Chronic Atrophic Gastritis- Lessons In Heterogeneity

Like I, you probably learnt in medical school that Chronic Atrophic Gastritis is associated with pernicious anaemia and possibly carries an increased risk of gastric cancer. This dogma is in fact true, but it leaves out more than it reveals. What about H.pylori? Which type of gastric cancers are increased? Is there a role for surveillance? It's time to take a look.

The stomach is probably best viewed as a viscus of two parts- the first- the acid and pepsin secreting part is called the oxyntic mucosa, and comprises the cardia, fundus and body of the stomach, the second- the non-acid secreting part, is the antrum, which of course leads to the pylorus.

The oxyntic mucosa contains two principal types of cells- the parietal cells, which secrete hydrochloric acid, and the chief or peptic cells, which secrete pepsin. The antrum does not contain either of these specialised cells, but has epithelial cells that can secrete mucin, like similar cells in the oxyntic mucosa. Gastric mucin differs from intestinal mucin in having neutral pH. On the other hand, intestinal mucin has acidic pH and can be sialo-mucin (containing N-acetyl muramic acid) or sulfo-mucin, depending on the negatively charged components that it comprises.

The stomach does not contain goblet cells, unlike the intestinal epithelium. Goblet cells are mucus secreting cells found in the intestine.

Pepsin is derived from a zymogen called pepsinogen. There are two isoenzymes of pepsinogen- types I & II. The oxyntic mucosa secretes both isoenzymes of pepsinogen, while the antral mucosa only secretes pepsinogen II.

It is widely believed that the sequence of change in the gastric mucosa in response to inflammation proceeds thus: gastritis-->atrophy-->metaplasia-->dysplasia-->cancer.

The three processes of gastritis, atrophy and metaplasia form a continuum and have been subsumed into one term- metaplastic atrophic gastritis or MAG. There are two principal triggers that drive MAG- autoimmunity (A) and environmental (E) factors- thus the two subtypes of MAG are described as AMAG and EMAG.

Metaplasia connotes a change in the type of gastric mucosal epithelium. The normal gastric mucosa may change into a pseudopyloric or an intestinal phenotype.

AMAG is due to an autoimmune attack on the resident cells of the oxyntic mucosa. This is typically accompanied by the presence of anti-parietal cell and anti-intrinsic factor antibodies in the serum. Autoimmune gastritis slowly destroys the parietal and chief cells, in a patchy manner at first, and more extensively as time wears on. There is therefore a lack of intrinsic factor, leading to pernicious anaemia, and if the process carries on, achlorhydria ensues. Achlorhydria leads to hypertrophy of G cells or gastrin secreting cells present in the antrum. Thus, hypergastrinemia is one of the key features of AMAG.

H.pylori infection of the oxyntic mucosa is uncommon in AMAG. This may be because the atrophic mucosa of AMAG may not form a good substrate for H.pylori, or because of colonisation by other bacteria.

On the other hand, the principal trigger of EMAG is H.pylori infection of the gastric mucosa. While AMAG involves the oxyntic mucosa, EMAG favours the antral mucosa. Diet is thought to be involved in some cases, particularly a high salt intake and a group of chemicals called nitrosoamines, that are produced in the stomach from dietary nitrates. Unlike in AMAG, complete acholrhydria rarely occurs in EMAG, pernicious anaemia does not occur, and hypergastrinemia is not a feature, as the G cells of the antrum are lost to the inflammatory process.

An useful differentiating feature between AMAG and EMAG is the ratio between serum Pepsinogen I & II. As Pepsinogen I is only secreted by the oxyntic mucosa, and and Pepsinogen II by both oxyntic and antral mucosa, a low Pepsinogen I : Pepsinogen II ratio is found in AMAG and in patients with pernicious anaemia. It can be used as a risk marker for the development of AMAG, pernicious anaemia and gastric adenocarcinoma in relatives of affected subjects.

In time, in some subjects, the mucosa becomes dysplastic, a precursor to development of gastric cancer. Gastric adenocarcinoma is the commonest malignancy and occurs more commonly in the antral mucosa than in the oxyntic mucosa. The principal risk factor for gastric adenocarcinoma is untreated H.pylori infection leading to EMAG. Gastric cancer can also arise post-pernicious anaemia in subjects with AMAG.

In subjects with AMAG, a second type of cancer- carcinoid tumour- may arise in the oxyntic mucosa. Hypergastrinemia in AMAG is a powerful trigger for hypertrophy of enterochromaffin type cells (ECL)present in the oxyntic mucosa. These cells are normally responsible for secreting histamine, an important secretagogue for acid (hence the role of H2 blockers in treating peptic ulcer disease). With continued stimulation from gastrin in AMAG, the ECL cells first form polyps, which may, in time, turn into carcinoid tumour. Thus, antrectomy is sometimes employed in treating gastric carcinoid to remove gastrin as a driver.

A third type of cancer can rarely arise from the gastric mucosa from mucosa associated lymphoid tissue (MALT). These lymphomas typically arise in subjects with Sjogren's syndrome and other allied conditions such as Rheumatoid arthritis, leading to a histological subtype of lymphoma called "Extranodal marginal zone lymphoma". The main driver for MALT associated lymphoma is again H.pylori infection. While extranodal marginal zone lymphoma may arise elsewhere, such as in the parotid glands, in a majority of cases, the stomach is also involved and should be examined through endoscopy and biopsy.

The role of surveillance in early diagnosis of gastric cancer is clearer in high risk subjects such as those from Far Eastern countries, and those with a family history of gastric cancer, but is less clear in Western subjects who generally have a lower risk of progression from MAG to gastric cancer. In general subjects with pernicious anaemia should have one endoscopic examination to look for AMAG and gastric cancer but repeat endoscopies are not advised. In contrast, subjects in or from high risk countries such as Japan, or those with a family history of cancer should have OGD every 2-3 years. Apart from biopsying abnormal lesions, non-targeted biopsies should be taken from the fundus, antrum and incisura- at least two each from the fundus and antrum and one from the incisura. These should be labelled in different containers and the pathologist should report the biopsy by area examined in his/her report. The incisura is usually involved in extensive EMAG.

Finally, some commonly used terms. The term Type I or "complete" gastric metaplasia is used to describe replacement of the gastric mucosa by small intestinal mucosa (containing goblet cells and brush border). Type III or "incomplete" gastric metaplasia describes replacement of the gastric mucosa by colonic mucosa- large droplets of mucin but no brush border. This classification is of some importance as Type III or "incomplete" metaplasia is associated with a higher risk of gastric adenocarcinoma than Type I or complete metaplasia. The pathologist can identify intestinal metaplasia by the presence of acid sialo-mucin or acid sulpho-mucin, as opposed to the presence of neutral mucin in normal gastric mucosa.

The terms "complete" and "incomplete" metaplasia do not desribe the extent of involvement. If only one area of the stomach is involved, say the antrum or fundus, this is described as "limited" involvement. With metaplasia of more than one area, say antrum, fundus and body of stomach, involvement is termed "extensive" and carries a higher risk of cancer.