The genetics of familial prostate cancer in Tasmania

Prostate cancer is the most frequently diagnosed non-cutaneous cancer in Australia and the second largest cause of male cancer deaths after lung cancer. One of the strongest risk factors for prostate cancer is family history and investigators have undertaken to identify susceptibility genes through linkage analysis of high-risk families. A number of putative prostate cancer susceptibility genes and candidate loci have been identified, however little consensus has been reached as to their contribution to disease. The objective of this study was to identify susceptibility genes and loci that contribute to familial prostate cancer in the population of Tasmania, Australia. The research takes advantage of 43 Tasmanian prostate cancer pedigrees and a second case control dataset comprising sporadic cases and age-matched controls. The familial dataset includes nine extended pedigrees, comprising at least 11 affected men. The first aim of the study was to evaluate the effects of previously reported putative prostate cancer susceptibility genes in the Tasmanian population. This was achieved by screening selected risk variants from ELAC2, MSR1, RNASEL, and AMACR in familial and sporadic Tasmanian prostate cancer cases and in controls. In addition, the families were used to perform linkage analysis in the vicinity of the above mentioned genes, as well as at BRCA1, BRCA2, and at chromosomal regions 16q23 and 20q13. Analyses did not provide evidence that any of these genes or chromosomal regions play a significant role in prostate cancer susceptibility in Tasmania. However, a BRCA2 mutation was found to partially segregate with disease in one large prostate cancer pedigree (p=0.035). A high-density 10K SNP genome-wide scan was performed on individuals from one large pedigree comprising 25 cases (PcTas9). A novel linkage analysis method was used for analysis of this family. Suggestive evidence for linkage to chromosome 5p13- q12 was obtained. This locus has been suggested previously as a candidate susceptibility region for prostate cancer. Follow up microsatellite analysis of an expanded number of cases from this family, including samples derived from pathology specimens, identified a shared haplotype in 8 cases (p=0.0017; Simwalk2 non-parametric B Statistic of 2.42, p=0.0038). The contribution of this putative susceptibility locus was investigated in the remaining large pedigrees, however no evidence of linkage was observed. A 10K SNP genome-wide scan was also performed on a second large pedigree (PcTas72). Because a smaller number of individuals were genotyped in this pedigree, it was analysed in a standard fashion using Merlin. Analysis revealed suggestive evidence for linkage to 10p14-15 (NPL=5.77, p=0.0009). KLF6, a previously proposed prostate cancer susceptibility gene, is located within this region. Screening of this gene did not reveal any mutations segregating with disease. A second approach to identifying genes that may be important in prostate cancer development and/or progression involved examining tumour samples for regional DNA sequence copy number abnormalities. The tumours of 20 cases in PcTas9 and PcTas72 were laser-dissected and studied using BAC array comparative genomic hybridisation (aCGH). To our knowledge, this is the first known aCGH analysis of multiple tumours from the same family. All tumours displayed multiple regions of loss and gain, and three common alterations were identified: seven cases displayed loss at 7p21, 5 cases displayed loss at 19p13.3, and 4 cases displayed gain at 17p13.3. This study has exploited a rare resource - a number of Tasmanian multiplex prostate cancer families - to examine the contribution of previously identified prostate cancer susceptibility genes in this population, and to identify novel putative loci. While two suggestive linkage loci and three regions of common chromosomal alterations were identified, the study has also highlighted the heterogeneous nature of prostate cancer even in large extended pedigrees from a genetically homogeneous island population