Methods for prospective studies of adverse effects as applied to prostate cancer patients treated with surgery or radiotherapy without hormones

Abstract

Recently two new methods for prospective studies of adverse effects after treatment have been developed: Proportions of patients regaining 90% of baseline function score (PBS-90) and Generalized Estimating Equation (GEE). We compared these methods to examine changes of sexual, urinary, and bowel functions after robot-assisted prostatectomy (RALP) and conformal external beam radiotherapy (EBRT) in patients without androgen deprivation therapy (ADT). The post-treatment functional course was studied prospectively in 254 patients (N = 150 RALP and N = 104 EBRT) with PBS-90 and GEE. The time points at which functions reached stability and significant associations with function at 24 months were examined with PBS-90, and predictors were identified with GEE. The patients filled in the UCLA-PCI questionnaire at baseline and at 3, 6, 12, and 24-month post-treatment. The proportions reaching PBS-90 at 24 months were 69% EBRT and 34% RALP patients for urinary function, 70% of EBRT and 7% of RALP patients for sexual function, and 70% of EBRT and 86% of RALP patients for bowel function. GEE showed that the function scores at 6 months were significantly associated with the functions at 24 months. PBS-90 found that stability of function was reached at 3 months for urinary and 6 months for sexual and bowel functions. In outcome assessment PBS-90 mainly demonstrates when post-treatment level become stabilized and GEE shows the time points at which final outcome can be predicted. The two methods therefore supplement each other. Changes of functions corresponded to those reported in samples including patients having ADT.