Molecular characterization of High Risk Neuroblastoma. Potential biomarkers for high-risk neuroblastoma classification

[eng] BACKGROUND: High-risk neuroblastoma (NB) represents a heterogeneous group of tumors, whereby patients can display response to treatment and long-term outcome or develop early progressive, chemoresistant disease with poor outcome. To date, high-risk NB patients are generally treated uniformly w...

ver descrição completa

Detalhes bibliográficos
Autor: Garrido Garcia, Alícia
Formato: tesis doctoral
Estado:Versión publicada
Fecha de publicación:2022
País:España
Recursos:Universidad de Barcelona
Repositorio:Dipòsit Digital de la UB
OAI Identifier:oai:diposit.ub.edu:2445/192108
Acesso em linha:https://hdl.handle.net/2445/192108
http://hdl.handle.net/10803/687394
Access Level:acceso abierto
Palavra-chave:Oncologia pediàtrica
Epigenètica
Metilació
ADN
Aprenentatge automàtic
Tumors in children
Epigenetics
Methylation
DNA
Machine learning
Descrição
Resumo:[eng] BACKGROUND: High-risk neuroblastoma (NB) represents a heterogeneous group of tumors, whereby patients can display response to treatment and long-term outcome or develop early progressive, chemoresistant disease with poor outcome. To date, high-risk NB patients are generally treated uniformly with no further stratification, as established in routinely used risk stratification systems. A revised molecular risk stratification has been proposed based on the analysis of telomere maintenance mechanisms, and RAS or TP53 pathway mutations. However, genetics underlying this aggressive subgroup is still greatly unknown and risk stratification of high-risk NB tumors is still challenging. AIM: To study high-risk NB tumors and to define a subgroup of high-risk NB patients with particularly poor outcome, with the aim of improving risk­ stratification of high-risk NB, of identifying altered biological pathways that may represent therapeutic options, and to learn more about the biology underlying this malignant pediatric tumor. METHODS: We analyzed DNA methylation microarray and gene expression data from nearly 700 high-risk NB samples obtained at diagnosis. Cox-regression models and Machine-Learning analysis were used for survival analyses. Survival curves were estimated by Kaplan-Meier method and compared by log-rank test. Pathway analysis was performed using R package KEGGREST, ConsensusPathDB-MaxPlanck and R package topGO. Pyrosequencing, phospho-kinase array, immunoblotting and immunohistochemical techniques were used for validation purposes. RESULTS: We identified distinct DNA methylation profiles within high-risk NB. Cox­ regression models and Machine Learning analysis, identified differentially methylated CpG sites that defined two subgroups of patients with substantially different overall survival (OS). Moreover, we identified methylation markers that could distinguish these clinically relevant subgroups of tumors. Integrative analysis of DNA methylation and matching gene expression data, identified differential expression of genes involved in cellular metabolism, purine biosynthesis and AKT/mTOR cell signaling. Protein expression analysis identified high levels of proteins involved in IMP metabolism and increased activation of AKT/mTOR pathways in highly aggressive NB. CONCLUSION: We have identified (epi)genetic changes underlying the heterogeneous behavior of aggressive NB, and revealed altered pathways of interest for potential therapeutic options. We identified a set of markers that enabled classification of high-risk NB into clinically relevant subgroups.