A comprehensive study analyzing genetic and health data from over 400,000 individuals has found that socioeconomic status and exposure risk, not genetic variations in immune response genes, are the most significant predictors of SARS-CoV-2 infection. The research, which examined the human leukocyte antigen (HLA) region’s relationship to COVID-19 test positivity, concludes that most previously reported genetic associations may be misleading due to confounding social and environmental factors.
The study, based on a dataset of 419,234 individuals with recorded HLA genotypes and COVID-19 PCR test results, is the largest of its kind. After accounting for variables such as estimated socioeconomic status (SES) and exposure risk based on geographic location, the researchers found that only eight HLA alleles were significantly associated with COVID-19 positivity, down from 41 before these adjustments.
Among the strongest findings was the impact of socioeconomic status on infection risk. The study used a neighborhood deprivation index (NDI) as a proxy for SES and found it had the largest effect size of all factors analyzed, including genetic ones. This supports prior research linking lower SES to increased infection rates, likely due to factors such as dense living conditions, frontline occupations, and limited access to healthcare.
The HLA region, long considered a key player in immune system response, has been the subject of many smaller studies since the beginning of the pandemic. However, these studies often struggled with inconsistent findings and small sample sizes. The new research suggests that many of those earlier associations may have been spurious, the result of failing to account for demographic and environmental differences.
Even among the eight HLA alleles that remained statistically significant after adjustments, effects varied across different ancestry groups and were sometimes contradictory. For instance, the allele HLA-DRB108:03 was associated with protection in individuals of Asian/Pacific Islander ancestry, while a related allele, DRB108:02, was linked to higher infection risk in Hispanic populations, mirroring earlier findings but raising questions about the biological consistency of such associations.
The study also examined paired allele effects and found only two combinations that enhanced individual allele effects, though one of these may reflect hidden population substructures rather than genuine genetic effects.
Other commonly discussed risk factors, including blood type and age, showed less relevance in the new analysis. While blood type O appeared protective in individuals of European ancestry, no such effect was observed in other groups. The study found no significant association with age, and only confirmed male sex as a risk factor in one ancestry group.
The researchers acknowledged the potential impact of changing viral variants and the rollout of COVID-19 vaccines during the study period (late 2019 to mid 2021). However, they noted that most T-cell immune targets remained stable across variants and that differences in infection risk before and after vaccine availability were not substantial in their models.
Still, the study’s authors caution that more granular data on individual behavior, occupation, and vaccination status could further refine the findings.