DEFENSA DE TESIS: Irene Pérez Díez

Defensa de Tesis: «Development and application of novel computational approaches for the
characterization of cancer subtypes».
María de la Iglesia-Vayá, Francisco García García.

Abstract

Cancer remains a global health crisis, demanding further research to understand its molecular basis. Even within the same cancer type, inter-patient variability is an obstacle to disease understanding and therapy development. Cancer subtyping, therefore, becomes essential to address cancer heterogeneity. One promising approach to unravel this heterogeneity is through cancer subtyping based on the transcriptomic landscape of patients. Through gene expression profiling, researchers can identify groups of patients that may represent distinct cancer subtypes with unique biological characteristics, response to therapies and clinical outcomes. However, this approach has its major challenge in requiring large sample sizes, which are crucial for the identification of meaningful subtypes.
This is where transcriptomics data meta-analysis emerges as a powerful statistical tool. By systematically collecting and re-analyzing data from public repositories, researchers can integrate sample sizes across multiple studies, overcoming the limitations of individual datasets. This approach allows for the identification of subtle yet critical differences in gene expression patterns that might be missed in smaller cohorts. In this thesis, we explored cancer heterogeneity in two distinct contexts: lung adenocarcinoma and pancreatic ductal adenocarcinoma. We used an in-silico approach, leveraging published data through meta-analysis, overcoming limitations
of individual studies, and driving novel discoveries. We have identified sex-specific transcriptomic differences in lung adenocarcinoma, particularly in the immune system, purinergic signaling, and lipid metabolism pathways. We have also characterized the transcriptomic landscape of pancreatic ductal adenocarcinoma and its links to patient survival, revealing two prognostic gene signatures associated with the immune system and the extracellular matrix.