Objectives The presence of autoimmune metaplastic atrophic gastritis (AMAG) may lead to an increased risk of associated gastric neoplastic lesions. This study aims to investigate the prevalence of gastric neoplasia in AMAG patients and to explore the possibility of PGI/II ratio as a predictor for AMAG diagnosis. Patients and methods Retrospective audit of 135 patients diagnosed with AMAG on endoscopic gastric biopsy between January 2017 and December 2020 at Beijing Friendship Hospital. The study was registered in Chinese Clinical Trial Registry (ChiCTR2000041163). Results A total of 135 patients (the mean age 61.9 +/- 10.9 years,109 female) had histologically confirmed AMAG. 31.1% (42/135) had AMAG without neoplasia on the initial biopsy; 37% (50/135) had multiple type 1 gastric neuroendocrine tumors (g-NETs), 36 grade 1 and 14 grade 2, the median diameter was 5 mm (range 1-25); 31.9% (43/135) had multiple gastric hyperplastic polyps (GHPs), including 15 cases of GHPs with neoplastic transformation, the median diameter was 14.5 mm (range 3-50). 3.7% (5/135) had single gastric low-grade dysplasia/adenoma, the median diameter was 5 mm (range 3-15). 5.9% (8/135) had single or double gastric high-grade dysplasia or adenocarcinoma, the median diameter was 15 mm (range 8-43). 40.7% (55/135) had pepsinogen (PG) I< 10 ng/ml, 45.9% (62/135) had PG I/II ratio <= 1 and each group had a median of PG I/II ratio Conclusions Lower serum PG I level and PGI/II ratio may be a predictor to indicate the diagnosis of AMAG. It's necessary to perform regular endoscopic surveillance for AMAG patients to recognize associated gastric neoplasia timely.