231 research outputs found
Hypertranscription in Human Cancer
Get PDFCancer is a disease driven by aberrant gene expression leading to uncontrolled cellular proliferation. To date, tens-of-thousands of patient tumors have undergone RNA-sequencing, creating a catalog of the genes and pathways differentially expressed in cancer tissues. Despite these significant advances, a fundamental aspect of gene regulation remains poorly characterized: the overall expression level across all genes. Recent work has demonstrated that certain oncogenes, such as MYC, might drive tumor growth by globally increasing transcription of all active genes, a phenomenon known as hypertranscription. While hypertranscription has been studied in model systems and cell lines, where drugs that dampen global transcription have shown promise against aggressive ‘transcriptionally addicted’ cancers, hypertranscription has never been characterized in human patients. Thus, we do not know hypertranscription’s prevalence across cancer types, its drivers, or its impact on patient outcomes. This gap in knowledge is driven in large part by the absence of appropriate methods to accurately measure global transcription. Nearly all reported gene expression estimates incorrectly assume relatively equal RNA output across samples.
In Chapter 2, a novel computational method is developed that allows joint measurement of global and focal gene expression changes in patient tumor RNA-sequencing data. Critically, this method accounts for differences in tumor purity and ploidy, providing a direct fold-change measure in overall cancer-cell transcription. In Chapter 3, this method is applied to 7,494 tumor samples spanning 31 types, revealing that hypertranscription is a hallmark of aggressive cancers, with over 40% of all cancers harboring hypertranscription levels of at least 2-fold. Investigation of single-cell RNA-sequencing data revealed hypertranscriptional clones that dominated transcript production regardless of their size. Exploration of transcription factors revealed that loss of transcriptional suppression may be fundamental to the hypertranscriptional phenotype. In Chapter 4, the clinical implications of hypertranscription are explored. Hypertranscription defined patient subgroups with worse survival across multiple cancers, even within well-established subtypes. Finally, patients with hypertranscribed mutations have improved response to immune checkpoint therapy. Taken together, this work provides fundamental insights into gene dysregulation across human cancers and may prove useful in identifying patients that would benefit from novel therapies.Ph.D
Mutation Evolution and Genomic Patterns of Recurrence in Ewing Sarcoma and Leiomyosarcoma
No full textSarcomas are a highly heterogeneous group of tumours of the bone and soft tissue whose mutation evolution has been unexplored in the majority of subtypes. From a molecular perspective, sarcomas are sub-classified in two categories: (1) those with disease-defining genetic alterations (commonly translocation-associated gene fusions) and overall quiet genomes and (2) those with no disease-defining alterations and numerous other genetic changes. The work in this thesis centered on a representative sarcoma from each category. First, the genomes of Ewing sarcoma (ES), the second most frequent bone cancer of childhood, were explored. ES represents a prototypical fusion-driven sarcoma as it is characterized and driven by the EWSR1-ETS fusion. Secondly, leiomyosarcoma (LMS), an adult soft-tissue cancer, was examined. LMS is a malignant neoplasm that affects smooth muscle tissue and has a high risk of metastatic relapse. The aim of this research was to advance our knowledge of the basic biology of these two sarcoma subtypes, including the initiating events of sarcomagenesis, the subsequent order of mutations, and the ongoing mutagenic processes in primary, relapse and metastatic ES and LMS.
Chapter 1 provides a comprehensive background on the current knowledge of cancer genomics and an introduction to general sarcoma biology, followed by an overview of the clinical and genomic intricacies of ES and LMS. Chapter 2 provides a synthesis of the informatics tools and approaches developed to detect and characterize mutations in the cancer genome. Chapter 3 describes the patterns of mutations in ES tumours. This chapter reports that in several sarcomas, canonical fusions frequently emerge from rearrangement bursts, also called ‘chromoplexy’, creating complex genomic loops and disrupting additional genes. The transcriptional heterogeneity and cellular lineages of LMS molecular subtypes are presented in Chapter 4. Additionally, the genomic mutation signatures and the mutation dynamics contributing to relapse and metastatic spread are described. Lastly, Chapter 5 examines future directions for ES and LMS genomics research. Overall, this thesis highlights recent advances in ES and LMS genomics and provides the molecular framework for future work in patient stratification and early cancer detection in these subtypes.Ph.D.2022-06-22 00:00:0
Studies on the Evolution of and Mutational Processes Driving Childhood Cancer in the Context of Genetic Predisposition
Get PDFCancer predisposition syndromes (CPSs) are caused by heritable mutations, often affecting DNA-repair pathways, which dramatically increase cancer risk. Once diagnosed, screening of additional family members combined with cancer surveillance protocols have shown significant survival benefits. However, CPSs are largely underdiagnosed due to clinical heterogeneity and variants of uncertain significance. In this thesis I explore the hypothesis that CPS-associated cancers exhibit characteristic DNA-repair-associated mutational signatures—patterns of somatic mutation related to mutation aetiology¬— and/or evolutionary dynamics, which may be exploited to aid in CPS diagnosis and management. To address this hypothesis, I studied two model CPSs—constitutional mismatch repair deficiency (CMMRD) and Li-Fraumeni syndrome (LFS).
CMMRD results from biallelic germline mutations in one of four mismatch repair genes, and is associated with childhood brain, colorectal and lymphocytic neoplasms. Recent work has shown cancers developing in CMMRD patients to possess recurrent somatic mutations in POLE/POLD1 and massively elevated numbers of somatic point mutations (hypermutation). To assess the frequency, timing, and aetiology of hypermutation in adult and childhood cancer, I performed a comprehensive analysis of hypermutation across >80,000 human cancers. Our work identified CMMRD in 15 patients, resulting in their enrollment on a surveillance protocol and immune checkpoint inhibitor trial, which has shown sustained responses for CMMRD patients.
Li-Fraumeni syndrome (LFS) is a cancer predisposition syndrome caused by germline mutations in the TP53 tumor suppressor gene, and is associated with a wide range of cancers, including sarcomas, breast cancers, adrenocortical carcinomas and brain tumours. To investigate somatic mutational events driving tumourigenesis in LFS, I performed whole-genome sequencing (WGS) analysis of bulk and multi-region dissections of tumours derived from childhood and young adult patients with germline TP53 mutations. Our analyses revealed that the life history of LFS cancers is marked by early loss of heterozygosity of TP53, mutational signatures related to homologous recombination repair deficiency and in some cases previous chemotherapeutic treatment. In summary, my thesis research demonstrated CPS-related malignancies are often mutationally and evolutionarily distinct entities. The unique molecular features of these tumours reveal aspects of their aetiology and in some cases can be used to aid in diagnosing and treating these patients.Ph.D.2022-11-30 00:00:0
Abstract LB-223: Whole genome sequencing of rhabdomyosarcoma germline cohort identifies low frequency of pathogenic mutations
No full textAbstract
Introduction: Rhabdomyosarcoma (RMS) is among the most frequently occurring tumors in Li-Fraumeni syndrome patients and is often observed in patients with a strong family history of cancer. Recent studies suggest that at least 10% of children with cancer harbor an underlying pathogenic germline mutation, although the frequency in children with RMS is not well characterized. In this study we sought to determine the rate of likely pathogenic germline mutations in rhabdomyosarcoma patients, unselected for family history.
Methods: Of 275 RMS patients enrolled on one clinical trial (COG ARST0531), whole-genome sequencing was performed on blood-derived DNA of an initial set of 50 using the Illumina HiseqX to an average sequencing depth of 44.9X [34.2-58.4]. Single nucleotide variants (SNVs) and Indels were called and filtered to select only those occurring in exons or splicing regions. Variants were further filtered to include only those with a frequency of <0.5% in the ExAC database and lying in one of 99 genes identified as a cancer predisposition gene in the COSMIC Cancer Gene census. Nonsynonymous SNVs were required to be predicted as deleterious by one or more of Polyphen, mutation assessor or sift.
Results: Across the 50 initial samples, a total of 290 variants ([0-17], median of 5) passed the described filters. Notably, no point mutations were identified in TP53 in the initial 50 samples. Two samples contained variants annotated as pathogenic in ClinVar - a frameshift deletion in CHEK2 and a rare SNP in MUTYH. Also notable was a potentially pathogenic frameshift deletion in MSH6 that lacked previous annotation.
Conclusions: Our preliminary analysis suggests a lower incidence of causative TP53 mutations in RMS than previously suggested. Whole genome sequencing of the remaining samples in the cohort along with analysis of copy number and structural variants is ongoing. Statistical analysis of the entire cohort of 275 patient samples will allow for the most complete characterization of the germline genomic landscape of rhabdomyosarcoma to date. This information will better inform clinical surveillance and management parameters for RMS patients and families.
Citation Format: Nicholas Light, Philip Lupo, Javed Khan, Joshua Schiffman, Douglas Hawkins, Adam Shlien, David Malkin. Whole genome sequencing of rhabdomyosarcoma germline cohort identifies low frequency of pathogenic mutations [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr LB-223. doi:10.1158/1538-7445.AM2017-LB-223</jats:p
Genomic DNA Copy Number Variations and Cancer: Studies of Li-Fraumeni Syndrome and its Variants
No full textCopy number variations (CNVs) are a major source of inter-individual genetic difference, accounting for a greater proportion of the human genome than other forms of variation. Recently, the identification of benign and pathogenic CNVs has improved due to arrays with increased coverage. Nevertheless, most CNVs have not been studied with great precision and questions persist regarding their exact breakpoint, gene content, frequency and functional impact. This is especially true in cancer, in which a role for CNVs as risk factors is under-explored.
Li-Fraumeni syndrome (LFS) is a dominantly inherited disorder with an increased risk of early-onset breast cancer, sarcomas, brain tumors and other neoplasms in individuals harboring germline TP53 mutations. Known genetic determinants of LFS do not fully explain its clinical phenotype. In this thesis we describe the association between CNVs and LFS. First, by examining DNA from a healthy population and an LFS cohort using oligonucleotide arrays, we show that the number of CNVs per genome is well conserved in the healthy population, but remarkably enriched in these cancer-prone individuals. We found a significant increase in CNVs among carriers of germline TP53 mutations with a familial cancer history. Second, we find that
ii
specific CNVs at 17p13.1 are associated with LFS or developmental delay, depending on the exact breakpoint with respect to TP53. Using a purpose built array with 93.75% accuracy, we fine-mapped these microdeletions and find that they arise by Alu-mediated non-allelic homologous recombination, and contain common genes, whose under-expression distinguishes the two phenotypes. Third, we explore somatic CNVs in choroid plexus carcinoma tumor genomes. We show that this tumor is over-represented in LFS, and the number of somatic CNVs is associated with TP53 mutations and disease progression. These studies represent the first genomic analyses of LFS, and suggest a more generalized association between CNVs and cancer.Ph
Genomic DNA Copy Number Variations and Cancer: Studies of Li-Fraumeni Syndrome and its Variants
Get PDFCopy number variations (CNVs) are a major source of inter-individual genetic difference, accounting for a greater proportion of the human genome than other forms of variation. Recently, the identification of benign and pathogenic CNVs has improved due to arrays with increased coverage. Nevertheless, most CNVs have not been studied with great precision and questions persist regarding their exact breakpoint, gene content, frequency and functional impact. This is especially true in cancer, in which a role for CNVs as risk factors is under-explored.
Li-Fraumeni syndrome (LFS) is a dominantly inherited disorder with an increased risk of early-onset breast cancer, sarcomas, brain tumors and other neoplasms in individuals harboring germline TP53 mutations. Known genetic determinants of LFS do not fully explain its clinical phenotype. In this thesis we describe the association between CNVs and LFS. First, by examining DNA from a healthy population and an LFS cohort using oligonucleotide arrays, we show that the number of CNVs per genome is well conserved in the healthy population, but remarkably enriched in these cancer-prone individuals. We found a significant increase in CNVs among carriers of germline TP53 mutations with a familial cancer history. Second, we find that
ii
specific CNVs at 17p13.1 are associated with LFS or developmental delay, depending on the exact breakpoint with respect to TP53. Using a purpose built array with 93.75% accuracy, we fine-mapped these microdeletions and find that they arise by Alu-mediated non-allelic homologous recombination, and contain common genes, whose under-expression distinguishes the two phenotypes. Third, we explore somatic CNVs in choroid plexus carcinoma tumor genomes. We show that this tumor is over-represented in LFS, and the number of somatic CNVs is associated with TP53 mutations and disease progression. These studies represent the first genomic analyses of LFS, and suggest a more generalized association between CNVs and cancer.Ph
FractalAnalyzer: A MATLAB Application for Multifractal Seismicity Analysis
No full textEarthquakes are seismic phenomena caused by the sudden release of energy in the Earth’s crust. Their effects range from ground shaking to faulting. Geological and geophysical studies, especially in light of plate tectonic theory have been used to explain the occurrence of earthquakes. Thus from the point of view of statistical fractals, earthquakes cannot be interpreted as random independent events (i.e., having Poisson distribution). Rather, it is observed that the events of the same sequence are clustered in time and space (Shlien and Toksoz, 1970; Vere?Jones, 1970; Smalley et al., 1987; De Natale et al., 1988; Roy and Mondal, 2012a,b).Precision and Microsystems EngineeringMechanical, Maritime and Materials Engineerin
Abstract 3409: Age of cancer onset differentiated by sex and TP53 codon change in Li-Fraumeni Syndrome patient population
No full textAbstract
Introduction:Li-Fraumeni Syndrome (LFS) is a highly penetrant autosomal dominantly inherited cancer predisposition disorder. Germline mutations of the TP53 tumor suppressor gene cause &gt;80% of LFS and confer an increased risk of a range of early onset cancers, as well as of second tumors even in the absence of a family history of cancer. For this reason, we previously reported the implementation of a comprehensive life-long clinical surveillance protocol for individuals with a germline TP53 mutation for early tumor detection. Here, we set out to build a predictive model of age of onset of cancer in LFS patients to inform this screening protocol aiming to make it more targeted. We identify characteristics that differentiate the age of cancer onset consistently, across multiple LFS patient cohorts.
Methods:The LFS cohort at Toronto’s Hospital for Sick Children (SickKids) (n = 171 patients) was used as a discovery set to identify factors that distinguish age of onset among LFS patients. This project focused specifically on patient characteristics such as sex and mutations within TP53, both as they appear on the genome and manifest in the protein, as predictors for age of onset. These predictors were tested in an exponential parametric survival model. Findings from the SickKids discovery set were tested for replication in the International Agency for Research on Cancer (IARC) TP53 database (n = 2374 patients).
Results:In the discovery cohort, female sex was associated with a 1.53 fold later age of cancer onset than in males (p = 0.019). This did not replicate in the IARC TP53 set with 0.99 fold earlier onset for females than males (p = 0.843). However, in the discovery set, there appears to be a point at which female and male age of onset converges at 43 years. Controlling for onset before vs after 43 years in our replication set shows 1.12 (p= 0.0204) times later age of cancer onset in females than in males which is the same direction and significance as in our discovery set. The discovery cohort also showed 2.23 (p = 0.08) later cancer onset for individuals with a germline Arginine to Cysteine (Arg&gt;Cys) codon change (model significance p = 0.047). This finding replicated in the IARC TP53 data set which showed individuals with an Arg&gt;Cys codon change having onset 1.29 (p = 0.043) times later than those with a TP53 mutation that did not result in this codon change.
Conclusions:Our study identified two LFS patient characteristics, sex and TP53 Arg&gt;Cys codon change, which consistently differentiate age of cancer onset within the LFS patient population. Females under the age of 43 when compared to males under the age of 43 appear to have later tumor onset, an effect which disappears after the age of 43. Individuals with an Arg&gt;Cys TP53 codon change are expected to have later onset cancer than those with TP53 mutations that do not result in this change. Future work will disentangle these findings further and build a more comprehensive predictive model of cancer onset in LFS patients.
Citation Format: Lauren Erdman, Ben Brew, Jason Berman, Adam Shlien, Andrea Doria, David Malkin, Anna Goldenberg. Age of cancer onset differentiated by sex and TP53 codon change in Li-Fraumeni Syndrome patient population [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3409. doi:10.1158/1538-7445.AM2017-3409</jats:p
Abstract 973: Methylation accurately predicts age of cancer onset in patients with Li Fraumeni Syndrome
No full textAbstract
Introduction: Li Fraumeni Syndrome (LFS) is a rare hereditary genetic cancer predisposition syndrome. Germline mutations of the TP53 tumor suppressor gene are the underlying cause in &gt;80% of patients with LFS, and are associated with an increased risk of second tumors and a spectrum of early onset cancers, even in the absence of a family history of cancer. We have previously developed and implemented a comprehensive life-long clinical surveillance protocol for individuals with a germline TP53 mutation. We set out to make this screening process more targeted by building a predictive model of age of onset. We accomplished this goal by implementing machine learning methods on germline methylation data.
Methods: We made use of the Toronto Hospital for Sick Children (SickKids) LFS family cohort in our predictive model of age of onset. In all, we have 74 patients with germline methylation data, consisting of ~450,000 probe sites. We subset this data by identifying probes that fall into differentially methylated regions between LFS and cancer patients with wild-type TP53. The probes identified in these regions were used in our predictive model of age of onset. Because age of sample collection was highly correlated with age of onset (r2 ~ .90), we corrected for confounding using a strategy that is two-fold: (1) we extracted the variation of each probe that is independent of the age of sample collection (the residual after regressing on the age of sample collection) and use these as predictors in our model, and (2) we test our models on the task of predicting the age of sample collection for LFS patients that do not have cancer. The former provided us with more robust predictions while the latter verified that we are in fact predicting age of onset, rather than simply predicting age at which the sample was collected.
Results: Our machine learning model was able to achieve 86% correlation between true and predicted values of the age of onset. Additionally, we have tested the ability of our models to predict whether an individual will be diagnosed before or after the age of 4. Our classification machine learning model achieved 91% accuracy on average. We verified that our model does not simply predict age of sample collection by using our cohort of LFS patients that do not have cancer (n = 37). The distribution of the age of sample collection matched those of the patients used in our model. The model has no predictive power on the age of sample collection, thus confirming that our model is highly predictive of the age of cancer onset in LFS TP53 Mutation patients.
Conclusions: We identified two predictive models for age of cancer onset in LFS patients that achieve high accuracy, both when predicting the age of onset as a continuous variable (86% correlation) and whether cancer onset will occur before or after the age of 4 (91% accuracy). Our model will assist clinicians in targeting high risk patients for screening, lower the cost of treatment, and raise the likelihood of survival among LFS patients.
Citation Format: Benjamin M. Brew, David Malkin, Lauren Erdman, Andrea Doria, Jason Berman, Adam Shlien, Tanya Guha, Ana Novokmet, Anna Goldenberg. Methylation accurately predicts age of cancer onset in patients with Li Fraumeni Syndrome [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 973. doi:10.1158/1538-7445.AM2017-973</jats:p
- …
