Gene variant linked to greater COVID-19 risk

The UK Biobank isn’t the place to go and get cash from. As the name suggests it’s a bank for biology. It contains samples and health information from around five hundred thousand volunteers and gives anonymous data and samples to researchers here and from overseas. The aim is to help them come up with treatments for a wide range of serious and life-threatening illnesses including cancer, heart diseases, stroke, diabetes, and so on. As you can probably tell Biobank’s data, gathered over a number of years, is potentially also very useful in the fight against Covid 19, and this week a study from the University of Exeter found that those with a certain genetic make-up... may have a worse time dealing with the disease. A gene which usually helps the body transport fats, called APOE, has emerged as a risk factor for the disease: if a person carries what’s referred to as the APO-E4 forms of the gene, they’re more than twice as likely to succumb to Covid-19. What’s interesting is that this genetic profile is also linked to developing dementia. And initially when the team looked at who was getting Covid, dementia sufferers were heavily represented. But when this was taken into account, there was still a strong link: there were people with Covid and APOE4 who didn’t have dementia. This means that it’s not the dementia itself, but something about the APOE gene that’s changing people’s risks. Adam Murphy’s been hearing from David Melzer how he spotted the link...

David - The disease that came out amongst the oldest sample as by far the biggest risk factor was dementia. People who had been diagnosed with dementia since baseline approximately 12 years ago were three times more likely to be admitted to hospital and test positive for COVID. And so the obvious thought was, "could this possibly be an effect of the APOE gene?" The APOE4 mutation has also been linked to a series of viral infections, including HIV. So it was the obvious question. The other possibility of course is the higher risk of dementia might be due to higher prevalence of dementia in care homes. So it seemed an important thing to explore. It turned out that people with a full 4 mutation were over twice as likely to test positive in hospital during that peak period. And we did a number of tests to check whether it was robust, for example. It was remarkably robust, really; the estimate didn't change at all. So we thought it important to publish so that other groups could look at it and check it in their own data.

Adam - How can you tell that the risk is based on the gene? How do you factor out things like age and that kind of thing?

David - We have 12 years of hospital records on the whole sample of over 300,000 people. The UK Biobank study has also linked up these diagnostic data to primary care records, so we can search their anonymised primary care records. Basically we identified the people who've been diagnosed before and exclude them from the analysis. And as I said, it basically made no difference. So it doesn't look as if the association between APOE, the risk gene, and COVID-19 positivity in hospital is due to dementia. People in Biobank with this mutation - double mutation - actually had dementia.

Adam - Are there any recommendations that come off the back of this?

David - Well, one of the things I've been concerned about is in the early phases of the pandemic, many countries introduced age-based cutoffs for risk. In the UK age, 70 was used as an indicator. And a lot of my group's work over the last 15 years has been how different individual older people are. I mean, some people die of age-related diseases in their sixties; others go on to be centenarians and remain active even beyond the centenarian age. So lumping everybody together on the basis of chronological age is convenient - I mean, I can see why you would do it at the height of the pandemic - but we were hoping to inform policies going forward to be more focused on what the actual risks are.