This article is more than 5 years old. Information may no longer be current.
Kallikrein markers improve high-grade prostate cancer detection, reduce unnecessary biopsies
Click Here to Manage Email Alerts
A statistical model using kallikrein markers better predicted high-grade prostate cancer in men with elevated PSA levels and reduced unnecessary biopsies compared with PSA level and age alone, according to the results of a prospective analysis.
“Risk of death from prostate cancer is strongly associated with levels of PSA in blood measured in middle-aged men,” Hans Lilja, MD, PhD, of Memorial Sloan Kettering Cancer Center, and colleagues wrote. “Evidence from randomized screening trials in Europe shows that PSA-based screening can reduce deaths from prostate cancer, but also leads to overdiagnosis and the risk of overtreatment among elderly men with a limited life expectancy.”
Previous retrospective studies identified four kallikrein markers — free PSA, intact PSA, total PSA and human kallikrein-related peptidase 2 (hK2) — that can predict biopsy outcomes, according to study background.
Lilja and colleagues evaluated the four kallikrein markers in cryopreserved blood samples from 6,129 men with elevated PSA (≥ 3 ng/mL) enrolled in the randomized, prospective ProtecT (Prostate Testing for Cancer and Treatment) Trial.
Researchers developed a statistical model to predict any-grade and high-grade (Gleason score of 7 or higher) prostate cancer at the time of 10-core biopsy using marker levels from the anticoagulated plasma of 4,765 men (without cancer, n = 3,032). Researchers then independently validated the model using the markers in the serum from 1,364 men (without cancer, n = 1,221).
Results showed the four kallikrein markers served as a better predictor for prostate cancer than PSA and age. The area under the curve (AUC) for the kallikrein markers was significantly higher than the AUC for age and PSA level for predicting any-grade cancer (0.719 vs. 0.634; P < .001) and high-grade cancer (0.82 vs. 0.738; P ˂ .001).
Using an illustrative cutoff of 6% for the risk for high-grade cancer, researchers determined that per 1,000 men biopsied men with high PSA levels, the four-marker model would reduce the risk for unnecessary biopsy in 428 men and detect 119 instances of high-grade cancer. The risk for a delay in diagnosing high-grade cancer was relatively low (14 of 133).
The secondary analysis indicated use of the four-marker model in serum samples raised the AUC associated with the PSA and age base model for evidence of any-grade cancers (0.665 to 0.757; P < .001) and high-grade cancers (0.785 to 0.859; P < .001).
The researchers acknowledged the use of cryopreserved samples vs. fresh samples and the high number of men in the ProtecT trial who underwent biopsy without undergoing a standard clinical workup (87%) may be limitations to these data.
“In a decision analysis we found that implementation of a statistical model based on the [kallikrein] markers would reduce by close to half the number of unnecessary biopsies undertaken, while delaying diagnosis of only a small number of high-grade cancers,” Lilja and colleagues concluded. “These findings need to be confirmed in prospective research using clinical cohorts.” – by Cameron Kelsall
Disclosure: Lilja reports stock ownership in OPKO and holds patents for free PSA, hK2 and intact PSA assays that have been licensed by OPKO Diagnostics. Lilja and an additional study researcher report a pending patent application related to this study that has been licensed by OPKO Diagnostics. Please see the full study for a list of all other authors’ relevant financial disclosures.
Click Here to Manage Email Alerts