Cancer risk by length of Barrett's esophagus in Japanese population: a nationwide multicenter retrospective cohort study.

ABSTRACT

BACKGROUND: The cancer risk for each length of Barrett's esophagus (BE) in Japanese is unknown. This nationwide, multi-institutional study aims to clarify the cancer risk by length of BE in the general Japanese population. METHODS: Consecutive subjects who underwent upper endoscopic screening at 17 centers between 2013 and 2017 and had at least one follow-up endoscopy by December 2022 were included. The presence/absence of BE and, if present, its length were retrospectively assessed using the retrieved endoscopic images recorded at baseline. Information on the subsequent occurrence of esophageal adenocarcinoma and other upper gastrointestinal cancers was also collected. Cancer incidence was calculated and expressed as %/year. RESULTS: A total of 33,478 subjects were enrolled, and 17,884 (53.4%), 10,641 (31.8%), 4889 (14.6%), and 64 (0.2%) were diagnosed as absent BE, BE < 1 cm, 1-3 cm, and ≥ 3 cm, respectively. During a median follow-up of 80 months, 11 cases of esophageal adenocarcinoma developed. The annual incidence of esophageal adenocarcinoma is 0%/year for absent BE, 0.0032 (0.00066-0.013)%/year for BE < 1 cm, 0.026 (0.011-0.054)%/year for 1-3 cm, and 0.58 (0.042-2.11)%/year for ≥ 3 cm, respectively. Meanwhile, the incidence of esophageal squamous cell carcinoma and gastric cancer were 0.039 (0.031-0.049)%/year and 0.16 (0.14-0.18)%/year, respectively. CONCLUSIONS: By enrolling a large number of subjects with long-term follow-up, this study demonstrated that the risk of cancer increased steadily with increasing length of BE in the Japanese population. Therefore, it is important to consider the length of BE when determining the management strategy for BE.