Autoimmune gastritis reveals distinct regional profiles and measurable neoplastic risk
International data on autoimmune gastritis are finally beginning to move beyond small single-center series. In a retrospective cohort spanning eight tertiary centers across Europe, Türkiye, Latin America, the United States and Japan, investigators assembled 1240 adults with histologically confirmed disease and tracked outcomes for a median of 68 months. The picture that emerged was not uniform. Presentation varied by geography, background autoimmune burden differed sharply across regions, and the incidence of gastric neoplasia was substantial enough to matter in routine follow-up.
Two neoplastic outcomes drove the analysis: gastric adenocarcinoma and type 1 gastric neuroendocrine tumors, or NETs. Over follow-up, 36 patients developed adenocarcinoma and 132 developed NETs. The cohort’s size and international spread make those events more than an academic footnote. They frame autoimmune gastritis as a chronic premalignant condition whose risk profile cannot be assumed to look the same in every clinic, let alone every country.
Anemia patterns tracked with geography and likely disease stage
The hematologic phenotype differed by region in a way that likely reflects both nutritional context and the stage at which patients enter care. Macrocytic anemia predominated in Europe, affecting 45.6% of patients there. In Türkiye and Latin America, microcytic anemia was more common, at 56.1% and 64.7%, respectively. Those patterns are clinically familiar to gastroenterologists: corpus-predominant parietal cell loss can produce vitamin B12 deficiency and macrocytosis, while iron deficiency may appear earlier and can coexist with chronic gastric atrophy, bleeding, or other nutritional constraints.
Autoimmune comorbidity also clustered unevenly. Latin America carried the highest burden, with autoimmune diseases present in 67.7% of patients. For clinicians, that matters because autoimmune gastritis rarely behaves as an isolated gastric disorder. It often sits within a broader immune-mediated phenotype that should prompt screening for thyroid disease, diabetes, and other autoimmune conditions already known to travel with it.
The symptom and laboratory profile therefore looks less like one disease and more like several regional entry points into the same pathobiology: autoimmune destruction of oxyntic mucosa, hypochlorhydria, hypergastrinemia, and metaplastic change that can later set the stage for neoplasia.
Adenocarcinoma risk rose with age, metaplasia and marked hypergastrinemia
Adenocarcinoma developed in 2.9% of the cohort overall. No incident adenocarcinomas were observed in the Latin America or Japan cohorts, even though the study still recorded meaningful event rates in the broader population. Crude incidence rates for adenocarcinoma ranged from 1.15 to 1.47 per 100 person-years, depending on region. The absence of observed cases in two regions should be interpreted cautiously, because event counts were low and follow-up patterns may differ. Still, the result raises the possibility that referral patterns, surveillance intensity, background prevalence of risk modifiers, and timing of diagnosis are not equivalent across settings.
Several factors were associated with adenocarcinoma. Age over 65 years carried an odds ratio of 4.50. Intestinal metaplasia was also associated with higher odds, with an odds ratio of 1.51. Both findings fit the expected biology of gastric carcinogenesis, where cumulative mucosal injury and metaplastic transformation mark a stomach that has moved farther along the precancerous pathway.
The strongest signal came from gastrin-17 levels above 1316 pg/mL, which carried an odds ratio of 15.52. In autoimmune gastritis, elevated gastrin is a direct consequence of acid loss and antral G-cell stimulation. Persistent hypergastrinemia is not merely a laboratory abnormality; it is a biological driver of mucosal proliferation, and in long-standing atrophic stomachs it may help explain why neoplastic risk rises in parallel with disease duration.
Prior proton pump inhibitor use was also associated with adenocarcinoma, with an odds ratio of 5.74. The association does not establish causality, and it cannot separate medication effect from indication bias. Patients already at higher risk may have been more likely to receive acid suppression. Even so, the finding is clinically uncomfortable enough to demand restraint when prescribing PPIs for nonspecific upper gastrointestinal symptoms in patients who may already have achlorhydria from autoimmune disease.
Type 1 gastric NETs were more frequent than adenocarcinoma
Type 1 gastric NETs occurred far more often than adenocarcinoma, affecting 10.6% of patients. Crude incidence rates ranged from 0.70 to 1.62 per 100 person-years. In practical terms, these lesions are the most common neoplastic complication of autoimmune gastritis and are tightly linked to the same hypergastrinemic drive that fuels enterochromaffin-like cell hyperplasia.
Risk factors for NETs partially overlapped with those for adenocarcinoma. Prior PPI use was associated with higher odds, with an odds ratio of 2.69. Smoking also tracked with increased odds, at 2.45, and intestinal metaplasia remained associated with risk, at 2.88. Gastrin-17 above 1316 pg/mL again carried a higher odds ratio, 3.25, reinforcing the central role of chronic gastrin excess in the neoplastic cascade.
A single protective association stood out. Helicobacter pylori eradication was linked to lower odds of NETs, with an odds ratio of 0.25. The mechanism is not fully resolved here, but the direction is biologically plausible. Removing an added inflammatory and atrophic insult may reduce the cumulative burden on the already compromised gastric mucosa, although patients with established autoimmune gastritis remain at risk even after eradication.
Follow-up strategy may need to reflect geography and biology
This cohort does not support a one-size-fits-all surveillance model. The disease presented differently across regions, and the observed neoplastic risk did not distribute evenly. A patient in Europe with macrocytic anemia, intestinal metaplasia and extreme hypergastrinemia is not clinically interchangeable with a younger patient in another region whose presentation is dominated by iron deficiency and autoimmune comorbidity. The underlying autoimmune process may be shared, but the trajectory is not.
For bedside practice, the most actionable signals are familiar but now backed by a large multicenter dataset: older age, intestinal metaplasia, very high gastrin-17, smoking, and prior PPI exposure all mark a higher-risk patient. Those features should sharpen decisions about endoscopic surveillance, especially when histology already confirms autoimmune gastritis and the stomach shows advanced atrophic change.
The study’s retrospective design leaves room for referral bias, differences in endoscopic practice, and unmeasured regional confounding. Even so, the event burden is large enough to justify region-sensitive follow-up research and more careful risk stratification in current care. The open question is not whether autoimmune gastritis can lead to cancer. It can. The harder task is deciding which patients should be followed most aggressively, and whether the answer changes depending on where they live and how they first enter the healthcare system.