Predicting cancer: it’s all in the numbers

Identifying the genetic risk factors that underlie cancer depends on a few variables. One is pinpointing the genes themselves, which has become easier as sequencing technology has advanced.

The second key stage is determining how much those genes influence risk, which has proven a more difficult challenge.

Now researchers from the Queensland Institute of Medical Research (QIMR) have developed a new statistical model that can help to more accurately identify those who are at risk of bowel and endometrial cancer.

QIMR’s Dr Amanda Spurdle has led research into the genetic variants responsible for Lynch Syndrome, an inherited form of cancer responsible for up to 5 per cent of cases of bowel cancer and 2 per cent of cases of endometrial cancer.

“Some patients who have genetic testing for Lynch Syndrome are told they have ‘variants of uncertain significance’ which is an inconclusive result,” Dr Spurdle said. “These people are left in a kind of genetic limbo, unsure of whether they will develop a second cancer, or whether their family members are also at risk. “We’ve developed a statistical model which means doctors can more conclusively say whether those patients have Lynch Syndrome, and therefore whether they are at a higher risk of getting another cancer. “The reverse of that is that we may also ease the worry of some people who’ve had inconclusive results.”

Dr Spurdle was involved in two separate studies into the genetic variants responsible for the syndrome. Together they showed it was possible to get a more accurate “grading” of a person’s risk, when researchers introduced more clinical data, including features of tumours, into the equation, and standardised existing computer modelling for genetic variants. “This classification model will be valuable for genetic testing services, to help investigate newly discovered variants in genes, and for doctors to determine the best clinical management strategies for patients and their families,” Dr Spurdle said.

The studies were published in this month’s Human Mutation Early View e-publication. The research was funded by Cancer Australia.