Putting this out there, with the notation that the main author of this article is the CEO of deCode Genetics, a company that is applying for a patent on the test...
The accuracy of testing for prostate specific antigen (PSA) levels to assess risk of prostate cancer -- long assailed as far from perfect -- could be improved in the future by using genotyping, a study from Iceland suggested.
A genome-wide association study, conducted among 4,500 men, found that PSA test results in 6% to 7% of those who were genotyped for variants associated with elevated PSA levels could have changed the decision on whether to do a prostate biopsy, according to Kari Stefansson, MD, of deCODE Genetics in Reykjavik, and colleagues.
"For a screening test as important and widely used as the PSA test, having a better way to interpret the measured serum PSA levels is likely to improve substantially the clinical utility of the test," Stefansson and co-authors wrote online in Science Translational Medicine.
PSA levels between 2.5 and 4 ng/mL have typically been used as the threshold for deciding if a patient should be referred for biopsy, but, in addition to cancer, infections and inflammation can boost PSA levels. Also, genetic factors may account for up to 45% in the variability in PSA levels. The researchers say correcting for the influence of inherited factors could enhance the PSA test's value.
To explore the extent of the genetic contribution to PSA variability, the group conducted a complex series of analyses.
Initially they analyzed data from two European cohorts -- one from Iceland consisting of more than 15,000 men and the other from the U.K. of about 450 men, all without cancer. Within these cohorts they identified six single nucleotide polymorphisms (SNPs) associated with elevated serum PSA.
They then looked at these loci in men who had prostate biopsies, and found that three were associated with high PSA levels -- and also with a higher probability of a negative biopsy result.
To assess the combined effects of these SNP variants on PSA levels, they found that for men with the strongest genetic association -- the top 5% -- PSA levels were elevated as much as 47% higher than average.
Similarly, men in the bottom 5% of PSA level distribution had PSA levels around 56% lower than average.
The researchers then calculated "personalized PSA cutoff values," by multiplying the standard cutoff of 4 ng/mL by the estimated relative effects of the individual SNPs.
They found among men with the strongest genotypic effects, the PSA threshold for referral for further evaluation rose from 4 ng/mL to between 4.9 and 5.9 ng/mL in the Icelandic cohort and to between 5.6 and 7.7 ng/mL in the U.K. cohort.
Conversely, for those with the lowest genetic effects, the cutoff value fell to between 1.7 and 2.8 ng/mL in the Icelandic group and to between 0.8 and 1.9 ng/mL in the U.K. group.
"These results demonstrate that for a substantial fraction of men undergoing PSA-based prostate cancer screening, the personalized PSA cutoff value is shifted following correction for the effect of the PSA sequence variants," the researchers explained.
They also noted that the markers they identified were primarily predictive of the outcome of screening -- the decision to biopsy -- rather than actually predicting the presence of prostate cancer.
Stefansson and colleagues concluded that the current study data such as these, in combination with information about age, family history, and ethnicity, "could lay a foundation for the development of individual prostate cancer screening strategies that would have the ultimate goal of reducing cost and improving quality of life."
Unneeded prostate biopsies and the resultant over-treatment of men based on PSA testing came under fire in a recent study and earlier this month, the U.K. Screening National Screening Committee recommended against routine PSA screening.
In an editorial accompanying the current article, John S. Witte, PhD, of the University of California San Francisco, described the study by Stefansson and colleagues as "a noteworthy advance toward clarifying the quandary surrounding PSA screening for prostate cancer."
Witte noted that previous studies have suggested that about 10% to 20% of men who are screened are referred for follow-up, but only about 20% to 30% of those who undergo biopsy -- with its attendant adverse effects -- turn out to have cancer.
However, in assessing the current study results from Stefansson's group, Witte explained that while the associations between the identified SNPs and PSA were statistically significant, confirmatory analyses did little to alter the tradeoffs between true positive and false positive results for cancer.
Consequently, he argued that SNP adjustment of PSA values is not yet ready for use in the clinic.
But the data represent an advance toward more individualized and meaningful PSA testing and interpretation, and further investigations should include the identification of other genetic markers and evaluation in larger trials, Witte advised.