Prostate-specific antigen (PSA) bounce and other fluctuations: which biochemical relapse definition is least prone to PSA false calls? An analysis of 2030 men treated for prostate cancer with external beam or brachytherapy with or without adjuvant androgen deprivation therapy.

PURPOSE: To determine the false call (FC) rate for prostate-specific antigen (PSA) relapse according to nine different PSA relapse definitions after a PSA fluctuation (bounce) has occurred after external beam radiation therapy (EBRT) or brachytherapy, with or without adjuvant androgen deprivation therapy. METHODS AND MATERIALS: An analysis of a prospective database of 2030 patients was conducted. Prostate-specific antigen relapse was scored according to the American Society for Therapeutic Radiology and Oncology (ASTRO), Vancouver, threshold + n, and nadir + n definitions for the complete data set and then compared against a truncated data set, with data subsequent to the height of the bounce deleted. The FC rate was calculated for each definition.
RESULTS: The bounce rate, with this very liberal definition of bounce, was 58% with EBRT and 84% with brachytherapy. The FC rate was lowest with nadir + 2 and + 3 definitions (2.2% and 1.6%, respectively) and greatest with low-threshold and ASTRO definitions (32% and 18%, respectively). The ASTRO definition was particularly susceptible to FC when androgen deprivation therapy was used with radiation (24%). DISCUSSION: New definitions of biochemical non-evidence of disease that are more robust than the ASTRO definition have been identified. Those with the least FC rates are the nadir + 2 and nadir + 3 definitions, both of which are being considered to replace the ASTRO definition by the 2005 meeting of the Radiation Therapy Oncology Group-ASTRO consensus panel.