The PSA test used to screen for prostate cancer misses as many as 82% of tumors in men under the age of 60 and 65% of those in older men, according to computer models developed by researchers at Harvard Medical School.
Their findings suggest that the chances of detecting a tumor at an early stage could be improved by lowering the threshold at which further testing, particularly biopsies, is performed. But critics argue that large numbers of men already receive unnecessary biopsies as a result of the PSA and that lowering the threshold would increase the number undergoing the painful procedure.
The PSA test measures levels of a protein called prostate-specific antigen in the blood. Most physicians now recommend further testing when a man has a PSA level above 4 and a rectal exam suggests the presence of a tumor.
Dr. Rinaa Punglia and her colleagues at Harvard studied 6,691 men who had undergone PSA screening in St. Louis. Only the men with PSA levels above 4 had undergone biopsies. Using mathematical modeling, the team estimated that large numbers of men who had lower PSA levels had tumors; those estimates were not confirmed by biopsies. The results are reported today in the New England Journal of Medicine.
Punglia said that if biopsies were given to men under 60 who had PSA levels of 2.6 or greater, 64% of tumors would be missed. But the number of men who would undergo unnecessary biopsies would triple.
The National Prostate Cancer Coalition recommends using the lower level in men under 60. Small increases in PSA levels can be more predictive in younger men, according to coalition spokesman Jamie Bearse.
But in an editorial in the same issue of the journal, Dutch researchers argued that there is no evidence that PSA screening actually reduces the risk of death without simultaneously reducing men's quality of life. They noted that many prostate tumors grow very slowly and can be safely ignored.
Prostate cancer strikes about 189,000 American men each year, killing 30,200.