The Gray laboratory explores mechanisms by which genomic, transcriptional and proteomic abnormalities occur in selected cancers, elucidates how these abnormalities contribute to cancer pathophysiologies and assesses the ways in which these abnormalities influence responses to gene targeted therapies. Current studies focus on developing: (a) integrated analyses of the spectrum of recurrent abnormalities that influence cancer behavior. (b) Mathematical models that describe how cancer-associated molecular abnormalities influence individual responses to therapeutic inhibitors. (c) Novel therapeutic approaches to treat breast or ovarian cancer subpopulations that do not respond well to current aggressive chemotherapeutic strategies. (d) Proteomic strategies for early detection of breast cancer related proteins in blood. (e) Automated functional assessment of genes deregulated by genomic abnormalities in cancers.
We are assessing abnormalities associated with clinical outcome in breast cancers using a combination of comparative genomic hybridization (CGH) with molecular inversion probe technology to assess allele specific genome copy number at ~10Kbp resolution, expression profiling using Affymetrix Exon 1.0 ST GeneChips to interrogate expression levels at ~1.5M known or predicted exons, and reverse phase protein lysate arrays and western analyses to assess protein and phosphoprotein levels in ~100 cancer related genes. These studies are being carried out in collaboration with investigators at UC San Francisco, UC Berkeley, the MD Anderson Cancer Center and the Buck Institute for Age Research. We are contributing to The Cancer Genome Atlas (TCGA) project managed by the NCI and NHGRI by assessing expression in ~1500 normal and 1500 glioblastomas, ovarian cancers and lung cancers using Affymetrix Exon 1.0 arrays. Work in this area is funded by the NCI Bay Area Breast Cancer SPORE and the NCI/NHGRI TCGA project.
We are developing mathematical methods to predict individual responses to therapeutic agents using information on responses to these agents in a collection of cell lines grown in vitro. Major emphasis in this project is on breast cancer. To date we have generated Boolean predictions of pathways that are active in breast cancer cell lines using the Pathway Logic (PL) model. We are attempting to modify the PL models to include concepts of quantitative signaling flux from a steady-state aspect using genetic algorithms to learn parameters for the network. We also have developed adaptive spline statistical association approaches to identify preexisting molecular abnormalities that predict individual responses. These studies are being carried out in collaboration with investigators at UC San Francisco, UC Berkeley, the Netherland Cancer Institute and SRI International. Work in this area is supported by an NCI Integrative Cancer Biology Project award and a grant from GalaxoSmithKline.
Novel therapeutic approaches.
We are assessing responses to NCI and private sector compounds measured for our collection of 50 breast cancer cell lines to identify therapeutic agents that will be highly effective against basal and luminal/amplifier breast tumor subtypes that do poorly on aggressive therapy. In addition, we are developing siRNA therapeutic approaches to treat breast and ovarian tumors that amplify and over express transcripts to which the tumors become “addicted”. Current emphasis is on development of strategies to inhibit the apparent non-coding RNA PVT1 encoded in an amplicon at 8q24 near MYC. This work is being carried out in collaboration with investigators at the MD Anderson Cancer Center, the University of British Columbia and UC San Francisco. It is supported by the NCI Ovarian Cancer SPORE and the NCI Bay Area Breast Cancer SPORE.
We are using information about genomic and transcriptional abnormalities in breast cancer to guide the development of mass spectrometric strategies that can detect breast cancer specific proteins in blood serum in order to enable early breast cancer detection. We are giving special attention to detection of aberrant proteins that result from cancer specific alternative splicing revealed using Affymetrix Exon 1.0 ST arrays as described. We also are employing scanned ion beam mass spectrometry to identify aberrant proteins or transcripts in tissue sections that are associated with cancer invasion. If found, these proteins will be included in the serum protein assessment study. These studies are being carried out in collaboration with investigators at UC San Francisco, the Buck Institute for Age Research and the Lawrence Livermore National Laboratory. They are supported by grants from the NCI Clinical Proteomic Technologies for Cancer (CPTAC) program and the NCI Early Detection Research Network (EDRN).
We are developing a suite of automated cell analysis procedures to facilitate analysis of cellular and molecular responses to pathway targeted therapeutics and/or pathway manipulation reagents such as siRNAs. Processes being automated in multi-well cell culture format include cell culture setup and feeding, molecular manipulation, RNA and protein harvesting and automated image analysis for assessment of motility, apoptosis, proliferation and pathway protein phosphorylation status. Work in this area is supported by an NCI Integrative Cancer Biology Project award and a grant from GalaxoSmithKline.