Introduction: To evaluate the association of total prostate specific antigen (T-PSA) and percent free PSA (%F-PSA) with prostate cancer outcomes in patients treated with radical prostatectomy (RP).
Methods: Pre-operative serum levels of T-PSA and F-PSA were prospectively measured in 402 consecutive patients treated with RP for clinically localized prostate cancer who had T-PSA levels below 10 ng/ml.
Results: T-PSA was not associated with any prostate cancer characteristics or outcomes. Lower %F-PSA was significantly associated with higher percent positive biopsy cores, extracapsular extension, seminal vesicle involvement, lympho-vascular invasion, perineural invasion, positive surgical margins, and higher pathologic Gleason sum. When adjusted for the effects of standard pre-operative features, lower %F-PSA significantly predicted non-organ confined disease, seminal vesicle involvement, lympho-vascular invasion, and biochemical progression. %F-PSA did not retain its association with biochemical progression after adjusting for the effects of standard post-operative features. Based on data from 22 patients with biochemical progression, lower %F-PSA was correlated with shorter T-PSA doubling time after biochemical progression (rho = 0.681, p = 0.010). %F-PSA was lower in patients who failed salvage radiation therapy (p = 0.031) and in patients who developed distant cancer metastases compared to patients who did not (p < 0.001).
Conclusions: Pre-operative T-PSA is not associated with prostate cancer outcomes after RP when levels are below 10 ng/ml. In contrast, pre-operative %F-PSA is associated with adverse pathologic features, biochemical progression, and features of aggressive disease progression in patients treated with RP and T-PSA levels below 10 ng/ml. %F-PSA may improve pre-operative predictive models for predicting clinical outcomes of patients diagnosed with prostate cancer nowadays.