How do we translate observations obtained in model systems, such as inbred mouse strains and cell lines, to improve patient outcomes? Can we develop predictive models to help tailor treatment strategies to the individual patient? These two questions help frame the research efforts in the Klinke lab. Within this context, we have focused on questions related to how tumors use direct and indirect methods to create a favorable environment for tumor growth. In particular, we use proteomics to profile the protein expression patterns in different cellular models of cancer and use computational tools drawn from chemical kinetics and Bayesian statistics to interpret these patterns. From the protein expression patterns we are able to identify altered signaling pathways that lead to resistance to molecularly targeted therapies in cancer. In addition, patients have their own genetic bias that alters an individual’s sensitivity to cancer therapy.
To address these problems, we use a combination of mathematical models, computationally intensive statistical methods, and wet biology tools to help understand the mechanisms by which biochemical signals and genetic bias influence the cellular response to molecularly targeted therapies. Research projects include:
Cell Heterogeneity and Emergent Trastuzumab Resistance in Breast Cancer
Yogesh Kulkarni, Vivian Suarez
Funding source: PhRMA Foundation, National Cancer Institute
Monoclonal antibodies, such as trastuzumab, are one of the largest categories of new drugs that target specifically molecules that differentiate cancer cells from normal cells. Despite the remarkable clinical efficacy and specificity of these molecularly targeted therapies, acquired and de novo resistance to therapy is an important clinical problem. Understanding emergent resistance to trastuzumab is inhibited by the inability to quantify aberrant cell signaling pathways among heterogeneous populations of breast cancer cells. Thus there is urgent need for multidisciplinary approaches to assess and interpret the clinical importance of cellular heterogeneity within breast cancer tumors. Our long-term goal is to improve the clinical management of cancer by establishing the scientific foundation for a prognostic technology that will identify individuals who will develop resistance to molecularly targeted therapies. The overall objective of this project is to identify unique patterns of signaling proteins associated with drug sensitivity and apply computational tools from chemical kinetics and Bayesian statistics to interpret the significance of these patterns of protein expression. Our central hypothesis is that breast cancer cells that overexpress ErbB2 exhibit heterogeneity in response to trastuzumab. Furthermore, this heterogeneity is due to variations in expression of proteins that influence the ErbB2 signaling pathway. Prior studies identify such proteins that individually correlate with trastuzumab resistance. The challenge is inferring how these proteins act in concert to influence trastuzumab resistance. The rationale that underlies the proposed research is that identifying patterns of signaling proteins that are correlated with sensitivity to trastuzumab will enable measuring these protein patterns at the single-cell level in tumor biopsy samples. The proposed research is innovative as it provides a novel approach that combines cutting-edge techniques in computational systems biology and proteomics to address the pressing issue of emergent resistance to trastuzumab in breast cancer patients.
Dendritic Cell Heterogeneity in Toll-like receptor 4 Signaling
Priyanka Dixit, Huanling Liu, Ning Cheng
Collaborators: Chris Cuff, WVU School of Medicine
Funding source: National Institute of Allergy and Infectious Disease
Understanding the basis for individual sensitivity to triggers of innate immunity is inhibited by the inability to interpret multivariate changes in quantitative signaling parameters that are related to Toll-like receptor signaling. Thus there is urgent need for multidisciplinary approaches to assess and interpret variability in Toll-like receptor signaling in terms of its impact on susceptibility to infectious agents. The overall objective of this project is to identify unique patterns of signaling proteins associated with sensitivity to infectious agents and to apply computational tools from chemical kinetics and Bayesian statistics to interpret the significant of these patterns of protein expression.