Researchers developing new breast cancer prediction model

Volume 4, Issue 3, 2015


Polygenic risk score and other factors may improve personalized estimates of breast cancer risk.

Photograph of Celine M. Vachon, Ph.D., of Mayo Clinic

Celine M. Vachon, Ph.D.

Photograph of Fergus J. Couch, Ph.D., of Mayo Clinic

Fergus J. Couch, Ph.D.

Recent large-scale genomic analyses have uncovered dozens of common genetic variants that are associated with breast cancer. Each variant, however, contributes only a tiny amount to a person's overall risk of developing breast cancer.

Mayo Clinic researchers were part of an international team that combined 77 of these common genetic variants into a single risk factor that can be used to improve the identification of women who are at an increased risk of breast cancer. This factor, known as the polygenic risk score, was built from the genetic data of more than 67,000 women.

The results of the research were published online on March 5, 2015, in the Journal of the National Cancer Institute (JNCI).

A companion study led by Mayo Clinic researchers now shows that this measure of genetic variation can be combined with other important risk factors for breast cancer, such as breast density, to improve personalized estimates of breast cancer risk. Those findings were published earlier this year in the journal Breast Cancer Research and Treatment

The same authors also showed that the polygenic risk score is a risk factor for breast cancer even in moderate-risk to high-risk patients treated with either tamoxifen or raloxifene for cancer prevention.

"This genetic risk factor adds complementary and valuable information to factors we already know are associated with developing breast cancer," said co-author Celine M. Vachon, Ph.D., a cancer epidemiologist at Mayo Clinic in Rochester, Minnesota. "We are developing a test based on these results, and although it isn't ready for clinical use yet, it's likely that within the next few years we will be using this genetic information to improve personalized breast screening and prevention strategies for our patients."

Scientists have known for decades that genetics can play a role in breast cancer. For example, inheriting a mutation in BRCA1 and BRCA2 genes greatly increases the risk of developing breast cancer. But these mutations are rare and account for less than 5 percent of all breast cancers.

More common genetic variations known as single nucleotide polymorphisms (SNPs) also contribute to cancer susceptibility, but the individual contributions are too small to predict breast cancer risk.

In the companion study, a large team of international researchers tested whether they could combine the effects of these individual SNPs into a single risk factor for breast cancer. The investigators combined the information on 77 SNPs from 33,673 breast cancer patients and 33,381 healthy participants to derive the polygenic risk score.

The study showed that the polygenic risk score could successfully place women into different categories of risk.

Compared with women with an average polygenic risk score, women in the top 1 percent were three times more likely to develop breast cancer. In addition, women in the lowest 1 percent of the score were at a 70 percent lower risk of developing breast cancer.

These results indicate that the polygenic risk score is as powerful as other known risk factors in predicting breast cancer.

"To do an even better job at risk prediction, we need to include this genetic profile into breast cancer risk models, along with other relevant information like family history, lifestyle risk factors, previous biopsies, and breast density," said study co-author Fergus J. Couch, Ph.D., a molecular geneticist and pathologist at Mayo Clinic in Rochester, Minnesota. "But first we need to make sure that each of the factors is independent, because if the polygenic risk score is simply repeating what was already accounted for by some of the other known risk factors, then it won't be valuable in a risk model setting."