Polygenic risk scores do still have a place in breast cancer screening and prevention strategies

According to a new study led by UCL (University College London) researchers, polygenic risk scores, which predict whether a person will develop a disease based on their genes, performed poorly for prediction and screening when holding them to the same standards as employed for other tests in medicine.

But polygenic risk scores are not designed to be used on their own. When used in combination with an analysis of lifestyle and other risk factors, polygenic risk scores produce an incredibly accurate prediction of future breast cancer risk.

Using polygenic risk scores for risk adapted screening

One of Prevent Breast Cancer’s aims is to personalise the breast screening programme by offering women a risk assessment when they first attend breast screening. A risk score can be calculated by analysing factors such as breast density, gene fragments, family history and lifestyle. Previous Prevent Breast Cancer research enabled women participating in the pilot study the opportunity to make proactive choices about their health to reduce the chances of the disease developing. Those found to be at higher risk were provided with advice and support and, where appropriate, offered options including more frequent mammograms, preventative drugs and in some cases surgery. Our goal is to see this risk adapted screening adopted by the NHS and available across the UK. But more research needs to be undertaken before this can become a reality.

Why this paper about polygenic risk scores is highly misleading

Upon reading the study’s findings, Prevent Breast Cancer Co-Founder and Honorary President Lester Barr and Scientific Patron Professor Gareth Evans released a joint statement:

“The headline that genetic risk scores ‘do not have a place in public health screening’ is akin to saying that a set of bicycle handlebars is useless because you can’t cycle to work on them. In the same way that handlebars are designed to be used in combination with a bicycle frame and two wheels, polygenic risk scores are not designed to be used on their own, but in combination with other things.

The authors of this study are being disingenuous in presenting results for risk stratification using a polygenic risk score and age alone. Many other known risk factors are easily available and can be used alongside a risk score in existing risk models, such as CanRisk and Tyrer-Cuzick. Assertions about not using a polygenic risk score are tantamount to saying you should not use any risk factor information to assess risk.

The huge amount of negative publicity for polygenic risk scores generated from this paper is highly misleading. In the case of breast cancer diagnosis, these scores produce an incredibly accurate prediction of future breast cancer risk when used in combination with an analysis of lifestyle and other risk factors such as breast density. This has been verified in a clinical study of over 50,000 women, called Predicting the Risk of Cancer at Screening, or PROCAS.

Medicine is always evolving, and we are moving on to using risk algorithms in clinical practice – of which polygenic risk scores are proving a crucial part. Don’t throw away the handlebars – we are going to need them – so we can confidently say we can predict an individual woman’s risk of developing breast cancer in their lifetime.”

Find out more about our research into the prediction and prevention of breast cancer.