Pathology - Theses

Permanent URI for this collection

Search Results

Now showing 1 - 7 of 7
  • Item
    Thumbnail Image
    The role of next generation sequencing in the management of haematological malignancies
    Corboy, Gregory Philip ( 2018)
    Next generation sequencing comprises a rapidly-evolving cohort of technologies which enable detection of genetic variants present in DNA or RNA. In the context of haematological malignancies, such variants may have diagnostic, prognostic or therapeutic relevance. The scope of diagnostic-grade genetic testing is increasing as the underlying molecular landscape of haematological malignancies is increasingly well-characterised and becomes further embedded in clinical management. Multiple conventional methods for the detection of somatic single-gene variants are already embedded in standard-of-care. The technical performance of next generation sequencing when compared to these existing methods is of interest, since this determines in which instances it can be used to supplement or replace the status quo. This work focuses on acute myeloid leukaemia and systemic mastocytosis, since detecting variants in these malignancies presents several technical challenges. Acute myeloid leukaemia demands detection of a broad spectrum of molecular lesions, whereas systemic mastocytosis requires high sensitivity testing. Chapter one reviews molecular testing for haematological malignancies, including aspects of next generation-based testing relevant to diagnostics. Molecular markers currently employed in the diagnosis of acute myeloid leukaemia and systemic mastocytosis are discussed. The second chapter details methods used in the generation of data for chapters three to five. In the third chapter the selection and use of a targeted sequencing panel is discussed, testing performance is compared to conventional molecular methods using data generated from 30 clinical samples. Limitations of panel-based testing are discussed, including sensitivity, detection of specific variant types, and relevant bioinformatic and analytical issues. In chapters four and five, proof of principle is demonstrated for two novel next generation sequencing library generation methods, addressing the issues of structural variant detection ‘HEPTAD’ and high sensitivity single nucleotide variant detection ‘LNA PCRbrary’, respectively. Theory, optimisation and testing results for a combination of 49 clinical samples and cell lines using HEPTAD are discussed. The technique is applied to detect chromosomal translocations involving RUNX1, BCR, ABL, RARA, MLL/KMT2A, a chromosomal inversion involving CFBF, and KMT2A partial tandem duplications. The method detects structural variants in diagnostic samples with 100% sensitivity but requires further optimisation for minimal residual disease testing. Successful prospective application in a clinical scenario where orthogonal FISH testing was performed is detailed. LNA PCRbrary’s application to the detection of KIT D816V single nucleotide variants in peripheral blood is discussed. Suboptimal results when testing RNA derived from clinical samples are detailed, but promising results when testing cell line DNA, and significant potential for further optimisation and testing. In conclusion this work shows that NGS does have a role in the management of haematological malignancies, but technical challenges remain, including high sensitivity detection of single nucleotide variants, minimal residual disease monitoring of structural variants and gene targets which are difficult to amplify. Some of these issues are in part addressed in this work, but careful validation of testing sensitivity and specificity is required when introducing such NGS assays into clinical diagnostics.
  • Item
    Thumbnail Image
    Investigation of clustered hypermutation in cutaneous melanoma
    Colebatch, Andrew James ( 2017)
    Cutaneous melanoma is a malignant neoplasm that arises from melanocytes in the skin, and is a significant cause of morbidity and mortality, especially in Australia. Next generation sequencing studies of cutaneous melanoma have confirmed the role of UV mutagenesis in its aetiology and demonstrated the high somatic mutation burden in this disease. To date, the majority of genomic analyses of cutaneous melanoma have limited their evaluation of somatic mutations to protein coding regions in order to detect cancer driver mutations. The recent discovery of driver mutations within the TERT promoter however has shown that biologically significant somatic events occur within non-coding regions of the melanoma genome. This thesis set out to further explore the presence, timing and biology of promoter-based somatic mutations in cutaneous melanoma. In this thesis, a systematic analysis of mutations in cutaneous melanoma demonstrated enrichment of somatic mutations within active promoters. Furthermore, somatic mutations were located within sequence motifs matching binding sites for ETS and Sp1 transcription factors. The number of promoter mutations within a tumour correlated with clinicopathological markers of chronic UV exposure and overall somatic mutation load. Strikingly, the same promoter mutations were detected in other cutaneous malignancies. In order to evaluate the potential of promoter mutations as biomarkers, a highly sensitive assay was developed to detect TERT promoter mutations, the most prevalent and biologically plausible promoter mutation. Using this assay on a series of benign melanocytic naevi identified TERT promoter mutations in 2/17 samples, providing evidence that these driver mutations occur in benign neoplasms with implications for the interpretation of TERT mutations in borderline clinical samples. Microdissection of melanomas with coexistent radial and vertical growth phases demonstrated that a majority of promoter mutations are early events in melanomagenesis. TERT promoter mutations are shown to sometimes occur late in melanoma development and show coexistent subclones. In addition, somatic promoter mutations were detected in non-neoplastic skin adjacent to melanoma. Finally, a somatic mutation load assay was designed and evaluated on clinical melanoma samples. An analysis of publically available melanoma exome data demonstrated a subgroup of highly mutated BRAF/NRAS wildtype cutaneous melanomas. The assay was tested on clinical samples and was shown to be incapable of accurate determination of mutation load in melanoma due to technical limitations. In conclusion, the results of this thesis demonstrate a novel somatic mutational process at work in cutaneous melanoma represented by hypermutation within specific promoters, and evaluates these, along with the previously described TERT promoter mutation, as potential biomarkers. In so doing, these findings contribute to understanding of the biology, aetiology and role of somatic promoter mutations, critical knowledge for their potential future use in clinical assays.
  • Item
    Thumbnail Image
    The genomic landscape of phaeochromocytoma
    Flynn, Aidan ( 2015)
    Phaeochromocytomas (PCC) and paragangliomas (PGL) (collectively PPGL) are rare neural crest-derived tumours originating from adrenal chromaffin cells or extra-adrenal sympathetic and parasympathetic tissues. More than a third of PPGL cases are associated with heritable syndromes involving 18 or more known genes. These genes have been broadly partitioned into two groups based on pseudo-hypoxic and receptor tyrosine kinase (RTK) signalling pathways. Many of these genes can also become somatically mutated, although up to one third of sporadic cases have no known genetic driver. Furthermore, little is known of the genes that co-operate with known driver genes to initiate and drive tumourigenesis. To explore the genomic landscape of PPGL, exome sequencing, high-density SNP-array analysis, and RNA sequencing was applied to 36 PCCs and four PGL tumours. All tumours displayed a low mutation frequency in combination with frequent large segmental copy-number alterations and aneuploidy, with evidence for chromothripsis seen in a single case. Thirty-one of forty (77.5%) cases could be explained by germline or somatic mutations or structural alterations affecting known PPGL genes. Deleterious somatic mutations were also identified in known tumour-suppressor genes associated with genome maintenance and epigenetic modulation (e.g. TP53, STAG2, KMT2D). A multitude of other genes were also found mutated that are likely important for normal neuroendocrine cell function (e.g. ASCL1, NCAM1, GOLGA1). In addition, the existing paradigm for gene-expression subtyping of PPGL was further refined by applying consensus clustering to a compendium of previously published microarray data, enabling the identification of six robust gene-expression subtypes and subsequent cross-platform classification of RNA-seq data. The majority of cases in the cohort with no identifiable driver mutation were classified into a gene-expression subtype bearing similarity to MAX mutant PPGL, suggesting there are yet unknown PPGL cancer genes that can phenocopy MAX mutations. The cross-platform classification model was then further refined to develop a 46-gene Nanostring-based diagnostic tool capable of classifying PPGL tumours into gene-expression subtypes. The strong genotype-to-subtype relationship in PPGL makes subtyping a powerful tool that can be used clinically to guide and interpret genetic testing, determine surveillance programs and aid in better elucidation of PPGL biology. In applying the diagnostic assay to a test set of 38 cases, correct classification into one of the six subtypes was achieved for 34 (90%) samples based on the known genotype to gene-expression subtype association. The observation that at least one of the six subtypes is likely defined by the presence of non-neoplastic cells led to further refinement into five, four, and three-class architectures, further improving classification accuracy. Increasingly tumour heterogeneity is being recognised as one of the most significant challenges facing modern oncology. Genomically diverse tumour regions create additional complexity in predicting treatment response and metastatic potential through biopsy. Multi-region sampling of multiple synchronous primaries from patients with a predisposing germline mutation was used to explore tumour evolution and heterogeneity in PPGL and concomitant medullary thyroid carcinoma. Evolutionary reconstruction of a single primary PPGL demonstrated periods of both branched and linear evolution resulting in a high degree of intratumoural heterogeneity. Comparison of multiple synchronous primaries provided strong evidence of convergent evolution through recurrent chromosomal aberrations, indicating these may be obligate events in tumourigenesis, and as such, may indicate potential novel therapeutic targets.
  • Item
    Thumbnail Image
    Investigating stemness and plasticity models as sources of intra-tumoural heterogeneity in Glioblastoma multiforme
    Brown, Daniel ( 2015-09-09)
    Glioblastoma mutiforme (GBM) is a heterogeneous tumour of the brain with a poor prognosis. Genome-wide profiling has revealed four molecular subtypes, yet there is no significant difference in long-term survival between subtypes. Recurrence and resistance to GBM therapy is believed to be due to an underlying Glioma Stem Cell (GSC) subpopulation. To identify gene expression differences with the ability to predict patient survival, the cancer stem cell subpopulation of Patient Derived Glioma Cells (PDGCs) was isolated based on the expression of the putative stem cell marker CD133. RNA-seq libraries were prepared from six different PDGCs with the identification of 37 differentially expressed genes. Downstream characterisation of the cellular phenotypes exhibited by CD133+ and CD133- cells indicated that CD133 does not enrich for stem and malignant phenotypes. Genes coexpressed with GSCs markers were used to build a gene signature that classified patients based on a CD133 coexpression module signature (CD133-M) or a CD44 coexpression module signature (CD44-M) subtype. CD133-M tumours were enriched for the Proneural GBM subtype and correspondingly CD44-M tumours were enriched for the Mesenchymal subtype. Gene set enrichment identified proliferative pathways as activated in CD133-M and invasion/ migration pathways as enriched in CD44-M. CD133-M patients benefitted most from radiotherapy while CD44-M classified patients responded equally well to temozolomide or radiotherapy. In different PDGCs there was a culture specific equilibrium of distinct PN and MES subpopulations, potentially due to the genetic background of the original patient. The distribution of Proneural and Mesenchymal cells in the same tumour was measured in GBM sections using the expression of Olig2 and CD44 proteins as markers. Heterogeneous expression of these markers in tumours was observed, consistent with the heterogeneity observed in cell cultures. The influence of oxygen tension and chemoradiotherapy on the intra-tumoural equilibrium of PN and MES cells in PDGCs was investigated, with hypoxia inducing a MES to PN shift and chemoradiotherapy inducing a PN to MES shift respectively. The results of this study favour a cellular plasticity model over a hierarchical cancer stem cell model and is in agreement with accumulating evidence that CD133 and CD44 expression are markers of PN and MES molecular subtypes respectively. Both PN and MES subtypes coexist in the same tumours while rare cells that transiently express both CD44 and CD133 (having PN and MES properties) may be cancer stem cells. The results of this thesis suggest the tumour specific proportion of these states is determined by a combination of genetic and environmental factors. Surveillance and modulation of intra-tumoural heterogeneity could be of benefit in the clinical management of GBM.
  • Item
    Thumbnail Image
    Genomic alterations in well- and de-differentiated liposarcoma
    Garsed, Dale William ( 2013)
    Well- and de-differentiated liposarcomas (WD/DDLPS) are tumours that arise from fat cells (adipocytes) and are characterised by the presence of supernumerary ring or ‘giant-rod’ marker chromosomes (neochromosomes; NCs). NCs are associated with several cancer types and are thought to harbour oncogenic mutations in the form of amplified oncogenes or fusion genes. However, knowledge of their structure and oncogenic properties is limited. The aim of this thesis is to comprehensively map genomic alterations in WD/DDLPS NCs, and to identify the mechanisms of pathogenesis in this disease. This report describes the first sequence-level map of cancer-associated NCs. A unique targeted genomics approach was developed, in which flow cytometry was used to isolate giant marker NCs from a panel of WD/DDLPS cell lines, and their content and structure was mapped using single nucleotide polymorphism mapping arrays and next-generation sequencing analyses. The NCs have a complex spatial architecture, with over 900 structural variations, the majority of which were formed by DNA translocations showing evidence of nonhomologous end-joining repair. I developed a novel classification system defining distinct patterns of breakpoints and fusions, which formed the basis for detailed analysis of this complex structure. In-depth characterisation of DNA fusions revealed the molecular history of NC evolution. Different patterns of rearrangement were observed, including evidence of localised shattering and random repair (chromothripsis) on the long arm of chromosome 12, which may be a fundamental mechanism in NC formation. Following initial assembly, DNA amplification occurs by breakage-fusion-bridge cycles in the ring conformation, a phase terminated by linearization and the acquisition of functional telomeres from normal chromosomal counterparts. A genome-wide survey of copy number variations (CNVs) was carried out in primary WD/DDLPS, to define regions most likely to harbour oncogenes and tumour suppressors. The most significant regions of CNV were determined by measuring the frequency of occurrence and amplitude of DNA gain or loss. The genes identified in these candidate regions were then triaged to a list of candidate genes suitable for functional validation, by (i) testing for copy number correlation with matched mRNA expression data, (ii) identifying biologically significant genes based on the literature, and (iii) comparing these findings to a previously published independent data set. A number of novel copy number alterations were identified, including amplification of NUP107, OS9 and CYP27B1, and deletion of DDX6, HRAS, and OTUB1. In order to refine the list of candidate oncogenes, a functional validation screen was performed using RNA interference (RNAi) to determine which amplified genes are necessary for WD/DDLPS cell growth. Short interfering RNA (siRNA)-mediated depletion of several amplified genes was anti-proliferative to WD/DDLPS cells, including the nucleoporin NUP107, recently reported to be over-expressed in some cancers. NUP107 is amplified in 53% of primary WD/DDLPS tumours, and its attenuation caused growth suppression and cell cycle arrest in vitro. In addition to copy number changes driving over-expression, transcriptome sequencing revealed alterations in mRNA splicing, cryptic exon usage, and fusion transcripts, presumably as a result of NC rearrangements. Alternatively spliced transcripts of the oncogene MDM2 are specifically expressed in WD/DDLPS tumours and are expressed at a low level or absent in normal fat tissue and other tumour types. Preliminary data indicates that the expression of altered mRNA may have a role in carcinogenesis. Taken together, the findings in this thesis provide exciting new insights into the structural complexity, evolution and pathogenesis of WD/DDLPS NCs.
  • Item
    Thumbnail Image
    Understanding mammographic density and breast cancer risk: from histology to genomics
    Lin, Suling Joyce ( 2012)
    This thesis describes a novel multi-disciplinary strategy that advances our understanding of mammographic density (MD) biology and genetics and its relationship to breast cancer risk. Studies have indicated that women with high MD are at 4 to 6 fold increased risk of getting breast cancer. MD is also highly heritable. Hence, a greater understanding of MD biology and genetics would improve our knowledge about breast biology and potentially identify new cancer predisposition genes. Nonetheless, MD biology remains controversial and much is unknown about the genetic basis of this trait. This thesis revisits MD histopathology using a novel methodology of sampling high and low MD tissue within the breast of an individual. This method samples tissue guided by its true MD, as determined by real-time mammography imaging - in contrast to previous studies that examined MD histopathology relying on “random” sampling of breast tissue. Studies have suggested that dense regions of the breast are potentially associated with breast cancer risk on the basis that percent MD (PMD) adjusted for body mass index (BMI) and age is a better measure of breast cancer risk. Hence, this method also allows assessment of tissue most pertinent to MD associated breast cancer risk. Furthermore, the use of sampling within an individual allows control of all potential confounders that helps improve the power of analysis. The histopathological analysis revealed that high MD regions were significantly associated with greater composition of dense connective tissue stroma and lesser composition of adipose tissue while no difference in glandular areas between high and low MD tissue was observed. However, it was noted that high MD tissue tended to have smaller and lower complexity glands compared to low MD tissue. This raises the possibility of high MD regions being associated with stem-like cells and their niche compared to low MD regions. Whole-genome expression profiling of accrued high and low MD tissue were interrogated to further our understanding of the genetic and biological bases of MD. Currently, only two other studies have investigated gene expression in MD tissue and these studies have generally failed to account for all potential unwanted variation in the data that could impact on the validity of the analysis outcomes. The present work differs from previous studies in that careful quality assessments and analyses were performed to improve sensitivity and power of the study. The use of precisely sampled tissue also allowed a better representation of the MD expression profile. Both single-gene and gene-set based analyses of adjusted MD expression dataset concurred with the histopathological correlates of high versus low MD tissue. Interestingly, these analyses also showed that high MD regions correlated with a cancer-signature and a CD24 (i.e. luminal epithelial)-signature; while the observed anti-correlation with the CD44-signature was postulated to be of stromal origin. The association of high versus low MD tissue with a cancer-signature agrees with the general direction of MD associated breast cancer risk. Furthermore, immunohistochemical examination of the tissue uncovered a trend of potential “stemness” in high versus low MD tissue that suggests the CD24-signature being of progenitor origin. This supports the hypothesis of high MD tissue being associated with “stemness” (and their niche), as proposed from the initial histopathological examination of MD. To further understand MD biology and its genetic basis, both univariate and pathway-based analyses of genome-wide association (GWA) study data were performed. One major issue with such pathway-based analysis is the assignment of SNPs to their respective gene(s). Increasingly, studies suggest that conventional assignment of SNPs to their nearest gene may be inaccurate. This was the impetus to develop a novel framework using expression profiling analysis of high and low MD tissue to guide mapping of SNPs to their genes. The methodology helped improve the biological relevance and reproducibility of the significant pathways, and also identified the mitogen-activated protein kinase (MAPK) signalling pathway as the most significantly associated with the MD trait. This outcome corresponds to the results obtained from a recent study of breast cancer GWA studies. Taken together, this multi-disciplinary approach has given potential insight into the biological and genetic basis of MD, allowing inference to be made about the increased risk associated with high MD.
  • Item
    Thumbnail Image
    Identification of novel growth promoters of ovarian cancer
    RAMAKRISHNA, MANASA ( 2011)
    Ovarian cancer is a heterogeneous, genomically unstable and highly lethal disease in women. Despite nearly half a century of research in ovarian cancer, few early detection markers, disease drivers or chemoresistance conferring genes have been discovered. The recent and rapid advances in microarray and sequencing technologies have provided new tools with which a thorough search can be performed to discover novel ovarian cancer associated genes. This thesis describes the use of two high resolution microarray platforms to identify novel growth promoters of ovarian cancer. This study investigated a cohort of 72 ovarian tumours using the Affymetrix Genome-Wide Human SNP Array 6.0 (SNP6) to assess copy number, and the Affymetrix GeneChip® Whole Transcript (WT) Sense Target Array 1.0 to study gene expression. The high quality copy number data generated using the SNP6 platform was used to define frequently gained regions of the ovarian cancer genome. Gene expression data was integrated with the copy number data using novel informatics approaches to identify genes whose expression was directly affected by copy number gain. Further integrative analyses and a search of current literature yielded seven candidate drivers of ovarian cancer - PRKCI, RAB2A, PUF60 (SIAHBP1), MYNN, PTK2, PLEC1 and TPX2. These seven candidate ovarian cancer driver genes were tested for functional effects using immortalised ovarian cancer cell lines. Nine cell lines were chosen in such a way that each of the seven candidate genes were represented by at least two cell lines that showed copy number gain and two cell lines that were copy number neutral at the candidate gene locus. Small-interfering RNAs were used to knockdown the expression of candidate genes in the selected ovarian cancer cell lines. Changes in cell proliferation upon gene knockdown were recorded for all cell lines and all genes. TPX2 and RAB2A, showed a significant decrease in proliferation following siRNA knockdown in some cell lines. The changes in proliferation for TPX2 were copy number independent while for RAB2A, changes were strongly driven by copy number. The functional assays validated the power of the analysis methods in identifying novel ovarian cancer drivers.