Search

Understanding the mutational intra-tumour heterogeneity within tumours is crucial to developing effective personalised cancer therapies. Bayesian Markov chain Monte Carlo (MCMC) sampling schemes have proven successful and trusted in reconstructing tumour progression histories, particularly mutation trees. To understand the effectiveness of mutation tree MCMC methods and their required runtimes, it is crucial to understand how quickly the empirical distribution of the MCMC converges to the posterior distribution. We quantify the MCMC exploration of the mutation tree space for the landmark inference scheme SCITE using tree similarity measures. In this simulation study, the tree similarities map features informative of a tumour’s clonal expansion from the mutation tree space to a scalar space, allowing the study of the MCMC exploration. Quantification of the exploration is provided by the novel application of convergence diagnostics established in continuous space to the discrete space of mutation trees via tree similarities. Consequently, we estimate the required runtime of SCITE for simulated data, which may imply significantly reduced runtimes for real-world datasets. Further, we find the dependence of the initial state of the MCMC to vanish quickly. We recommend trialling the significant reduction of the warm-up period for real-world datasets, implying another reduction in required runtime. In the process of exploring initialisation strategies, we validated the performance of the fast heuristic inference method HUNTRESS. Lastly, we investigate the topology of the Bayesian tree posterior, which is thought to contain multi-modalities potentially. For simulated data, we did not find evidence for any multi-modalities justifying the design of SCITE as a single-chain MCMC scheme.

Cancer has been the second leading cause of death in the US[1]. To provide care for cancer patients and retrospectively study this disease, clinicians and researchers need to manually analyze patient-level medical history to determine whether a tumor exists, has the state of existence changed, and does the change implicate disease progression. With the growing adoption of the electronic health records (EHRs), it is now possible to access these data and automate the discourse-level analysis on unstructured clinical texts using natural language processing (NLP) techniques.This thesis focuses on developing, training, and evaluating a transformer-based text classification algorithm that will capture contexts from unstructured radiology reports and output the discourse-level analysis on the tumor status and its progression through three conceptual frames: existence of a tumor, change of existence, and significance of change. This is the first clinical NLP work that conceptualize these representations using a wide range of systemic inferences, including contexts from presuppositions. The model shows promising results and can be extended to improve on casual reasoning, logical reasoning, numerical reasoning, and temporal reasoning in the future.

The 15-30% of human breast cancers that have upregulated HER2/ErbB2/Neu are highly aggressive and resistant to traditional treatments, resulting in poor prognosis. To identify novel therapeutic targets, we derived the transcriptomes associated with tumor progression in two independent mouse models of ErbB2/Neu-induced tumorigenesis. From MMTV-Neu mice, we identified 324 candidate genes unique to ErbB2/Neu-induced tumors relative to wild-type mammary glands. A subset of these genes also changed expression levels in preneoplastic mammary glands, indicating a pivotal role early in ErbB2/Neu-initiated tumorigenesis. Downregulation of several known transforming growth factor (TGF)-β target genes in the ErbB2/Neu molecular signature suggested attenuation of the TGF-β signaling cascade in these tumors. Analysis of TGF-β-Receptor-I/ALK5 by western blot and immunohistochemistry confirmed that Smad-dependent TGF-β signaling was inactive in these tumors. Although absent in most of the tumor, colocalization of phosphorylated Smad2 and Activin-Receptor-IB/ALK4 at the tumor periphery suggested functional Activin signaling at the leading edge of these tumors. Collectively, these data indicate intrinsic TGF-β pathway suppression in ErbB2/Neu tumors via loss of TGF-β-Receptor-I/ALK5. Recent studies have shown that pregnancy can accelerate ErbB2/Neu tumor development, inducing a susceptible cell population in MMTV-Neu mammary glands. The stochastic pattern of tumor development in multiparous MMTV-Neu mice suggests additional events are required for ErbB2/Neu oncogenesis. It remains unclear whether such events are genetic or reflective of the dynamic, pregnancy-associated hormonal control of the gland. Bitransgenic mice generated by breeding MMTV-Neu mice with a model of ovarian hyperstimulation developed multifocal mammary tumors in an accelerated, synchronous manner compared to virgin MMTV-Neu animals. This synchrony of tumor development suggests that trophic maintenance of the mammary gland provides the additional events required for tumor formation and maintains the population of cells targeted by ErbB2/Neu for transformation. The synchronous nature of tumor development in this bitransgenic mouse model permitted characterization of a window of commitment to tumorigenesis and subsequent identification of approximately 2800 genes that demonstrate altered expression with oncogenic commitment. Thirty-two of these genes are also changed in the preneoplastic glands of MMTV-Neu mice, suggesting that this subset of genes may be regulated by ErbB2/Neu and/or contribute to early events of ErbB2/Neu-induced tumorigenesis.

Astrocytomas are the most frequent group of intracranial tumors. Among them, glioblastoma is the most aggressive one. In this review we will describe the most common genetic abnormalities found out in astrocytomas. We will refer to the epidermal growth factor receptor (EGFR), p53, p16, PTEN and DMBT1 genes. We will also present certain genetic aspects that influence the progression to glioblastoma.

Primary and lung metastatic tumor cells were evaluated for the state of the viral-Kirsten-ras oncogene (v-Ki-ras) by DNA and mRNA blot analyses after injection of Kirsten murine sarcoma virus (KiMSV)-transformed BALB/c 3T3 cells into athymic nude mice by four routes. All tumors showed v-Ki-ras gene amplification and increased mRNA expression when compared with the originally-injected KiMSV late-passage cells. Single-cell clonal analyses of these tumor populations showed all KiMSV early-passage clones contained v-Ki-ras at high dosage, but analyses of KiMSV late-passage clones revealed a loss of v-Ki-ras in >95% of this cell population (v-Ki-ras- revertants). However, all clones of all tumors (0.25 to >2.0 cm tumor diameter) contained v-Ki-ras at high dosage levels, demonstrating an early and specific in vivo selection for a v-Ki-ras bearing cell in tumor formation in this system. Flow cytometric DNA content and chromosome analyses confirmed these observations. Furthermore, v-Ki-ras- revertant subclones showed a reduced tumorigenic potential, indicating loss of this gene is concimitant with reduced tumorigenesis. Mixture experiments with v-Ki-ras- revertant cells and pSV2neo genotypically-marked v-Ki-ras containing clones at different ratios demonstrated a clone-specific, early in vivo selection for v-Ki-ras containing cells over v-Ki-ras- revertant subclones in both induction/clonal origin of tumors and their malignant progression. These experiments taken together demonstrate the effectiveness and significance of the ras oncogene for competence in tumor progression in this system. Characterization of the fibronectin-mediated cell attachment and spreading responses of the BALB/c 3T3 cells and its KiMSV-transformed, tumorigenic, metastatic, and revertant derivatives on thermolysin-generated fibronectin fragments led to the isolation of a potential new "cell binding/spreading" activity specific for primary and metastatic tumor cells and to a lesser degree the original KiMSV transformants. BALB/c 3T3 cells and the v-Ki-ras- revertant cells did not respond to these sequences in the same manner. The molecular basis between the v-Ki-ras+ and v-Ki-ras- cells appears to result from the extensive catabolism of cell surface heparan sulfate-proteoglycan following oncogenic transformation. These studies suggest that following v-Ki-ras mediated transformation, cell surface alterations with respect to fibronectin-mediated adhesion occur that correlate with aggressive in vivo tumor progression

Clinically, the superficial spreading form of human cutaneous malignant melanoma (HCMM) progresses from a precursor nevus, to a metastatic incompetent primary radial growth phase (RGP), followed by a metastatic competent primary vertical growth phase (VGP), and finally metastasis. The early stages of this disease have not been readily amenable to current genetic analysis because the biopsies are fixed, paraffin embedded, and sectioned for clinical diagnosis. Therefore, molecular and cytogenetic studies have been primarily confined to metastatic melanoma which revealed frequent alterations involving chromosomes 1, 6, 7, 9, 10, 11, and 17. Localization of candidate loci has been difficult because of the complex genetic heterogeneity in melanoma. Cytogenetic analyses in conjunction with molecular cytogenetic techniques, including fluorescent in situ hybridization (FISH), chromosome microdissection, and comparative genomic hybridization (CGH) were employed to further define candidate loci in uveal and cutaneous melanoma samples. The investigations showed loss of 6q and gain of 6p in both types of melanomas, identified marker chromosomes, and potential candidate loci. However, discerning primary and secondary alterations remained difficult. A new procedure combining tissue microdissection and CGH (MD-CGH) was developed to investigate the DNA sequence copy number changes associated with both early and advanced stages of HCMM. Twenty-three archived paraffin embedded tissue sections containing normal, precursor nevus, primary tumor, and metastatic cells in a single biopsy were analyzed. Loss of 11q13, 12q, and 17p loci were implicated in the development of RGP while loss of 9q and 10 in addition to a previously undetected amplification of 7q32$\to$q34 were associated with the transition from RGP to VGP. Furthermore, linear progression and clonal genetic derivation of clinical HCMM stages were directly demonstrated for the first time. Finally, CGH analysis of metastatic melanoma cell lines revealed copy number changes similar to published cytogenetic reports. In contrast to the alterations identified in melanoma biopsies, alterations involving 1p, 4, 8p, 9p, 11, and 16 loci were significantly associated with cultured cells. These studies named chromosome alterations associated with specific stages of melanoma development providing insights into the temporal pattern of progressive genetic changes involved in the pathogenesis of HCMM.