFN-ω. Indeed, ELISA-based detection of aAbs to IFN-ω pointed out the presence of naAbs to IFN-ω (10 ng/mL condition) with a sensitivity of 10% and specificity of 98.4% (PPV of 9.1% and NPV of 98.5%) (Fig. 4D). Regarding the 100 pg/mL condition, aAbs to IFN-ω only indicated naAbs to IFN-ω with a sensitivity of 9.5% and a specificity of 98.5% (PPV of 18.2% and NPV of 96.9%) (Fig. 4F).
Sera from COVID-19 Patients with naAbs to IFN-α2 Show Low Concentrations of IFN-α2
We analyzed the concentration of IFN-α2 using 269 samples for which the exact time of specimen collection could be determined with the ProQuantum™ Human IFN alfa Immunoassay, which is a qPCR-based technique. The level of IFN-α2 in sera in patients with naAbs was significantly lower compared to those without naAbs. The serum IFN-α2 levels were below detection limit (< 4 pg/mL) in all but one 1 patient with naAbs detected by the high sensitivity condition (Fig. 5A, B). However, there is no correlation between disease severity and the concentration of IFN-α2 (P = 0.2238). We also compared the level of IFN-α2 between the samples collected from onset to day 4 and those from day 5 to day 7 after onset. We found that the concentration of IFN-α2 were significantly higher in samples from onset to day 4 compared to those from day 5 to day 7 (P = 0.0009) (data not shown). These results are consistent with a previous report .
Prevalence of aAbs to IFN-α2 in Uninfected Individuals from the General Japanese Population
In order to understand the risk of the general Japanese population to severe COVID-19 and other viral infections, we sought to determine the prevalence of naAbs to type I IFNs in the Japanese population by detecting aAbs to IFN-α2 via ELISA. We studied 3456 Japanese individuals aged 20–91 years and unaffected by COVID-19. In this population, 3 individuals had aAbs to IFN-α2 (0.087% (95% CI: 0.0295–0.255%)) (Fig. 5C). These 3 individuals consisted of an 86-year-old female, a 78-year-old male, and a 42-year-old male. These data suggest that the prevalence of aAbs, and by inference, that of naAbs, is low in the healthy general Japanese population.