Researchers at the Broad Institute of MIT and Harvard, Massachusetts General Hospital and Harvard Medical School have found that genetic background can alter the risk of developing colorectal cancer, breast cancer or heart disease in people that carry a high-risk genetic variant.
'The traditional approach is to focus on a single base-pair mutation linked to disease, but there are three billion base pairs in the genome' said co-first author Dr Akl Fahed. 'So we asked whether the rest of your genome can help explain the differing rates of disease we see in these patients, and the answer was a clear yes.'
Some diseases are associated with a change in a single gene which disrupts a specific biological pathway, but with many such conditions, not everyone who carries these variants will develop the associated disease, making predictions difficult. Many other variants within the genome can have small effects on disease risk. A polygenic score collates the effects on risk from changes in many genes which individually have insignificant effects, but in combination can substantially affect the risk of developing the disease.
The team set out to understand this further by studying three diseases: familial hypercholesterolaemia where a single-gene variant prevents cholesterol clearance which increases the risk of heart disease; Lynch syndrome, where a faulty DNA-repair gene often leads to colorectal cancer; and inherited breast cancer caused by changes in the BRCA1 and BRCA2 genes.
The team identified people with one of these genetic variants and calculated their polygenic score by analysing the rest of the genome. They found that polygenic score can be used to predict whether these individuals develop the disease. For example, computed risk of developing heart disease by the age of 75 can be as low as 17 percent in individuals with a familial hypercholesterolaemia variant but a low polygenic score, but if polygenic score is high, risk can go up to 78 percent.
The team suggest that incorporating polygenic scores into genetic tests could make predictions more accurate and genetic counselling more useful for patients. However, they acknowledge limitations to this approach:
'The ability to reliably classify monogenic variants as high-risk and stratify the population using polygenic scores is higher in people of European ancestry than other groups, just because that is where most of our training data comes from. So, we need to diversify datasets and improve the models so that they work well for people from different ancestries, ensuring that genomic risk stratification benefits everyone,' said senior author Dr Amit Khera, a cardiologist from Massachusetts General Hospital.