Background
Helicobacter pylori is a human pathogen that colonizes gastric mucosa and afflicts approximately half of the world’s population [
1].
H. pylori infection is acquired mainly in the first years of life and persists for decades, causing chronic gastritis, duodenal ulcers, and gastric ulcers, and is a significant risk factor for the development of gastric adenocarcinoma [
2]. The nature of gastroduodenal pathologies depends on the anatomical site of
H. pylori infection in the stomach. We previously showed that the
antrum and the
corpus are the major anatomic sites colonized by
H. pylori in patients with gastric cancer [
3]. However, only a third of the gastric biopsies were positive for
H. pylori, and its colonization was higher inside the tumor lesion compared with the surrounding non-tumor tissue. Therefore, it is tempting to speculate that vigorous abnormal immune responses at the local level are associated with the clearance of
H. pylori infection and with gross pathology. Gastric adenocarcinoma develops as a consequence of chronic inflammation of the stomach lining caused by persistent infection with
H. pylori [
4]. Gastric carcinogenesis progresses through a sequence of preneoplastic lesions that manifest histologically as atrophic gastritis, intestinal metaplasia, and dysplasia [
5]. Although a minority of infected people develops gastric cancer, this disease is the second leading cause of cancer death worldwide, partly because patients are not diagnosed until late-stage cancer is present and a poor prognosis [
2].
H. pylori bacteria persist in spite of activation of the host’s innate and adaptive immune response [
6]. Antibody production and cellular immune responses are not concordant with immunological memory against
H. pylori infection [
7]. Moreover, the bacteria seem to actively dampen the T-helper 1 (Th1) response, which is characterized by T cell activation (CD8 and CD4 positive T cells) and IFN-γ production, leading to considerable tissue damage [
8,
9]. Factors other than
H. pylori infection, which can predispose an individual to gastric cancer have been identified, among them are achlorhydria and oxyntic atrophy [
10]. However, the relationship between gastric cancer development and the strength of local humoral immune responses against
H. pylori is poorly understood.
IgA and IgG are the main effectors of the humoral immune responses against
H. pylori infection in the gastric mucosa [
4,
11,
12]. Unlike IgA, IgG is not actively secreted through the gastric mucosa; thus its protective function in the gastric lumen is limited [
13]. Although IgA is actively secreted to the gastric lumen, where its effectors function is achieved, it is also present in the systemic circulation [
14,
15]. Previous studies have shown that elevated serum levels of anti-
H. pylori IgA is a sensitive indicator of gastric cancer risk [
16,
17].
To determine the influence of humoral immune responses against H. pylori infection on gastric carcinoma, we assessed the presence of anti-H. pylori IgG and IgA levels in gastric adenocarcinoma patients and non-cancer patients by ELISA. We used tissue homogenates of different anatomical areas of the stomach and at the mid-lesion and marginal areas of the carcinoma lesion as well as nearby tumor-free tissue.
Results
Of the 30 patients that were recruited for the study, 16 were diagnosed with gastric cancer after endoscopic and histological examination, and the remaining 14 were positive for dyspepsia. A maximum of three biopsies of specific gastric anatomical sites were obtained per patient. As previously reported, sampled of gastric cancer patients were obtained from the tumor site, the tumor margin, and at least two and five centimeters beyond the margin [
3]. We found that 64.3 % of the patients of the normal anatomical and 93.8 % of the gastric cancer patient areas were colonized by
H. pylori. For their part, biopsies from the middle lesion, the margin, at least 2 and plus 5 cm, were colonized in 68.8, 68.8, 62.5 and 56.3 %, respectively. We found a discrete, but consistent relationship between the tissue dysplasia and the grade of colonization by
H. pylori (Table
1). On the other hand, the rest of patients were negative for both the culture and PCR.
The results of tissue immunoglobulin assays in patients with and without gastric cancer are shown in Table
2. The more striking differences are evident in the significant IgA increase in the predetermined sites of stomach in the patients with gastric cancer compared with tumor-free patients (Optical Density median, IQR:
antrum 0.868, 0.578–0.945
vs. 0.176, 0.129–0.867;
p = NS;
angular portion 0.802, 0.637–1.051
vs. 0.275, 0.135–0.945,
p = NS;
corpus 0.836, 0.688–1.039
vs. 0.413, 0.134–0.737,
p = 0.006;
fundus 0.772, 0.668–1.115
vs. 0.267, 0.160–0.675,
p = NS). Additionally, it is clear that a differential distribution within tumor sites was observed in the patients with gastric cancer; the center samples showed the lowest values (Optical Density median, IQR: 0.419, 0.152–0.736) compared with the rest of the stomach (Optical Density median, IQR: tumor margin 0.902, 0.536–0.975; at least 2 cm 0.976, 0.606–1.220; at least 5 cm 0.919, 0.753–1.293), having significant differences (
p = 0.001) even with the predetermined normal anatomical sites (
p = 0.004). Although IgG levels were also higher in the anatomical sites of the stomach in patients with gastric cancer compared with controls, the differences were not statistically significant. However, the center of the tumor had the lowest IgG values (Optical Density median, IQR 0.193, 0.119–0.311) while higher levels were found farther away from the center of the tumor (Optical Density median, IQR: tumor margin 0.300, 0.138–0.463; at least 2 cm 0.276, 0.165–0.631; at least 5 cm 0.215, 0.164–0.445), a difference that was statistically significant (
p = 0.005).
Table 2
Determination of IgA and IgG antibodies against H. pylori in gastric tissue by sampling site and presence of gastric cancer
Immunoglobulin A |
| Non-cancer n (14) | Gastric cancer n (16) | |
Sampling site | OD Median (IQR) | OD Median (IQR) | p value* |
Antrum | 0.176 (0.129–0.867) | 0.868 (0.578–0.945) | NS |
Corpus | 0.413 (0.134–0.737) | 0.836 (0.688–1.039) | 0.0068 |
Fundus | 0.267 (0.160–0.675) | 0.772 (0.668–1.115) | NS |
Angular portion | 0.275 (0.135–0.945) | 0.802 (0.637–1.051) | NS |
p value (predetermined sites) | NS | NS | |
Mid-lesion | . | 0.419 (0.152–0.736) | |
Tumor margin | . | 0.902 (0.536–0.975) | |
At least 2 cm | . | 0.976 (0.606–1.220) | |
At least 5 cm | . | 0.919 (0.753–1.293) | |
p value (tumor sites) | | 0.001 | |
p value (predetermined sites + tumor sites) | | 0.0048 | |
Immunoglobulin G |
| Non-cancer n (14) | Gastric cancer n (16) | |
Sampling site | OD Median (IQR) | OD Median (IQR) | p value* |
Antrum | 0.125 (0.107–0.615) | 0.200 (0.160–0.375) | NS |
Corpus | 0.199 (0.115–0.326) | 0.335 (0.226–0.627) | NS |
Fundus | 0.135 (0.116–0.462) | 0.342 (0.161–0.527) | NS |
Angular portion | 0.151 (0.103–0.608) | 0.247 (0.157–0.349) | NS |
p value (predetermined sites) | NS | NS | |
Mid-lesion | . | 0.193 (0.119–0.311) | |
Tumor margin | . | 0.300 (0.138–0.463) | |
At least 2 cm | . | 0.276 (0.165–0.631) | |
At least 5 cm | . | 0.215 (0.164–0.445) | |
p value (tumor sites) | | 0.005 | |
p value (predetermined sites + tumor sites) | | NS | |
When cancer patients were further divided into early (I and II) or advanced (III and IV) stages, differences were also found in immunoglobulin distribution across the tumor sites in the “advanced” gastric cancer patients (Table
3). The tumor center remained the site with the lowest values, both for IgG (Optical Density median, IQR 0.151, 0.103–0.233) and IgA (Optical Density median, IQR 0.273, 0.150–0.632). The comparison of patients in the early stage of gastric cancer with those without cancer was not statistically significant (
p = 0.08 and
p = 0.06 for IgA in the
antrum and
corpus, respectively), although the early gastric cancer was small (
n = 5).
Table 3
Determination of IgA and AgG antibodies against H. pylori in gastric tissue from gastric cancer group by sampling site and cancer stage
Immunoglobulin A |
| Early gastric cancer n (5) | Advanced gastric cancer n (12) | |
Sampling site | OD Median (IQR) | OD Median (IQR) | p value* |
Antrum | 0.913 (0.578–0.937) | 0.825 (0.689–0.945) | NS |
Corpus | 0.870 (0.688–0.957) | 0.823 (0.735–1.039) | NS |
Fundus | 0.668 (0.668–0.772) | 0.851 (0.717–1.115) | NS |
Angular portion | 0.637 (0.637–0.833) | 0.881 (0.744–1.075) | NS |
p value (predetermined sites) | NS | NS | |
Mid-lesion | 0.811 (0.474–0.827) | 0.273 (0.150–0.632) | NS |
Tumor margin | 0.941 (0.762–1.147) | 0.889 (0.484–0.965) | NS |
At least 2 cm | 1.175 (0.606–1.220) | 0.869 (0.628–1.225) | NS |
At least 5 cm | 0.919 (0.753–0.928) | 0.915 (0.642–1.308) | NS |
p value (tumor sites) | NS | 0.0023 | |
p value (predetermined sites + tumor sites) | NS | 0.0083 | |
Immunoglobulin G |
| Early gastric cancer n (5) | Advanced gastric cancer n (12) | |
Sampling site | OD Median (IQR) | OD Median (IQR) | p value* |
Antrum | 0.375 (0.192–0.563) | 0.182 (0.156–0.289) | NS |
Corpus | 0.473 (0.273–0.513) | 0.285 (0.175–0.627) | NS |
Fundus | 0.516 (0.340–0.527) | 0.299 (0.159–0.441) | NS |
Angular portion | 0.258 (0.236–0.550) | 0.222 (0.142–0.348) | NS |
p value (predetermined sites) | NS | NS | |
Mid-lesion | 0.285 (0.281–0.534) | 0.151 (0.103–0.233) | NS |
Tumor margin | 0.508 (0.463–0.565) | 0.221 (0.119–0.348) | NS |
At least 2 cm | 0.631 (0.296–0.926) | 0.212 (0.133–0.501) | NS |
At least 5 cm | 0.439 (0.215–0.558) | 0.188 (0.161–0.372) | NS |
p value (tumor sites) | NS | 0.0097 | |
p value (predetermined sites + tumor sites) | NS | NS | |
Serum determinations of IgG1 and IgG2 showed no difference in medians in both groups studied (data not shown). Correlation of serum and tissue immunoglobulins did not show a significant trend in either for the whole group or for the subgroups of presence of gastric cancer or of
H. pylori (Table
4).
Table 4
Correlation of mean tissue IgG at predetermined sites with serum IgG 1 and 2a
All | 21 | - 0.0753 | NS | 0.0492 | NS |
Non-cancer | 10 | - 0.1619 | NS | 0.2953 | NS |
Gastric cancer | 11 | - 0.0302 | NS | - 0.0812 | NS |
H. pylori- negative | 6 | - 0.6473 | NS | - 0.0588 | NS |
H. pylori-positive | 17 | 0.2112 | NS | 0.1306 | NS |
Discussion
The systematic analysis of IgA and IgG levels across different normal regions of the stomach (antrum, angular portion, corpus, and fundus), and the primary tumor and its surrounding tissue, allow us to define the dynamics of the humoral immune response against H. pylori and its association with the tissue pathology.
Topographic analysis showed that IgA levels were higher than IgG levels, except in the region of
corpus where, coincidently,
H. pylori colonized in higher frequency [
3,
18]. Patients with gastric cancer had twice as much anti-
H. pylori IgA than control patients, although IgG levels were similar in all patients. It is clear that the primary tumor expressed less IgA and IgG than the rest of the tissue, which had an abnormal higher production of IgA compared with controls. This increase in IgA secretion in the infected tissue correlates with previous studies that demonstrated serum IgA values might indicate gastric cancer risk development.
On the contrary, reduced levels in the center of the tumor led us to speculate that the damage to the tissue induced by chronic infection with
H. pylori affects the production of immunoglobulin. Quiding-Järbrink
et al., [
21] shown a decrease in IgA production in the stomach of patients with gastric cancer, suggesting that low levels of antibody production could be an indicative of risk for gastric cancer development in the case of precancerous atrophic gastritis caused by
H. pylori. On the other hand, Adamsson
et al., [
22] found reduced IgA levels in the non-tumor tissue from gastric cancer patients, suggested that this must be used as a marker for the detection of risk cancer group and early stage of gastric cancer. This study confirmed de early study of Quiding-Järbrink
et al. Furthermore; we found similar results in the IgA antibody levels in the patients with GC as in the control group
H. pylori positive (data not shown), as in Adamsson
et al., study [
22].
Contrary our findings disagree with this previous reports in which cancer patients in various stages of gastric cancer progression were examined; in as much as more advanced cancers were associated with decreased antibody production. Conversely, we found an increase of IgA in the tissue of patients with gastric cancer, probably this is due by the changes in the
H. pylori phenotypes during the development of gastric cancer, as we previous report [
3], we found high genotypic diversity in the gastric cancer group. Quiding-Järbrink
et al., [
21] suggested that a shift in antigens expression would probably lead to production of antibodies to the newly expressed antigens.
Previous studies have shown that IgA antibodies against
H. pylori are detected in gastric tissue and saliva [
23,
24]. Together with this data, we found a higher production of IgA in the tissue of patients with early and advanced gastric cancer stages compared with those patients without cancer. Then, detection of IgA antibodies against
H. pylori in saliva should be increased at least two-fold. This will be a useful method to detected patients in early stages with increased risk of gastric cancer. This must be corroborating with a conducted study in future.
Conclusions
In conclusion, gastric cancer is characterized by progressive accumulation of a concentrated, specific IgA response against H. pylori, beginning with an abnormal increase in the entire stomach but particularly in the adjacent non-tumor tissue. Thus, this strong immune response may also take part in the damage to some degree as suggested by the higher levels of humoral immune responses nearest to the tumor as compared to the adjacent normal tissue.
Acknowledgments
This work was partially funded by Consejo Nacional de Ciencia y Tecnología (CONACyT) CB2005-24779-50099M, CB2007-78787M and SALUD2009-C01-112588, and the operative budget of the Facultad de Medicina, Universidad Nacional Autónoma de México.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (
http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Competing interests
The authors declare that we have no competing interests.
Authors’ contributions
YLV and SPLR contributed in conception and design. RBZ and ATD contributed in sample recruitment. RIAF, HES, GCR contributed in development of methodology and data acquisition. GCR, YLV and SPLR contributed in analysis and interpretation of data, writing, review, and/or revision of the manuscript. All authors read and approved the final manuscript.