[Primary circulating prostate cells, age and prostatic volumen predict the presence of prostate cancer in men with indication of a second prostate biopsy.]

Introduction: The limitations of serumPSA as a screening test to detect prostate cancer remainproblematic, especially after an initial negative prostatebiopsy. Detection of primary circulating prostate cells(CPCs) has been reported to be useful in the detectionof prostate cancer in men with a serum PSA>4.0ng/ml.We present a prospective study comparing the detectionof CPCs, total PSA, percent free PSA, digital rectal examination(DRE) and prostate volumen (PV) to establisha predictive model for the detection of prostate cancerin men with an indication for a second prostate biopsy.
Objective: To establish a predictive model for the detectionof prostate cancer using the number of CPCs detectedper sample, DRE, age, total serum PSA, percentfree PSA and PV in men with an indication for a secondprostate biopsy.
Methods and Patients: A prospective, observationalstudy carried out in the Hospital de Carabineros deChile, between 2006 and 2014 including 199 menundergoing a second prostate biopsy. The variables,number of CPCs detected (nCPC), DRE, age, PSA, percentfree PSA and PV were registered for each patientand based on these findings and comparing them withthe results of the prostate biopsy a multivariate logistic regressionanalysis incorporating forward predictors. Themodel was evaluated for co-lineal tendency, reliabilityand error specificity, and analyzed using non-parametricreceiver operating characteristics and area under thecurve decision for the combined model and for eachvariable separately.
Results: The single variable nCPC had a superior predictivevalue with an area under the curve of 0.89 (95%CI 0.83-0.94). The final model incorporated nCPC (OR:2.03 95% CI 1.63-2.53) age (OR: 1.1 95% CI 1.04-1.17) and PV (OR: 0.96 95% CI 0.93-0.99) with anarea under the curve for the combined model of 0.92(95% CI 0.88-0.97). The combined model performedbetter than the variables used alone.
Conclusions: The model incorporating nCPC, ageand PV had a greater diagnostic yield for the predictingprostate cancer at second biopsy.