Identifying genetic susceptibilities underlying familial haematological malignancies in a Tasmanian family resource

Haematological malignancies (cancers of the haematopoietic and lymphoid tissues) are collectively one of the most frequently diagnosed cancers in Australia. Family history is one of the strongest risk factors for disease. Evidence for this derives from large population-based studies that have identified an increased risk of haematological malignancies in first degree relatives of cases, as well as studies of individual families where analyses have identified genes where family specific germline mutations predispose to these malignancies. Despite intensive research into the genetic predisposition to these cancers, the known genes account for only a small portion of the overall inherited component of haematological malignancies, leaving a significant gap in our understanding of the genetic basis of disease. Earlier studies used candidate gene approaches or sparse sets of genome wide markers to identify predisposition genes. Such approaches have a limited capacity for disease gene identification. Now, application of innovative technologies, such as next generation sequencing, to familial datasets with multiple cases of haematological malignancies presents an ideal opportunity to identify new predisposing germline mutations and other genetic factors contributing to disease development. The aim of this study was to identify the genetic architecture of disease susceptibility in large families affected by multiple subtypes of haematological malignancies. This study takes advantage of a collection of extended Tasmanian haematological malignancy pedigrees comprising 48 families, as well as 84 additional Tasmanian haematological malignancy cases with no known family history of disease. This resource is particularly valuable due to the recognised stability and relative genetic homogeneity of the island population of Tasmania. Next generation sequencing approaches were employed to identify novel, rare and shared predisposing mutations in affected family members. This was achieved through a combination of whole exome and whole genome sequencing in five prioritised families. Genome and exome alignment and variant calling were conducted using BWA and SAMtools. High-quality single nucleotide and small insertion / deletion variants identified were then annotated with information from public data sources using ANNOVAR. Variants were filtered to focus in on rare variants (with population frequency estimates of 1% or less) using frequencies in Caucasian population data from the 1000 Genomes Project and the UK10K consortia dataset. A large number of rare shared genetic mutations were identified between related haematological malignancy cases in these families. A tiered prioritisation strategy was developed and employed to identify the top preferred candidates for further followup. This strategy incorporated variant-based prioritisation, using in silico predictions of variant effect, and gene-based prioritisation using known gene biology. For genebased prioritisation a literature curated network analysis tool (Ingenuity Pathway Analysis) and an ontology-based tool (Phevor) as well as publically available tissue expression profiles of the mutated genes were used. Genes prioritised for further follow-up include examples such as TNFSF9, TDP2, MMP8, and NOTCH1. These genes have not been previously implicated in the familial risk for haematological malignancies, although some have previously established roles in malignancy. For example, TNFSF9 is a gene with clear connections to both T-cell and B-cell biology and there is evidence from a mouse knockout model that disruption to this gene can contribute to malignancy development. A subsequent aim of this study was to explore the role of telomere biology in familial haematological malignancies. Telomere biology has a well-characterised role in cancer development. Disruption of key telomere biology genes has been shown to lead to a spectrum of syndromes of which haematological malignancies are a feature such as dyskeratosis congentia and aplastic anaemia. To examine whether disrupted telomere biology was detectable in haematological malignancies, an analysis of telomere length was conducted using a PCR-based assay measuring across the familial resource, non-familial cases and population controls. Telomere length was analysed as a quantitative trait using variance components modelling, adjusting for age, sex and importantly kinship. The key finding from this analysis was that telomere length was highly heritable at 62.5% (P=4.7‚àöv=10-5) indicating a strong genetic effect driving variation in telomere length and that both familial and non-familial haematological malignancy cases had shorter telomeres (P=2.2‚àöv=10-4 and 2.2‚àöv=10-5 respectively). These results indicate that telomere length contributes broadly to haematological malignancies. Genetic variation in some of the known telomere biology genes was examined, however the underlying genetic contribution to the observed shortened telomere length remains to be determined. This thesis describes the genetic analysis of a rare resource, providing evidence for several novel genes with possible roles in the development of haematological malignancies. As expected next generation sequencing of these families has further highlighted the multigenic contribution to risk in this complex disease.