Phenotypic heterogeneity among cancer cells, observed even within single tumors, presents enormous challenges for developing "optimal" targeted treatment plans. In practical terms, heterogeneity can translate into varying degrees of tumorigenicity and drug response among tumor cells. Our hypothesis is that the characterization of a small number of subpopulations and their responses to drugs will lead to significant improvements in the diagnosis, prognosis, and treatment of lung cancer. Thus, it is our long-term goal to identify clinically important tumor phenotypes that are predictive of therapeutic outcome. To identify cellular subpopulations, we make use of high-content image-based platform for obtaining large number of immuno-fluorescence images of cancer cells exposed to varying drug treatments. Image processing tools are used to extract quantitative and multi-dimensional single-cell phenotypes. Subsequent analytical techniques are applied to determine the most informative cellular descriptions, to identify phenotypically distinct subpopulations, and to correlate with single-cell drug responses. This image-based approach does not require genetic or biochemical manipulation and can translate directly to disease-relevant primary cell samples. Taken together, this approach will initiate the development of databases for correlating quantitative descriptions of tumor heterogeneity to drug sensitivity and therapeutic outcome. In this study we develop our methodology on a progression of model systems, starting from cell lines, xenografts and finally moving to tissue sections of primary patient tumor samples. The proposed research has three goals.
The first aim develops and optimizes experimental assays to capture signaling heterogeneity from models of non small cell lung cancer.
The second aim develops and optimizes computational methodology to test whether patterns of signaling heterogeneity correlate with drug sensitivities.
The final aim tests the feasibility of translating image-based assays to frozen and formalin fixed, paraffin embedding (FFPE) primary patient samples.
In this proposal, we will study non small cell lung cancer heterogeneity and its implications for chemotherapy. We will develop experimental and computational approaches to capture and characterize lung cancer heterogeneity, identify their correlation to drug responses, and test the feasibility of applying our approach to clinically relevant samples. Ultimately we aim to provide a deeper understanding of mechanisms involved in the progression of lung cancer, the identification of new targets, and the potential for more effective therapies for many types of cancer.
|Pavie, Benjamin; Rajaram, Satwik; Ouyang, Austin et al. (2014) Rapid analysis and exploration of fluorescence microscopy images. J Vis Exp :|
|Coster, Adam D; Wichaidit, Chonlarat; Rajaram, Satwik et al. (2014) A simple image correction method for high-throughput microscopy. Nat Methods 11:602|
|Wang, Yanqin; Ku, Chin-Jen; Zhang, Elizabeth R et al. (2013) Identifying network motifs that buffer front-to-back signaling in polarized neutrophils. Cell Rep 3:1607-16|
|Zhang, Elizabeth R; Wu, Lani F; Altschuler, Steven J (2013) Envisioning migration: mathematics in both experimental analysis and modeling of cell behavior. Curr Opin Cell Biol 25:538-42|