Polygenic risk scores: how useful are they?

Being able to predict an individual’s risk of common conditions is regarded by many as the holy grail. So, where are we now?

Many of the serious health problems people face in our society, such as heart disease, cancer, diabetes and arthritis, are influenced by multifactorial inheritance. This means that, rather than being the result of a single gene variant, they are influenced by a large number of genetic variants, and by environmental factors as well.

What is a ‘polygenic score’?

The likelihood of a single gene variant causing a specific outcome can be shown in terms of percentage, and this is known as the penetrance of the gene. Polygenic scores aim to quantify the cumulative effects of a number of genes, which may individually have a very small effect on susceptibility. They can be used to predict a person’s likelihood of displaying any trait with a genetic component – including traits such as height or weight – and are known as polygenic risk scores when they refer to disease predisposition.

Polygenic scores aren’t new, but as they rely on large databases of genome sequences with accompanying phenotypic data, they have become more prevalent (and more powerful) as sequencing has become cheaper and faster. Initiatives such as the 100,000 Genomes Project provide ideal databases for this type of research: the bigger the dataset, the more variants associated with a particular trait can be identified.

What’s the use?

The use of polygenic scores to measure inherited predisposition to breast cancer could, for example, be very useful. The impact of variations in the BRCA genes is well known: they can have a large impact on a person’s risk. But over a hundred other genes have been identified (for example, here and here) as playing a role in an individual’s chance of developing breast cancer. Access to polygenic risk scores could be helpful in targeting higher-risk people, for example to offer more screening opportunities such as mammograms. A test which looks at 18 variants is being trialled at the Manchester University NHS Foundation Trust.

The tests have the potential to be very affordable. A recent study using UK Biobank data developed a test called the Genomic Risk Score for coronary heart disease which the authors estimate would cost around £40.

Polygenic scores may also be a useful alternative to genetic family history for those who do not have access to medical information about their relatives, such as people who are adopted or donor-conceived.

What are the challenges?

It’s important to be clear that polygenic risk scores are not diagnostic. A high risk-score does not mean that a person will definitely develop a condition, and a low score does not mean that they will not. It does not replace screening, such as mammograms, and this needs to be carefully communicated to patients. Polygenic risk scores have not been used widely in the clinic, and it isn’t yet clear how useful they are compared to more traditional approaches like taking a family medical history.

Although they are comparatively cheap to run, tests to determine polygenic risk scores do require the patient’s consent to genomic testing, which some people will be uncomfortable with, particularly if they are not unwell.

Another difficulty is that the accuracy of a polygenic risk score for an individual will depend on how closely that person’s DNA resembles the DNA of the people whose genomic data was used to develop the score. A test developed by a group of researchers in Massachusetts has been criticised because it appears to be less accurate in people of non-European heritage. Although the researchers are based in the US, much of the genomic data used came from the UK Biobank, and as a result reflects the UK’s ethnic mix. The resulting test is therefore significantly less accurate for African-American and Latino populations.

Please note: This article is for informational or educational purposes, and does not substitute professional medical advice.