Expression-Based Multi-Gene Signatures for CRC Recurrence and Chemoselection


Principal Investigator: Christopher Sears
Abstract: Summary Colorectal cancer (CRC) is the second leading cause of cancer deaths in the U.S and Western world. Despite increased screening and advances in treatment, the mortality rate (~50,000/year) and high national health-care burden for CRC is likely to remain high unless more effective methods are developed to predict recurrence and response to chemo-radiation therapy. For example, the five-year survival rate of CRC is 90% for the over 70% of patients who do not recur (even without therapy). However, this rate for non-local metastasized CRC (mCRC) drops to only 10%. CRC treatment and the patient outcome could be significantly improved if it could be focused on only those patients who need and are most likely to benefit from the therapy. For this reason the NCI has made as a critical goal the [unreadable]Discovery of new indicators of prognosis and the likelihood of response to chemotherapy and radiation[unreadable] (Recommendations of the NCI Colorectal Cancer Progress Review Group (PRG), 2000). As a result of the Phase II, AmberGen has developed a promising CRC tumor assay based on multi-gene expression profiling, termed the Praxis-CRCTM. This assay is designed to accurately predict the likelihood of CRC recurrence and response to therapies used to treat post-surgical CRC patients including major chemotherapies and external beam radiotherapy. The initial assay developed and evaluated in Phase II requires only 4 genes to predict recurrence (mCRC) with over 95% accuracy and 4 genes to predict response to 5-fluorouracil (5-FU) therapy with over 91% accuracy as described in the proposal. This compares favorably to commercial assays developed for breast cancer such as OncoDxTM (Genomic Health) which utilize a panel of over 25 genes with much less accuracy. Novel steps in the development of the Praxis-CRCTM signature include: i) Acquisition of high-quality fresh-frozen CRC tumor samples with known clinical outcome data going back 5 years, ii) Measurement and analysis of low-noise gene expression microarray data, iii) Generation of priority gene list by Meta-Analysis, iv) Validation of the priority gene list by RT-PCR on FFPE samples with known outcome for over 5 years, v) Development of the final assay on an industry standard 384-well RTPCR platform. This Phase II Bridge project is aimed at accelerating the commercialization of Praxis-CRCTM. This will involve further refinement and evaluation of signatures to predict outcome of common therapies such as folfox, folfiri and external radiation using 300 well-documented clinical FFPE tumor samples. The project involves a partnership between AmberGen and a global leader in biomedical assay products with over $1.7 billion in revenue in 2008 and a total of 9,000 employees. As a result of the success of the Phase II project and the opportunity to receive SBIR Phase II Bridge funding, this company will provide at least $6M in funds for the additional costs of commercialization including a large-scale multi-institution validation study, FDA regulatory approval and world-wide marketing of PRAXIS-CRCTM. The project also includes participation by the Ontario Institute of Cancer Research (OICR), Dana Farber Cancer Institute and the UCSD Morris Cancer Center, major sources of high quality CRC tumor samples, Ananomouse, Corp, (subcontractor) the developer of PRAXISTM, advanced software for the collection and analysis of microarray gene expression data, and WaferGen Biosystems, Inc., developer of SmartChip[unreadable] Real-Time PCR System which will be used for improved initial biomarker discovery.
Funding Period: 2006-09-01 - 2014-08-31
more information: NIH RePORT