Identification of molecular targets for therapy and prognostic classification of disease are the goals of this project, which includes functional characterization of potentially relevant genes.

Our strategy is to identify, prioritize, validate and develop genes and gene products that might serve as both prognostic markers and as potential targets for treatment of ovarian malignancy, using a global and systematic approach similar to that used in our early detection work but including a focus on functional characterization of the genes.  The project is consistent with our general strategy for achieving translational research goals, in this case including 1) exploitation of emerging molecular technologies to identify biologically relevant genes as candidate markers of prognosis and targets for therapy, 2) a systematic approach to prioritizing genes for functional characterization, 3) a collaborative approach to evaluating candidate targets that involves investigators from 3 institutions, and 4) use of novel statistical methods to use markers to predict biologic phenotype.  In this project we have progressed through the first phase of the translational research process, are actively engaged in the second phase, and propose to complete the third and fourth within the next 5 years.

Project investigators are making progress towards identifying a set of markers that are associated with resistance of ovarian cancer to standard combination paclitaxel and carboplatin chemotherapy.   Once validated these markers could be used to predict chemoresistance, thereby sparing patients the toxicity and burden of ineffective chemotherapy and assisting in selecting appropriate patients for clinical trials of novel agents.  Ultimately some of these markers may also prove to be useful therapeutic targets.

Investigators are working toward confirming their initial findings of an expression profile which predicts therapy responsiveness in a subset of ovarian cancers and generate some of the resources necessary to continue the project into the coming year (to correlate the transcriptome profiles displayed by different ovarian cancers with their clinical /biological behavior).  Investigators also anticipate mass producing a third generation, customized oligo chip.  Included on these chips will be 70-mers to the 260 genes previously identified as differentially expressed in therapy response vs nonresponse cancers and 1000 related genes that either belong to the same “GO” ontology functional category (e.g. DNA repair), are in the same pathway based on existing knowledge from public domains or are binding partners of the genes.  We hypothesize that these related genes may be as good or better markers of therapy responsiveness than the previously identified 260 genes.  We will also include 70 genes whose expression was previously identified to have prognostic significance in a retrospective study in early stage breast cancer and any genes described in the public domain as being related to therapy responsiveness in any cancer type, including OPCML, a gene recently shown to function as a tumor suppressor in epithelial ovarian cancer (Sellar et al. Nat Genet. 2003 Jul;34(3):337-43).  Taken together these  microarrays will include approximately 2500 oligos and should provide a more efficient and cost effective way to characterize expression patterns in a relatively large number of clinical specimens than commercially available oligonucleotide chips.  We will hybridize the arrays using tissue targets generated from 32 chemosensitive and 25 chemoresistant serous tumors containing at least 50% malignant epithelium which are immediately available in our tumor repository.  Data will be analyzed to identify the top ranking differentially expressed genes.