Three Prognostic Categories for High-Risk Prostate Cancer

Becky McCall

February 29, 2012

February 29, 2012 (Paris, France) — Patients with high-risk prostate cancer can be stratified into 3 prognostic subgroups using a novel investigational classification system that takes into account prostate-specific antigen (PSA) level, Gleason score (GS), and tumor stage. A follow-up of patients has shown that 95% of the good-prognosis subgroup were still alive after 10 years, suggesting that they are excellent candidates for aggressive treatment or surgery, researchers reported here at the European Association of Urology 27th Annual Congress.

Steven Joniau, MD, from the Department of Urology, University Hospitals Leuven, Belgium, leads the ongoing study, and he presented the results to date of the European retrospective multi-institutional study.

His team looked at the heterogeneity of 1632 patients with high-risk prostate cancer who underwent radical prostatectomy, who were collectively termed high-risk, and stratified them into groups, depending on whether they had 1 or more defining characteristics of high-risk disease.

"Unfortunately, most people think...[patients] with a high PSA [>20 ng/mL] and clinical stage T 3–4 disease or a Gleason score of 8 to 10 on biopsy are all the same and defined as high-risk, but there are combinations of these factors. Some patients have 2 or 3 of these factors," he explained.

Dr. Joniau's group defined 7 prognostic subgroups, according to a combination of factors (PSA > 20 ng/mL; GS, 8 to 10; and clinical stage T 3-4 disease). This system "is complex and not practical for use by clinicians in the everyday setting," he cautioned.

With that in mind, the investigators compared the predictive accuracy of this complex model with a simplified model, and found that some groups could be combined without any loss of accuracy.

"We found that...the accuracy of the simplified model was exactly the same as the accuracy of the more complex model," he added.

The Belgian team established 3 groups on the basis of the simplified model: good-prognosis patients had a PSA above 20 ng/mL, or clinical stage T3–4 disease, or a GS of 8 to 10; intermediate-prognosis patients had a PSA above 20 ng/mL and clinical stage T3-4 disease; and poor-prognosis patients had a GS of 8 to 10 plus a PSA above 20 ng/mL and/or clinical stage T3–4 disease.

Ten years after surgery, the researchers looked at survival outcomes to the 3 subgroups.

"In the first group, with a good prognosis, 95% of patients were still alive, but in the poor-outcome group, more than 20% of patients had died of their disease at this same timepoint, despite receiving a range of therapies," Dr. Joniau reported.

"This is the first paper with hard end points," he stated.

"We now have a model that is suitable for use in clinical practice by everybody — urologists, radiologists, and oncologists," he pointed out. This is the first time such a model has been shown to be as accurate as more complex models, he explained

Dr. Joniau believes that in time, urologists and oncologists will realize that patients in the high-risk group with only 1 risk factor are actually excellent candidates for surgery.

"We can show this because they have very good cancer-specific survival 10 years down the line. Many urologists who still hesitate to operate on these patients can actually be guided toward more aggressive local treatment," he pointed out.

He acknowledged that some clinicians have always treated these patients aggressively, but most have treated them with nonaggressive therapies, such as hormones. "If you treat these patients aggressively, the outcome is good," he confirmed.

The study is ongoing and is rapidly recruiting new centers and patients. Initially, 1700 to 1800 patients were included in the analysis, but this has expanded significantly to a database of 6000 patients. Patients are now drawn from centers such as the Mayo Clinic in the United States and clinics in Hamburg, Milan, Munich, and Paris.

Dr. Joniau pointed out that another group had published a paper looking at similar combinations of high-risk factors, but they looked at biochemical progression — rising PSA levels after treatment — as an outcome.

"Rising PSA is not a good outcome [measure]; these 3 groups would have completely different cancer-specific survival but the same biochemical profile. Our data are stronger with harder end points."

"These data fill a black hole in urology," concluded Dr. Joniau.

Manfred Wirth, MD, professor of urology at the Technical University in Dresden, Germany, commented on these findings. He questioned whether it is correct to consider a death rate of 20% in the poor-prognosis, high-risk group as a bad outcome.

"If only 20% of this high-risk group die, the multimodality treatment seems to me very effective," Dr. Wirth explained.

"These patients are very high risk. In our clinic, we treat them with surgery, followed by radiotherapy and hormone therapy if necessary," he said. "Using this approach, long-term survival is good and the mortality rate is low — almost zero — even if these patients are older."

"The work presented by Joniau and coworkers is of the utmost importance for the urologists, radiation oncologists, and oncologists involved in the management of patients with locally advanced prostate cancer. The definition of 'high risk,' according to the D'Amico criteria, is very vague and [includes] a broad heterogeneity of patients [who could have] completely different prognoses," said Axel Heidenreich, MD, from Aachen University in Germany.

As has been shown recently, "patients who harbor more than 1 of the poor risk factors have to be treated in a different manner than those with only 1 risk factor. Patients with only 1 risk factor are good candidates for a local surgical approach, with high long-term cure rates. High-risk patients with only 1 risk factor should not be excluded from local surgical therapy with a curative intent," Dr. Heidenreich explained.

Dr. Joniau, Dr. Wirth, and Dr. Heidenreich have disclosed no relevant financial relationships..

European Association of Urology (EAU) 27th Annual Congress. Presented February 25, 2012.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.