Dr. Catherine M. Tangen

Dr. Tangen's research interests include the design and analysis of clinical trials and non-parametric covariance adjustment, particularly for survival data. She is the statistician for Phase II, Phase III and ancillary studies for genitourinary cancers (e.g., prostate, bladder, renal, testes) in the Southwest Oncology Group. In addition to therapeutic trials, she is also the primary statistical investigator for the Prostate Cancer Prevention Trial (PCPT). The PCPT is a large (18,800 men), randomized, double-blind trial whose primary objective is to test the difference in biopsy-proven period prevalence of carcinoma of the prostate between a group of participants treated with finasteride and a group treated with placebo for seven years. The primary results, which were reported in June 2003, showed a 25% reduction in the period-prevalence of prostate cancer among those on finasteride relative to placebo. However, there was also a small but statistically significant increase in the rate of high grade cancer on the finasteride arm. A subsequent paper related to the prevalence of prostate cancer in men with normal PSA levels was published in the summer of 2004. A number of other manuscripts addressing the high grade issue and other topics are planned or currently underway.

The “Biology of PCPT” program project (P01) was submitted in October 2003 and resubmitted in June 2004. This P01 provides each project access to the invaluable repository of PCPT biospecimens and data. Each project is closely linked by interactive specific aims and planned collaborations. The five projects deal with (1) androgen metabolism, (2) diet and diet-related factors, (3) insulin-like growth factor axis and insulin resistance, (4) genotypic and phenotypic studies of inflammation, and (5) oxidative damage and DNA repair. These projects will use a nested case (n=1800) –control (n=1800) design to develop the P01 theme which is the genetic, metabolic and environmental factors associated with the risk of prostate cancer overall or high-grade disease and the effects of these factors on finasteride preventive efficacy. The mechanisms underlying these risk-factor associations also will be assessed. There will also be three cores which provide services: administrative, pathology and genotyping, and biostatistics. Dr. Tangen will serve the role as Core Director of Biostatistics for this program project.

The results of Dr. Tangen’s statistical research methods provide non-parametric treatment parameter estimates which clarify the components of bias adjustment and variance reduction. There are no formal assumptions required for how a response variable is related to the covariables. The rationale for non-parametric covariance analysis is based on the randomization in the study design. Computations for these methods are through the application of weighted least squares to fit linear models to the differences between treatment groups for the means of the response variable and the covariables jointly with a specification that has null values for the differences in covariates. Dr. Tangen is currently working on applying her non-parametric covariance methods to a survival function to describe the prevalence of a transient condition. Non-parametric extensions to the work of Pepe, Longton, and Thornquist are being explored.