Abstracts

The pharmaceutical and biotech industries are under extreme pressure to improve the processes by which new drugs are invented and converted to commercial products. It is widely believed that biomarkers, defined loosely as individual or combinations of molecular, cel-lular or clinical indicators that correlate with disease progression and/or response to therapeutic intervention, will play a critical role in improving efficiencies, low-ering costs, and shortening timelines across the entire spectrum of drug discovery and development. While the future of biomarkers seems assured, there remains but a single problem in the present: how to discover them. We have taken an integrated approach to biomarker discovery that builds on expertise in clinical medicine, bioanalytical chemistry, and bioinformatics. Specif-ically, we have focused on longitudinal, comprehen-sive phenotyping studies of well-defined patient pop-ulations in disease areas of interest. The foundations of this approach lie in the ability to design and im-plement clinical protocols that address key issues in pharmaceutical development (e.g. responders vs. non-responders), to perform state-of-the-art measurements on proteins, metabolites, and cell populations in a high-throughput clinical laboratory, and to efficiently mine broad, disparate data sets from numerous sources, including mass spectrometry, cytometry, immunoas-says, electronic case report forms, electronic medical records, and patient-based questionnaires. This presentation will address the current status of this biomarker discovery platform. Oligonucleotide microarray analysis of gene expres-sion profiles in lung adenocarcinomas and matched uninvolved lung tissue sample