NONLINEAR MODELS OF MRNA EXPRESSION IN CANCER BY RT-PCR

The overall goal of this project is to improve the data analytic procedures associated with quantitative real-time polymerase chain reaction (Q-RT-PCR), with particular focus on measuring mRNA expression of molecular biomarkers and their association with colorectal cancer. Colorectal cancer is the third most common malignancy in the United States, and the third most common cause of death among cancer-related mortalities. While resection therapy is the main course of treatment for these patients, over 50% of patients presumed cured will recur within three to five years. It is hypothesized that these recurrences are actually undetected micrometastases. We propose to adapt Q-RT-PCR technology by changing the methods to quantify the amount of mRNA expression based on nonlinear models of the kinetic reaction obtained from Q-RT-PCR experiments. The current approach to quantitation does not make use of most of the data from the kinetic RT-PCR reaction, and the assumptions of the current model do not match the reality of the experiments. Thus, we also propose extensions of the error structure of these models to account for serial dilution of samples, heterogeneity across concentrations, and dependence of replicate samples. It is our hypothesis that improved measurement of the quantity of molecular biomarkers will offer more accurate prognostic information for colorectal cancer patients. Data from two NCI funded multi-center trials of guanylyl-cyclase-C (GCC) are available for these studies to demonstrate the clinical utility of our proposed methods.