staged. Inappropriate surgery can have significant effects on quality of life because the potential side effects of radical prostatectomy include permanent urinary and sexual dysfunction.

Numerous studies have illustrated the prognostic usefulness of pretreatment PSA, clinical stage, and Gleason score in predicting posttreatment outcomes such as risk of recurrence (American Joint Committee on Cancer, 1997; Pisansky et al., 1997). Partin et al. (1993) have developed tabulated estimates for risk of the tumor spreading (called extracapsular extension) using the Gleason score (indicating tumor differentiation) and PSA. The goal of this approach is to attempt to improve the prediction of pathological stage for patient counseling and treatment planning. Estimates from their original tables have been improved by pooling data from patients across multiple facilities (Partin et al., 1997), but concerns about patient representativeness may limit the use of these tables as decision aids for physicians.

In addition to PSA, Gleason score, stage, and patient comorbidity can provide independent prognostic information about treatment outcomes. Experts in urology and radiation oncology at academic treatment centers around the United States agree about the importance of comorbidity assessment as part of the pretreatment workup, but there is considerable variation in the methods used for such assessment (Schuster et al., 1998). Suggested information to be used includes: Karnofsky performance status; patient self-reported activity levels; obesity; and history of cardiac disease, vascular disease, pulmonary disease, hypertension, diabetes, and surgeries. Pretreatment urinary, bowel, and sexual functioning have most commonly been assessed by patients' verbal reports. Some physicians have reported using the American Urological Association symptom score to assess obstruction; formal assessment of potency, voiding symptoms, or continence is rarely performed on a routine basis.

Although there is evidence in the literature that PSA, stage, Gleason score, and patient comorbidity provide useful prognostic information when treating patients, there is no evidence indicating whether performing these assessments prior to initiating treatment improves patient outcomes. Given the absence of process-outcomes links for the pretreatment evaluation, developing process measures for this aspect of prostate cancer care would have to be based completely on expert opinion. At present, there are no specific guidelines for the staging, workup, or pre-treatment assessment of patient comorbidity.

Choice of Treatment Modality

The modality used for primary treatment of prostate cancer varies depending on stage of disease, age or life expectancy, and patient preference. Treatment of localized prostate cancer (T1 or T2) can include surgery (radical prostatectomy), radiation therapy (external beam, brachytherapy, or conformal radiation therapy), or expectant management (watchful waiting). However, surgical treatment is not recommended for patients whose life expectancy is less than 10 years because the risks of surgery outweigh the survival benefit (Talcott, 1996). In addition, conformal radiation therapy is still being studied for efficacy and side effects (compared to standard external beam radiation therapy), but it has not yet been widely adopted as standard practice among radiation oncologists. Definitive evidence is lacking about the comparative efficacy of alternative treatment modalities for treating early-stage prostate cancer. The information used to make such decisions may have varying accuracy depending on its source: for example, clinicians' assess-

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement