2848 The Impact of Multitarget Stool DNA (Cologuard) Testing in a Real World Population

Abstract

INTRODUCTION: Colorectal cancer (CRC) remains the second leading cause of death for carcinomas. Effective screening for CRC is paramount to prevent morbidity and mortality. Several options for screening alternatives exist with the newest, as of 2014, being the recommendation for multitarget stool DNA testing. A cross-sectional study at 90 sites, including private and academic centers, throughout the US and Canada evaluated average-risk patients utilizing stool specimens and follow-up screening colonoscopy. This 2014 study found that sensitivity for stool DNA testing of detecting colorectal cancer was 92.3%, advanced precancerous lesions 42.4%, and high-grade dysplasia 69.2%. METHODS: We collected real-world prospective data from January 31, 2017 through May 31, 2019 of 157 patients who had positive stool DNA testing and documented their colonoscopy findings. The patients all underwent colonoscopy at the same outpatient endoscopy center which has four board certified gastroenterologists, all more than 10 years out of training. Patients were all average-risk prior to receiving stool DNA screening. RESULTS: The mean age of patients was 68.8 years, with 99 female patients and 66 male patients. Three patients were found to have invasive adenocarcinoma which included the cecum, sigmoid and rectum. A total of 81 patients had no pathology or hyperplastic polyps and 73 patients had polyps that were either tubular adenomas or sessile serrated adenomas. CONCLUSION: In our study, 48% of patients had a positive predictive value (PPV) indicating those who had stool DNA testing followed by positive initial screening colonoscopies. We compared our data to the initial 2014 clinical trial by Imperiale et al with a 41.37% PPV. Interestingly, the number needed to treat in the clinical trial found that 166 individuals would have to undergo multitarget DNA testing for any colorectal cancer to be found and 31 for advanced precancerous lesions. Our real-world study determined that it took about 52 patients (3 in 157) to identify colorectal cancer and 1 in 2.15 patients for advanced precancerous lesions. Our study highlights the continued necessity of colonoscopies as the first line therapy for screening, even among average risk populations.