Cancer and COVID-19 Susceptibility and Severity: A Two-Sample Mendelian Randomization and Bioinformatic Analysis

The clinical management of patients with COVID-19 and cancer is a Gordian knot that has been discussed widely but has not reached a consensus. We introduced two-sample Mendelian randomization to investigate the causal association between a genetic predisposition to cancers and COVID-19 susceptibility and severity. Moreover, we also explored the mutation landscape, expression pattern, and prognostic implications of genes involved with COVID-19 in distinct cancers. Among all of the cancer types we analyzed, only the genetic predisposition to lung adenocarcinoma was causally associated with increased COVID-19 severity (OR = 2.93, β = 1.074, se = 0.411, p = 0.009) with no obvious heterogeneity (Q = 17.29, p = 0.24) or symmetry of the funnel plot. In addition, the results of the pleiotropy test demonstrated that instrument SNPs were less likely to affect COVID-19 severity via approaches other than lung adenocarcinoma cancer susceptibility (p = 0.96). Leave-one-out analysis showed no outliers in instrument SNPs, whose elimination rendered alterations in statistical significance, which further supported the reliability of the MR results. Broad mutation and differential expression of these genes were also found in cancers, which may provide valuable information for developing new treatment modalities for patients with both cancer and COVID-19. For example, ERAP2, a risk factor for COVID-19-associated death, is upregulated in lung squamous cancer and negatively associated with patient prognosis. Hence, ERAP2-targeted treatment may simultaneously reduce COVID-19 disease severity and restrain cancer progression. Our results highlighted the importance of strengthening medical surveillance for COVID-19 deterioration in patients with lung adenocarcinoma by showing their causal genetic association. For these patients, a delay in anticancer treatment, such as chemotherapy and surgery, should be considered.