Abstracts & Commentary
Synopsis: Radical prostatectomy reduces the risk of death from prostate cancer when compared to watchful waiting, but it does not confer a survival advantage.
Sources: Holmberg L, et al. N Engl J Med. 2002;347:781-789; Steineck G, et al. N Engl J Med. 2002;347:790-796.
Holmberg and colleagues in Scandinavia enrolled men with early prostate cancer and then randomized them to either radical prostatectomy or watchful waiting. These men were, by protocol, younger than 75 (average age, 64.7 years) and were expected to have a life expectancy of an additional 10 years (to allow for follow-up). They had no other cancers. Their prostate cancers had to be newly diagnosed and previously untreated. Cytology, histology, or both confirmed the diagnoses. The definition of "early" prostate cancer was tumor in stage T0d or T1 (clinically unapparent), T2 (confined to the prostate), or T1c (diagnosed by needle biopsy prompted by an elevated prostate-specific antigen [PSA] level). The tumors had to be well- or moderately well-differentiated with a low Gleason score (< 10). Additionally, before a patient could be enrolled, his bone scan had to show no sign of metastases, there could be no sign of urologic obstruction, and his PSA had to be less than 50 ng/mL.
The watchful waiting group received no immediate therapy, although some of the men were diagnosed after receiving a transurethral resection of prostate (TURP). Patients in the intervention group first had pelvic lymph node dissection. If, on frozen section, the nodes were positive (ie, metastatic disease), no further surgery was performed. No one received adjuvant or systemic therapy. If there was symptomatic local progression, men in the watchful waiting group could get a TURP. Men in the intervention group could have chemical or surgical castration. Men in both groups received the same treatment for disseminated disease. All patients had an examination and laboratory work (hemoglobin, creatinine, PSA, and alkaline phosphatase) twice in years 1 and 2, and then yearly. They also had annual chest x-rays and bone scans. Cause of death, distant metastases, local progression, and local recurrence were rigorously defined. The study had 3 end points: disease-specific mortality (the time to death from prostate cancer), the rate of distant metastasis (the time to diagnosis of distant metastases), and death from causes other than prostate cancer.
Six hundred ninety-eight men, who were similar at baseline, were enrolled with equal numbers in both groups. After randomization, 2 men were excluded because the prostate cancer diagnoses were wrong and one man was excluded because he previously had Hodgkin’s disease. The patients were followed for up to 8 years (median, 6.2 years). No one was lost to follow-up. Analysis was by intention-to-treat. Thirty watchful-waiting men eventually had treatment with intent to cure, and 25 men in the intervention group did not have prostatectomies. In all, 115 men died, 62 from the watchful waiting group and 53 in the intervention group. The relative hazard of death from any cause was 0.83, a nonsignificant finding because the 95% confidence interval (CI) was 0.57-1.2. The deaths due to prostate cancer (disease-specific mortality) were 31 and 16, respectively. The relative hazard was 0.5, which was significant (CI, 0.27-0.91). The absolute difference at 8 years was 6.6% (number needed to treat [NNT] = 17). Only 1 of these 47 men did not have clinically apparent distant metastases. Only 2 men died within 1 month of randomization, 1 in each group. This effectively rules out operative or postoperative complications as the cause of the higher death rate from other causes in the intervention group. Men in the intervention group were also significantly less likely to develop distant metastases and local progression.
Comment by Allan J. Wilke, MD
The decision to undergo radical prostatectomy is a very personal one. As primary care physicians, our patients turn to us for guidance and advice. However, we have not had good information on which to base our counsel. Thus, the results of this trial were eagerly awaited. The answers are not all in, but at this point we can tell our patients that radical prostatectomy will probably reduce their chances of death from prostate cancer, developing distant metastases, and local progression, but will not reduce their chance of death overall. We will have to treat 17 men with radical prostatectomy and wait 8 years to prevent 1 death from prostate cancer.
Will they feel better? In a companion article, Steineck and colleagues looked at quality of life in about half of the men in the study. Not surprisingly, men from the intervention group had worse erectile dysfunction (ED; 80% vs 45%, number need to harm [NNH] = 3), better urinary bladder function (28% with weak stream vs 40%, NNT = 8), and worse urinary leakage (49% vs 21%, NNH = 4). Bowel function and psychological symptoms showed a nonsignificant trend toward advantage in the intervention group.
Now, the fine print. This study enrolled Scandinavians exclusively. There was no racial breakdown given, but the number of men of African ancestry undoubtedly was very small. In this country, black men are twice as likely to die of prostate cancer as white men1 (48.7 vs 19.6 per 100,000). Only lung cancer causes more cancer deaths. The men in this study were followed for a median 6 years; the mortality curves separated at about 2½ years and grew farther apart as the study progressed. What will the next 6 years show? These men had the very best kind of prostate cancer—well or moderately well differentiated with no evidence of local or distant spread—and they were healthy enough to be expected to live 10 more years. The results for patients with more advanced disease or other co-morbidities are likely to be worse. The participants did not receive any adjuvant local or systemic therapy. Holmberg et al note in their discussion that they did not stress nerve-sparing prostatectomy. Had they, it is possible that there might have been less ED.
More studies need to be published before we have definitive answers. Until then, this study gives us information we can share with our patients. Some men will consider ED a fair trade for easier bladder emptying. Others will weigh urinary incontinence against distant metastases and roll the die. We will be there for them in any case to support their decisions.
Dr. Wilke, Assistant Professor of Family Medicine, Medical College of Ohio, Toledo, is Associate Editor of Internal Medicine Alert.
1. MMWR Morb Mortal Wkly Rep. 2002;51:49-53.