Genomic Analyses Reveal Dozens More Prostate Cancer Susceptibility Loci

A large genomic analysis of more than 140,000 men identified more than 60 new prostate cancer susceptibility loci, including one locus significantly associated with early onset of the malignancy. The study could improve risk prediction and help elucidate the underlying biology of prostate cancer.

“Although prostate cancer is the most common noncutaneous cancer among men in the Western world, and one in seven men will be diagnosed during their lifetime, very few modifiable risk factors have been established,” wrote study authors led by Fredrick R. Schumacher, PhD, MPH, of Case Western Reserve University in Cleveland. Previous genome-wide association studies have identified more than 100 gene variants associated with prostate cancer.

For the new studies, the researchers developed a custom high-density genotyping array (OncoArray). First, they applied this to a dataset from previously published studies of 46,939 prostate cancer cases and 27,910 controls with no known prostate cancer diagnosis. They then compared this with previously genotyped data from 32,255 prostate cancer cases and 33,202 controls, for a total of 79,194 cases and 61,112 controls. Results were published in Nature Genetics.

The analysis identified a total of 64 loci associated with overall prostate cancer susceptibility, though two of these were deemed likely to be false positives; this left 62 novel susceptibility loci, and one locus associated with onset of prostate cancer at age 55 or younger.

Among the loci were several strong candidate genes, the authors reported. This included variations in ATM, which is a key gene in the DNA-damage response pathway; a missense variant (rs1800057) was found, which exerted what the authors called a “modestly increased risk” of prostate cancer, with an odds ratio of 1.16 (P = 8.15 x 10^–9). Other potentially important variants were found in CDKN1B, RFX7, and INCENP.

A further analysis found that these new loci along with 85 previously reported loci together capture 28.4% of the familial risk for prostate cancer; the new 62 loci increased the risk by 4.4%.

Using the full genetic information to create a polygenic risk score, they found that men in the top first percentile had a relative risk for prostate cancer of 5.71 compared with those between the 25th and 75th percentiles (95% CI, 5.04–6.48). For those in the 90th to 99th percentile for risk score, the relative risk was 2.69 (95% CI, 2.55–2.82).

“Our novel associations highlight several biological pathways that warrant further investigation,” the authors concluded. “The increased polygenic risk score can be used to improve the identification of men at high risk for prostate cancer and therefore inform PSA guidelines for screening and management to reduce the burden of over-testing.”