Abstract A53: Comprehensive population-specific biobehavioral marker panel for early prostate cancer diagnostics and risk assessment: A DETECT study

Abstract

Abstract Prostate cancer remains the most common cancerous malignancy and the second leading cause of cancer death among men of all races in the United States. However, the men of African descent are 1.4 times more likely to be diagnosed, and two to three times more likely to die of prostate cancer, compared to their counterparts of European descent. Even though the multifaceted nature of the striking racial disparity has been recognized, the majority of research efforts geared towards its resolution is focusing on exploring a discrete contribution of a single factor or a small group of similar factors. This approach has failed to comprehensively elucidate the etiology or the mechanism(s) responsible for the disparity, neither it has been utilized for an efficient prostate cancer screening and early detection in the high risk group of African American men. Therefore, we hypothesize, that a combination panel (DETECT) of biobehavioral population-specific biomarkers and factors, associated with prostate cancer in African Americans, will provide a valuable screening and risk assessment tool. The purpose of this study is to expound a prostate cancer screening tool that is based on a collective, rather than individual, assessment of prostate cancer bio-behavioral diagnostic markers and risk factors. The bio-behavioral markers are grouped into four general categories: genetic, biochemical, lifestyle and socio-cultural. The category “genetic markers” refer to factors associated with any bodily molecules containing genetic information. We include the measure of family history in this category since it gives an indication of the underlying genetic factors. Age is also included in this section, as the mechanism of aging has genetic background. The group “biochemical markers” refers to biomolecules of any nature (other than nucleic acids, which are classified as genetic markers) found anywhere in the body. The category “lifestyle markers” references the personal habits which heavily dependent upon individual choice. Finally, “sociocultural factors” reflect societal-and cultural-level influence on an individual's risk of prostate cancer. The choice of biobehavioral markers, together with the descriptive statistics, used to summarize distribution of each marker, is based on the review of current relevant literature. Two approaches are applied to evaluate the potential combined markers. The markers which are marginally significantly associated (p<0.1) are selected and evaluated using a logistic regression model. With this approach, pure interactions with no or weak main effects may be overlooked. Thus, we use Classification and Regression Trees (CART) as the second approach. Finally, logistic regression model is used to suggest a prototype prediction model of prostate cancer risk, based on the data, specified above. This study contributes into effort to eliminate the prostate cancer racial disparity by offering a prototype of comprehensive prostate cancer screening tool. This may result in higher levels of compliance, lower cost and higher effectiveness of screening, and decreased overall burden of prostate cancer in the high risk African American population. Citation Information: Cancer Epidemiol Biomarkers Prev 2010;19(10 Suppl):A53.