A qPCR gene expression module test to predict resistance to cetuximab in colorectal cancers independent of KRAS mutation status.

Abstract

e14164 Background: While KRAS mutation predicts resistance to anti-EGFR therapy in colorectal cancer, not all KRAS wild-type patients benefit from such therapy, suggesting that complementary biomarkers capable of identifying additional non-responsive patients would have clinical utility. Methods: To search for such a biomarker, we studied the relationship of cetuximab response with twelve gene expression modules, derived from an unsupervised analysis of 20 independent microarray datasets comprising more than 2,000 colorectal cancer patients. Each module represents a set of highly co-expressed genes related to an important aspect of colorectal cancer variability. Two cetuximab-treated cohorts were studied. The first was a Phase II clinical trial (Khambata-Ford et al, J Clin Oncol, 2007) with accompanying microarray data from pre-treatment biopsies. The second was a single-institution study of cetuximab response from which formalin-fixed paraffin-embedded primary tumor specimens were available. Results: In the first study, module scores were computed by averaging co-expressed module genes in the microarray data. In the second study, module scores were generated from a qPCR gene expression module test, OncoScore Colon, which quantifies modules by averaging three representative module genes relative to housekeeping controls. Notably, in both studies, the mesenchymal module was significantly associated with cetuximab resistance, with module positive patients tending to progress on cetuximab within 10 weeks. Additionally, the status of this module was independent of KRAS mutation status—KRAS mutations occurred in both module-positive and -negative patients. Future clinical studies will continue to test the predictive capacity of the module in regards to cetuximab resistance and other mechanisms. Conclusions: In summary, this study demonstrates the value of a gene expression module-based qPCR panel for stratifying colorectal cancer patients for treatment response, and suggests that our approach may have immediate utility for cetuximab treatment response prediction.