The ongoing COVID-19 pandemic has increased awareness about sex-specific differences in immunity and outcomes following SARS-CoV-2 infection. Strong evidence of a male bias in COVID-19 disease severity is hypothesized to be mediated by sex differential immune responses against SARS-CoV-2. This hypothesis is based on data from other viral infections, including influenza viruses, HIV, hepatitis viruses, and others that have demonstrated sex-specific immunity to viral infections. Although males are more susceptible to most viral infections, females possess immunological features that render them more vulnerable to distinct immune-related disease outcomes. Both sex chromosome complement and related genes as well as sex steroids play important roles in mediating the development of sex differences in immunity to viral infections.

Background. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) continues to spread worldwide and pose a major public health burden. There is increasing evidence that men are more likely to die from SARS-CoV-2 infection than women. However, underlying factors that mediate the observed sex bias in coronavirus disease 2019 (COVID-19) remain unknown. Methods. In this retrospective cohort, we included COVID-19 patients who were admitted to an Intensive Care Unit at the University Hospital Hamburg-Eppendorf, Germany. We obtained demographic data of all patients who were discharged or had died by 29th April 2020. We systematically analyzed sex hormones as well as cytokine and chemokine responses in male and female patients with laboratory-confirmed SARS-CoV-2 infections upon hospital admission. We used uni- and multivariable linear regression methods to identify potential risk factors for disease severity in males and females. Findings. All enrolled patients (n=45; n=35 males and n=10 females) presented comorbidities with hypertension being the most common (45.7% in males; 40% in females), followed by cancer (35% in males; 40% in females), obesity (34.3% in males and 30% in females), type II diabetes (25.7% in males and 20% in females) and chronic heart diseases (8.6% in males and 0% in females). We detected that the vast majority of male COVID-19 patients present low testosterone (68.6%) and low dihydrotestosterone (48.6%) levels. In contrast, most female COVID-19 patients have elevated testosterone levels (60%) without alterations in dihydrotestosterone levels. Both, female and male COVID-19 patients may present elevated estradiol levels (45.7% in males and 40% in females). Disease severity defined by SOFA score correlates with elevated cytokine responses (e.g. IL-6) in males and IL-2 in females. In male COVID-19 patients, testosterone levels negatively correlate with inflammatory IL-2 and IFN-γ, whereas estradiol levels positively correlate with the inflammatory cytokine IL-6. Vice versa, in female COVID-19 patients, testosterone levels positively correlate with inflammatory cytokines (e.g. IL-6). Interpretation. We here show that critically ill male COVID-19 patients suffer from severe testosterone and dihydrotestosterone deficiencies. Both androgens are required to mount antiviral immune responses to combat infection in males.

No abstract

The immunogenicity of current influenza virus vaccines is assessed by measuring an increase of influenza virus-specific antibodies in a hemagglutination inhibition assay. This method exclusively measures antibodies against the hemagglutinin head domain. While this domain is immunodominant, it has been shown that hemagglutination inhibition titers do not always accurately predict protection from disease. In addition, several novel influenza virus vaccines that are currently under development do not target the hemagglutinin head domain, but rather more conserved sites, including the hemagglutinin stalk. Importantly, antibodies against the hemagglutinin stalk do not show activity in hemagglutination inhibition assays and will require different methods for quantification. In this study, we tested human serum samples from a seasonal influenza virus vaccination trial and an avian H5N1 virus vaccination trial for antibody activities in multiple types of assays, including binding assays and also functional assays. We then performed serum transfer experiments in mice which then received an H1N1 virus challenge to assess the in vivo protective effects of the antibodies. We found that hemagglutinin-specific antibody levels measured in an enzyme-linked immunosorbent assay (ELISA) correlated well with protection from weight loss in mice. In addition, we found that weight loss was also inversely correlated with the level of serum antibody-dependent cellular cytotoxicity (ADCC) as measured in a reporter assay. These findings indicate that protection is in part conferred by Fc-dependent mechanisms. In conclusion, ELISAs can be used to measure hemagglutinin-specific antibody levels that could serve as a surrogate marker of protection for universal influenza virus vaccines.

Pregnant women are at high risk for severe influenza disease outcomes, yet insights into the underlying mechanisms are limited. Here, we present models of H1N1 infection in syngenic and allogenic pregnant mice; infection in the latter mirrors the severe course of 2009 pandemic influenza in pregnant women. We found that the anti-viral immune response in the pregnant host was significantly restricted as compared to the non-pregnant host. This included a reduced type I interferon response as well as impaired migration of CD8 T cells into the lung. The multi-faceted failure to mount an anti-viral response in allogenic pregnant mice resulted in a less stringent selective environment that promoted the emergence of 2009 H1N1 virus variants that specifically counteract type I interferon response and mediate increased viral pathogenicity. These insights underscore the importance of influenza vaccination compliance in pregnant women and may open novel therapeutic avenues.