39 - Incidence of Young-Onset Barrett’s Esophagus is Increasing: A Population-Based Study From a Multi-Institutional National Database

Anila Vasireddy, MD¹, Apoorva K. Chandar, MBBS, MPH², Mart Andrew Maravillas, MS³, Jaime A. Perez, PhD⁴, Sandra Algarin Perneth, MD⁵, Cadman Leggett, MD⁵, Prasad G. Iyer, MD, MS, FACG⁶, Srinivas Gaddam, MD, MPH⁷, Amrit K.. Kamboj, MD^7
1University of Pennsylvania Health System, Philadelphia, PA; ²Case Western Reserve University School of Medicine, Cleveland, OH; ³University Hospitals Cleveland Medical Center, Akron, OH; ⁴University Hospitals Clinical Research Center, Cleveland, OH; ⁵Mayo Clinic, Rochester, MN; ⁶Mayo Clinic, Phoenix, AZ; ⁷Cedars-Sinai Medical Center, Los Angeles, CA

Introduction: Esophageal adenocarcinoma (EAC), a cancer with rising incidence and presenting at advanced stages in younger patients (< 50 years) compared to > 50 years. We aimed to examine the epidemiology of young-onset Barrett’s esophagus (BE) (< age 50), including its incidence, trends, and risk factors, using a large national database.

Methods: Data was extracted from the TriNetX database, a composite of electronic health records from 88 healthcare organizations, from inception to November 2023. BE was defined using ICD-9/10 and encompassed non-dysplastic BE (NDBE), dysplastic BE, and EAC. BE patients were required to have a negative EGD prior to index BE diagnosis. Controls were patients without BE. For both cases and controls, data was extracted on conventional BE risk factors including age, sex, race, gastroesophageal reflux disease (GERD) symptoms, body mass index (BMI), smoking, hiatal hernia, and obstructive sleep apnea (OSA). Patients were further stratified by age (young, < 50 years; old, ≥50 years). Categorical and continuous variables were analyzed using the Pearson’s chi-squared test and Welch’s t-test respectively. Joinpoint regression was used to evaluate trends. Logistic regression was used to evaluate for independent predictors of young-onset BE.

Results: A total of 2,670,143 patients were included (Incident BE - 102,598 (3.8%) [young, < 50 years, n=19,901 (1.9%); old, ≥50 years, n=79,686 (5.2%)]. Young-onset BE accounted for 20% of all incident BE cases. Majority had NDBE at initial diagnosis (n=18,612, 94%). Incidence of BE in 45-49 year and 50-54-year age groups were not clinically significant (2.3% vs. 2.6%, p< 0.05). A significantly increasing trend of young-onset BE was seen from 2014 to 2023 on Joinpoint regression (p< 0.05) (Figure 1B). Hiatal hernia (OR 2.6), smoking (OR 2.3), White race (OR 2.3), OSA (OR 2.2), male sex (OR 2.0), GERD symptoms (OR 1.2), and BMI (OR 1.1) were independent predictors of young-onset BE (Table 1).

Discussion: In a large population-based study, we show that 20% of patients had BE incidence before the age of 50, with an increasing trend over the last decade. An alarming 6% of young-onset BE had BE-related neoplasia. The incidence of BE in the 45 – 50 age-group is nearly the same as that of 50 – 55 age-group (2.3% vs 2.6%). Our study validates that conventional risk-factors of BE in older patients also predict young-onset BE. Screening for BE before the age of 50 may have a significant impact on early detection of EAC.