Welcome to the CJU website » LOG IN


Risk stratification in clinically localized prostate cancer
Montreal General Hospital, McGill University, Montreal, Quebec, Canada
Jun  2002 (Vol.  9, Issue  31, Pages( 18 - 20)


Text-Size + 

  • Clinical outcomes in patients with localized prostate cancers are heterogeneous. In recent years, analyses of large datasets from multiple centres have yielded a better understanding of how to measure risk in localized prostate cancer. Regardless of whether patients are treated with prostatectomy, radiotherapy, brachytherapy, or expectant management, three factors appear correlated with clinical outcome: biopsy Gleason score, clinical T stage, and serum prostate-specific antigen (PSA). Partin Tables, derived from these parameters and recently updated and refined, may be used to estimate the risk of metastasis and to assess certain aspects of surgical management in clinically localized disease. Partin tables, however, are limited by the fact that pathologic stage does not always predict clinical outcome. Nomograms that employ serum PSA, biopsy Gleason score, and clinical T-stage have been developed with the aim of predicting clinical recurrence after radical prostatectomy or radiation therapy. Three risk categories for clinically localized prostate cancer have recently been developed by the Canadian Genito-Urinary Radiation Oncologists Consensus Conference, which group cases according to serum PSA, T-stage, and biopsy Gleason score. Additional factors have been assessed in the hopes of improving the prediction of outcome in clinically localized disease, but none of these has consistently been demonstrated to add independent value to the principal parameters of serum PSA, T-stage, and Gleason score. Virtually all predictive nomograms, algorithms, and tables incorporate a combination of these three parameters. While these tools may be useful in prognosticating an individual case, several limitations preclude their widespread use. The greatest benefit to date of risk stratification is its use in comparing outcomes of series of patients treated with various modalities, and in clinical trial design and analysis.