The impact of different prostate-specific antigen (PSA) testing intervals on Gleason score at diagnosis and the risk of experiencing false-positive biopsy recommendations: a population-based cohort study

ObjectiveGiven a man’s current prostate- specific antigen (PSA) level, age and family history of prostate cancer, what are the benefits (decreased risk of higher Gleason score [GS] cancer at diagnosis) and harms (increased risk of false-positive biopsy recommendation) of waiting 1, 2, 3, 4 or 5–8 years until the next PSA test?DesignProspective cohort.SettingAll PSA tested men in Stockholm, Sweden, between 2003 and 2015.ParticipantsMen aged 50–74 years with at least two PSA tests between 2003 and 2015 (n=174 636).Main outcome measuresLog-binomial regression to calculate the risk ratio (RR) of GS ≥7 and GS 6 versus benign outcome at prostate biopsy and 12-year cumulative probability of experiencing a false-positive biopsy by testing interval, age, PSA level and first-degree family history.ResultsMen with PSA ≤1 ng/mL had low risk of GS ≥7 prostate cancer irrespective of testing interval; 3 at the next testing occasion, and of the 663 men biopsied after the next PSA test only 32 (5%) had GS ≥7 cancer. Men with PSA >1 ng/mL had increased risk of being diagnosed with GS ≥7 prostate cancer when screened with longer than annual intervals (RRs ranged from 1.4 to 3.2 depending on PSA level and testing interval). The results were consistent across age groups and family history status. This benefit needs to be balanced against the increased risk for false-positive biopsy recommendation with shorter testing intervals (twofold for annual vs biennial and threefold for annual vs triennial).ConclusionsMen aged 50–74 years with PSA ≤1 ng/mL can wait 3–4 years before having a new PSA test. For men with PSA >1 ng/mL, we observed an increased risk of being diagnosed with GS ≥7 prostate cancer with longer than annual testing intervals. This benefit needs to be balanced against the markedly increased risks for false-positive biopsy recommendations with shorter testing intervals recommendations.