Studies Shed More Light on Value of "PSA Kinetics" in Prostate Cancer
Two new retrospective studies reinforce earlier findings that rate of change in prostate-specific antigen (PSA) values both before and after treatment for localized prostate cancer can provide important information about patients' prognoses.
The studies, published in the July 27 Journal of the American Medical Association, indicate that such changes - often dubbed "PSA kinetics" - either alone or in combination with other clinical markers, can identify some patients at high risk for disease recurrence and death who otherwise would have been considered at low risk.
Although both studies were retrospective, they add to the mounting evidence that suggests PSA kinetics can be a valuable tool in predicting the course of localized disease, along with absolute PSA level, Gleason score, and clinical stage, says Dr. Alison Martin of the NCI Cancer Therapy Evaluation Program. "PSA screening has shifted the profile of prostate cancer in the United States to lower risk disease, so it's important to develop prognostic models that can accurately distinguish between aggressive cancers that should be treated versus those that could be managed conservatively," she says. (See NCI Cancer Bulletin, June 14).
In the first study, lead author Dr. Stephen J. Freedland of Johns Hopkins Medical Institutions and colleagues described a new, three-factor model that identifies which patients are at risk of dying from recurrent prostate cancer after radical prostatectomy. The model uses PSA doubling time - the time it takes for PSA values to double after PSA returns to measurable levels following surgery (also known as biological recurrence) - Gleason score, and the time from surgery to biological recurrence.
The study included 379 men who had undergone radical prostatectomy, with 16 years of follow-up. Differences in disease-specific survival were independently associated with each risk factor. But assessing patients based on the combination of all three risk factors offered a more potent prognostic tool: patients who had the lowest scores on all three indicators had a median survival of only 3 years, compared with those who scored highest, all of whom were alive after 16 years of follow-up. The research team combined the three variables into tables that estimate disease-specific survival at 5, 10, and 15 years.
The second study considered the change in PSA levels among men with localized prostate cancer in the year prior to treatment with radiation therapy, also known as "PSA velocity." Lead researcher Dr. Anthony V. D'Amico of the Dana-Farber Cancer Institute and his colleagues looked at a group of 358 patients whose cancer had recurred. They found that men whose PSA velocity was greater than 2.0 ng/mL in that year were more likely to have recurrence than were those whose PSA level rose more slowly.
Patients with higher PSA velocities also were more likely to die sooner. Among men whose prostate cancers were initially classified as low risk, 19 percent with high PSA velocity were likely to die in 7 years, compared with none of the patients with a PSA velocity at or below 2.0 ng/mL per year. In men initially classified as high risk, 24 percent with high PSA velocity were likely to die in 7 years versus 4 percent of high-risk men with lower PSA velocity.
It's still too early to make treatment decisions based on these studies, cautions Dr. Mitchell S. Anscher of the Department of Radiation Oncology at Duke University Medical Center, who wrote an editorial accompanying the studies.
Although the potential of PSA kinetics as a prognostic marker is important, he argues, the greater value of PSA kinetics may be as a surrogate marker for survival "as the ultimate endpoint" in prostate cancer studies.
"So if you can find that correlations in PSA kinetics can be translated into relationships to deaths from prostate cancer, we may be able to complete studies in a much shorter period of time," he says. "Then we can move the field forward more rapidly than we've been able to in the past."
A number of prostate cancer treatment trials involving the NCI Cooperative Groups that have already opened or are set to open in the coming year will begin to address these issues, Dr. Martin explains.
"Our NCI programs and the cooperative groups are working together to nest promising surrogate or biomarker questions into our therapeutic trials," she says. "The trials also will involve standardized collection of essential data, including serum and tissue where possible, which will allow us to create a comprehensive database or platform for subsequent analyses of promising biomarker questions across trials and agents. Hopefully, this will both aid clinical management and speed drug development."
By Carmen Phillips