Researchers identify seven independent risk factors and four risk groups for cancer detection after a false-negative biopsy
Biopsies detect prostate cancer in about one-fourth of men with abnormal screening tests. However, a negative biopsy does not ensure the absence of disease. Repeat prostate biopsies are positive for cancer in about 25 percent of patients.
"Men whose prostate screening tests are abnormal often continue to worry they have prostate cancer despite negative results from an initial biopsy," said Mark Garzotto, M.D., director of urologic oncology at the Portland Veterans Affairs Medical Center, assistant professor of surgery (urology) in the OHSU School of Medicine, and member of the OHSU Cancer Institute.
During a biopsy procedure, less than 1 percent of the entire prostate gland is sampled, so men can harbor prostate cancer in spite of having a negative initial biopsy. "Ultrasound also has some limitations in seeing cancers," Garzotto said.
In a new study looking at predictors of prostate cancer in men undergoing repeat biopsies, Garzotto and colleagues identified four distinct groups of men in whom prostate cancer is present but undetected. Within these groups, risk for cancer detection within two years ranged from 3 percent to 67 percent.
"The results of this study now allow urologists to determine an individual patient's risk for harboring cancer within the prostate after the initial biopsy is negative," said Garzotto, the study's principal investigator. "This information will be useful in counseling patients if and when another biopsy is needed."
Researchers also identified seven independent risk factors for a positive repeat prostate biopsy after a false-negative biopsy. They include: abnormal prostate exam findings; age; family history of prostate cancer; prostate specific antigen (PSA) density; the length of time it takes for PSA level to double; number of samples taken during the initial biopsy; and the presence of abnormal cells known as HGPIN (high-grade prostatic intraepithelial neoplasia), which are believed to be early precursors of prostate cancer.
"Identifying seven risk factors is significant because most earlier studies have identified one or at most two risk factors for cancer detection after an initial negative biopsy," Garzotto said. The study was presented on June 6, 2004, at the American Society for Clinical Oncology annual meeting in New Orleans, La.
In a retrospective study at a single institution, data were reviewed from the medical records of all men undergoing at least one repeat prostate biopsy after an initial negative biopsy from 1992 through 2003. Variables recorded included age, race, family history of prostate cancer, body mass index, referral indication, all PSA values, digital rectal exam, PSA density, ultrasound findings, and HGPIN.
During the study period, 373 patients underwent 977 biopsy procedures. Prostate cancer was detected in 107 (28.9 percent) of patients by biopsy or other means. Median time to cancer detection was 10.3 years.
"This was an excellent data set as we were able to record a host of factors and to follow these men for an average of nearly four years," Garzotto said.
Once the seven risk factors were identified, researchers used a statistical analysis method called recursive partitioning to stratify the data into four distinct risk groups. "Recursive partitioning has the advantage over other statistical methods in that it unveils unique relationships between risk factors," Garzotto said. "In our study, it identified unique associations between PSA doubling time, PSA density and HGPIN."
PSA density, PSA doubling time and presence of HGPIN were most significant in determining the risk groupings of cancer detection within two to five years. Researchers also found that while HGPIN is an important and independent factor in predicting cancer, it loses significance when either the PSA density is high or the PSA doubling time is low.
"HGPIN has been a subject of controversy for the prediction of cancer in the field," Garzotto said. "Our study may help to explain why HGPIN is useful in some settings and not others. Thus, HGPIN is useful when observed as a single factor, but it should not drive biopsy decisions when other factors are present."
"Patients with high-risk features remain at risk for the detection of prostate cancer despite multiple negative biopsy procedures. Vigilant follow-up is warranted in high-risk groups despite an initial negative biopsy," Garzotto said.
Particulars: ASCO Abstract No. 4589, Predictors of prostate cancer in men undergoing serial prostate biopsies: Results from 371 patients