Appropriate interval of surveillance endoscopy for early detection of gastric cancer.

Abstract

Surveillance endoscopy is conducted for the prevention and management of gastric cancer by continuously investigating and monitoring its occurrence. However, evidence regarding the appropriate intervals for surveillance endoscopy is insufficient. One of the reasons for this is that the risk stratification for gastric cancer has become more complex as the Helicobacter pylori infection status has diversified due to improvements in water and sewage systems and its widespread eradication therapy. Therefore, it is essential to determine surveillance endoscopy intervals based on appropriate risk stratification to efficiently detect gastric cancer while simultaneously reducing endoscopy-related adverse events, including cost, burden, and complications. Yasuda et al.1 investigated the frequency of pT1b or deeper gastric cancers and the prognosis at each surveillance interval in a retrospective, multicenter, cross-sectional study. They defined patients with a history of endoscopic resection for gastric cancer as a high-risk group, those with past or existing H. pylori infection without history of gastric cancer, as a medium-risk group, and those with a history of gastric surgery or H. pylori uninfected, as a low-risk group. The low-risk groups were excluded from their study. Their study involved 1085 gastric cancers detected by surveillance endoscopy, among which 199 (18.3%) had a depth of invasion of T1b or deeper and 883 (81.4%) were successfully treated with curative endoscopic resection. With a reference surveillance interval of ≤0.5 years, the risk of T1b or deeper gastric cancer exhibited a gradual increase as the surveillance interval increased. The proportion of T1b or deeper gastric cancers was found to be 11.6%, 15.5%, 21.9%, and 24.1% for surveillance intervals of ≤1.0 year, ≤1.5 years, ≤2.0 years, and ≤2.5 years, respectively, with a significant increase of gastric cancer at surveillance intervals of ≤2.0 years. Risk stratification revealed no significant difference in the proportion of T1b or deeper gastric cancers in the high-risk group with a surveillance interval of ≤2.0 years. Currently, the Japanese Gastric Cancer Treatment Guidelines recommend surveillance endoscopy once or twice annually following a successful curative endoscopic resection.2 However, there are no studies comparing the proportion of T1b or deeper gastric cancers, or prognosis rates between surveillance endoscopy intervals of 0.5 vs. 1.0 years. The results of the study by Yasuda et al.1 suggest that a surveillance endoscopy interval of 1.0 year is sufficient to ensure the patient's quality of life and prognosis following successful endoscopic curative resection. Similarly, Nakajima et al.3 reported that more than 95% of patients with metachronous gastric cancers could be treated by curative endoscopic resection, even if the interval between surveillance endoscopies was 1.0 year. Conversely, Yasuda et al.1 reported that the proportion of T1b or deeper gastric cancers increased with increasing surveillance intervals in the medium-risk group. They further examined 5-year cancer-specific survival rates (5y-CSS) and found that the 5y-CSS for surveillance intervals of ≤1.5 years groups was significantly higher than that of >1.5 years groups. The Japan Gastroenterological Endoscopy Society Guideline recommends performing surveillance endoscopy for patients with some risk factors for gastric cancer, even if they do not have a history of gastric cancer; however, the interval and duration are not clearly defined.4 Yasuda et al.1 recommended performing surveillance endoscopy with a maximum interval of 1.5 years for those deemed at risk for gastric cancer. However, caution is needed when translating these results into real-world clinical practice, because 90% of their cohort had open-type atrophy. Gastric mucosal atrophy is an established risk factor for gastric cancer. Uemura et al.5 reported the relationship between risk of gastric cancer and spread of gastric mucosal atrophy assessed by the Kimura–Takemoto classification. The risk of gastric cancer was 1.7 (95% confidence interval [CI] 0.8–19.2) for moderate atrophy and 4.9 (95% CI 2.8–19.2) for severe atrophy when no to mild atrophy was set as 1.5 In the cohort of Yasuda et al.,1 90% of patients had open-type atrophy, who may have had a relatively high risk of gastric cancer. Based on their results, a surveillance interval of ≤1.5 years may be recommended for those with severe atrophy to reduce the risk of surgical resection for gastric cancer.1 Comparing the risk of gastric cancer in the initial 10 years and the second 10 years after eradication, Take et al. reported that with a longer observation period, the risk of developing diffuse-type gastric cancer, such as poorly differentiated adenocarcinoma and signet-ring cell carcinoma, increased in patients with mild to moderate gastric atrophy at the time of eradication.6 These data suggest that, even if gastric mucosal atrophy is mild or moderate after eradication, the risk of gastric cancer is prevalent in the long term, suggesting the need for continued surveillance. Furthermore, they reported that the annual rate of gastric cancer was 0.15%, 0.29%, and 0.67% in the mild, moderate, and severe atrophy groups, respectively, at a maximum follow-up period of 21.4 years (mean 7.1 years).6 Based on the report by Yasuda et al.,1 a surveillance interval of >1.5 years would be acceptable for those with mild to moderate endoscopic mucosal atrophy. This interval is a subject for future research; however, existing evidence suggests that an interval of <2.0 years would be appropriate. It is worth noting that, in the study of Yasuda et al.,1 H. pylori uninfected cases were excluded as a low-risk group. Gastric cancers not associated with H. pylori include gastric cancer associated with autoimmune gastritis, gastric cancer due to CDH1 gene mutations, fundal adenocarcinoma, signet-ring cell carcinoma, and cardiac cancer. Recently, gastric cancer occurring in H. pylori uninfected cases has attracted attention, but its incidence is extremely low (0.66% of all gastric cancers), and routine surveillance is not recommended unless the malignancy and risk factors are certain.7 In contrast, regarding postoperative surveillance for gastric cancer, the cumulative risk of remnant gastric cancer in patients who underwent pyloric gastrectomy for early gastric cancer was reported to be 2.6–6.1% at 10 years, suggesting that postoperative surveillance endoscopy is necessary.8 Evidence regarding the appropriate duration and interval for remnant gastric cancer is limited; nevertheless, given the cumulative risk of developing the disease, we suggest endoscopy once every 1.5 years. In addition, the risk of developing remnant gastric cancer remains high even more than 20 years after surgery, due to gastroduodenal reflux, suggesting the need for long-term surveillance.9 The H. pylori infection rate in Japan is constantly decreasing, with 30% for those born in the 1960s, 20% for those born in the 1970s, and less than 10% for those born after 1980. The rates of gastric mucosal atrophy have also decreased from 82% in the 1970s to 19% in the 2010s.10 Cost-effective surveillance endoscopy requires an accurate and simple classification of gastric cancer risk stratification. Endoscopy is essential for risk stratification of gastric cancer, as it can assess the degree of endoscopic atrophy, which is an accurate predictive marker of gastric cancer,6 and truly exclude H. pylori-negative individuals. The current evidence suggests that, under the most cost-effective surveillance strategy, those with no history of H. pylori infection and no other risks would not undergo surveillance, those with a history of gastric cancer should undergo surveillance endoscopy annually, those with no history of gastric cancer but with H. pylori infection and a severe endoscopic mucosal atrophy once every 1.5 years, and those with mild to moderate mucosal atrophy once every 2 years. In conclusion, we believe that an accurate and simple risk classification of gastric cancer requires a gastric cancer-specific interview, a H. pylori test, and one endoscopy to evaluate the status of H. pylori infection and the degree of mucosal atrophy. Author H.I. is the Deputy Editor-in-Chief of Digestive Endoscopy and has received research grants from FUJIFILM Co., Ltd., Tokyo, Japan. The other authors declare no conflict of interest for this article. None.