Gastric cancer (GC) is one of the most prevalent cancers worldwide, with nearly half of the world's new cases and deaths occurring in China [1]. Chronic atrophic gastritis (CAG) is an extremely important precancerous in the evolution of gastric cancer [2]. Early diagnosis and monitoring of CAG is of great significance for early detection and treatment of gastric cancer.

In recent years, deep learning (DL), a branch of Artificial intelligence (AI), has been used for CAG detection and has obtained reliable accuracy [3]. However, endoscopic biopsy histology remains the gold standard for diagnosing CAG [4]. The use of gastroscopy is limited by its invasiveness, high costs, and insufficient supply of skilled endoscopists and endoscopic equipment. Therefore, there is an urgent need to develop a highly reliable and reasonably accurate method to identify CAG in general individuals.

In recent decades, several gastric biomarkers have been used for non-invasive diagnosis of CAG, including pepsinogen I (PGI), PGII, the PGI/II ratio, anti- H. pylori antibodies, and gastrin-17 (G-17) [4], [5], [6], [7], [8], [9]. Miki established the ABC method, which combines anti- H. pylori antibodies with serum PG, to identify individuals at higher risk for imminent GC [10]. Tu et al also evaluated a serological biopsy composed of the five stomach-specific circulating biomarkers to stratify individuals’ risk of developing GC [11]. However, the practical utility is controversial due to its highly varied accuracy in different regions of the world.

Therefore, to better understand the relationship between these gastric serological biomarkers and CAG, this study examined the levels of PGI, PGII, the PGI/II ratio, anti- H. pylori antibodies, and G-17 in patients with CAG and chronic non-atrophic gastritis (CNAG). Preliminary analysis of the value of these serum markers in the diagnosis of CAG could provide further clinical evidence for better prediction rules.