Virtual Library

Start Your Search

M. Mosslemi



Author of

  • +

    P1.02 - Poster Session 1 - Novel Cancer Genes and Pathways (ID 144)

    • Event: WCLC 2013
    • Type: Poster Session
    • Track: Biology
    • Presentations: 1
    • +

      P1.02-004 - Differential pathway disruption in lung adenocarcinomas from current and never smokers - A multi-omics data integration analysis (ID 1072)

      09:30 - 16:30  |  Author(s): M. Mosslemi

      • Abstract

      Background
      Lung cancers in smokers and never smokers (NS) are distinct clinical diseases. Specific molecular differences identified in these two groups include: EGFR and KRAS mutation, DNA methylation levels at specific loci, and most recently, global mutation spectra. However, much remains to be understood about the biology driving lung tumourigenesis in smokers and NS in order to improve treatment outcome. To date, no multi-dimensional integrative genomics (i.e. multi-omics) analysis designed to specifically compare current (CS) and NS lung tumours has been performed. We hypothesize that a multi-omics analysis which considers each tumour as its own unique perturbed system (as opposed to a grouped approach) will reveal molecular mechanisms of lung adenocarcinoma (AC) biology that are common or different in CS and NS.

      Methods
      Copy number, DNA methylation, and gene expression profiles were generated for lung AC and matched non-malignant lung tissues from 34 CS and 30 NS. PCR was performed to determine EGFR and KRAS mutation status. Copy number, methylation and expression alterations were integrated for 14,000 genes on an individual tumour basis. Disrupted genes were ranked according to the magnitude of alterations they exhibited using a novel algorithm we developed denoted MITRA. Of the genes scored by MITRA, those ranking in the 99th and 1st (top) percentiles for up- and downregulation, respectively, were subjected to Ingenuity Pathway Analysis (IPA). IPA was performed separately on all 64 lung tumours and pathway results for CS and NS were compared.

      Results
      We identified 361 genes that ranked in the top percentiles for up- or downregulation in at least 20% of the lung ACs we assessed. Identification of recurrent RASSF1A downregulation, and EGFR upregulation predominantly in NS demonstrates the ability of our ranking algorithm to prioritize genes known to be involved in lung tumour biology using multi-dimensional genomics data. To determine cellular pathways and functions likely deregulated as a consequence of gene disruption, we performed IPA on each tumour and determined the frequency of individual pathway disruption across tumours. This analysis revealed 88 annotated pathways with a minimum disruption frequency of 15% in either or both CS and NS. Commonly affected pathways involved: adhesion and extravasation implicating tumour invasion and migration; various catabolic and anabolic processes implicating cell metabolism; and several specific signaling pathways including atherosclerosis and Wnt/β-catenin signaling implicating inflammation and cell proliferation. Comparison of the pathways identified in CS and NS revealed 13 differentially disrupted pathways (Fisher's Exact test p < 0.05 and disruption frequency difference > 15%). Eleven pathways were preferentially disrupted in CS and affected metabolic, immune response, and inflammatory pathways. Anandamine degradation and ephrin receptor signaling were preferential to NS.

      Conclusion
      Our novel, multi-omics tumour system based approach revealed genes prominently disrupted in CS and NS lung AC which were associated with several cellular pathways commonly or differentially disrupted in these two groups. Pathways affected by genes disrupted at both the DNA and RNA level may contribute to the distinct clinical characteristics associated with CS and NS lung cancer and may serve as targets for intervention.