- L O Reis: Departamento de Urología y Oncología, Universidad Estatal de Campinas, Brasil. reisleo@unicamp.br
INTRODUCTION: Prostate specific antigen (PSA) and digital rectal examination (DRE) are the main tests for initial prostate investigation; there is no consensus about the best criterion for prostate biopsies. We aim to check the accuracy of different criteria in this context including PSA derivatives to detect prostate cancer.
MATERIAL AND METHODS: Four different criteria for indication of prostate biopsy were compared: (A) PSA-density (>15 ng/ ml/ cc); (B) PSA > 2,5 ng/ml; (C) PSA-velocity (> 0.7 ng/ ml/ year); (D) free/total PSA ratio (<15%). All biopsies and histopathological examinations were performed by the same urologist and pathologist, respectively.
RESULTS: The study was performed on 180 consecutive biopsies with 37.7% overall cancer detection rate: 29 (16.1%) performed following criterion A, 42 (23.3%) criterion B, 65 (36.1%) criterion C and 44 (24.4%) criterion D. Based on PSA criteria alone, the predictive positive value (PPV) was 37.9% for criterion A, 33.3% for B, 32.3% for C and 50.0% for criterion D, respectively, (p > 0.05). Associating positive DRE with changed PSA, the PPV increased to 50%, 50%, 43.9% and 68.2% for criteria A, B, C and D, respectively (p>0.05). In univariate analysis, DRE (positive versus negative), PSA level (>10 ng/ ml versus <4.0 ng/ ml), free/total PSA ratio (<10% versus >15%) and age were associated with PC. In multivariate analysis only positive DRE was associated with prostate cancer.
CONCLUSIONS: All the criteria of PSA derivatives are complementary and useful predictors of cancer risk. However, a positive DRE increased the PPV of PSA derivatives. New tools are needed to improve the accuracy of prostate cancer detection.