Impact of Endocrine Therapy on Survival in Men with Local or Regional Prostate Cancer - Feasibility Study

Background

Recent trends in prostate cancer survival illustrate that mortality from prostate cancer is declining. While some of this decline can be attributed to PSA screening, the survival patterns do not fully support PSA as the only reason for the trend. Initial management of localized or regional prostate cancer usually involves surgical excision, radiation therapy or watchful waiting. Although hormonal therapy is also available, this treatment is mostly used only after a recurrence is detected.

Study Aim

The goal of this study is to determine whether early use of hormone therapy among men with local or regional prostate cancer has an effect on survival. A feasibility study was undertaken as an initial step to gather information about the validity and availability of important variables. If warranted by the results of the feasibility study, the CRN will continue with a full-scale study using automated data to ascertain the impact of early hormonal therapy on prostate cancer survival.

Methods

Automated data from four CRN sites were used to identify clinically localized or regional prostate cancer cases occurring between 1990 and 1998. Treatment variables and other key data elements such as tumor characteristics were also collected from the sites' computerized data systems. To validate quality and completeness of the critical automated data elements, chart abstraction was performed for a sub-sample of cases.

Significance

This collaborative effort includes more than 20,000 patients with localized or regional prostate cancer, 15 percent of whom received early hormonal therapy. This study is designed to efficiently evaluate the association between early hormonal treatment of prostate cancer and mortality by using automated data sources, and will provide important information about prostate cancer treatment.

Project Status

This study was funded as an administrative supplement to the CRN parent grant, and is under the leadership of Marianne Ulcickas Yood, DSc, MPH. This work has confirmed that automated data can be used to classify hormone treatment exposure among men diagnosed with prostate cancer, although variability exists across HMO and over calendar time. Further analysis will determine whether a full study of hormone therapy and prostate cancer, using only automated data with these four CRN sites, is warranted and the parameters with which such a study should be undertaken.

Return to top