Health 2020 data collected decades ago is yielding useful results – in ways we couldn’t have foreseen
The Human Genome Project was launched in 1990. It took 13 years and US$3 billion to complete mapping out the billions of DNA bases that make up a full set of a person’s genes and everything in between. That is a long time to wait and a huge price to pay for just one genome. But the Human Genome Project led to technological advances that found better, faster and cheaper ways of getting the job done. Today, analysing an entire human genome costs less than $US 5,000 and only takes a day or two.
Collecting genetic data in HEALTH 2020
HEALTH 2020 was launched in 1990, the same year as the Human Genome Project. The blood samples we collected were originally used to measure blood levels of sugar and cholesterol, and the results were given to participants for their information. In our early studies within Health 2020 we also used the stored blood samples to measure levels of dietary markers, inflammation, lipids, vitamins and hormones.
We had not anticipated being able to ever measure genes, but the fast progress in gene-mapping technology handed us the ability to use these blood samples in a new way, to generate useful knowledge for controlling cancer.
We don’t sequence whole genomes of HEALTH 2020 participants; that would be unnecessary and far too expensive. Instead, we perform a genome-wide association study: we extract DNA from the stored blood samples and scan selected locations on the genome (usually around 500,000 out of the ~3 billion DNA bases). These locations host common genetic variants (not rare mutations that put people at high risk of cancer), that make us unique individuals, much like the variants that influence our height or hair colour. These data are used by Cancer Council Victoria researchers to search for new variants associated with cancer risk.
What can genetic data tell us?
When we compare the DNA scan results for people who have had a certain type of cancer to those from people who don’t, we look for patterns: Do any genetic variants crop up more often in people with cancer? That way, we can identify ‘suspect’ variants that could be influencing the risk of cancer. In the last few years, HEALTH 2020 participants’ data have been used to successfully identify genetic markers associated with prostate, breast, bowel, ovary and many other cancers.
Any single genetic variant does not make much of a difference to a person’s chance of getting cancer but, collectively, the small risks associated with a large number of these variants can add up to something substantial.
Our researchers usually work in international collaborations, often combining results drawn from hundreds of thousands of participants.
This is because the wider the net we cast – looking at data from as many people as we can, from varied populations – the better our chances of detecting the small risks we search for.
Our ultimate goal is to develop models based on this genetic information that would give a person an indication of whether they were in the low-, mid- or high-risk category for a type of cancer. This information could inform their prevention and screening decisions, and in the future may lead to more effective treatments.
To maintain the privacy of HEALTH 2020 participants, their data is always anonymised, so that collaborating researchers cannot access personally identifying information.