Each year, over 1,000,000 needless prostate biopsies are performed on men who present with elevated levels of prostate specific antigen (PSA) but show no evidence of cancer upon biopsy. Whether one believes that widespread PSA testing leads to overtreatment of indolent cancer or is an important tool to reduce prostate cancer mortality, screening for prostate cancer is unlikely to go away and it is clear that reducin these false positive findings will greatly reduce unnecessary medical procedures and their attendant adverse consequences. The long-term goal of this research project is to develop improved screening tests that will identify men at risk for prostate cancer, especially potentially lethal disease, with high sensitivity and specificity. The overall objective of this application isto determine if consideration of an individual's genetic makeup can improve the accuracy of screening tests based on PSA and other prostate-produced biomarkers. The central hypothesis underlying this application is that consideration of SNPs that are associated with levels of prostate cancer biomarkers and their interaction in predictive models will improve model performance. The rationale behind this project is that if SNPs do influence biomarker levels independent of disease status, then personalized biomarker evaluation that takes into account SNP genotype is necessary to maximize accuracy of the test. This approach to prostate cancer screening will be developed by: 1) Identifying SNPs associated with levels of PSA and other biomarkers in men without prostate cancer. A panel of SNPs previously associated with prostate cancer risk and/or levels of prostate secreted biomarkers will be tested for association with a panel of five isoforms of prostate secreted proteins in healthy young men and healthy older men. 2) Determining the ability of SNPs to improve accuracy of PSA-based predictive models. Using a nested case- control study, those SNPs associated with biomarker levels in aim 1 will be included, along with their interaction with biomarker levels, in a predictive model for prostate cancer. 3) Determining the generalizability of such models in diverse populations. The best model from aim 2 will be externally validated in three separate studies from the US and Sweden representing a variety of ethnicities and both nested case-control designs and a biopsy cohort. This proposal is innovative because it investigates the association of SNPs with levels of prostate biomarkers in healthy young men who can be presumed to be cancer-free;because it suggests interpretation of biomarkers in light of SNP genotype rather than simply combining biomarker and SNP information;and because it is not restricted to a population with a single ancestry. The expected outcome of this research is a predictive model for prostate cancer that integrates genetic variation with prostate secreted protein biomarkers. The positive impact of such a model will be the reduction of unnecessary biopsies while still facilitating the early detection of prostate cancer.
While measurements of levels of the prostate-specific antigen (PSA) in the blood of healthy men is routinely used to screen for prostate cancer, this test is les than ideal as elevated levels of PSA result in many men undergoing needless prostate biopsies that reveal no cancer. To reduce the number of unnecessary biopsies conducted, a method of integrating genetic information with blood levels of PSA and related proteins will be developed under this grant. This method will enable early prostate cancer to still be detected while reducing the number of men without cancer who undergo needless biopsies.
Showing the most recent 10 out of 18 publications
