The health secretary, Matt Hancock, this week shared his shock at discovering that he is at greater than average risk for prostate cancer, despite having no family history of the disease.
The revelation came after he took a predictive genetic test that assesses risk for 16 common diseases, including coronary artery disease, type 2 diabetes, asthma and breast and prostate cancers.
Hancock said the test might have “saved his life” and that such tests should be urgently rolled out on the NHS, to guide screening programs and the age at which drugs like statins are prescribed. However the suggestion was immediately met with controversy. Some claim that the usefulness of predictive DNA tests has been overstated, pointing to the fact that, while Hancock described his own 15% lifetime risk of prostate cancer as “high”, it was only marginally above the population average risk of 12%. So are predictive genetic tests going to revolutionise prevention and treatment of common diseases or should we be sceptical?
Until now, genetic testing for disease risk has largely focused on versions of single gene that confer a substantial amount of risk, such as BRCA for breast cancer.
The latest predictive tests for a range of common diseases take a different approach: they aggregate the tiny contributions to risk made by hundreds or even thousands of genes to give a personalised score. Because the risk is spread out over many genes, people can end up at the very high-risk end of the spectrum by chance, without having a family history of a particular illness.
Our genes are made of deoxyribonucleic acid, DNA, which forms double helix strands inside the nuclei of our bodies’ cells.
DNA contains information that is passed from generation to generation and directs the development of our bodies.
Scientists began studying ancient DNA 20 years ago when Svante Pääbo used gene amplification techniques to extract and analyse genetic material from Neanderthals.
At first only fragments of Neanderthal DNA could be studied, but by developing techniques to piece together small overlapping fragments, it was possible to recreate the entire set of genes, or genome, of a Neanderthal.
Subsequent research showed that most people outside Africa contain small numbers of Neanderthal genes, the result of interbreeding between the two species as modern humans emerged from their African homeland about 70,000 years ago.
In 2010 scientists discovered the genome of a completely new human species in bone and tooth fragments found in the Denisova cave, Siberia. These people are now known as Denisovans.
Research published last year suggests that modern humans interbred with both Neanderthals and Denisovans on numerous occasions over the past 250,000 years.
Genomics Plc, the Oxford-based data science company that provided the health secretary with personalised risk scores, has just released results showing the extent to which predictive DNA tests can reveal the likelihood of diseases from cancers to asthma and coronary heart disease.
Based on data from 160,000 men (UK Biobank participants), it found that those who scored in the top 1% in terms of risk for heart disease had the same risk at 45 years old as an average man at 60-65 and one of the lowest risk men in their mid-70s.
When the scientists looked more closely at common characteristics of the high-risk men, they found they had slightly higher cholesterol, BMI and blood pressure – but not extraordinarily high.
Prof Peter Donnelly, a statistical geneticist and CEO of Genomics Plc, said: “Many of them are individuals who would be invisible to the health system currently. They’re happily wandering around the streets at much higher risk of heart disease but not aware of it. If you were one of these men that’s potentially helpful to know.”
Similarly for breast cancer, the top 1% of women – based on genetic risk score – had a 30% lifetime risk of breast cancer, while the lowest 5% had a 2%-3% lifetime risk. Genomics said its tests cost £20-£40 per person, much lower than the £500-£1,000 cost to sequence a whole genome.
“The world we’re in, where we screen women just based on age and offer mammography at 50, just doesn’t make sense,” said Donnelly. “We should be screening these women much earlier.”
Prof John Bell, a professor of medicine at Oxford university who led a recent government-commissioned review of the life sciences industry, said the approach could have a “quite profound” effect on the ability to manage disease.
Currently, he said, screening programmes face huge problems: they are expensive, they give lots of false positives and miss people such as women in their early forties who never enter routine screening for breast cancer, but who have a high genetic risk. “It’s all slightly hopeless,” he said. “This is exactly what we need.”
David Spiegelhalter, professor for the public understanding of risk at the University of Cambridge, agrees that genetic tests could allow the NHS to rapidly identify those who may need closer monitoring. However, he said care would need to be taken in how risk scores were communicated to individual patients, most of whom (as in the case of Matt Hancock) fall somewhere in the mid-range for most illnesses.
“Even the most extreme 3% are only at around 2-3 times average risk, and so most people will only be given a moderately raised or lowered risk,” said Spiegelhalter. “It is essential that these results are communicated properly, to avoid any suggestion that they are predicting the future.”
A significant concern is that the vast majority of data that has been used to identify risk genes has come from people of white European heritage, who are significantly overrepresented in genetics studies and databases such as UK Biobank. The extent to which genes confer risk also depends on the overall genetic backdrop, which varies across ethnicities. This means that currently the tests developed by companies such as Genomics Plc work less well for people of non-white European heritage.
Donnelly said the company is currently assessing the extent of the performance gap and looking for ways to improve the technology so it works better for everyone. “One has to even think really hard about whether it should be rolled out now when it’s more useful for some individuals than others,” he said, adding that if there was a significant gap then he would be against rolling it out quickly.