Early detection practices, decreasing age at diagnosis, and a linkage analysis in prostate cancer families.

Abstract

Among United States men, prostate cancer is the most commonly diagnosed cancer, and the second most frequent cause of cancer death. Despite these facts, prostate cancer etiology is not well understood. Established risk factors include age, race, and family history. Multiple studies provide evidence that genetic risk factors for prostate cancer exist, and several susceptibility loci have been proposed. Nevertheless, many questions remain about familial and genetic aspects of the disease, and this thesis addresses several of these issues among subjects sampled from the University of Michigan Prostate Cancer Genetics Project (PCGP). The first paper examines early detection practices among 126 men to determine whether PCGP participants who report having never been diagnosed with prostate cancer are likely to be truly unaffected. Results from a self-administered questionnaire suggest that over 95% of unaffected men had undergone a prostate-specific antigen (PSA) test and/or a digital rectal examination (DRE). There was no significant difference between affected and unaffected men in age at first PSA or DRE. Therefore, it is unlikely that misclassification of unaffected men occurs systematically in familial prostate cancer studies. The second paper examines the hypothesis that age at prostate cancer diagnosis is decreasing over successive generations, a hallmark of genetic anticipation, within prostate cancer families. Survival analysis in 1,345 men from 489 families estimates that members of a particular generation are at 1.3 times higher risk of being diagnosed with prostate cancer at a given age compared to the previous generation, after controlling for birth decade (p = 0.0005). While statistical evidence supports the hypothesis, interpretation requires caution due to the potential for ascertainment bias. The third paper is the first replication study examining a prostate cancer genetic susceptibility locus on chromosome 20 in 172 families with a history of prostate cancer. Although some positive results were observed, this linkage study does not provide statistically significant support for a prostate cancer susceptibility gene <italic>HPC20</italic> at 20q13. A better understanding of genetic and behavioral risk factors for prostate cancer through these and other studies may contribute to eventual improvements in prostate cancer prevention, detection, and treatment.