Age-dependent impact of the major common genetic risk factor for COVID-19 on severity and mortality

Abstract

BackgroundThere is considerable variability in COVID-19 outcomes among younger adults, and some of this variation may be due to genetic predisposition.MethodsWe combined individual level data from 13,888 COVID-19 patients (n = 7185 hospitalized) from 17 cohorts in 9 countries to assess the association of the major common COVID-19 genetic risk factor (chromosome 3 locus tagged by rs10490770) with mortality, COVID-19-related complications, and laboratory values. We next performed metaanalyses using FinnGen and the Columbia University COVID-19 Biobank.ResultsWe found that rs10490770 risk allele carriers experienced an increased risk of all-cause mortality (HR, 1.4; 95% CI, 1.2-1.7). Risk allele carriers had increased odds of several COVID-19 complications: severe respiratory failure (OR, 2.1; 95% CI, 1.6-2.6), venous thromboembolism (OR, 1.7; 95% CI, 1.2-2.4), and hepatic injury (OR, 1.5; 95% CI, 1.2-2.0). Risk allele carriers age 60 years and younger had higher odds of death or severe respiratory failure (OR, 2.7; 95% CI, 1.8-3.9) compared with those of more than 60 years (OR, 1.5; 95% CI, 1.2-1.8; interaction, P = 0.038). Among individuals 60 years and younger who died or experienced severe respiratory failure, 32.3% were risk-variant carriers compared with 13.9% of those not experiencing these outcomes. This risk variant improved the prediction of death or severe respiratory failure similarly to, or better than, most established clinical risk factors.ConclusionsThe major common COVID-19 genetic risk factor is associated with increased risks of morbidity and mortality, which are more pronounced among individuals 60 years or younger. The effect was similar in magnitude and more common than most established clinical risk factors, suggesting potential implications for future clinical risk management.

Keywords: COVID-19; Genetic variation; Genetics.

Figure 1. Associations with mortality.

The results described here were restricted to 9699 COVID-19 patients of European ancestry with available follow-up and cause of death information. (A) Survival analysis using Cox’s proportional hazard model. Kaplan-Meier curves stratified by rs10490770 risk allele carrier status. (carriers: n = 1469 vs. noncarriers: n = 8,230). HRs were calculated by adjusting for age, sex, and genetic PCs 1 to 5 as fixed effects and a dummy variable representing the participating studies as random effects. (B) Cumulative incidence curves for COVID-19–related death and COVID-19–unrelated death among the same individuals as described in A.

Figure 2. Associations between rs10490770 risk allele carrier status and COVID-19 severity and complications.

The results described here were restricted to COVID-19 patients of European ancestry. Logistic regressions were fit to assess the associations of rs10490770 risk allele carrier status with COVID-19 severity and complications, adjusting for age, sex, and genetic PCs 1 to 5 as fixed effects, and a dummy variable representing the participating studies as random effects. Red: All participants (n = 12,091); blue: hospitalized participants only (n = 6054). The case counts demonstrated as Ncase are the case counts in the analyses of all participants. The full list of case and control counts in the analyses of all participants and those hospitalized only are described in Supplemental Table 5.

Figure 3. Influence of age and clinical risk factors for the effect of rs10490770 risk allele carrier status on death or severe respiratory failure.

(A) Age distribution in COVID-19 patients of European ancestry who died or experienced severe respiratory failure (n = 2005). Median (IQR) age was 67.2 (range, 59–76) years in carriers (n = 506) and 72 (range, 63–78) years in noncarriers (n = 1499). (B) ORs of rs10490770 risk allele carrier status for death or severe respiratory failure. Regressions were performed within subgroups stratified by age (age ≤60 years and age >60 years) (cases/controls = 2005/7047) or by the number of established risk factors (0, 1, or ≥2); BMI ≥30, smoking, cancer, chronic kidney disease, COPD, chronic heart failure, transplantation, and DM (cases/controls = 898/6454). All analyses were adjusted for age, sex, genetic PCs 1 to 5 as fixed effects, and a dummy variable representing the participating studies as random effects.

Figure 4. Multivariable regression models and risk prediction estimates for death or severe respiratory failure.

Multivariable regression analyses for death or severe respiratory failure were restricted to European-ancestry individuals with complete information of demographic variables (green), comorbidities (blue), and rs10490770 risk allele status (red). n = 7352 for all and n = 2499 for age ≤60. CKD, chronic kidney disease; CHF, chronic heart failure. Error bars indicate 95% CIs. (A) Forest plots comparing ORs from multivariable regression models. The size of each dot represents the frequency of the risk factors. (B) Comparison of AUCs of predictions for COVID-19 outcomes. rs10490770 risk allele and nongenetic clinical risk factors were included separately in addition to age and sex in multivariable regression models.