SAN DIEGO – A blood test that detects the –2proPSA isoform of prostate-specific antigen may provide a way to reduce the number of unneeded prostate biopsies, results from a multicenter study showed.
Using a Prostate Health Index (phi) level of 27 as a threshold for selecting men for prostate cancer could eliminate unnecessary biopsies in 26% of men when total PSA is 4-10 ng/mL, said Dr. Martin G. Sanda, chief of urology at Emory University in Atlanta, during a press briefing at the annual meeting of the American Urological Association.
Dr. Martin Sanda
"This is a substantial portion of the population who may undergo PSA testing. [The index] would allow the ability to detect aggressive prostate cancer while having an acceptable false-negative rate. The Prostate Health Index has the potential to mitigate harms of overdetection/overtreatment of indolent cancers while retaining benefits of detecting aggressive prostate cancer which warrants treatment," he said.
The Prostate Health Index (phi), developed by Beckman Coulter and granted premarket approval from the Food and Drug Administration in June 2012, is a simple, noninvasive blood test that is 2.5 times more specific in detecting prostate cancer than PSA in patients with PSA values in the 4- to 10-ng/mL range and is shown to reduce the number of prostate biopsies.
"The Achilles’ heel of PSA detection in its current form is the overdetection and subsequently the downstream overtreatment of indolent prostate cancers," Dr. Sanda said. "The phi is a manner of reporting the detection of the –2proPSA isoform of PSA. This is a small subset of the PSA molecules, as opposed to the routine total PSA test that we are familiar with."
For the current study Dr. Sanda and his associates investigated whether the use of phi, compared with total PSA and the ratio of free to total PSA (%fPSA), could reduce unnecessary biopsies and overdetection of indolent prostate cancer while improving the detection of aggressive prostate cancer. He reported results from 658 men whose PSA was 4-10 ng/mL. Of these 658 men, 324 had prostate cancer. Among these 324 cancers, 160 were aggressive (meaning a Gleason score of 7 or greater) and 164 were indolent cancers.
Dr. Sanda reported that at 90% sensitivity, the specificity of phi was 31.1%, compared with 19.8% for %fPSA (P = .024) and 10.8% for PSA (P less than .001). When the phi ranged from 0 to 26.9, the probability of significant prostate cancer was 3.9% and rose sequentially with increasing range of phi. Specifically, the probability of significant prostate cancer was 8.5% for those with a phi of 27.0-35.9, 14.4% for those in the range of 36.0-54.9, and 28.9% for those with a phi level of 55 or higher.
"When phi is less than 27, the probability of one of these cancers being a Gleason score of 7 or higher was under 4%," said Dr. Sanda, who also directs the university’s Prostate Cancer Center. "With that particular threshold, we would be able to retain the benefits of being able to detect aggressive cancers in patients who had a biopsy when their phi was higher than 27 while avoiding unnecessary [biopsies] in about 26% of the men, substantially reducing the number of indolent cancers diagnosed and the number of unnecessary biopsies performed."
The false-positive rate was "in an acceptable range," he added. Only 4 out of 109 Gleason 3 + 4 cancers were missed (3.7%), and only 1 out of 35 Gleason 4 + 3 cancers was missed (2.9%).
"Because this is a straightforward serum assay, phi does have the potential to have a favorable cost profile relative to some of the genetic marker testing that’s coming down the pipeline," Dr. Sanda commented. "The next step is to validate these findings in a larger and separate cohort." That effort is currently underway with the Early Detection Research Network, a cohort study funded by the National Cancer Institute.
The study was funded by Beckman Coulter. Dr. Sanda disclosed that he is an investigator for the company. He also reported affiliations with Medicametrix, Accuray, and other companies.