Abstract B1-17: Airway gene-coexpression network rewiring in the presence of bronchial dysplasia and lung cancer

Abstract Lung cancer is a deadly disease, killing 160,000 a year in the US. The 5-year survival rate of lung cancer patients is invariably low, partially due to a lack of early diagnoses and limited access to CT screening programs during the early, most-curable stages of the disease. The urgent need for alternative early-stage diagnostics has been addressed in previous work, which focused on smoking- and lung cancer-specific gene expression alterations in cytologically normal airway epithelial cells. These lung cancer associated alterations have been shown to serve as a clinically relevant biomarker for the early detection of lung cancer. In this study, we have expanded upon this work by leveraging gene expression from airway epithelial cells collected in the presence of bronchial premalignant lesions. Specifically, we conducted network-based analyses to further explore alterations in bronchial gene coexpression (GCE) patterns in the uninvolved large airway of patients with tobacco-induced premalignancy and those with lung cancer. By studying airway GCE patterns present in patients with lung cancer and premalignancy, but absent in disease-free smokers, we hope to discover the earliest changes in process of lung carcinogenesis that may serve as novel targets for lung cancer treatment and prevention. Coexpression networks use expression-profile correlations to cluster genes to provide system-wide context for single genes implicated in a disease and elucidate the influence of interconnectedness of cellular components on phenotype. Here, we employed Weighted Gene Coexpression Network Analysis (WGCNA) to build GCE networks derived from cytologically normal airway epithelial cells obtained from current and former smokers with lung cancer (n=409), dysplastic airway lesions (n=121), and neither condition (n=412). Samples were collected from subjects undergoing white-light bronchoscopy as part of the AEGIS trial by Allegro Dx, or from subjects undergoing autofluorescence bronchoscopy in the British Columbia Lung Health Study. RNA was extracted from cells and profiled on microarrays or by mRNA-Seq. In order to detect and quantify reconfiguration of networks under different disease states, the Modular Differential Connectivity (MDC) measure was calculated to determine the degree of topological similarity between pairs of GCE modules. In addition, Fisher's Exact Test (FET) was used to calculate significance of gene-member overlap between pairs of GCE modules originating from disease networks to distinguish modules conserved in premalignancy and cancer. To identify modules conserved between disease states, but dissimilar from the healthy smoker modules, we focused on those with (a) significant overlap by FET, (b) non-differentiating premalignancy vs. cancer MDC and (c) differentiating normal vs. disease MDC. Hub genes within each network were identified based on intra-modular connectivity, a measure of module essentiality. We prioritized genes with increased connectivity upon entering the premalignant state that remained strong in the lung cancer network. We identified 16 module pairs conserved between disease states, i.e. cancer modules similarly connected (MDC FDR<0.05) and significantly overlapping (pFET<0.01) with premalignancy modules, one of which was weakly connected in the disease-free networks (MDC FDR<0.05). Functional annotation of these modules revealed associations with pathways and transcription factors implicated in carcinogenesis and cell proliferation. Furthermore, we identified two hub genes linked to pathways regulating cigarette smoke-induced signaling and inhibition of lung cancer cell growth. Integration of differential connectivity and GCE network analyses represents a novel approach to discovering pathways associated with early lung carcinogenesis. Our results suggest the occurrence of transcriptomic alterations that arise in patients with premalignancy and remain present in patients with lung cancer. Future work will include experimental validation of these findings. The computational approach outlined above may aid in the identification of candidate master regulators of early lung carcinogenesis that have potential utility both as indicators of increased lung cancer risk and as early-stage drug targets for chemoprevention or therapy. Citation Format: Anna Tassinari, Bin Zhang, Katrina Steiling, Duncan Whitney, Kate Porta, Stephen Lam, Marc Lenburg, Avrum Spira, Jennifer Beane. Airway gene-coexpression network rewiring in the presence of bronchial dysplasia and lung cancer. [abstract]. In: Proceedings of the AACR Special Conference on Computational and Systems Biology of Cancer; Feb 8-11 2015; San Francisco, CA. Philadelphia (PA): AACR; Cancer Res 2015;75(22 Suppl 2):Abstract nr B1-17.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.