Prediction of locoregional extension and metastatic disease in prostate cancer: a review

Abstract

Despite efforts to enhance the accuracy of prediction of extraprostatic disease, approximately 40% of the men undergoing radical prostatectomy are found at surgery to have non-organ-confined cancer. Predictive algorithms based on multivariate regression analysis and neural networks are widely available and are superior to our standard empirical methods of clinical staging. These algorithms have been validated in diverse and well-characterized patient groups. For enhancement of the predictive value, data input must be standardized and improved input variables must be incorporated. In addition to the three “classic” staging parameters, i.e., pretreatment prostate-specific antigen (PSA), biopsy pathology, and digital rectal examination, new variables now show promise in predicting disease extent and may be integrated in future predictive models. This review focuses on our present methods for prediction of locoregional spread and distant metastases in men with clinically localized prostate cancer.