Quality-of-life outcomes in men treated for localized prostate cancer.

OBJECTIVE

To assess health-related quality of life (HRQOL) in men treated for clinically localized prostate cancer.

DESIGN

A cross-sectional analysis of HRQOL after treatment with radical prostatectomy, pelvic irradiation, or observation alone for clinically localized prostatic adenocarcinoma, and in age-matched comparison patients.

SETTING

A large managed care population in California.

SUBJECTS

A total of 528 men, including 214 treated for clinically localized prostate cancer (41 with evidence of metastatic disease were excluded from this analysis) and 273 age-matched, ZIP code-matched comparison patients without prostate cancer. Cancer patients were analyzed in three treatment groups: radical prostatectomy (n = 98), primary pelvic irradiation (n = 56), and observation alone (n = 60).

MAIN OUTCOME MEASURES

General HRQOL was measured with the RAND 36-Item Health Survey 1.0. Cancer-specific HRQOL was measured with the CAncer Rehabilitation Evaluation System-Short Form and the Functional Assessment of Cancer Therapy-General form. Disease-targeted quality of life was measured with a new instrument assessing function and bother in three organ systems: sexual, urinary, and bowel.

RESULTS

No differences among treatment groups were seen in comparisons of general HRQOL: Significant differences among treatment groups were seen in both function and bother in the prostate-targeted measures of sexual, urinary, and bowel domains. When cancer patients were compared with men of similar age without prostate cancer, differences were seen in the sexual, urinary, and bowel function and bother but not in general HRQOL measures. Although cancer-free men were found not to have full potency or continence, prostate cancer patients treated with surgery or radiation reported significantly worse sexual, urinary, and bowel function than men without cancer. Men who had undergone nerve-sparing prostatectomy did not differ from those who had undergone standard prostatectomy, but the power to detect a difference was low.

CONCLUSIONS

Although no differences were seen in general HRQOL, three disease-targeted domains were found to differ significantly among the treatment groups and comparison patients. Even after controlling for the sexual and urinary dysfunction experienced by older men without cancer, those receiving therapeutic interventions for their prostate cancer were found to have poorer disease-targeted HRQOL: We conclude that in addition to general HRQOL, disease-targeted measures must be used to assess outcomes of care in men treated for localized prostate cancer.