Five-SNP test may predict the risk of prostate cancer

In a group of 2,893 Swedish men with prostate cancer, five gene variants plus family history accounted for 46% of cases, reported S. Lilly Zheng, MD, of the Center for Human Genomics, Wake Forest University, and her colleagues (N Engl J Med online at 10.1056/NEJMoa075819, and in print February 28).

The researchers selected for testing the most significant single-nucleotide polymorphism (SNP) found at each of five chromosomal regions previously reported to be linked with prostate cancer—three at 8q24 and one each at 17q12 and 17q24.3. The study included 1,781 control subjects as well as the subjects with prostate cancer.

After adjustment for age, geographic region, and family history, each of the five SNPs selected was associated with prostate cancer risk, but the association grew stronger as the SNPs were combined (from an odds ratio of 1.50 for one SNP up to 4.47 for four or more).

Family history added

When family history was included as a sixth risk factor, the effect was even stronger. Men who had any five or more of these six risk factors had an odds ratio of 9.46 for prostate cancer, compared with men with none of the six factors. This combination of factors was found in 1.4% of the prostate cancer patients and 0.3% of the controls.

"It may be possible to use the combined information from the five SNPs and family history to assess an individual patient's risk of prostate cancer, but this strategy will have to be tested in a prospective study before proceeding with any such risk assessments," the authors emphasized.

None of the five SNPs was significantly associated with the aggressiveness of prostate cancer, Gleason score, family history, serum PSA at diagnosis, or age at diagnosis. This suggests that some men with low PSA levels may have an increased risk of prostate cancer if they carry one or more of the prostate-cancer-associated SNPs, the authors said.

The mechanism by which the SNPs may affect the risk of prostate cancer is not known. But the authors said that since the five SNPs appear to be associated with prostate cancer risk in general, rather than with a more or less aggressive form of the disease, "we suspect that the genetic variants act at an early stage of carcinogenesis."

In an editorial, Edward P. Gelmann, MD, of the Herbert Irving Comprehensive Cancer Center, Columbia University Medical Center, said that the study does not provide "an easily applicable test" to aid in identifying men at high risk for prostate cancer.

A man with four or five of the SNPs studied and no family history would have a more than fourfold increased risk of prostate cancer, a circumstance found in 5.4% of the prostate cancer patients. However, 2.2% of the control subjects also had four or more of the five SNPs.

"If these percentages are representative of the accuracy of the SNP profile as a screening test, we would falsely assign more than 2% of men to a high-risk population," Dr. Gelmann said.

He noted that although a positive five-SNP test increases a man's likelihood of having prostate cancer, it doesn't predict whether the tumor is indolent or aggressive. Thus, he said, the test is of "great mechanistic importance but of less clinical consequence." Even so, he said, "the ability to quantify risk before the family history becomes apparent or before the tumor is identified by a rising PSA level may be important in some cases."