S671 Increased Incidence of Upper Gastrointestinal Cancers in Patients With Positive Stool-Based Studies: A Multicenter Study

Introduction: Stool-based screening tests have been approved for colorectal cancer screening in average-risk patients and have gained popularity due their accessibility, convenience, and non-invasive nature. While its use is primarily indicated to detect colorectal neoplasms, its utility in detection of upper gastrointestinal malignancies (UGM) remains unclear. Our study investigated the incidence of UGM in patients with positive stool-based studies (SBS) and colonoscopy who were negative for polyps. Methods: We performed a retrospective cohort study using International Classification of Diseases. Ten codes using TriNetX, a healthcare database with over 119 million patients in the United States. Out of 1,854,754 patients with positive SBS (ICD.10 R19.5 and LOINC 29771-3), including Cologuard or Fecal Immunochemical Test after January 1, 2014, we excluded those with history of colorectal cancer and colon polyp (adenomatous and/or serrated) in both groups. We identified 118,049 patients in the case group with positive-stool study who underwent a follow-up colonoscopy within one year of positive SBS without polyps. The control group consisted of 955,648 patients with an encounter of screening colonoscopy without history of positive SBS who met study criteria. Cohorts were propensity score matched (PSM) on age, sex, race, tobacco use, alcohol use, Helicobacter pylori infection, gastroesophageal reflux disease, Human Papillomavirus, Barrett’s esophagus with dysplasia, obesity, esophagitis, and hemorrhoids. Outcomes were the incidence of UGM (esophageal, gastric, and small intestine). Odds ratios and 95% confidence intervals were reported and P-values < 0.05 were statistically significant. Results: After PSM, we identified 116,859 patients in both cohorts. Within a ten-year follow up period, in patients with positive SBS, the incidence of UGM including esophageal, gastric, and small intestinal malignancies was significantly increased compared to patients in the control group (0.28% vs 0.21%, odds ratio [OR] 1.34 [1.14, 1.58]). Among UGM, gastric cancer had the highest incidence compared to the control group (0.14% vs 0.09%, OR 1.41 [1.11, 1.79]). Conclusion: Among patients with positive SBS, a subset of this population may be at risk for UGM. Though the overall risk remains low, patients with additional risk factors and/or alarm symptoms should be further evaluated for UGM. Future studies are needed to determine high-risk features for this population and guidelines for further diagnostic assessments.