Compared with observation, radical prostatectomy failed to reduce either disease-specific or all-cause mortality significantly among men with clinically localized prostate cancer that was diagnosed early in the era of PSA testing, according to a report published online July 18 in the New England Journal of Medicine.
The lack of benefit from prostatectomy was "particularly robust" among men with a PSA value of 10 ng/mL or less, and in those with low-risk tumors. "Our findings support observation for men with localized prostate cancer, especially those who have a low PSA value and those who have low-risk disease," said Dr. Timothy J. Wilt and his associates in the PIVOT (Prostate Cancer Intervention Versus Observation) study.
PIVOT was a randomized trial designed specifically to compare outcomes after radical prostatectomy with those after observation. It involved 731 patients who were diagnosed as having clinically localized prostate cancer in 1994-2002, early in the PSA-testing era.
The participants were treated at 44 Veterans Affairs centers and eight National Cancer Institute sites across the country. Investigators had planned to randomize 2,000 men, but scaled down the trial because of "recruitment difficulties." In all, 364 men were randomly assigned to radical prostatectomy using a technique chosen at the surgeon’s discretion, and 367 were assigned to observation.
Of these, 281 in the surgery group and 36 in the observation group underwent radical prostatectomy, whereas 53 and 292, respectively, underwent observation. Other interventions included external-beam radiotherapy and brachytherapy. Any additional interventions were decided upon by the patient and his physician.
Investigators followed participants every 6 months for 8-15 years. Patients (mean age, 67 years) underwent bone scans every 5 years. Approximately one-third of the patients were black, and 85% were fully independent in performing the activities of daily living.
Some 40% of the men were found to have low-risk tumors, 34% had intermediate-risk, 21% had high-risk, and 5% had unknown risk because of missing data. Findings were reported on an intention-to-treat basis.
During a median follow-up of 10 years, 47.0% of the surgical group and 49.9% of the observation group died from any cause, a difference that was not significant. Median survival was 13.0 years with radical prostatectomy and 12.4 years with observation, also an insignificant difference.
"The absolute reduction in mortality with radical prostatectomy was not significant at any interval and declined over time, from 4.6 percentage points at 4 years to 2.9 percentage points at 12 years," the investigators said (New Engl. J. Med. 2012 July 19 [doi:10.1056/NEJMoa1113162]).
This trend suggests that longer follow-up would not alter the results, added Dr. Wilt of the Center for Chronic Disease Outcomes Research of the Minneapolis VA Health Care System and his coauthors.
The results were similar with prostate cancer–specific mortality.
Death that was definitely attributable to prostate cancer or its treatment occurred in 4.4% of the men in the surgical group and 4.9% of those in the observation group, a nonsignificant difference. Disease-specific mortality was identical between the two groups at 4 years and was not significantly different at 12 years.
Subgroup analyses showed that radical prostatectomy did not improve all-cause mortality according to patient age, race, performance status, or comorbidities, nor did it vary by tumor score on the Gleason histologic scale. However, the surgery was associated with a slight (13.2%) decline in all-cause mortality in the subgroup of men who had PSA values greater than 10 ng/mL.
The finding of no mortality benefit was particularly strong among men with low-risk cancers (defined as those with PSA values of 10 ng/mL or lower, a score of 6 or less on the Gleason scale, and a stage T1a-c or T2a tumor). In this subgroup, prostatectomy was actually associated with a nonsignificant 15% increase in mortality, Dr. Wilt and his associates said.
Perioperative complications, including one death, developed in 21.4% of men in the radical prostatectomy group. Wound infection was the most common complication, occurring in 4.3% of the men. Other problems included urinary tract infection, bleeding requiring transfusion, and the need for urinary catheterization for more than 30 days.
At 2 years, urinary incontinence and erectile dysfunction were more common among patients in the surgical group than in those in the observation group.
This study was supported by the Department of Veterans Affairs, the National Cancer Institute, and the Agency for Healthcare Quality and Research. Dr. Wilt reported no financial conflicts of interest, and his associates reported ties to numerous industry sources.