The personal genomics and biotechnology company 23andMe is perhaps best known for its home DNA testing kits, whereby consumers can learn about their genetic make-up or ancestry.
However, after the COVID-19 outbreak was declared a global pandemic, the company switched gears and looked to see how their gene testing services could offer a helping hand in understanding the disease’s pathophysiology.
One aspect of COVID-19 that has puzzled scientists and clinicians alike is why, when infected with SARS-CoV-2, do some patients display mild to moderate symptoms (or no symptoms at all), whereas other patients develop severe symptoms that can prove fatal?
23andMe questioned whether genetics could play a role here. Why? Well, we know that specific genetic variants can render individuals as less or more susceptible to developing other infectious diseases such as HIV, malaria and norovirus. 23andMe has contributed to this field of research, publishing a study in 2017 which identified almost 60 genetic variants associated with susceptibility to one of 17 infectious diseases.
To cause infection, SARS-CoV-2 must enter cells, and does so by attaching to a human protein known as ACE2. Genetic variants can exist in and around the gene that encodes this protein, which might impact exactly how much of the protein is produced, or how it functions. It's therefore plausible to consider the fact that genetic variation might play a role in COVID-19 severity. But, a large-scale study in a range of populations would be required to gather enough data to prove or negate this.
In early April, 23andMe announced its plans to conduct such a study utilizing its large bank of genetic data from its customers who had consented to their information being used for research purposes.
The study takes the form of a GWAS, or genome-wide association study, whereby the aim is to identify genetic variants in an individual's genome that is associated with differences in COVID-19 severity. The announcement of the research study was received well, with over half a million participants enrolling within the first few weeks.
As such, on May 13, 23andMe announced that it was expanding the project and opening enrolment to individuals that had been hospitalized with severe COVID-19 but were not 23andMe customers.
“Opening up the research to individuals with more severe symptoms will increase our power to learn how genes play a role in the severity of this disease,” said Joyce Tung, PhD, 23andMe’s Vice President of Research.
Analyzing genetic association
Fast-forward to the present day, and 23andMe are sharing their preliminary results, which are certainly interesting. The data seems to lend further evidence to the notion that an individual's blood type, determined by the ABO gene, is associated with differences in COVID-19 susceptibility.
After analyzing over 750,000 individuals' data, the company states that when comparing the research participants who reported that they had tested positive for COVID-19 to those who tested negative, variants in the ABO gene were associated with a lower risk.
Of course, it is important to note here that this is a self-reported measure, which is vulnerable to validity issues. The genetic variant in the ABO gene is a single nucleotide polymorphism, rs505922 – a T nucleotide at this location is associated with a lower risk. The P value here is measured at 0.88.
23andMe state that these preliminary findings support at least two recently published pre-print studies, one by Zhao et al and another by Ellinghaus et al that explore the role of the ABO gene in COVID-19.
Looking at blood types
Several studies have also suggested that blood group could be implicated in susceptibility and severity for COVID-19, and as such, the 23andMe study is also exploring this.
The company estimated the contribution to risk of blood group by comparing each blood group against each of the others in the data from the 750,000 enrolled participants.
In the dataset available from the survey, the percentage of respondents that reported a positive test for COVID-19 was lower for individuals that are O blood type (1.3%), compared to A (1.4%), B (1.5%) and AB (1.5%) blood types. Individuals within groups A, B, and AB were not statistically different from one another, and this relationship stands when adjusting the measures for age, sex, BMI, race, ethnicity and co-morbidities.
A closer look at blood group O
From 23andMe's reported data set, there were no statistical differences between self-reported rhesus factor (blood type + or -) in blood group O individuals, challenging the findings of previous research.
To assess blood group with regards to risk of acquiring infection, the company restricted the data to individuals with a high probability of exposure, such as health care professionals, essential workers and individuals with close contact with known cases. The trend in results reflected the original dataset, with the reporting of a positive COVID-19 test being 3.2% for blood group O, 3.9% for A, 4.0% for B and 4.1% for AB.
Blood type O showed a protective effect against both acquiring (OR = 0.86, p < 0.0001), and being hospitalized for the infection (OR = 0.81, p = 0.05). The protective effect of acquiring the infection strengthened in models restricted to the “exposed” population (OR = 0.81, p < 0.0001).
In percentages, in the entire population, individuals with blood group O were 9-18% less likely to test positive when compared to other groups. "Exposed" individuals with blood group O were 13-26% less likely to test positive.
The research study is still ongoing, and enrolment can be completed online. In a press release the company says: "Ultimately, we hope to publish our research findings in order to provide more insight into COVID-19 for the scientific community."