PSA density improves prediction of prostate cancer

Abstract

Introduction: Prostate-specific antigen (PSA) and the digital rectal exam (DRE) have moderate sensitivity but low specificity for cancer diagnosis, potentially causing unnecessary treatment complications with prostate biopsy. Transrectal ultrasound (TRUS) to evaluate prostate size and calculate PSA density can improve the specificity of PSA in predicting cancer. We evaluated the sensitivity and specificity of different pre-biopsy tests to detect prostate cancer. Materials and methods: Pre-biopsy data were collected from 521 men referred for biopsy from January-December 2011 and cancer aggressiveness data from 96 men who had radical prostatectomy. Model predictors included total PSA, DRE, the ratio of free to total PSA (PSAf/t), and PSA density. We used logistic regression and ROC curve analyses to compare the accuracy of different models to predict positive biopsy. Results: The area under the curve (AUC) for model A (PSA total, DRE, PSAf/t) was moderate, but significant (AUC = .59, p < .05), PSA density was the only strong predictor (OR = 1067.93, p