Age and PSA predict likelihood of organ-confined disease in men presenting with PSA less than 10 ng/mL: implications for screening.

Overview

abstract

OBJECTIVES: To examine age and pretreatment prostate-specific antigen (PSA) level as predictors of organ-confined disease in a large multinational database of men with pretreatment PSA levels less than 10 ng/mL who were treated with radical prostatectomy. The optimal PSA and age cutoffs for triggering prostate biopsy are controversial. Although a PSA level greater than 4.0 ng/mL is generally accepted, recent observations suggest that PSA values between 2.5 and 4.0 ng/mL are associated with a 20% detection rate. Furthermore, age has been shown to represent an independent predictor of outcome after radical prostatectomy. METHODS: We analyzed a cohort of 3198 patients from five institutions, who underwent radical prostatectomy between 1985 and 2001 for clinically localized prostate cancer. All presented with a pretreatment PSA level of less than 10 ng/mL. Multivariate analyses addressed age, preoperative PSA, year of treatment, biopsy Gleason sum, and clinical stage as independent predictors of non-organ-confined disease (NOCD). RESULTS: Twenty-nine percent of patients had NOCD. All tested variables were independent, multivariate predictors of NOCD (age, P = 0.004; year of treatment, P <0.0001; PSA, P <0.0001; Gleason sum, P <0.0001; clinical stage, P <0.0001). A linear relationship between age and the risk of NOCD was noted. The likelihood of NOCD remained constant for PSA levels between 0.1 and 4.0 ng/mL and rose substantially for PSA levels greater than 4.0 ng/mL. CONCLUSIONS: After controlling for other covariates, the rate of NOCD increased in proportion to age and pretreatment PSA level in men initially diagnosed with a serum PSA level less than 10 ng/mL. These observations suggest that younger men with a lower PSA are more likely to have organ-confined and curable disease at diagnosis. This information has important implications for counseling and screening.