Low Preoperative Prolactin Levels Predict Non-Organ Confined Prostate Cancer in Clinically Localized Disease

Abstract

<b><i>Introduction:</i></b> To evaluate the association between preoperative serum prolactin (PRL) levels and risk of non-organ confined prostate cancer (PCa) in clinically localized disease. <b><i>Materials</i></b> <b><i>and Methods:</i></b> From December 2007 to December 2011, 124 patients with clinically localized PCa were retrospectively evaluated. Non-organ confined disease in the surgical specimen was defined according to extra-capsular extension, seminal vesicle invasion, positive surgical margins, and lymph node invasion. The association between clinical factors and serum levels of pituitary-testis hormones with the risk of non-organ confined disease was evaluated. <b><i>Results:</i></b> Perioperative factors associated with non-organ confined disease include prostatic-specific antigen (OR 1.144; <i>p</i> = 0.025), proportion of biopsy positive cores (BPC, OR 36.702; <i>p</i> = 0.007), bioptical Gleason Score > 6 (OR 2.785; <i>p</i> = 0.034), and PRL (OR 0.756, <i>p</i> < 0.0001). The association was strong for BPC (area under the curve [AUC] 0.704; <i>p</i> < 0.0001) and PRL (AUC 0.299; <i>p</i> < 0.0001). When we dichotomized according to median value, PRL ≤7.7 µg/L was an independent predictor of extraprostatic disease (OR 6.571; <i>p</i> < 0.0001) with fair discrimination power (AUC 0.704; <i>p</i> < 0.0001). <b><i>Conclusion:</i></b> Low preoperative PRL levels predict the risk of non-organ confined PCa in clinically localized disease.