Transforming Colorectal Cancer Screening through Multimodal Spectral Markers
Colorectal cancer (CRC) remains the second leading cause of cancer death among Americans largely because colonoscopic screening for all the >100 million Americans over age 50 is unfeasible from both a patient (non-compliance) and societal (inadequate endoscopic capacity/funding) perspective. This is juxtaposed by the fact that average risk screening and surveillance (repeat colonoscopy for a previous adenoma history) is remarkably unproductive with advanced adenoma yields of only 6-10%. Thus, pre- colonoscopic risk-stratification development is of crucial. We have developed a suite of novel light-scattering technologies to sensitively detect microvascular (MVS) and micro- architectural (via low-coherence enhanced backscattering, LEBS) manifestations of field carcinogenesis. Published results from >500 patients indicated that interrogation of the endoscopically normal rectum for MVS or LEBS reliably predicted advanced adenomas elsewhere in the colon (AUROC 0.88 and 0.89, respectively). We have recently achieved a technological breakthrough that allows, for the first time, concurrent measurement of MVS/LEBS markers via a single fiber-optic probe. Importantly, on a pilot dataset, the diagnostic performance was synergistic improving AUROC by ~7-10% thereby enabling detection of all adenomas > mm. Moreover, the rectal fiber-optic probe can be employed without a bowel purge. This proposal will test the hypothesis that combined rectal MVS/LEBS markers analysis will enable accurate risk stratification. We will first identify the precise location of the MVS/LEBS alterations and optimize the fiber- optic probe penetration depth by performing biological/imaging studies on ex vivo tissue and in vivo. Secondly, we will evaluate the ability of this optimized rectal fiber-optic probe to predict advanced adenomas in average risk patients using a training (n=500) and testing set (n=750) in both prepped and unprepped patients. Finally, we will validate this approach on patients (n=400) undergoing surveillance (previous neoplasia). These studies will confirm that MVS/LEBS fiber-optic probe rectal examination provides sensitive, minimally-intrusive risk analysis technique thereby heralding the era of personalized medicine for CRC population screening. PUBLIC HEALTH RELEVANCE: CRC impacts 150,000 Americans per year despite being eminently preventable by colonoscopy. However, colonoscopy for the entire population is impractical (not enough funding or endoscopists) and inefficient (most screening colonoscopies are negative). Our approach is to develop a minimally intrusive, highly accurate pre-screen that would allow tailoring timing/intensity of screening and surveillance.