Letter to the Editor: “Parietal Cell Antibody Levels Among Chronic Gastritis Patients in a Country With Low Helicobacter pylori Infection: Epidemiology, Histopathological Features, and H. pylori Infection”

Abstract

We read with great interest the article titled “Parietal Cell Antibody Levels Among Chronic Gastritis Patients in a Country with Low Helicobacter pylori Infection: Epidemiology, Histopathological Features, and H. pylori infection” by Amalia et al. [1]. In this study, the authors investigated the prevalence of autoimmune gastritis (AIG), and its association with Helicobacter pylori (H. pylori) infection status and gastric histopathological features in the Indonesian population. Given the previously reported high incidence of gastritis and relatively low H. pylori infection rates in this region, exploring other causes of gastritis, such as AIG, is essential for improving diagnostic accuracy and guiding clinical risk management [2].

While the authors conducted a comprehensive analysis of parietal cell antibody (PCA) levels and its association with clinicopathological features, several aspects of the study's methodology and interpretation raise concerns and warrant further clarification. A key observation of this study is the reported 78.99% prevalence of PCA positivity among 113 Indonesian patients with chronic gastritis and 25 healthy controls, while only one patient was diagnosed with AIG. The authors suggest that this discrepancy might be due to H. pylori-associated immune activation. However, we think it may also reflect the combined impact of overly strict diagnostic criteria for AIG and the inappropriate interpretation of PCA results.

In the diagnostic methods, AIG was defined based on the PCA positivity, absence of H. pylori infection, histopathological features associated with AIG, and sparing of the antrum. We understand the concern that coexisting H. pylori infection may complicate the diagnosis of AIG, as it can cause histopathological changes that resemble AIG. But we note that neither antrum sparing nor negative H. pylori-negative status should be a required criterion for AIG diagnosis. Our previous findings have shown that AIG can coexist with H. pylori infection, and hence patients may present with antral inflammation or atrophy. Such patients still show characteristic serological features similar to AIG patients without H. pylori infection, including decreased pepsinogen (PG) I levels and elevated gastrin 17 [3]. Additionally, they may exhibit pathological findings such as enterochromaffin-like (ECL) cell hyperplasia [4], which was not investigated in the study by Amalia et al. Therefore, we think the criteria used may have led to underdiagnosis of AIG in the Indonesia cohort. Rather than excluding patients with H. pylori infection or antrum inflammation/atrophy, more specific diagnostic tests should be considered to improve the diagnosis of AIG, especially given this is the first study reporting AIG prevalence in this region.

Adding to these diagnostic challenges Beyond these diagnostic criteria, the method used for detecting PCA in the study also needs careful consideration. The presence of PCA was detected by Amalia et al. using an indirect immunofluorescence antibody test (IFT), a semi-quantitative method which has high sensitivity but limited specificity. Notably, nearly half of the patients had borderline PCA levels (10 units/mL, in an assay ranging from 0 to > 300 units/mL, with 10 units/mL being the first discrete value), which warrants careful interpretation. Our own studies showed that 81 of 256 (31.6%) patients tested positive for PCA using IFT, but only 18% were confirmed positive by the H⁺/K⁺-ATPase-specific EliA, an automated quantitative enzyme fluoroimmunoassay, and pathology [3]. Moreover, studies showed that PCA is present at lower rates in the general healthy adult population, who may never develop AIG or pernicious anemia [3, 5]. While Amalia et al. showed a correlation between gastric body inflammation score and PCA levels, it is unclear from their report whether controls exhibited a lower prevalence of PCA compared to gastritis patients. Thus, additional tests may be required to prove the presence of PCA.

The authors suggest a potential link between the high PCA prevalence observed in this cohort and H. pylori infection. While the hypothesis that H. pylori could trigger autoimmune responses via molecular mimicry is biologically plausible, the current PCA data presented in this study may be somewhat limited in supporting this interpretation with confidence. To explore this further, we examined data from 81 patients with gastric atrophy and found no significant differences in PCA levels (determined by EliA) between patients with and without H. pylori infection (36.1 vs. 39.4, p = 0.24, Figure 1A). Further correlation analysis showed that the height of the serum level of PCA is not associated with the H. pylori antibody titer in patients with H. pylori infection (r = 0.08, p = 0.66, Figure 1B).

Finally, the authors examined H. pylori antibody levels and PCA levels in relation to clinicopathological features, such as inflammation, atrophy, and PG I/II ratio. Results of their study indicate that levels of H. pylori antibody and PCA were higher in patients with gastric antrum or body inflammation, as well as in those with atrophy. This seems expected given that H. pylori infection and AIG are known to cause inflammatory and atrophic mucosal changes. However, the height of antibody titers in H. pylori-infected patients and those with AIG may be affected by patient genetics, BMI, bacterial load, mounted immune response, antibody decay, and many other factors [6]. Thus, it remains unclear what a correlation between IgG levels and atrophy stage would biologically mean, and it would be of interest to see whether the correlation remains significant after the exclusion of H. pylori seronegative and PCA negative patients from their linear regression analyses. Our own data show that, at least in our cohort of patients with premalignant lesions of the stomach, the level of atrophy as determined by OLGA score is not correlated with IgG levels of anti-H. pylori antibodies (r = −0.05, p = 0.70, Figure 1C) or PCA (r = 0.09, p = 0.69, Figure 1D).

In summary, we think the diagnostic criteria used by Amalia et al. may have resulted in underestimation of AIG prevalence in the Indonesian cohort, while the prevalence of PCA may have been overestimated. More specific serological assays (e.g., EliA, anti-intrinsic factor antibody, pepsinogen I, and gastrin-17) and additional histological features like enterochromaffin-like (ECL) cell hyperplasia should be incorporated to improve the detection of PCA and diagnosis of AIG.

Xiaopei Guo: conceptualization, writing original draft. Manon C W Spaander: conceptualization, review and editing. Gwenny M Fuhler: conceptualization, writing and editing.

The authors have nothing to report.

The study was approved by the Medical Ethical Review Committee of Erasmus MC (MEC-2009-090), and informed consent was obtained from all included participants.