Colorectal cancer (CRC) remains the second leading cause of cancer deaths among Americans largely due to inability to screen the ~100 million Americans >/=50 years old with colonoscopy because of the expense, discomfort, potential complications and lack of adequate endoscopic capacity. This is juxtaposed with the inefficiency of screening colonoscopies in that only ~5-6% yield "screening relevant neoplasia" (advanced adenomas or early CRC). Even in patients with a personal history of adenomas and thus deemed to be at higher risk, the yield of surveillance colonoscopy is still < 10%. Thus, there is an urgent need for a minimally intrusive risk stratification technique. Our biomarker development laboratory has focused on developing innovative light-scattering technologies to identify the micro-architectural correlates of the genetic/epigenetic changes in field carcinogenesis. We have used early techniques to prove that interrogation of the microscopically-normal rectal mucosa can predict concurrent neoplasia throughout the colon. In this competitive renewal, we plan to develop a novel technique, partial wave spectroscopy (PWS) that enables unprecedented quantitative characterization of the nanoscale architecture through analysis of brushed or extruded cells. Our preliminary data suggests that PWS signatures paralleled neoplastic progression in experimental models. In humans, rectal brushings were able to accurately predict concurrent neoplasia. These nanoscale alterations appear robust in that they represent a final common manifestation for heterogeneous genetic/epigenetic alterations. We propose to develop novel PWS signatures for prediction rule optimization and then use a training/testing set design to assess the ability to rectal PWS analysis to predict concurrent neoplasia in 1000 patients. We will assess another 500 patients to predict recurrence of personal history of adenomas (surveillance). Finally, we will explore novel approaches such as PWS analysis of fecal-derived colonocytes and compare the rectal PWS with molecular marker Muc 4 (developed in our BDL) and cellular markers (proliferation versus apoptosis). This should provide a highly accurate, minimally intrusive technique to allow risk-stratification and hence tailoring of CRC screening regimen.