We have implemented a systems approach for the discovery of lung cancer biomarkers. Applications include risk assessment, early detection, and predicting response to treatment. The approach which includes genomic, proteomic, metabolomic, and immune profiling integrates data from human biospecimens, a large number of lung cancer cell lines grown under various conditions and several genetically engineered mouse models of lung cancer. An important component of the human studies consists of profiling of pre-diagnostic plasmas available through longitudinal cohort studies, particularly useful for the discovery and validation of blood based risk and early detection markers. Findings include molecular signatures and networks for lung cancer sub-types. Mouse model studies yielded an NKX2.1 signature relevant to early detection, an EGFR network signature for EGFR driven lung cancer and a neuroendocrine signature for small cell lung cancer which were validated in human samples. Integration of proteome and transcriptomic profiling of lung cancer cell line yielded several signatures and a marker panel that predict survival for early stage lung cancer. Mining of the large and continuously expanding sets of data currently being collected benefits from an organized collaborative effort.

This content is only available via PDF.
©2013 American Association for Cancer Research.
2013