Genetically determined serum testosterone level and Covid-19 illness level: A mendelian randomization study

Background: It is hypothesized that different levels of hormones especially serum testosterone level could explain the sex differences between men and women on the susceptibility and case fatality rate of COVID-19. However, traditional observational studies that support this hypothesis could not effectively establish the causal effects. Objective: Utilizing recently published genome-wide associations studies (GWAS) on serum Testosterone level and on COVID-19 related phenotypes, we sought to assess the causality through Mendelian Randomization (MR) analyses. We further applied a suite of statistical genomics methods to further explore the biological mechanisms. Results: We found that testosterone level is significantly associated with Covid-19 critical illness. All six MR methods yielded significant associations. There is no significant association between Testosterone and COVID-19 respiratory failure or COVID-19 susceptibility. Conclusion: Based on the GWAS currently available, we provide support for a causal role of Testosterone on COVID-19 critical illness. Nevertheless, we recognize that the COVID-19 susceptibility GWAS effort is still ongoing and there is no such strong locus as CCR5 for HIV discovered for COVID-19.