Prospective follow-up of chronic atrophic gastritis in a high-risk population for gastric cancer in latin america

Background. Gastric adenocarcinoma (GA) is preceded by premalignant conditions such as chronic atrophic gastritis (CAG) with or without gastric intestinal metaplasia (GIM). Endoscopic follow-up of these conditions has been proposed as a strategy for the detection of early-stage GA. Aim. To describe the risk of progression to gastric dysplasia (GD) and early-stage GA of patients who underwent esophagogastroduodenoscopy (EGD) with gastric biopsies obtained following the updated Sydney System biopsy protocol (USSBP). Methods. We conducted a real-world, multicenter, prospective cohort study. Patients undergoing EGD surveillance with USSBP were enrolled between 2015 and 2021 from three endoscopy units at Santiago, Chile. Patients with prior history of GA or gastric resection were excluded. Follow-up surveillance schedule was determined by gastroenterologist in accordance with the Chilean Digestive Endoscopy Association Guidelines. CAG was confirmed by two expert GI pathologists and categorized by the Operative Link on Gastritis Assessment as stage 0 (normal) through stage IV (advanced stage). The primary endpoint was a composite of GD (low-grade, LGD or high-grade, HGD) or GA, while secondary endpoints were progression in OLGA and separate outcomes of LGD, HGD or GA. Multivariable Cox regression analysis was used to estimate the association between CAG +/- GIM and the outcomes, adjusted for age, sex and Helicobacter pylori (Hp) infection. Results. 600 patients were included in the cohort (64% female; mean age 58 years). At baseline 32.3% (n=194) had active Hp infection. OLGA stage was: 31% (n=184) OLGA 0, 48% (n=291) OLGA I-II and 21% (125) OLGA III-IV. GIM was identified in 52% (n=312) and autoimmune gastritis in 6.2% (n=37). Median follow-up was 28 months (IQR 17-42). During follow-up, 6 early-stage GA, 3 HGD and 6 LGD were observed. No advanced-stage GA was diagnosed. Only 19% (n=35) of baseline OLGA 0 patients progressed to OLGA I-IV, with <2% progressing to OLGA III/IV (Figure 1). Persistence of Hp infection (aOR 2.1; 95%CI 1.1-4.0) was independently associated with increase of at least 1 point in the OLGA scale during follow-up. GA/GD free survival at 3- years for OLGA 0, I-II and III-IV was 99.4%, 97.1% and 91.7%, respectively (p=0.0015) (Figure 2). Based on multivariable Cox regression, OLGA III-IV (vs. OLGA 0) was associated with a 12.1-fold (95%CI 1.5-97.4) higher risk of GA, while GIM was associated with a 13.0-fold (95%CI 1.7-101.2) higher risk, although the CI was wide; this was particularly between 2 and 3 years of follow-up. Discussion: These findings, including the observation that all GAs were early-stage, support endoscopic/histologic surveillance for patients with advanced OLGA stages or GIM, which is a common finding in patients with advanced CAG. Further studies are needed to determine the optimal time interval for surveillance.