The focus of our research is to investigate core signaling pathways that contribute to cancer growth, and to develop models to accurately determine optimal therapeutic regimens for cancer patients. Recent results from clinical trials using targeted therapies for solid tumors have shown that drug response is oftentimes not driven by one mutation or pathway alone. Instead, response is confounded by interactions between the target gene and deregulation of downstream and alternative pathways. Therefore, our studies aim to model how signaling pathways work in relation to others in human tumors, and to identify patterns that correlate to drug response. We hypothesize that integrated 'omic'pathway models composed of multiple components of the growth factor receptor pathways will define biologically distinct subtypes of breast cancer and will accurately predict drug response in patient tumors. Specifically, we will develop and use genomic signatures centered on multiple levels of the growth factor receptor networks (GFRNs) to investigate how these pathways signal in human tumors. Novel statistical modeling approaches, including probabilistic barcode data standardization and Bayesian factor analysis for prediction of pathways and pathway interactions in tumors will move beyond individual pathway predictions to instead profile multi-pathway models in human tumors. Further, these models will integrate 'omic'data types, including RNA-sequencing, mutation status, and proteomic data, enabling a more comprehensive analysis of GFRN deregulation. GFRN pathway activity predictions and sensitivity/resistance to drugs that target the respective pathways will be validated in both cell lines as well as in "fresh" human tumor cells grown in 3-dimensional culture. Importantly, clinical validation of the pathway profiles will be carried out with I-SPY 2 (Investigation of Seril Studies to Predict Your Therapeutic Response with Imaging and Molecular Analysis 2) clinical trial data, which uses targeted therapies directed at GFR pathway components in the treatment of breast cancer. Ultimately, our studies will generate a series of well-validated pathway based biomarkers for individualized assessment of drug responsiveness, as well as interrogation of the coordinate deregulation of specific GFRN components in human tumors.

Public Health Relevance

At the end of this U01, we will have developed a series of novel and well-validated genomic tools to directly interrogate discrete growth factor receptor (GFR) signaling pathways within tumors and for prediction of response to targeted agents. Novel statistical modeling approaches will move beyond individual pathway analysis to instead leverage gene expression, mutation and proteomic data to provide multi-pathway models in human tumors. The proposed research will provide the following deliverables: (SA1) validated biologically relevant gene expression signatures, (SA2) novel multi-pathway models that integrate 'omic'data and identify signaling networks deregulated in human tumors, and (SA3) investigation of relationships between network deregulation and drug response. Cumulatively, these research efforts aim to positively impact patient treatment strategies and GFR network biology comprehension for human tumors.

Agency
National Institute of Health (NIH)
Institute
National Cancer Institute (NCI)
Type
Research Project--Cooperative Agreements (U01)
Project #
1U01CA164720-01A1
Application #
8366165
Study Section
Special Emphasis Panel (ZCA1-SRLB-C (M1))
Program Officer
Li, Jerry
Project Start
2012-08-08
Project End
2017-07-31
Budget Start
2012-08-08
Budget End
2013-07-31
Support Year
1
Fiscal Year
2012
Total Cost
$624,117
Indirect Cost
$111,837
Name
University of Utah
Department
Pharmacology
Type
Schools of Pharmacy
DUNS #
009095365
City
Salt Lake City
State
UT
Country
United States
Zip Code
84112
Bild, Andrea H; Chang, Jeffrey T; Johnson, W Evan et al. (2014) A field guide to genomics research. PLoS Biol 12:e1001744
El-Chaar, Nader N; Piccolo, Stephen R; Boucher, Kenneth M et al. (2014) Genomic classification of the RAS network identifies a personalized treatment strategy for lung cancer. Mol Oncol 8:1339-54
Piccolo, Stephen R; Withers, Michelle R; Francis, Owen E et al. (2013) Multiplatform single-sample estimates of transcriptional activation. Proc Natl Acad Sci U S A 110:17778-83