Because of the relative ineffectiveness of systemic chemotherapy, the majority of therapeutic agents are delivered to many patients to actually benefit only a few. This defensive treatment strategy is also driven by the paucity of effective surrogate markers predicting response. Even molecularly targeted therapies may have relatively poor response rates in the presence of the known target. In order to change the current practice, effective predictors of response are needed. We have recently developed a family of molecular classifiers capable of predicting histological origin of 21 different tumor types with a high degree of accuracy. We hypothesize that, using a genome-wide gene expression profiling strategy similarly effective classifiers can be built to recognize chemosensitive tumors. This proposal will fund a Phase II trial of standard chemotherapy for metastatic colorectal cancer that will seek to identify molecular fingerprints common to chemosensitive tumors. Because there are now two equally effective regimens (FOLFOX vs FOLFIRI) available for first line therapy of metastatic colon cancer, we will seek to derive an expressed genetic profile that will predict response to a more practical modification of each of these regimens (XELOX vs CAMCAPE) with a high degree of accuracy. Furthermore, we will determine if the subset of patients responding to CAMCAPE is the same as the subset responding to XELOX. Finally, we will validate the expression and characterize the function of informative genes useful in tumor classification. ? ?
Showing the most recent 10 out of 12 publications
