PSA Velocity Does Not Improve Prostate Cancer Detection
Adapted from the NCI Cancer Bulletin.
A rapid increase in prostate-specific antigen (PSA) levels is not grounds for automatically recommending a prostate biopsy, according to a study published online February 24, 2011, in the Journal of the National Cancer Institute.
The rate of change in PSA levels, referred to as PSA velocity, has been studied as a marker for the presence of prostate cancer. Clinical recommendations from two organizations, the National Comprehensive Cancer Network (NCCN) and the American Urology Association (AUA), suggest that men with a PSA velocity that exceeds a certain threshold (0.35 ng/ml per year) should consider having a needle biopsy, even if their overall PSA levels are below the standard cutoff for the procedure and they have a normal result on a digital rectal examination (DRE).
The findings from this new study indicate that these recommendations should be revised, the authors concluded.
“Overall, PSA velocity did not importantly add predictive accuracy to a standard predictive model or to just PSA alone,” wrote Andrew Vickers, Ph.D., and his colleagues from Memorial Sloan-Kettering Cancer Center (MSKCC). “We found no reason to believe that implementation of the guideline would improve patient outcomes; indeed, its use would lead to a large number of unnecessary biopsies.”
PSA velocity falls under the general umbrella of what is often called PSA kinetics—assessing PSA levels dynamically over time, explained Howard Parnes, M.D., of NCI’s Division of Cancer Prevention. In addition to PSA velocity, PSA kinetics also includes measures such as how long it takes for PSA levels to double.
“Intuitively, you would think you could get more useful information by looking at a change in a biomarker over time,” Dr. Parnes said. The study results indicate that this is not the case with PSA velocity. “You can understand why urologists feel that if the PSA is going up, a biopsy should be offered, because they don’t want to miss a cancer,” he added.
To conduct the study, the researchers used data from more than 5,500 men in the placebo arm of the NCI-funded Prostate Cancer Prevention Trial (PCPT). In that clinical trial, men ages 55 or older with a PSA level below 3.0 ng/ml and a normal DRE result at study entry were randomly assigned to receive either the drug finasteride or a placebo.
In addition to receiving regular PSA tests and a prostate biopsy recommendation for PSA levels of 4.0 ng/ml or higher, the majority of these men also underwent end-of-study biopsies irrespective of PSA levels. As a result, the study authors wrote, “the PCPT provides a perfect test case for PSA velocity” throughout the entire range of PSA values.
Researchers looked at whether a PSA velocity above the 0.35 ng/ml per year threshold—when added to a standard risk model that includes age, PSA level, DRE result, family history of prostate cancer, and history of a prior prostate biopsy—improved the model’s predictive accuracy. The analysis was broken out by different methods for measuring PSA velocity as well as by whether PSA velocity might improve the detection of more aggressive cancers (defined as those with a Gleason score of 7 or higher) and “clinically significant” cancers (defined by the commonly used Epstein criteria).
In all cases, the addition of PSA velocity yielded only a slight difference in the model’s predictive accuracy. In fact, the analysis indicated that lowering the PSA threshold for a biopsy from 4.0 ng/ml to 2.5 ng/ml would be a more effective means of directing prostate biopsies.
Adding PSA velocity to the model would have identified 115 additional cancers (although not necessarily fatal cancers) but also resulted in 433 “unnecessary biopsies” that would have shown no cancer. Lowering the PSA threshold for biopsy to 2.5 ng/ml would have led to nearly the same number of unnecessary biopsies but identified 24 more cancers.
“The decision to proceed to prostate biopsy should be based not only on an abnormal PSA and/or digital rectal examination results, but should take into account multiple factors,” said Peter Carroll, M.D., chair of the Department of Urology at the University of California School of Medicine, in a statement from the AUA on the study findings. PSA velocity can “on occasion” be one of those factors, he said.
In clinical practice, for men who have previously had a prostate biopsy, an increase in PSA velocity often leads to a repeat biopsy, Dr. Parnes explained. In such situations, he recommended using the risk calculator developed by PCPT investigators using data from the trial. (The risk model used in the JNCI study was based on that tool.) The online calculator estimates the likelihood that a biopsy will detect prostate cancer in general and high-grade prostate cancer in particular.
The study’s findings are helpful, Dr. Parnes said, because it’s clear that PSA velocity is actively being used to guide biopsy decisions. “They allow for discussions between doctors and patients that are based on data,” he continued, “and an understanding that we’re dealing with risk estimates.”