| The long-term goal of our group has been the development of non-invasive airway gene expression profiles to be used as diagnostics for airway diseases including lung cancer. The three major steps we have undertaken to achieve this goal are: (1) Developing the infrastructure needed to support high-throughput, online data-mining, (2) Profiling the gene expression effects of cigarette smoke on airway epithelium and (3) Identifying a gene expression signature capable of clinically diagnosing lung cancer.; The basic hypothesis underlying our research is that cigarette smoke inhalation causes detectable expression changes in human airway epithelial tissue. We believe these gene expression differences provide insight into the physiological conditions that exist deeper in the pulmonary system i.e. lungs. This "field defect" theory has been proposed for several pulmonary diseases including lung cancer. Using microarray technology, we identified the set of gene transcripts expressed in an individual sample and defined the core "transcriptome" of each smoker class: non-smokers, former and current smokers. Comparative statistical analyses also enabled us to identify genes with differential expression between smoker classes.; We further hypothesized that clinical covariates, such as race and cumulative smoke exposure, modulate the gene expression behaviour of the smoker airway transcriptome. In addition, the "field defect" theory of lung cancer led us to believe that expression profiles derived from airway epithelial cells could serve as diagnostic biomarkers in smokers with suspect lung cancer. To support this hypothesis, we identified differential gene expression between smokers with and without lung cancer and defined a set of genes exhibiting high sensitivity and specificity for lung cancer diagnosis. Combining this biomarker with traditional bronchoscopy results yielded a 2-component test that demonstrated 95% sensitivity and 95% negative predictive value in a study set of 164 patients with suspect lung cancer (78 with lung cancer, 86 without lung cancer).; Validation of the lung cancer biomarker was achieved by using an alternative mRNA quantitation platform. Gene expression measurements returned by the iPLEX(TM) assay showed high levels of correlation with corresponding microarray data. Furthermore, the iPLEX(TM) derived measurements also demonstrated a highly accurate ability to predict the cancer status of assayed samples. |
