Polygenic risk scores and the path to prevention

The Human Genome Project produced the first sequence of the human genome more than two decades ago, and Sir Peter reflects that initial expectations of the emergent genomics field were probably a little high.

“While its scientific achievement was undeniable, the sense, back then, of the timescale over which it would have an impact on people’s lives was way too optimistic,” he says, explaining that the practical value of genome sequencing was relatively limited at first. “Until a few years ago, if I had the entire DNA sequence of a healthy 40-year-old, I’d only learn something medically actionable in about 1% of cases.”

Actionable results in 70% of cases

This was largely because genetic risk was initially understood through the lens of rare mutations – single changes in DNA that have significant consequences, such as those responsible for cystic fibrosis, Huntington’s disease or muscular dystrophy. In addition to being rare, these conditions typically present early in life, meaning that if a person reached middle age without symptoms, their genome revealed little of clinical significance.

“The landscape has changed dramatically over the last few years,” says Sir Peter. “Now it’s the case that if I had the entire DNA sequence or substantive genetic information on a healthy 40-year-old, I’d learn something medically actionable in about 70% of cases.”

The difference lies in how genomics now addresses common diseases such as heart disease, diabetes, breast and prostate cancer – conditions that are not only leading causes of morbidity and mortality but also the greatest drain on healthcare budgets. For these common diseases, risk is not dictated by a single gene, but by the complex interaction of many genetic variants.

“If you take heart disease as an example, there are something like a million different places in your DNA, each of which affect your risk,” says Sir Peter. “While each has a minuscule effect on its own, collectively they can be combined into a polygenic risk score, which is a summary of a person’s genetic predisposition to a given condition. Genetics doesn’t change in a person’s lifetime – it gives us a fixed component of risk that we can measure – and now, finally, act on.”

Data drives genomics advances

The risk prediction accuracy of PRS is impressive. Sir Peter says that 40% of men with a high polygenic risk for heart disease will be diagnosed with the disease by the time they’re 70, compared with just 2-3% of men with a low PRS. Similar patterns are seen in other diseases. In prostate cancer, genetic risk can increase the likelihood of diagnosis by up to 40 times, and genetic predisposition plays a measurable role in conditions like obesity.

“People in the top 10% of genetic risk for obesity are, on average, obese by the time they’re in middle age,” says Sir Peter.

So what has changed to make genomics so much more powerful today than it was before? The primary driver is data.

“We’ve now got very large datasets on people with diseases and people without diseases,” says Sir Peter. “These massive datasets enable the development of more accurate prediction tools, including those that go beyond disease risk to forecast who might benefit most from specific therapeutic drugs. Even when drugs work, they often work for some people but not others, and genetics will be part of solving those mysteries.”

‘Momentum is building’

Despite all the advances, the integration of genomics into mainstream healthcare has been slower than some might hope.

“Health systems are often inherently conservative and not necessarily quick to adopt innovation,” acknowledges Sir Peter. “But momentum is building. For example, the UK government has embraced the potential of using algorithms based on AI and genomics to better understand disease risk and focus more on prevention.”

While many diseases already have screening programs targeting high-risk individuals, Sir Peter argues that the key is to direct those interventions more precisely.

“For example, breast screening in the NHS is currently offered to all women when they turn 50,” he says. “But quite a lot of women have the same level of polygenic risk as a 50-year-old by the time they’re 40, so it makes no sense to just ignore that and wait till they get to 50.”

Part of Genomics’ focus is on making this kind of precision medicine a reality, and the company has developed polygenic risk scores for over 50 diseases and physiological traits, including things like blood pressure and cholesterol levels. The company is already involved in supporting large-scale population health programs, such as the UK’s Our Future Health, which aims to recruit five million participants.

“We’re providing the risk scores for that initiative,” says Sir Peter. “In a couple of years’ time, there will be millions of people who know about their genetic risk, and that’ll just change the whole conversation.”

The other side of Genomics’ business supports drug developers in identifying and validating drug targets, with the goal of improving the efficiency of clinical trials and ultimately bringing new treatments to market more rapidly.

“At the moment, 95% of drug targets fail when they get to clinical trials,” says Sir Peter. “But there is very good evidence that if you have human genetic support, drug targets are more than twice as likely to succeed.”

Healthcare needs to be bold on prevention

Of course, PRS still captures only a portion of overall genetic risk and does not account for environmental or lifestyle factors, and this factors into Genomics’ future roadmap. The company is focused on two main directions: expanding the range of conditions for which risk can be predicted, and integrating genetic risk with clinical and other molecular data, an area where it already has validated technology and products

“Over time, we’ll be using other sorts of information, whether it’s from proteomics or metabolomics to give a snapshot of what’s happening,” says Sir Peter. “While genetics provides insight into baseline predispositions, these additional biomarkers can capture dynamic, real-time aspects of a person’s health.”

This multi-layered approach aims to make individuals “copilots” in their health journey, not just passive recipients of care. Sir Peter believes that most people are unaware of the specific conditions they’re at highest risk for.

“There will be two or three diseases that you’re at high risk for, and you have no idea what they are, and your doctor has no idea what they are,” he says. “By identifying these risks, individuals and health systems can take preventive action early – ideally before symptoms arise.”

However, understanding risk and acting on it are two different challenges.

“If the only thing we had to rely on was people changing their behavior, it would be less effective,” admits Sir Peter. “Personalized risk information might motivate some, but others will still struggle, which is why health systems must also play a role, such as inviting high-risk individuals into screening programs earlier.”

Genomics envisions a future where genetic data becomes a routine component of health records and primary care, but Sir Peter says the biggest hurdle remains structural.

“While the technology is ready and the evidence is promising, healthcare systems often lack the organizational and financial infrastructure to prioritize prevention,” he says. “Everyone talks a good game about prevention, but without senior leadership, accountability, and resources, it will remain on the periphery.”

“If health systems keep focusing on service delivery, they’ll keep talking about prevention for a long time, and it won’t have the impact it needs to have. We need to be bold and audacious, and not just talk about prevention being important, but to put real resources behind it.”

Photographs courtesy of Genomics Ltd.