The third analysis on prostate cancer mortality in the ERSPC trial, published in the Lancet August 2014, shows stable relative (RR=0.79) but larger absolute benefit in follow-up extended to 13 years (two additional years from the previous report).
The ERSPC is the largest randomized trial of screening for prostate cancer with 162,388 men and 900 prostate cancer deaths. Screening is based on regular prostate-specific antigen (PSA) testing every 2-4 years in the intervention arm and usual care with no screening offered in the control arm. The ERSPC trial has previously demonstrated significant reductions in prostate cancer mortality after 9 years and 11 years of follow-up of our trial (Schröder and colleagues publications in the New England Journal of Medicine 2009 and 2012). We now provide updated results truncated at 13 years for men in the core age group of the study, aged 55-69 years at randomization.
In this update of the ERSPC with follow-up truncated at 13 years, a significant 21% relative reduction in prostate cancer was found in intention to screen analyses, and 27% in men who actually attended screening. The absolute risk reduction of death from prostate cancer at 13 years, with 1.28 fewer prostate cancer deaths per 1,000 men randomized was increased by a quarter compared with analyses limited to 11 years. A total of 781 men needed to be invited to screening (“NNI”=number needed to invite) and 27 to be diagnosed (“NND”=number needed to diagnose) with prostate cancer to avert one death from the disease. These numbers are substantially lower compared to previous reports at 9 years (NNI 1410, NND 48) and 11 years (NNI 979, NND 35) demonstrating a larger effect.
The main drawback of screening is the increased risk of overdiagnosis of prostate cancer due to screening, meaning detection of cancers that may not give rise to symptoms or lead to death during the lifetime of a typical man.
Despite showing a clear prostate cancer mortality reduction, the findings are not sufficient to justify population-based screening. We still need further quantification of harms of screening and better strategies to overcome overdiagnosis and overtreatment for both more targeted screening and assessment of prostate cancer risk, such as multivariate risk stratification. In the meantime, well-informed men suitable for screening should have access to PSA-testing if they wish after careful consideration of the pros and cons. Various decision support tools are available online developed for communication of the pros and cons for men considering screening to discuss with their doctors. Multiparametric magnetic resonance imaging holds promise in evaluation of screen-positive men and is being tested in large scale trials.