Proteomic analysis of biological specimens offers an enormous potential for the early detection and prediction of cancer. Before this approach moves toward clinical applications, population-based studies addressing research gaps and methodological concerns are needed. We are proposing a population-based epidemiological study to examine proteomic detection and prediction of colon cancer, one of the major life threatening tumors in the U.S. The study will use a population-based nested case-control design, the unique resources of serial pre-diagnostic serum samples stored at the Department of Defense Serum Repository, collected epidemiological information, and state-of-the-art proteomic and bioinformatic technologies, a combination of which will make the study unique in addressing important scientific questions and overcoming major methodological limitations in serum proteomic research. Cases will be 445 colon cancer patients aged 17-79 years. Controls will be 445 individuals without a history of colon cancer of the same age, ethnicity and time at the collection of each serum sample. Serum proteomics and glycoproteomics will be analyzed using the Bruker Daltonic UltraFlex MALDI-TOF/TOF for serial serum samples from each study subject. We will also measure serum insulin-like growth factor I (IGF-I) and IGF binding protein-3. Telephone interviews will be conducted to collect information on medical history, personal habits, occupational experiences, family history, reproductive history, and demographic characteristics. Information on pathologic diagnosis of colon cancer, medical history, occupational experiences, and other factors will also be extracted from the computerized military health databases. Advanced bioinformatic and statistical methods will be used to identify cancer proteomic profiles and related factors.
The specific aims of the project are: (1) to investigate whether there are proteomic profiles that discriminate between cases and controls;(2) to investigate the earliest time point at which identified proteomic profiles appear, and assess whether proteomic profiles vary with time interval from the diagnosis;and (3) to evaluate whether there are epidemiological factors that may be related to the identified proteins. Relevance: The National Cancer Institute's National Investment in Cancer Research (Fiscal Year 2006) lists cancer prevention/early detection/prediction as the first investment area, and indicates proteomic advances as new hope for the early detection and prediction of cancer. Proteomic early detection and prediction of colon cancer will contribute to the cure of the disease and the optimization of prevention/intervention strategies to reduce colon cancer occurrence.