NEW YORK (GenomeWeb) – Over the past few years, sensitivity problems in the prostate-specific antigen screening test have led the US Preventive Task Force to rate the test a "D" as an indicator of possible prostate cancer, causing some insurers to revoke coverage. Now, a large genome-wide association study from the University of California, San Francisco and Kaiser Permanente suggests the test might be improved by taking into account natural genetic variation in PSA levels.
In a study published in Nature Communications today, the researchers used long-term data from 28,503 men in a Kaiser Permanente dataset and 17,428 men from additional cohorts to identify 40 loci that predicted about 10 percent of the normal variation in PSA levels in men who do not have cancer.
The authors also found the remaining genotyped and imputed SNPs in the study accounted for 32 percent of the variation, suggesting more genetic predictors of normal PSA levels could be revealed in future studies that employ even larger sample sizes.
Overall, the authors uncovered SNPs that influenced PSA levels but not prostate cancer (PCa) as well as SNPs that appeared to influence both, highlighting "the complexity of using conventional PSA levels as a screening tool for PCa." This would indicate that elevated PSA levels alone may not necessarily be a cancer risk factor, given the existence of SNPs that affect PSA levels but that appear to have no impact on prostate cancer.
However, knowing whether a patient has a genetic predisposition to high PSA may allow doctors to reduce overtreatment by normalizing test results to an individual patient or modifying threshold levels for follow-up testing, such as biopsies. "Despite the flaws of PSA testing, it does have value as a diagnostic tool," John Witte, co-senior author of the study and a UCSF professor of epidemiology and biostatistics and of urology, said in a statement. "If we want to improve PSA as a useful diagnostic, we need to fold personalized genetics into medical decision-making. We need to be able to tell patients – this is a high PSA value for you, not just for the average person."