New research reveals how specific genetic variants and chronic conditions like depression and fibromyalgia increase the risk of long-COVID, offering insights into potential treatments.

Study: ‭Multi-ancestry GWAS of Long COVID identifies‬ ‭immune-related loci and etiological links to chronic fatigue‬ syndrome, fibromyalgia and depression‭. Image Credit: Lightspring / Shutterstock

*Important notice: medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.

In a recent research paper uploaded to the medRxiv preprint* server, researchers at 23andme conducted the largest meta-analysis of genome-wide association studies (GWAS) comprising more than 174,000 participants from ethnically diverse backgrounds, including European, Latinx, and African-American cohorts, to identify genetic loci or phenotypic traits demonstrating an increased risk of long-COVID.

Study findings revealed that three specific genetic loci, HLA-DQA1–HLA-DQB1, ABO, and BPTF–KPAN2–C17orf58, and three phenotypes were at significantly heightened risk, highlighting high-priority populations for interventions against this poorly understood disease.

Background

Long-COVID is the condition of prolonged coronavirus disease of 2019 (COVID)-like symptoms that persist or develop for months or even years following survival from a severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) infection.

An estimated 10-80% of COVID survivors suffer from the condition, whose symptoms include severe fatigue, cognitive ‘brain fog,’ and constant breathlessness.

More than 65 million people have been confirmed as long-COVID patients. Unfortunately, while recent genome-wide association studies have provided an understanding of the immune responses underpinning acute SARS-CoV-2 infections, little is known about the biology and risk factors contributing to long-COVID.

Long-COVID is a multiorgan condition, with several patients reporting multiple, frequently unrelated symptoms, complicating studies aimed at validating the roles of autoimmune responses, viral load, and inflammation in long-COVID incidence.

Epidemiological investigations have established links between long-COVID occurrence and population phenotypes, suggesting that preexisting diseases may increase long-COVID risk. The study identified 13 phenotypes linked to long-COVID, including chronic pain, chronic fatigue, and fibromyalgia, with varying association strengths.

Leveraging large, multi-ethnic genome-wide association study (GWAS) datasets provides a means to test this hypothesis while further allowing for identifying genetic predispositions (loci) at heightened risk.

About the study

The present study leverages extensive genotype data from the 23andMe database (n = 174,432) across diverse ethnic backgrounds to identify genetic variants at heightened risk of long-COVID and elucidate phenotypic clusters with similar hierarchies in disease susceptibility.

Additionally, it employs Mendelian randomization models to explore causal relationships between genetics and phenotypes (e.g., COVID-19 severity) and their combined impacts in determining a COVID-19 survivor’s long-COVID risk.  

Study participants were identified from online repositories recording confirmed COVID-19 survivors. Potential participants were invited via email and requested to provide demographics, socio-behavioral, and health characteristics.

Additionally, detailed medical reports on participant age, sex, chronic disease, education level, and health behaviors (smoking and alcohol consumption) were obtained from participants or previous 23andMe publications.

Participants without medically diagnosed SARS-CoV-2 infections or those reporting first COVID-19 infections in the preceding three months before recruitment were excluded from the study cohort.

Participant genotyping was carried out using saliva-derived DNA by the Laboratory Corporation of America, with data used for participant genetic ancestry classification (hidden Markov model [HMM] classifier) and GWAS loci characterization.

All analyses were run separately for each genetic ancestry category (European, Latinx, African-American). Identified genetic variants were clustered based on human leukocyte antigen (HLA)-imputed allelic similarity, and age- and sex-adjusted logistic regression were carried out to establish associations with long-COVID.

Finally, genetic causal liability between phenotypes and long-COVID was computed using two-sample Mendelian randomization (MR). The study also explored a more specific “Long-COVID Impact” phenotype, focusing on cases where long-COVID significantly impaired daily living activities.

“To limit our analyses to unrelated individuals, we selected participants such that no two individuals shared more than 700cM of DNA identical by descent. In a scenario where a case and a control have at least 700cM of DNA identical by descent, we removed the control from the analytical sample. After excluding unrelated individuals while prioritizing the retention of cases, the analytical cohort for GWAS consisted of 53,764 cases and 120,688 controls for long-COVID, and 32,087 cases and 121,298 controls for long-COVID Impact.”

Study findings

Initial database screening revealed 332,638 COVID-19 survivors, of which 179,167 met the study inclusion criteria and were included in subsequent analyses.

Most participants were identified as being of European ancestry (78.9%), with Lantinx (16.6%) and African-American (4.5%) comprising the other ethnic cohorts.

Participants from other ancestry groups were excluded from further analyses due to insufficient sample size for robust statistical analyses.

Comparisons between long-COVID (cases) and non-long-COVID (controls) survivors revealed that females (66.7%) and non-tobacco consumers (37%) were more likely to develop the condition compared to their male and tobacco-using counterparts.

After adjusting for covariates (age, ancestry, sex), preexisting high blood pressure (odds ratio [OR] = 1.52), depression (OR = 1.98), autoimmune conditions (OR = 1.55), and cardiometabolic conditions (OR = 1.60) were more frequently observed in cases compared to controls.

GWAS analyses using ~110 million imputed genetic variants revealed three distinct loci—HLA-DQA1–HLA-DQB1, ABO, and BPTF–KPAN2–C17orf58—significantly associated with long-COVID across all three ancestry cohorts. Additionally, the rs78794747 variant was found to be significant in participants of European ancestry.

Notably, analyses of phenotype-long-COVID associations revealed 13 phenotypes with varying association strengths, with chronic pain, chronic fatigue, and fibromyalgia being the most common co-occurrences observed. Since chronic pain is a participant-provided, non-specific 23andMe database definition, it was excluded from MR modeling.

“We found strong evidence of a potential causal effect to each of these three conditions on long-COVID (chronic fatigue: OR=1.59 (95%CI: 1.51, 1.66), fibromyalgia: OR=1.54 (95%CI: 1.49, 1.60), and depression: OR=1.53 (95%CI: 1.46, 1.61); estimates from IVW-MR). These effects persisted when employing robust MR approaches including weighted median and MR Egger.”

Conclusions

The present study utilizes the largest multi-ethnicity GWAS dataset hitherto used in long-COVID investigations and provides evidence that individuals with genetic predispositions to chronic fatigue, depression, and fibromyalgia, as well as other phenotypes such as autoimmune conditions and cardiometabolic conditions, are at significantly higher risk of long-COVID than individuals without these conditions.

The study’s two-tier approach, focusing on both general long-COVID and a more severe “Long-COVID Impact” phenotype, helps refine our understanding of long-COVID’s broader and more debilitating forms.

Since these conditions also result in higher probabilities of hospital visits following COVID-19 infections, it corroborates reports highlighting higher long-COVID rates in hospitalized COVID-19 patients.

“Together, these findings can help identify at-risk individuals for Long COVID, as well as provide novel insights that support developing therapeutic options for both long-COVID and symptomatically similar conditions.”

*Important notice: medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.