Evaluation of the Houston biochemical relapse definition in men treated with prolonged neoadjuvant and adjuvant androgen ablation and assessment of follow-up lead-time bias.

PURPOSE: To validate the Houston prostate-specific antigen relapse definition in a mature cohort of men treated with external beam radiotherapy (EBRT) and adjuvant androgen ablation (AA) and men treated with EBRT monotherapy, and to compare these results with the American Society for Therapeutic Radiology and Oncology (ASTRO) and Vancouver prostate-specific antigen relapse (biochemical no evidence of disease) definitions. METHODS AND MATERIALS: A prospective database of 1490 men treated with EBRT, with or without AA, was examined. The impact on hazard proportions, as well as the predictive ability, of the Houston, ASTRO, and Vancouver definitions was tested.
RESULTS: For all patients, the Houston definition was more accurate (79.5%) than the ASTRO (76.7%) or Vancouver (77.2%) definitions in predicting subsequent clinical relapse. The Houston definition was superior to the ASTRO definition in those treated both with and without AA and equivalent to the Vancouver definition in those receiving AA. The Houston definition demonstrated proportional hazards when categorized for the use of AA, unlike the ASTRO and Vancouver definitions. The effect of inadequate follow-up on the projected relapse rates was negligible with the Houston definition.
CONCLUSION: The Houston relapse definition is favored after EBRT monotherapy or combined EBRT and AA. Use of the Cox proportional hazard multivariate analysis is appropriate with the Houston definition, but not with the ASTRO or Vancouver definitions if AA and non-AA patients are combined.