Use this URL to cite or link to this record in EThOS:
Title: Use of multiple platform 'omics' datasets to define new biomarkers in oral cancer and to determine biological processes underpinning heterogeneity of the disease
Author: Saeed, Anas Amjad Mohammad
Awarding Body: University of Edinburgh
Current Institution: University of Edinburgh
Date of Award: 2013
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Oral cancer in early stages (I and II) may be curable by surgery or radiation therapy alone but advanced stage disease (III and IV) has a relatively low survival rate. The pathogenic pathways that contribute to Oral Squamous Cell Carcinoma (OSCC) remain poorly characterised and the critical factor in the lack of prognostic improvement is that a significant proportion of cancers initially are asymptomatic lesions and are not diagnosed or treated until they reach an advanced stage. Hence, a clinically applicable gene expression signature is in high demand and improved characterization of the OSCC gene expression profile would constitute substantial progress. For OSCC, possible themes that might be addressed using microarray data include distinguishing the disease from normal at the molecular level; determining whether specific biomarkers or profiles are predictive for tumour behaviour; and identifying biologic pathways necessarily altered in tumourigenesis, potentially illuminating novel therapeutic targets. However, OSCC is a heterogeneous disease, making diagnostic biomarker development difficult. Although this phenotypic variation is striking when one compares OSCC from different geographic locales, little is known about the underpinning biological mechanisms. Cancer may be accompanied by the production and release of a substantial number of proteins, metabolites and/or hormones into the blood, saliva, and other body fluids that could also serve as useful markers for assessing prognosis, metastasis, monitoring treatment, and detecting malignant disease at an early stage. The primary aim of this thesis is to investigate metabolomic and transcriptomic profiles using multiple bioinformatics approaches and biological annotation tools in an attempt to identify specific biomarkers and prediction models for OSCC from each profile as well as from the interface outcomes of integrating the two platforms. Additional aims of the thesis go further to identify the mechanisms underlying the biological changes during tumorigenic transformation of OSCC, as well as to determine biological processes underpinning the heterogeneity of the disease among populations. Two review studies were carried out in this thesis. The review study of published transcriptomic profiles of OSCC specified several genes and pathways exhibiting substantially altered expression in cancerous versus noncancerous states across studies. However, the result of the review suggests not relying on the final set of genes published by the individual studies, but to access the raw data of each study and start subsequent analysis from that stage using unified bioinformatics approaches to acquire useful and complete understanding of the systems biology. The other review study focused on the metabolic profiles of OSCC and revealed a systemic metabolic response to cancer, which bears great potential for biomarker development and diagnosis of oral cancer. However, the metabolic signature still needs to improve specificity for OSCC from other types of cancer. In an attempt to detect a robust gene signature of OSCC overcoming the limitation of the transcriptomic review in accessing the raw data from the previous works, four public microarray raw datasets (comprising 365 tumour and normal samples) of OSCC were successfully integrated using ComBat data integration method in R software, determining the common set of genes, biomarkers, and the relative regulatory pathways possibly accountable for tumour transformation and growth in OSCC. Examination of the meta-analysis datasets showed several discriminating gene expression signatures for OSCC relative to normal oral mucosa; with a signature of 8 genes (MMP1, LAMC2, PTHLH, TPBG, GPD1L, MAL, TMPRSS11B, and SLC27A6) exhibiting the best discriminating performance and show potential as a diagnostic biomarker set. In addition, 32 biomarkers specific to OSCC and HNSCC were identified with the majority involved in extracellular matrix (ECM), interleukins, and peptidase activity where around 2/3 of them are located in the extracellular space and plasma membrane. Additionally, investigation of the interactive network created by merging metabolic and transcriptomic profiles highlighted the significant molecular and cellular biofunctions, pathways, and biomarkers distinguishing OSCC from normal oral mucosa. The results highlighted interactions of significantly altered expression of Dglucose, ethanol, glutathione, GABA, taurine, choline, creatinine, and pyruvate metabolites with the expressed PTGS2, IL1B, IL8, IL6, MMP1, MMP3, MMP9, SERPINE1, COL1A1, COL4A1, LAMC2, POSTN, ADAM12, CDKN2A, PDPN, TGM3, SPINK5, TIMP4, KRT19, and CRYAB biomarkers of OSCC. Such a pattern may represent a clinically useful surrogate for the presence of OSCC which might help in deciphering some of the obscure multifaceted mechanisms underlying carcinogenesis of OSCC which emerged from dysregulated genetic and metabolic system of the body. In an attempt to define pathways of importance in two phenotypically different forms of OSCC, transcriptomic analysis of OSCC from UK and Sri Lankan patients was undertaken. The development of OSCCs in UK and Sri Lankan populations appears largely mediated by similar biological pathways despite the differences related to race, ethnicity, lifestyle, and/or exposure to environmental carcinogens. However, results revealed a highly activated “Cell-mediated Immune Response” in Sri Lankan tumour and normal samples relative to UK cohorts which may reflects a role in resistance of patients to invasiveness, metastasis, and mortality observed in Sri Lankan relative to UK patients. In conclusion, multiple molecular profiles were able to identify a unique transcriptomic profile for OSCC and could further discriminate the tumour from normal oral mucosa on the basis of 8 genes. Altered expression of several metabolic and transcriptomic biomarkers specific for OSCC were identified, along with several dysregulated pathways and molecular processes found common in patient with oral cancer. Integrating both metabolomic and transcriptomic signatures revealed a promising strategy in analysing the concurrent perturbation in both genetic and metabolic systems of the body. Additional results revealed possible impact of specific supplementary dietary components in boosting the immune system of the body against invasion, progression, and metastasis of the disease. Further clinical studies are required to confirm and validate the current results.
Supervisor: Lopes, Victor Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: OSCC ; transcriptomics ; metabolomics