Radical prostatectomy vs watchful waiting in early prostate cancer


e-mail: jrichie@partners.org

The New England Journal of Medicine rarely publishes follow-up studies for articles previously published in that prestigious journal. The excellent randomized prospective study of radical prostatectomy (RP) vs watchful waiting from the Scandinavian Prostate Cancer Group Study [1], with an additional 3 years of follow-up, has yielded statistically significant differences in both primary endpoint of death from prostate cancer and the secondary endpoints of death from any cause, distant metastases, and local progression. This study, which accumulated 695 men from 14 centres from 1989 to 1999, included patients with clinical stage T1 or T2 prostate cancer, a PSA level of <50 ng/mL and negative bone scans. The patients were stratified according to tumour grade and randomization centre, and were randomly assigned to undergo either RP or watchful waiting. Analysis was by intention to treat, with a 5% crossover in the RP group and a 10% crossover in the watchful-waiting group.

There were significant advantages in the RP group in terms of death from prostate cancer (30 vs 50 men, P = 0.01) and deaths from any cause (83 vs 106 men, P = 0.04). Importantly, although there was no difference in the incidence of distant metastases in the two groups during the first 5 years of follow-up, an additional 3-year follow-up yielded an absolute risk reduction of 10% in favour of the RP group, with a relative risk of 0.60. Likewise, the difference in cumulative incidence of local progression, although statistically significant at 5 years of follow-up, increased markedly in the additional 3 years, with a relative risk of 0.33 in the RP group. These differences cannot be explained by hormonal therapy, as this was used less often in the RP group than in the watchful-waiting group.

This randomized prospective study is the first to show a clear advantage to RP over watchful waiting in a cohort of patients with clinically localized prostate cancer, either well or moderately differentiated. Several lessons can be learned from this important study. First and most important, the 5-year follow-up data in treatments for prostate cancer have limited value; 8- or preferably 10-year data are necessary to discern important differences. The greater incidence of local progression and distant metastases in the watchful-waiting group would also suggest that relative risks may be further improved in the RP group by a longer follow-up. The subgroup analysis would suggest that the reduction in disease-specific mortality was greatest among patients aged <65 years. This observation is hypothesis-generating because subgroup analyses are not powered to ensure a balance of all known and unknown prognostic factors between the age cohorts. Nonetheless, this subgroup analysis would suggest that younger patients would benefit more from intervention rather than watchful waiting.

Should this study sway the clinician to abandon watchful-waiting techniques? Clearly, selected patients will benefit from active surveillance or watchful waiting and avoid the potential side-effects from RP or other interventions with curative intent. Generally, older patients, especially those with comorbid diseases and lower Gleason sum prostate cancers, would seem to be reasonable candidates for active surveillance protocols. The use of PSA velocity, in particular a >2-point rise during the year before diagnosis, will be an important adjunct to identify those patients at risk of death from prostate cancer, in whom intervention may be preferable [2]. Information can be gleaned from two important recent studies; that by Albertsen et al.[3] on causes of death after observation in the Connecticut series suggested that younger patients, especially with higher Gleason sum carcinoma of the prostate, had a greater likelihood of prostate cancer mortality with conservative management. Much of the emphasis on watchful waiting came from the original study by Johansson et al.[4]. In their 10-year follow-up of 223 patients, cause-specific survival from prostate cancer was excellent. Interestingly, in their 20-year follow-up, recently published [5], the mortality from prostate cancer increased dramatically, indicating the pitfall of a shorter follow-up in a disease such as prostate cancer.

Although the present study consisted of patients with clinical stage T2 disease (palpable nodules), widespread use of screening with PSA will identify patients at earlier stages, most of whom have stage T1 disease, with potentially greater lead-time bias. Nonetheless, this study [1], and that of Johansson et al.[5] with a 20-year follow-up, suggests that with a longer follow-up, watchful waiting has a significant disadvantage compared to surgical intervention, especially in younger healthier patients with a presumed greater longevity. Further advantages of surgical intervention include a lower incidence of local progression with attendant need for intervention, and lower likelihood of hormonal treatment, with its attendant morbidity.

This study adds credence to the concept that active treatment with effective methods can significantly reduce cause-specific mortality and reduce later morbidity for younger patients with prostate cancer. Much credit should go to the Scandinavian Prostate Cancer Group for the careful design, recruitment, and follow-up of a large number of patients in a prospective randomized study adequately powered to detect differences in RP vs watchful waiting.