Increasing Specificity to Reduce Prostate Biopsies
Two promising markers could mean more accurate prostate cancer diagnosis
By Julie McDowell
Every year, approximately one million men in the U.S. undergo a prostate biopsy, primarily because they have elevated prostate specific antigen (PSA) levels, according to the Urological Sciences Research Foundation. However, only 25% of these biopsies are positive for prostate cancer. Many clinicians and laboratorians have long wished for a more specific prostate cancer diagnostic test, especially for those men who undergo repeated biopsies due to elevated PSA levels. Two markers—the prostate cancer gene 3 (PCA3) and early prostate cancer antigen-2 (EPCA-2)—might lead the way to more accurate diagnoses and reduce the number of unnecessary biopsies, recent research indicates. This issue of Strategies examines these findings, and what they mean for expanding the options for prostate cancer diagnosis.
This year, an estimated 219,000 U.S. men will be diagnosed with new cases of prostate cancer, according to the American Cancer Society. But accurate diagnosis has continued to be a challenge for clinicians and clinical laboratorians. Their primary diagnostic tool—PSA testing—has specificity problems that can lead to false positive results and unnecessary biopsies. But recent research indicates that early prostate cancer antigen-2 (EPCA-2) might be useful as a more specific marker. Results from a study published in the April issue of Urology by researchers at Johns Hopkins University (JHU) School of Medicine in Baltimore found that when EPCA-2 levels were measured in 385 patients, the test was negative in 97% of the cancer-free patients (Urology 69(4): 714–720, 2007).
“The specificity is what we went after when we designed this study,” said lead author Robert Getzenberg, PhD, Professor of Urology and Director of Research at JHU’s James Buchanan Brady Urological Institute. “At the tissue level, we knew PSA was very specific, but we wanted to see if that specificity would extend to the serum test.”
The patients in the study were separated in to several groups—some had normal PSA levels with no evidence of disease, while others had elevated PSAs and negative biopsies. Other patient groups included those with normal PSA levels who were determined to have prostate cancer, those with a non-cancerous prostate condition known as benign prostatic hypertrophy (BPH), and those with other cancer types. “We looked at these other conditions that we know PSA does not detect well,” explained Getzenberg. “We even studied a set of men who have prostate cancer with normal PSA levels, a situation which occurs in about 15% of patients tested using PSA. We always talk about the specificity issue, but PSA does miss some cancers as well.” The study results showed that men with no evidence of cancer and patients with other cancer types and benign conditions had EPCA-2 levels below the cutoff level, which was established at lower than 30 ng/mL.
Among those patients in the cohort who had cancer, the researchers also found that EPCA-2 levels were significantly higher in patients whose cancer had already spread outside the gland, compared to those with disease confined to the gland. This is important because EPCA-2 could also distinguish the more aggressive forms of cancer, said Getzenberg. “In many ways, we are treating men halfway—we’re over-treating some men and under-treating others who may require earlier chemotherapy or hormonal intervention than we are doing right now,” he explained. “I think finding a marker that can clearly distinguish between the more aggressive forms of prostate cancer is one of our most important duties.”
Targeting the Dilemma Group
In another recent study, investigators with the Culver City, California-based Urological Sciences Research Foundation (USRF) evaluated the utility of a prostate cancer gene 3 (PCA3) urine assay in the “PSA dilemma group” of 233 male patients (Urology 69(3): 532–535, 2007). Gen-Probe (San Diego, Calif.) developed the investigational PCA3 assay used in the study. It is currently available to laboratories in the form of an analyte specific reagent, or ASR.
“We focused on men who were undergoing repeat prostate biopsies because of elevated PSA levels, and who had had a previous negative biopsy,” explained lead author Leonard Marks, MD, USRF founder and Clinical Professor in the Department of Surgery/Urology at the University of California (Los Angeles) School of Medicine. “In these men, PSA has essentially no diagnostic value at all, because they are already known to have elevated PSA levels that don’t tell us anything.”
Marks and his co-authors evaluated the PCA3 assay by performing a receiver-operating-curve (ROC) analysis. For the PCA3 score, the area under the curve was 0.678—an improvement over the 0.524 score for the PSA test. The ROC curve was also used to determine the sensitivity and specificity of the PCA3 assay at different score cutoffs. The PCA3 score was calculated using a ratio of PCA3 mRNA levels over PSA mRNA levels (PCA3 mRNA/PSA mRNA x 1000). The researchers determined that a PCA3 score of 35 corresponded to the greatest diagnostic accuracy, with a sensitivity of 58% and specificity of 72%.
These findings indicate that a PCA3 assay is more sensitive and specific than PSA. “The PSA test is going to be mostly a manifestation of the benign cells, or BPH cells, in the prostate,” said Marks. “BPH cells contain more PSA than cancer cells, therefore PSA is really a better marker for volume of BPH than it is for cancer. But PCA3 has a different dynamic altogether. It is a highly specific genetic marker for prostate cancer, and in fact has nothing to do with BPH.” However, Marks doesn’t believe that the PCA3 test will replace PSA testing. Rather, in the future, he believes that the PCA3 test could complement a PSA test to better identify biopsies that will be positive for prostate cancer.
Dr. Getzenberg has patented the diagnostic method used in his study on EPCA-2. The patent is owned by the University of Pittsburgh and Johns Hopkins University and has been licensed to Onconome, Inc. (Seattle, Wash.). Dr. Getzenberg is also a consultant to Onconome and has received a research grant from the company.
Dr. Marks has served as a paid consultant, speaker, and study investigator for Gen-Probe and Beckman-Coulter, Inc. He is also a consultant and advisor to Onconome.
Please help us improve Strategies by rating how useful this article is to your job.