Pinning Down Cancer Risk in Barrett Esophagus with LGD

The estimated rate of progression of Barrett esophagus (BE) with low-grade dysplasia (LGD) to esophageal adenocarcinoma (EAC) ranges widely from <0.2% to >3% annually, in large part because of interobserver variability among pathologists. In the current systematic review and meta-analysis, investigators estimated the incidence of EAC using cohort studies with a minimum mean follow-up of 2 years after diagnosis of BE with LGD and LGD duration >6 months before incidence of high-grade dysplasia (HGD) or EAC.

In 24 studies comprising 2694 patients and 16,672 patient-years of follow-up, the pooled incidence rate (IR) of EAC was 0.54% annually, ranging from 0.02% to 11.43% in individual studies. Significant heterogeneity was observed. The pooled IR of HGD, EAC, or both was 1.73% per year. A ratio of patients with LGD to all patients with BE in each cohort was used as a surrogate for selection bias and quality of pathology reporting. In the 14 studies with LGD:BE ratios <0.15 (the estimated prevalence of BE with LGD), the pooled IR of EAC was 0.76%, with low heterogeneity between studies. In the six studies with two expert GI pathologists, the pooled IR of EAC was 1.6%. Estimated incidence was similar with use of prevalent versus incident BE, unifocal versus multifocal LGD, and long-segment versus short-segment BE. In four studies, cause-specific mortality rates were between 1.0% and 2.2% for EAC and 28.3% for nonesophageal disease.