Using Genomics to Predict Prostate Cancer

A team of genomics researchers at the School of Medicine report that a simple blood test can help predict which men are likely to develop prostate cancer.

The team’s work, reported in the New England Journal of Medicine, showed that five genetic variants previously associated with prostate cancer have a strong cumulative effect. For example, men with four of the five variants had an increased risk of 400 to 500 percent. By adding in a sixth risk factor of family history to the equation, men with five of the six risk factors had increased risk of more than 900 percent of getting prostate cancer.

The information can substantially improve physicians’ ability to assess risk and determine the need for more aggressive screening or even a biopsy, said Jianfeng Xu, MD, and director of the school’s Center for Cancer Genomics. It was only in 2007 when any specific genetic variants had been consistently identified as markers for prostate cancer.

Xu, a senior researcher on the project, said although each variant alone was associated with moderate risk, the effect wasn’t significant enough to justify testing individuals.

The team’s research was the first to evaluate the cumulative effect from having multiple variants, and showed that genome-wide screening could be useful in clinical practice.

In mid-2012, Xu was among a large team of School of Medicine researchers (led by Karim Kader, MD, a former School of Medicine team member who had since joined the University of California San Diego Health System) to report development of a successful genetic test to predict prostate biopsy outcomes in men after an initial negative biopsy.