dc.description.abstract
Human cancer arises as a result of genomic alterations that transform cells and make them to grow without control and to pathological levels. The characterization of such genomic changes has enabled understanding tumor development and identifying clinical biomarkers for prognosis and therapy. Many of the genomic and epigenomic alterations in cancer can also be observed through the analysis of the transcriptome, which gives a more functional approach. Next-generation sequencing technologies, such as whole- genome sequencing (WGS) and RNA-seq, have provided the opportunity to assess molecular characterization of distinct tumors, leading to the discovery of several molecular aberrations linked to the biology of tumors. These molecular alterations include cancer-driving mutations, complex chromosomal rearrangements, atypical transcriptional profiles, gene fusions, among others.
Despite the efforts to generate a comprehensive molecular atlas of tumor biology, there are still many unknown aspects, mainly due to technical and methodological limitations. In particular, regarding the genomic characterization of cancer genomes, structural variation is known to play a crucial role in carcinogenesis as a major component of cancer genome architecture. However, the full spectrum of complex rearrangements is still largely unknown due to the lack of straightforward frameworks that allow their identification, classification, and comprehensive characterization.
On the other hand, regarding the transcriptomic characterization in tumor progression, changes in gene expression have been found to be related to metastasis. Nevertheless, the entire landscape of mRNA alterations has not been explored, opening a gap in the discovery of new transcripts, such as fusion transcripts, that might be associated with metastasis and evolve into predictive markers.
In this thesis, we address these two limitations through two studies to better characterize structural genomic and transcriptomic alterations that contribute to tumor development.
With the particular aim to classify and isolate complex somatic rearrangements in the cancer genome, with a potential molecular mechanism behind, we first investigated the landscape of chromosomal rearrangement from 2,586 tumor genomes across 40 cancer types (from the ICGC-TCGA Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium). We developed a novel framework for clustering, classification, and characterization of SV patterns with distribution that match with potential specific molecular mechanisms behind, rather than with random distribution. As a result, we identified complex rearrangements likely triggered by single catastrophic events, such as Chromoplexy or Chromothripsis. Among these, we identified a new pattern that we named Chromotrikona, which involves reciprocal translocations between different chromosomes. These findings contribute to the understanding of the genomic structural processes leading complex genomic reorganizations with impact in cancer.
To search for RNA alterations associated with metastasis, we further conducted a study of RNA alterations in metastatic breast cancer, specifically in differential gene expression and fusion transcripts. We collaborated with Dr. De Mattos from IrsiCaixa, analyzing RNA-seq datasets of 82 metastatic breast samples obtained from 10 different patients, to search for evidence that could determine the metastasis in various tissues. Running in parallel, each one of these lines has provided interesting results. In particular, in the context of fusion transcripts, we have generated and applied a comprehensive strategy to identify and characterize transcript fusion events. The study has revealed a new pattern of massive multi-fusions of different transcripts in most of the metastases analyzed. This pattern has been experimentally validated and characterized. Surprisingly, it has been found beyond metastatic samples, being also present in normal breast cells. Although it is not clear the origin and the real nature of this event, we hypothesize that it might occurs at RNA level and be tissue specific. Further analyses will be needed to provide a deeper insight of the impact of this event in human cellular phenotypes.
en_US
