Is It Necessary to Detect All Prostate Cancers in Men with Serum PSA Levels <3.0 ng/ml? A Comparison of Biopsy Results of PCPT and Outcome-Related Information from ERSPC

Context: The European Randomized Study of Screening for Prostate Cancer (ERSPC) section Rotterdam was initiated in 1993. Men who initially presented with prostate-specific antigen (PSA) values <3.0 ng/ml were not biopsied (with few exceptions). In the Prostate Cancer Prevention Trial (PCPT) eligible men who initially presented with PSA values <3.0 ng/ml were all biopsied during or at the end of a 7-yr study period. Objective: To compare biopsy rates in PCPT and cancer detection rates, interval cancers, and prostate cancer deaths in ERSPC. Report the number of additional biopsies needed to detect these cases using PCPT policy. Evidence acquisition: 21,210 men, aged 55-74 yr, were randomised to screening; 19,970 were actually screened between November 1993 and December 1999. A total of 15,852 initially presented with PSA values of <3.0 ng/ml; after excluding 79 detected at first screens, 15,773 remained as the study population. A second and third screening round followed after 4- and 8-yr intervals. All cancers found in three rounds of screening or as interval cancers during the 12-yr interval were identified and characterised. Evidence synthesis: Screening for prostate cancer and routine clinical management, comparison of detection rates and outcome data. During the 12-yr observation period, which may be too short, 700 cancers were found, 620 through screening and 80 as interval cancers. None of the screen-detected cases but 6 of the 80 men with interval cancers died of prostate cancer. Applying the positive predictive value of 21.7% of the PCPT trial 3472 cancers would have been detected in ERSPC Rotterdam had all men with PSA values <3.0 ng/ml been biopsied. Assuming 80 interval cancers and 6 deaths from prostate cancer might have been prevented if all 15,773 eligible men had undergone biopsy. Conclusions: The present data suggest a very much unfavourable "number needed to be biopsied" to find one missed cancer or to detect the deadly interval cancers.

• View at Publisher
• View at Pubmed
• View at Scopus
• View at Web of Science