NEOPLASIA DETECTION RATE (NDR) AND RISK OF POST-ENDOSCOPY ESOPHAGEAL ADENOCARCINOMA (PEEC) AND POST-ENDOSCOPY ESOPHAGEAL NEOPLASIA (PEEN) IN A POPULATION-BASED COHORT STUDY – THE NORDIC BARRETT’S ESOPHAGUS STUDY (NORDBEST)

Background:
Similar to post-colonoscopy colorectal cancer, PEEC and PEEN undermine early cancer detection in BE. Analogous to adenoma detection rate, NDR (defined as rate of high-grade dysplasia (HGD)/EAC detected during index endoscopy associated with BE diagnosis), has been proposed as a quality measure in BE. Little is known about population-based estimates of NDR and the association between NDR and risk of PEEC/PEEN. Using a population-based cohort study, we evaluated the association between facility NDR (F-NDR) and the risk of PEEC/PEEN among newly diagnosed BE patients.
Methods:
This population-based cohort study was conducted in Finland and Sweden during the study period 2006-2020. Data were retrieved from the national and complete patient, cancer, and cause of death registries. PEEC and PEEN were defined as esophageal adenocarcinoma (EAC) or HGD/EAC, respectively, diagnosed 30-365 days from BE diagnosis (index endoscopy). NDR-1 was defined as EAC alone and NDR-2 as HGD/EAC diagnosed from 0-29 days and F-NDR was calculated as a 2-year moving average. Incident HGD/EAC was diagnosed >365 days from BE diagnosis. Patients were followed until HGD/EAC, death or end of study period. Incidence rates (IR) and mortality rates (MR) per 100,000 person-years were calculated using Poisson regression. The F-NDR was log transformed due to data skewness and association between F-NDR and PEEC/PEEN were assessed using the Poisson model using quadratic term for log(F-NDR).
Results:
15,126 newly diagnosed BE patients (mean age 64.6 yrs, 66% men) were included. Among 198 patients diagnosed with EAC: NDR-1 (n=36, 18.1%), PEEC (n=44, 22.2%) and incident EAC (118, 59.5%) (Table 1). The mean F-NDR-1 was 0.23 (SD 1.1). IRs/100,000 person-years for PEEC and incident EAC were 341 (95% CI 254-458) and 193 (161-231). All-cause and EAC specific MR/100,000 person-years were 3782 (95% CI 3645-3924) and 121 (98-149). Unadjusted and adjusted hazard ratios (HR) for PEEC based on F-NDR using quadratic term were 0.16 (0.05-0.52) and 0.18 (0.06-0.56), respectively. Only Sweden reported HGD rates and in this database, 249 patients were diagnosed with HGD/EAC: NDR-2 (n=47, 18.8%), PEEN (n=47, 18.8%) and incident HGD/EAC (n=155, 62%). The mean F-NDR-2 was 0.39 (SD 1.8). IRs/100,000 person-years for PEEN and incident HGD/EAC were 474 (356-630) and 291 (248-340). The adjusted HRs for PEEN based on F-NDR-2 was 0.08 (0.02-0.36), respectively. Figure highlights the significant variability in F-NDR across participating sites and association between F-NDR and IRs of PEEC/PEEN.
Conclusions:
Results from this large population-based study demonstrate an inverse association between NDR and the risk of PEEC/PEEN. These findings support the validity of NDR and PEEC/PEEN as quality measures in BE and intervention studies to determine whether improving NDR leads to a reduction in PEEC/PEEN are warranted.