Systematic Review and Meta-analysis of Anti-interferon Auto-antibodies in Infectious Diseases

Abstract Purpose To perform a systematic review and meta-analysis of prevalence and function of anti-interferon auto-antibodies in acute infectious diseases. Methods We performed a search on the following electronic bibliographic databases: Medline, Embase, Web of Science and Cochrane. Eligible studies generated a systematic review and random effects model meta-analysis of pooled seroprevalence and neutralisation status of anti-interferon auto-antibodies in acute infectious diseases. Results Thirty-seven studies including 12,629 individuals with SARS-CoV-2 infection were analysed. There are insufficient data for meta-analyses of auto-antibodies in non-SARS-CoV-2 infectious diseases, with crude seroprevalence of auto-antibodies varying widely (0.0–77.0%). The pooled seroprevalence of anti-IFNɑ and/or anti-IFN⍵ auto-antibodies in individuals with SARS-CoV-2 infection was 14% (95%CI 9–18%, I 2  = 92%, $$\:\tau\:$$ 2  = 0.0151, p  < 0.01). Pooled prevalence of neutralising auto-antibodies were slightly lower (12%; 95%CI 8–17%, I 2  = 77%, $$\:\tau\:$$ 2  = 0.0027, p  < 0.01). The high heterogeneity may reflect divergent SARS-CoV-2 disease severity across studies, and methodological diversity for measurement and definition of auto-antibody binding and neutralisation. Pooled seroprevalence for anti-IFNɑ auto-antibodies in uninfected healthy controls were < 1% (95%CI 0–1%, I 2  = 90%, $$\:\tau\:$$ 2  = 0.0028, p  < 0.01). Conclusion Anti-interferon auto-antibodies may impair immune responses to diverse infections leading to life-threatening disease. These auto-antibodies have not been studied at all in most infectious diseases, most notably for bacterial infection. This study illustrates the urgent need to standardise methodology and reporting of auto-antibodies in the setting of infectious diseases so that the immunopathology and translational impact of these novel biomarkers can be realised.