LONDON – Leading genome sequencing groups are launching the first meta-analysis in the hunt for genetic factors that explain why some people have worse COVID-19 symptoms than others, after agreeing to share patient sequence data from around the world.
The COVID-19 Host Genetics Initiative (CHGI), set up by scientists at the Institute for Molecular Medicine Finland (FIMM), now includes 151 registered studies that are searching for genetic variation associated with severity and outcomes. The findings will be a potential source of drug targets, both for de novo discovery and repurposing, and also could form the basis of prognostics for identifying people at unusually high risk. The aim of the initiative is to make it possible to work cooperatively, for example, by agreeing to standard protocols; to increase the statistical power of the various studies by organizing analyses across datasets; and to provide a platform to share research.
To date, most investigations into the host genomics of coronavirus infections have used mouse models infected with the SARS-CoV virus that was responsible for the 2003 epidemic.
But now dozens of studies are being set up using SARS-CoV-2 patient samples.
“The variability observed during exposure and infection to SARS-CoV-2 makes it very likely that there are human genetic factors influencing the response to the virus,” said Laurent Abel, of the Imagine Institute of Genetic Diseases in Paris, France. Abel previously has uncovered some 100 human mutations that cause unusual susceptibility to infectious diseases, including tuberculosis and herpes.
Naveed Aziz, chief administrative and scientific officer at CGEN, Canada’s national genome sequencing and analysis platform, agreed. “We realized, as have a lot of geneticists globally, that there seems to be a strong genetic association to severity of illness.”
It was evident from the earliest information from China that people with co-morbidities experience more severe disease, Aziz said. “But even in people with the same underlying conditions the response was different,” he told BioWorld.
On April 24, CGEN was awarded CA$20 million (US$14.2 million) in federal funding for Canada’s national COVID-19 host genome initiative, which aims to sequence DNA from 10,000 patients. That will provide much-needed data to support diagnosis, prognosis and treatment of COVID-19 infections, said Aziz. “As we take this step forward, it is important to build a database for Canadian collaborators of genome data and associated records, but also for collaboration across the world.”
The groups lined up to take part in CHGI range from large, established biobanks of DNA samples and medical records, such as UK Biobank, Finland’s Finngen and Japan Biobank, to new research projects looking to recruit 100 subjects that are just getting off the ground.
Similarly, studies registered in CHGI vary in their objectives, from genome-wide association research to projects focusing on human leukocyte antigen genes that influence the body’s immune response, and the search for variants in genes coding for the ACE-2 receptor via which the virus enters host cells.
Meanwhile, other researchers are looking at specific populations, for example, people intubated in intensive care, and fit, younger people who suffer severe disease.
CHGI was able to get off the ground quickly and get early initial agreement on what analyses should be performed, because it piggy backs on the International Common Disease Alliance, launched in Washington last September.
The alliance, including academics and industry, was set up to accelerate understanding of how tens of thousands of genetic variants picked up in genome-wide associations studies contribute to the biology of common diseases. CHGI has been able to build on the relationships and online communication channels established by the alliance.
Further boosting CHGI’s aim of carrying out large enough studies to generate statistically rigorous results, several research institutes and companies are offering to carry out genotyping and sequencing for hospitals that do not have their own capabilities. Those include FIMM, the Human Genomics Facility at the Erasmus Medical Center in Rotterdam, the Netherlands, Illumina Inc. and the Wuxinextcode subsidiary, Genomics Medicine Ireland.
The sharing of anonymized datasets will be via the European Bioinformatics Institute’s European Genome-phenome database or the U.S. National Institutes of Health Human Genome Research Institution’s Anvil (Analysis, Visualization and Informatics Labspace) system.
The first new study to get up and running under the CHGI banner is Gen-COVID, a consortium of more than 20 hospitals in Italy, which on March 16 began collecting samples from 2,000 patients, with the aim of identifying genetic causes of clinical variability that could be applied in diagnosis, prognosis and personalized treatments.
A genome-wide association study of the 2,000 samples will be carried out at FIMM, while the University of Sienna, Italy, will conduct whole exome sequencing of the protein coding regions of the genome. The patient registry and biobank generated as a result will be open to academic and industry partners. The project has backing from Astrazeneca plc, Glaxosmithkline plc, Regeneron Pharmaceuticals Inc. and Illumina.
UK Biobank, already recognized as one of the world’s leading resources for genomics research, has made moves to tune its data for COVID-19 host genomics studies, with the results of diagnostic tests, primary care, hospital and death records from any of the 500,000 participants who have been tested and/or treated for the virus being uploaded to the database.
Apart from understanding why severity varies so much, that will make it possible to explore how co-morbidities, and different therapies and treatments, affect outcomes.
“This data will help researchers understand the differences between individuals,” said Rory Collins, chief executive of UK Biobank. “What are the differences in their genetics? Are the differences in their genes related to their immune response? Are there differences in their underlying health?”
Russ Altman, professor of biomedical data science at Stanford University, one of 15,000 researchers in 85 countries approved to use UK Biobank, said adding the COVID-19 data “provides an unprecedented opportunity to study how clinical and genetic factors affect the spread and outcome of this disease.”
Other, smaller host genomics studies also are shifting focus in response to the pandemic. One example is the Global Genetics of Susceptibility and Mortality in Critical Care, which since 2016 has been recruiting patients with emerging infections, including MERS and SARS, who need intensive care. The U.K. arm of the study now aims to recruit every patient in the country with COVID-19 infection who is intubated.
Meanwhile, Abel and his colleague Jean-Laurent Casanova, who have joint postings in Paris and at Rockefeller University, New York, have set up a France/U.S. collaboration to study DNA of outliers – patients who had unexpectedly severe COVID-19 infections, or people who are known to have had multiple exposures but remain free of disease.
On a similar theme, researchers at École polytechnique fédérale de Lausanne, Switzerland are to sequence the genomes and blood transcriptomes of 100 patients under the age of 50 who developed severe or fatal infections.
In Japan, a group at the Center for Genomic Medicine at Kyoto University is conducting a 1,000-subject study to find out if variants in HLA genes are associated with risk of the onset and progression of COVID-19 infections.
For Aziz, the importance of collaboration between all those various studies is illustrated by the way in which international communication of genome sequences of SARs-CoV-2 made it possible to rapidly set up vaccines projects, start the hunt for therapies and track the evolution of the pandemic.
“The biggest advantage we see is just sharing data and information,” he said. The tools now available to do that mean it is possible to open up data without compromising confidentiality. “We can take the question to the data, rather than giving data [away].”