Genomic Classification of Lung Cancer for Clinical Use
Published reports suggest that DNA microarrays identify clinically meaningful subtypes of lung denocarcinomas not recognizable by other routine tests. I will describe work done in my lab to validate the reproducibility of the reported tumor subtypes. I will focus primarily on three independent cohorts of patients with lung cancer were evaluated using a variety of DNA microarray assays. Using the integrative correlations method, a subset of genes was selected whose reliability was acceptable across the different DNA microarray platforms. Tumor subtypes were selected using consensus clustering and genes distinguishing subtypes were identified using the weighted difference statistic. Gene lists were compared across cohorts using centroids and gene set enrichment analysis (GSEA). Particular attention will be focused on the importance of defining cohorts of similar composition before entertaining unsupervised analyses. Having defined reproducible subtypes, attention will be given to extending the classification to the clinical setting.