UBC Theses and Dissertations
Development and application of an integrative genomics approach to lung cancer Chari, Rajagopal
Lung cancer has the highest mortality rate amongst all diagnosed malignancies with adenocarcinoma (AC) being the most commonly diagnosed subtype of this disease in North America. The dismal survival statistics of lung cancer patients are largely due to the detection of the disease at an advanced stage and to a lesser extent, the limited efficacy of current front line treatments. Genomic approaches, namely gene expression analysis, have provided tremendous insight into lung cancer. While many gene expression changes have been identified, most changes are likely reactive to changes which have a primary role in cancer development. Moreover, one feature which can discern primary from reactive changes is the presence of concordant DNA level alteration. Many well known genes involved in cancer such as TP53 and CDKN2A have been shown to be affected by multiple mechanisms of alteration such as somatic mutation in or loss of DNA sequence. For a given gene, one tumor may be affected by one mechanism while another tumor may be affected by a different mechanism. Although this level of multi-dimensional analysis has been performed for specific genes, such analysis has not been done at the genome-wide level. This thesis highlights the development and application of a multi-dimensional genetic and epigenetic approach to identify frequently aberrant genes and pathways in lung AC. I present, first, the design and implementation of the system for integrative genomic multi-dimensional analysis of cancer genomes, epigenomes and transcriptomes (SIGMA²). Next, analyzing a multi-dimensional dataset generated from ten lung AC specimens with non-malignant controls, I identified novel genes and pathways that would have been missed if a non-integrative approach were used. Finally, examining genes involved with EGFR signaling, I identified a gene, signal receptor protein alpha (SIRPA), which had not been previously shown to be associated with lung cancer. Taken together, these findings demonstrate the power of a multi-dimensional approach to identify important genes and pathways in lung cancer. Moreover, identifying key genes using a multi-dimensional approach on a small sample set suggests the need of large datasets may be circumvented by using a more comprehensive approach on a smaller set of samples.
Item Citations and Data
Attribution-NoDerivs 3.0 Unported