Outcomes Based Guideline Development for Prostate Cancer Screening and Treatment

The broad, long-term objective of this study is to reshape the process by which guidelines for prostate cancer screening are produced. Almost 20 years have passed since PSA was first used for early detection. Yet there is still enormous uncertainty about the right PSA-based criteria for biopsy referral and whether to treat localized tumors detected solely on the basis of a positive test. Many groups have been charged with issuing guidelines for prostate cancer screening, but their recommendations are not consistent.

The central premise of this application is that knowledge of the likely benefits and harms associated with candidate policies is critical in the formulation of good practice guidelines. However, in the case of PSA screening, there is no accepted quantitative framework for projection of these outcomes. Most decision models of prostate cancer screening do not explicitly model PSA growth and therefore cannot compare policies that vary the PSA-based criteria for biopsy. There are no published models that jointly evaluate screening and treatment decisions, but liberal screening policies may necessitate adoption of selective treatment policies.

The specific objective of the proposed work is to take a recently developed model of PSA growth and prostate cancer progression and detection, and turn it into an outcomes calculator for use in developing guidelines for PSA screening and curative treatment. The calculator will include an interface that will allow users to specify a range of policies of interest and produce a series of outcomes including benefits, harms, and economic costs. This activity is supported by a grant from the National Institute of Health/National Cancer Institute grant (R01 CA131874)

Publications