© Georg Thieme Verlag KG Stuttgart · New York
Ulcers and gastritis
10 November 2009 (eFirst)
High-risk population for gastric cancer development based on serum pepsinogen status and lifestyle factors (Yamaji et al., Helicobacter 2009 )
This is one of a number of recent papers that provide new insights into the relationship between gastritis and gastric cancer, the effect of Helicobacter pylori eradication on progression of the severity and extent of gastritis, and on the pathogenesis of H. pylori-related gastric inflammation.
Chronic gastritis is the “soil” in which gastric cancer develops. It is now accepted that the risk of developing gastric cancer is related to the extent and severity of gastritis and that the primary cause of progressive gastritis leading to gastric cancer is H. pylori infection. The natural history of H. pylori gastritis is to progress both in severity and proximally from the antrum into the corpus. The outcome of a particular infection is thought to reflect the interactions between the virulence of the infection organism, host genetics, and environmental factors. The risk of gastric cancer varies both within and between populations with a high prevalence of H. pylori infection. The fact that cancer risk varies significantly with geography and that it can decrease rapidly when environmental factors change (e. g. immigration or changes in the method of food preservation), suggests that environmental factors are the dominant element in the cancer equation. Nonetheless, because H. pylori infection is the necessary but not a sufficient cause of gastric cancer, elimination of the bacteria from mankind will therefore lead to a virtual eradication of gastritis. By contrast, eradication of an ongoing infection will have a variable effect on gastric cancer risk depending on how extensive and severe the gastritis is at the time eradication is accomplished.
More than 100 years ago, it was shown that there was an association between achlorhydria and gastric cancer. During the first half of the 20th century numerous studies examined the relationship between gastric acid secretion, the extent and severity of gastritis, and gastric cancer . Although histology is the final arbitrator, endoscopic appearance with or without chromoendoscopy, and the maximum stimulated acid secretion, have all been shown to provide good estimates of the extent and severity of gastritis. The most widely used surrogate marker is the measurement of pepsinogen levels, which has proven to be especially amenable for use in large epidemiology studies and cancer screening . Pepsinogen I (PGI) is produced exclusively by the chief cells in the gastric corpus. PGII is produced in the corpus, cardiac, pyloric, and duodenal Brunner gland cells. Destruction of the parietal cell-containing proximal part of the stomach results in low levels of PGI without a major change in PGII levels, thus a low PGI/II ratio .
Watabe et al. previously presented a prospective evaluation of cancer risk among 6958 Japanese individuals, mean age 49 years, who were undergoing regular gastric cancer screening . Participants were divided into four groups depending on the results of H. pylori serology and serum pepsinogen levels. Serum pepsinogen status was defined as ”atrophic” when serum PGI level of ≤ 70 ng/mL and a PGI/II ratio of ≤ 3.0 were simultaneously present; all other cases were classified as ”normal.“ In this study they extended those findings to search for patient characteristics and lifestyle factors (age, sex, body mass index [BMI], smoking, and alcohol consumption), that might modify cancer risk .
Cancer risk assessed per 100 000/year varied from 30 for those without H. pylori infection and non-atrophic pepsinogen levels (46.5 % of the total), to 60 for those with positive H. pylori serology and non-atrophic pepsinogen levels (31.3 % of the population), to 370 for those with H. pylori and atrophic pepsinogen levels (15.7 %), and finally to 530 for those with loss of H. pylori positivity and atrophic pepsinogen levels (6.3 %). In contrast to esophageal adenocarcinoma, there was no association between an increased BMI and gastric cancer. Gastric cancer was significantly higher among regular drinkers compared with non-regular drinkers (0.19 % vs. 0.06 %; P = 0.002); the type of alcohol was not specified. Approximately 18 % of gastric cancer cases were attributed to smoking. In the present study, the incidence of gastric cancer was 0.19 % per year in current smokers and in past smokers, significantly higher than that of 0.03 % per year in non-smokers. This was consistent with previous studies showing a relationship between tobacco smoking and gastric cancer, including a meta-analysis of 42 studies, which estimated that the risk was increased by approximately 1.60-fold and was higher in men . In the current study, the incidence of gastric cancer was higher (1200/100 000/year) in the older patients aged ≥ 60 years with ”atrophic” pepsinogen status, and increased when stratified by drinking and smoking habits. Multivariate analysis demonstrated that old age, alcohol consumption, smoking habits, and an ”atrophic” pepsinogen status were independent risk factors for gastric cancer development.
There are a number of important conclusions that can be derived from this study. Despite being enrolled in a gastric cancer surveillance program, the majority of participants (78 %) were at very low risk and received minimal or no immediate benefit from having enrolled in a surveillance program (i. e. cancer was discovered in less than 6 per 10 000 procedures). Eradication of H. pylori in patients positive for the infection with non-atrophic pepsinogen levels would be expected to prevent the ”natural” progression of atrophy and thus cancer risk and break the time-dependent increase in cancer risk in the general population. As such there would be no need for further surveillance following H. pylori eradication .
The fact that some individuals with serologic tests negative for H. pylori and who were nonatrophic went on to develop gastric cancer was thought to be related to the fact that serologic tests rarely have specificities and sensitivities of 90 % or greater, making the presence of false positives and false negatives inevitable. Those from the low-risk groups typically did not have normal pepsinogen levels but had levels higher than those used to define ”atrophicr” and were thus not those with normal mucosa. Additional testing for those with abnormal pepsinogens and negative serology is likely indicated (i. e. urea breath test or stool antigen testing).
Clearly, only the minority of individuals (22 %) (i. e. those in the two groups with atrophic pepsinogens), would benefit from continued cancer surveillance post H. pylori eradication, as H. pylori eradication would be expected to reduce but not eliminate risk in this group . The second important lesson is that there are clearly modifiable lifestyle factors that can decrease gastric cancer risk. However, as H. pylori eradication among those with an already increased cancer risk is expected only to reduce risk, it is important to study whether such modifications (e. g. stopping smoking) after H. pylori eradication would reduce risk further.