Unintended consequences of decreased PSA-based prostate cancer screening.

BACKGROUND: In May 2012, the US Preventive Services Task Force issued a grade D recommendation against PSA-based prostate cancer screening. Epidemiologists have concerns that an unintended consequence is a problematic increase in high-risk disease and subsequent prostate cancer-specific mortality. MATERIALS AND METHODS: To assess the effect of decreased PSA screening on the presentation of high-risk prostate cancer post-radical prostatectomy (RP). Nine high-volume referral centers throughout the United States (n = 19,602) from October 2008 through September 2016 were assessed and absolute number of men presenting with GS ≥ 8, seminal vesicle and lymph node invasion were compared with propensity score matching. RESULTS: Compared to the 4-year average pre-(Oct. 2008-Sept. 2012) versus post-(Oct. 2012-Sept. 2016) recommendation, a 22.6% reduction in surgical volume and increases in median PSA (5.1-5.8 ng/mL) and mean age (60.8-62.0 years) were observed. The proportion of low-grade GS 3 + 3 cancers decreased significantly (30.2-17.1%) while high-grade GS 8 + cancers increased (8.4-13.5%). There was a 24% increase in absolute numbers of GS 8+ cancers. One-year biochemical recurrence rose from 6.2 to 17.5%. To discern whether increases in high-risk disease were due to referral patterns, propensity score matching was performed. Forest plots of odds ratios adjusted for age and PSA showed significant increases in pathologic stage, grade, and lymph node involvement. CONCLUSIONS: All centers experienced consistent decreases of low-grade disease and absolute increases in intermediate and high-risk cancer. For any given age and PSA, propensity matching demonstrates more aggressive disease in the post-recommendation era