Radical prostatectomy offers no significant mortality reduction, according to study

September 1, 2011

After a median follow-up of 10 years, the intent-to-treat with radical prostatectomy did not significantly reduce all-cause or prostate cancer-specific mortality compared with active surveillance.

All of the men in the study had clinically localized prostate cancer detected during the early PSA era.

"Cumulative prostate cancer mortality in the overall population of enrolled men with localized prostate cancer was <10% at 12 years, and the intent-to-treat with RP compared to observation produced absolute reductions in all-cause and prostate cancer mortality rates that were less than 3% and not statistically significant," said Timothy Wilt, MD, MPH, core investigator in the section of general medicine at the Minneapolis Veterans Affairs Medical Center and professor of medicine at the University of Minnesota, Minneapolis.

Dr. Wilt noted that PIVOT is the only randomized study comparing surgery to active surveillance in men with prostate cancer detected since PSA testing became common and is the only study of its kind conducted in the U.S.

"As such, PIVOT information can guide informed health care decision making between physicians and their patients with clinically localized prostate cancer across a wide range of clinical and tumor characteristics," he told Urology Times.

"PIVOT results add to growing evidence that support the use of observation in most men currently diagnosed with early-stage prostate cancer, especially men with PSA values of 10.0 ng/mL or less or men with low-risk disease. Any survival benefit due to radical prostatectomy appears to be limited to men with PSA levels greater than 10.0 ng/mL or having higher-risk prostate cancer. PIVOT findings support a greater awareness of the long-term favorable prognosis of low-risk disease as well as the comparative effects of observation or active surveillance versus radical prostatectomy."

Between 1994 and 2002, 5,023 men screened for PIVOT met the eligibility criteria of age ≤75 years, life expectancy of at least 10 years, and prostate cancer diagnosis within the previous 12 months with cT1-2 disease of any histologic grade and PSA <50.0 ng/mL. A total of 4,292 declined randomization, and the remaining 731 men were randomized 1:1 to radical prostatectomy or observation. Prostatectomy was performed in 77% of the 364 men randomized to the prostatectomy arm, while 84% received some form of definitive therapy; 10% of the observation arm underwent prostatectomy. Baseline characteristics were similar for the men randomized and those who declined as well as between the two randomized study groups.

The randomized men had a mean age of 67 years, about one-third were African-American, and they were healthy based on self-reported health status and Charlson comorbidities. Baseline PSA averaged 10.0 ng/mL, was <4.0 ng/mL in 11% of men and ≤20.0 ng/mL in 10%. Half of the men had T1c disease, 42% had palpable disease, and Gleason score was ≤6 in 70% according to local histology.

Nearly half of the men died during the course of the study, but only 52 (7.1%) died of prostate cancer. Prostatectomy was associated with a 2.9% absolute reduction in the rate of all-cause mortality compared with active surveillance (47% vs. 49.9%) and a 12% relative reduction that was not statistically significant. Similarly, there was a small, 2.7% absolute reduction in prostate cancer mortality among men assigned to prostatectomy compared to active surveillance (5.8% vs. 8.5%) that translated into a 37% relative reduction in favor of surgery that was not statistically significant.

No variation in RP effect on mortality

The potential benefit of surgery for improving survival in men with worse tumor characteristics was identified in additional analyses conducted to assess whether treatment effect varied in subgroups based on seven predefined baseline patient and tumor attributes. The data showed that the effect of RP on both all-cause mortality and prostate cancer mortality did not vary and was not significant according to categories of age, race, health status, Charlson score, or Gleason score (<7 vs. ≥7). However, a significant interaction of treatment effect on all-cause and prostate cancer mortality was identified in the subgroup defined by PSA ≤10.0 ng/mL vs. >10.0 ng/mL and borderline significant effect according to D'Amico tumor risk categories.

For men with PSA 10.0 ng/mL or less and those with low tumor risk characteristics, surgery did not reduce all-cause or prostate cancer mortality. Surgery did reduce both overall and prostate cancer mortality in men with PSA values >10.0 ng/mL, but benefits of surgery were less consistent when evaluating tumor risk categories. For example, in men with intermediate-risk category tumors, surgery reduced overall mortality by 12.6% in absolute terms, but did not significantly reduce prostate cancer mortality. Mortality differences were not significant when using central histopathology rather than local histopathology to classify tumors as intermediate risk.

"In men with high-risk prostate cancer, compared with observation, the intent-to-treat with RP was associated with statistically significant relative risk reductions for prostate cancer mortality, but not all-cause mortality, of approximately 60% among men with high-risk disease (8.4% absolute reduction) regardless of whether local or central histopathology was used to classify tumor risk categories, and 64% among men with PSA >10.0 ng/mL (7.2% absolute reduction)," Dr. Wilt reported.