Autoantibodies in hospitalised patients with COVID-19

Abstract

Objectives: CD209L and its homologous protein CD209 act as alternative entry receptors for the SARS-CoV-2 virus and are highly expressed in the virally targeted tissues. We tested for the presence and clinical features of autoantibodies targeting these receptors and compared these with autoantibodies known to be associated with COVID-19.

Methods: Using banked samples (n = 118) from Johns Hopkins patients hospitalised with COVID-19, we defined autoantibodies against CD209 and CD209L by enzyme-linked immunosorbent assay (ELISA). Clinical associations of these antibodies were compared with those of patients with anti-interferon (IFN) and anti-angiotensin-converting enzyme-2 (ACE2) autoantibodies.

Results: Amongst patients hospitalised with COVID-19, 19.5% (23/118) had IgM autoantibodies against CD209L and were more likely to have coronary artery disease (44% vs 19%, P = 0.03). Antibodies against CD209 were present in 5.9% (7/118); interestingly, all 7 were male (P = 0.02). In our study, the presence of either antibody was positively associated with disease severity [OR 95% confidence interval (95% CI): 1.80 (0.69-5.03)], but the association did not reach statistical significance. In contrast, 10/118 (8.5%) had IgG autoantibodies against IFNα, and 21 (17.8%) had IgM antibodies against ACE2. These patients had significantly worse prognosis (intubation or death) and prolonged hospital stays. However, when adjusting for patient characteristics on admission, only the presence of anti-ACE2 IgM remained significant [pooled common OR (95% CI), 4.14 (1.37, 12.54)].

Conclusion: We describe IgM autoantibodies against CD209 and CD209L amongst patients hospitalised with COVID-19. These were not associated with disease severity. Conversely, patients with either anti-ACE2 IgM or anti-IFNα IgG antibodies had worse outcomes. Due to the small size of the study cohort, conclusions drawn should be considered cautiously.

Keywords: COVID‐19; anti‐ACE2 IgM; anti‐CD209 IgM; anti‐CD209L IgM; autoantibodies.

Figure 1

Autoantibodies against CD209 and/or CD209L and their associations with disease characteristics. (a) Prevalence of anti‐CD209 IgM and IgG or anti‐CD209L IgM and IgG autoantibodies in disease cohort. (b) Median and interquartile range of patient age at admission categorised based on the presence of anti‐CD209 and/or anti‐CD209L IgM antibodies. (c, d) Percentage of patients with listed comorbidities (c) or given antihypertensive medications (d) categorised based on their anti‐CD209 and/or anti‐CD209L IgM autoantibody status. (e) Percentage of males and females grouped according to anti‐CD209 and/or anti‐CD209L IgM positivity. (f) Median and interquartile range of admission laboratory items of patients with and without anti‐CD209 and/or anti‐CD209L IgM autoantibodies. The normal reference range of item criteria is shown as grey bars. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001.

Figure 2

Anti‐IFN‐α IgG antibodies and their associations with disease characteristics. (a) Prevalence of anti‐IFN‐α IgG in disease cohort. (b) Max WHO status category and Max WHO severity of patients categorised according to the presence of anti‐IFN‐α IgG antibodies. (c) Median patient LOS in days grouped according to anti‐IFN‐α IgG positivity. (d, e) Percentage of patients with listed comorbidities (d) or given anti‐inflammatory medications (e) categorised based on their anti‐IFN‐α IgG autoantibody status. (f) Median and interquartile range of admission laboratory items of patients with and without anti‐IFN‐α IgG autoantibodies. The normal reference range of item criteria is shown as grey bars. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0 0.001, ****P ≤ 0.0001.

Figure 3

Anti‐ACE2 IgM antibodies and their associations with disease characteristics. (a) Prevalence of anti‐ACE2 IgM and IgG in disease cohort. (b) Max WHO status category and Max WHO severity of patients categorised according to the presence of anti‐ACE2 IgM antibodies. (c) Median patient LOS in days grouped according to anti‐ACE2 IgM positivity. (d, e) Percentage of patients with listed comorbidities (d) or given anti‐inflammatory medications (e) categorised based on their anti‐ACE2 IgM autoantibody status. (f) Median and interquartile range of admission laboratory items of patients with and without anti‐ACE2 IgM autoantibodies. The normal reference range of item criteria is shown as grey bars. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0 0.001, ****P ≤ 0.0001.

Figure 4

Heatmap depicting autoantibody profiling of cohort.