Objective Early detection and treatment of prostate cancer is common and may be influenced by commercial and media determinants that increase overdiagnosis and overtreatment. We describe the long-term effects of surgery versus observation in early prostate cancer (CaP) and incorporate evidence to develop strategies to reduce overdiagnosis and overtreatment harms.

Methods Long-term follow-up of a randomized trial (PIVOT) of surgery versus observation in 731 men with early CaP detected during the early Prostate Specific Antigen (PSA) era. We compare PIVOT findings to randomized trials of: 1) surgery versus observation in men diagnosed prior to PSA screening and 2) surgery or radiation versus PSA-based monitoring in PSA screen detected disease to describe the magnitude of overdiagnosis and overtreatment that occur with increasingly indolent cancers detected through PSA testing.

Results In PIVOT, death from any-cause occurred in 246 of 364 (67.6%) men assigned to surgery compared to 269 of 367 (73.3%) assigned to observation through 22 years follow-up; absolute risk difference 5.7% (95% CI; -0.9 to 12.3%), relative risk: 0.92 (95% CI: 0.84 to 1.05). Results did not vary by patient (age, race, comorbidities, performance status) or tumor (PSA, histology, risk status) characteristics. Through 20 years follow-up death attributed to CaP or treatment occurred in 7.4% of men assigned to surgery and 11.4% of men assigned to observation; risk difference 4.0% (95%CI: -0.2% to 8.3%). CaP death occurred in 5% of men with low risk disease with nonsignificant absolute risk differences of 1.4% (95% CI: -3.9% to 6.7%). Among men with low risk disease, symptomatic or systemic progression as well as associated reatment was infrequent and did not differ between groups. In comparison to our findings, among men with CaP detected prior to PSA screening, 23 year CaP mortality was 31% in men assigned to observation and 20% with surgery. In men with PSA screen detected disease CaP mortality was 1% and did not differ between surgery, radiation, or PSA-based monitoring. Symptomatic and systemic progression was infrequent through 10 years.

Conclusion CaP mortality and symptomatic progression is infrequent in with low risk and PSA detected CaP treated with observation or PSA-based monitoring. All-cause or CaP mortality was not significantly reduced with early surgery or radiation. While early intervention appears to be effective and necessary in some men, the risk of overdiagnosis and overtreatment increases with enhanced screening intensity, revised disease risk classifications, and widespread early treatment. Our findings, combined with PSA screening trials demonstrating at most small reductions in CaP mortality and no reduction in all-cause mortality, indicate that methods to reduce overdiagnosis and overtreatment should include decreasing screening intensity by reducing populations screened, increasing screening intervals among well informed men who request screening, and increasing test thresholds that define screening abnormality and lead to further diagnostic testing and detection of indolent disease. Reducing overtreatment through greater use of observation or PSA-based monitoring in men with screen detected and low-risk CaP will result in similar length of life, avoid prostate cancer morbidity and mortality, and decrease treatment harms.