Grant Number:	5R21CA202477-02 Interpret this number
Primary Investigator:	Xirasagar, Sudha
Organization:	University Of South Carolina At Columbia
Project Title:	Reducing Colorectal Cancer Disparities: Racial Differences in Colorectal Polyp Profile
Fiscal Year:	2017

 DESCRIPTION (provided by applicant): Colorectal cancer (CRC) is the second leading cause of cancer and cancer deaths in the United States, about 136,000 new cases diagnosed and 50,000 dying from the disease each year. CRC racial disparities (26% higher incidence among Blacks and 52% higher mortality) have persisted at pre-2000 levels despite significant strides in population-wide colonoscopy screening. Nearly 60% of both Whites and Blacks are now covered by colonoscopy screening. Current screening recommendations (initiation of screening at 50 years of age regardless of race) are driven by an assumption that the prevailing CRC disparities are largely due to differences in access to screening and the quality of post diagnostic care. However, in younger age groups, Blacks have twice the CRC incidence rate as Whites, and their stage- adjusted CRC survival rates are lower than Whites. Assuming that most CRCs arise from precancerous polyps that evolve into cancer, it is plausible that Blacks experience earlier polyp initiation and more rapid polyp progression to cancer. If such is the case, one potential factor in the persistent CRC disparities may be the race-neutral age guideline for screening initiation. However there is no direct evidence regarding age-related Black-White differences in precancerous polyps and the indicators of progression through the intermediate stages (polyp size and dysplastic histology). To establish such evidence, it requires a data source representing the polyp prevalence and profile in the population. In the absence of a large enough autopsy study, the next best alternative is a colonoscopy series validated to have achieved near-complete polyp clearance, and having complete polyp documentation. No such data source has been available so far. Our study will leverage a unique polyp database generated by a previous NIH-funded study. It represents the nearest possible proxy to the population polyp prevalence and profile, and has sufficient statistical power to compare Blacks and Whites on polyp prevalence and cancer-relevant features within age strata. It is a large and well-documented database of a single-center colonoscopy series which is externally validated to have achieved near-complete polyp clearance: this cohort experienced 83% CRC incidence reduction and 89% CRC mortality reduction relative to the general South Carolina population over a 4.9-year of follow-up period. The rates are similar to those of the only colonoscopy clinical trial so far. We will use data on ~17,681 colonoscopy patients with ~32,900 polyps (54% Black), and CRC data on this cohort obtained from the South Carolina Central Cancer Registry. We will study Black-White differences within 5-10 year age strata in: the presence of polyps, polyp number and features (size, histology, anatomic location, dysplasia status, and cancer), as well as combinations of these factors. We will use univariate statistical tests, multiple regressio (binary and ordered polynomial logistic regression, linear and Poisson regression) and cluster analysis. We will use sensitivity analyses to account for potential bias in the sample.





None