Systematic review: Comparative effectiveness and harms of treatments for clinically localized prostate cancer

Background: The comparative effectiveness of localized prostate cancer treatments is largely unknown. Purpose: To compare the effectiveness and harms of treatments for localized prostate cancer. Data Sources: MEDLINE (through September 2007), the Cochrane Library (through Issue 3, 2007), and the Cochrane Review Group in Prostate Diseases and Urologic Malignancies registry (through November 2007). Study Selection: Randomized, controlled trials (RCTs) published in any language and observational studies published in English that evaluated treatments and reported clinical or biochemical outcomes in localized prostate cancer. Data Extraction: 2 researchers extracted information on study design, sample characteristics, interventions, and outcomes. Data Synthesis: 18 RCTs and 473 observational studies met inclusion criteria. One RCT enrolled mostly men without prostate-specific antigen (PSA)-detected disease and reported that, compared with watchful waiting, radical prostatectomy reduced all-cause mortality (24% vs. 30%; P = 0.04) and prostate cancer-specific mortality (5% vs. 10%; P = 0.01) at 10 years (P = 0.04). Effectiveness was limited to men younger than age 65 years but was not associated with Gleason score or baseline PSA level. An older, smaller trial found no significant overall survival differences between radical prostatectomy and watchful waiting (risk difference, 0% [95% Cl, -19% to 18%]). Radical prostatectomy reduced disease recurrence at 5 years compared with external-beam radiation therapy in 1 small, older trial (14% vs. 39%; risk difference, 21%; P = 0.04). No external-beam radiation regimen was superior to another in reducing mortality. No randomized trials evaluated primary androgen deprivation. Androgen deprivation used adjuvant to radical prostatectomy did not improve biochemical progression compared with radical prostatectomy alone (risk difference, 0% [Cl, -7% to 7%]). No randomized trial evaluated brachy-therapy, cryotherapy, robotic radical prostatectomy, or photon-beam or intensity-modulated radiation therapy. Observational studies showed wide and overlapping effectiveness estimates within and between treatments. Adverse event definitions and severity varied widely. The Prostate Cancer Outcomes Study reported that urinary leakage (>= 1 event/d) was more common with radical pros-tatectomy (35%) than with radiation therapy (12%) or androgen deprivation 0 1 %). Bowel urgency occurred more often with radiation (3%) or androgen deprivation (3%) than with radical prostatectomy (1 %). Erectile dysfunction occurred frequently after all treatments (radical prostatectomy, 58%; radiation therapy, 43%; androgen deprivation, 86%). A higher risk score incorporating histologic grade, PSA level, and tumor stage was associated with increased risk for disease progression or recurrence regardless of treatment. Limitations: Only 3 randomized trials compared effectiveness between primary treatments. No trial enrolled patients with prostate cancer primarily detected with PSA testing. Conclusion: Assessment of the comparative effectiveness and harms of localized prostate cancer treatments is difficult because of limitations in the evidence.