DAY 2: New Research on Colorectal Cancer (CRC) Prevention

Dr. Thomas Imperiale, MD, from Indiana University School of Medicine, in Indianapolis, IN, presented research examining whether the performance of fecal immunochemical testing (FIT), the most used non-invasive CRC screening test, varies by sex. This was a retrospective analysis of data from two prospective cross-sectional studies (DeeP-C and BLUE-C) of the previous and new versions of the Cologuard FIT. The studies included more than 30,000 patients (approximately 14,000 males and 16,000 females) who underwent FIT and colonoscopy. A total of 163 CRCs and 2901 advanced adenomas were detected. The study found that, compared to females, males had higher fecal hemoglobin levels. The tests also had higher sensitivity for CRC and advanced adenomas in males, specifically for proximal lesions. Sex, smoking status, BMI and lesion size, location, and histology were independently associated with FIT sensitivity for advanced neoplasias. The study suggested that the FIT threshold for females would need to be reduced to 12 ng/mL to match test sensitivity in males, resulting in 76% specificity.

Dr. A. Mark Fendrick from the University of Michigan in Ann Arbor, MI, presented findings on the implementation and impact of a 2023 CMS policy change that expanded the definition of CRC screening to include stool-based testing. Administrative claims from Optum on commercially insured adults was linked with laboratory data for adults 45 to 75 years of age between November 2020 and October 2024, before and after the policy change. A total of 16,859 patients pre-policy and 22,360 patients post-policy were included who had results from an mt-sDNA test or FIT. The median age was 57-58 years. Median out-of-pocket costs dropped to $0 for follow-up colonoscopy with and without polypectomy. Approximately 30% of patients incurred out-of-pocket costs averaging $879. Follow-up colonoscopy completion within 180 days of a positive screening test did not change after the policy implementation, however. The study identified a number of patient-, provider-, and system-level barriers to follow-up colonoscopy, including limited capacity for the additional colonoscopies. Previous modeling studies suggest that shifting initial screens to more stool-based testing will allow screening of all eligible patients, prevention and detection of more CRC cases, maximization of colonoscopy efficiency, an increase in colonoscopy revenue due to more polypectomy, and lower total CRC costs.

Dr. Stuart Akerman from Digestive Health Associates of Texas, Plano, TX, shared the results of a retrospective cohort study on the impact of AI-assisted polyp detection on quality metrics and efficiency in two community ambulatory surgery centers. The performance of computer-assisted detection (CADe) has been tested in academic centers but not in private practice settings. The study included 2756 colonoscopies before and 2410 colonoscopies after implementation of CADe. The two cohorts were well matched, and the centers had a 35% adenoma detection rate before implementation. After implementation, every endpoint improved, including polyp detection rate (65.1%), adenoma detection rate (42.8%), tubular adenoma detection rate (37.6%), and sessile serrated lesion detection rate (9.2%). These gains were similar by sex, indication, and physician experience level. 

Irene Chang, BS, from Dartmouth Hitchcock Medical Center, Lebanon, NH, described a single-center study of endoscopist adherence to a 2020 task-force recommended 7-  to 10-year interval timing of surveillance colonoscopy for non-advanced tubular adenomas. The study compared adherence before and after the guidelines and the patient, procedure, and endoscopist factors predicting non-adherence. The study included 1334 colonoscopies. Before the recommendation, >90% of endoscopists recommended a 5-year interval after detection of a non-advanced adenoma; this number dropped to 50% after the 2020 recommendation. Family history of CRC, good or fair bowel prep quality, number of adenomas, and longer endoscopist experience were predictors of non-adherence to the longer 2020 surveillance interval recommendation. 

Dr. Yizhong Wu from Baylor Scott & White Health in Round Rock, TX, reported on the results from a National Surgical Quality Improvement Program examining the use of colectomy for non-malignant polyps, despite the availability of advanced resection techniques. The study was a retrospective analysis of the ACS-NSQIP database from 2012 to 2024. They found that the proportion of colectomies performed for non-malignant polyps decreased significantly over time, in both sexes, and across all age and race groups. Dr. Wu speculated that the trends are driven by the safety and efficacy of advanced methods and their adoption in guidelines, patient preference, and selective referral of patients to colectomy.