Early gastric cancer is defined as adenocarcinoma confined to the mucosa or submucosa irrespective of lymph node involvement. In Japan, mucosal high-grade neoplasia is diagnosed as intramucosal early gastric cancer. Some early gastric cancers progress to advanced gastric cancer after several years of follow-up. Image-enhanced endoscopy (chromoendoscopy), narrow-band imaging, and magnifying endoscopy increase the diagnostic yield in characterizing early gastric cancer. Endoscopic resection of intramucosal early gastric cancer with endoscopic mucosal resection or endoscopic submucosal dissection is currently performed in East Asian countries to prevent the development of advanced gastric cancer and preserve patients’ quality of life after treatment.
Key points
- •
Early gastric cancer is defined as adenocarcinoma confined to the mucosa or submucosa irrespective of lymph node involvement. In Japan, mucosal high-grade neoplasia is diagnosed as intramucosal early gastric cancer.
- •
Some early gastric cancers progress to advanced gastric cancer after several years of follow-up.
- •
A proper endoscopic screening procedure would increase the detection of intramucosal early gastric cancer.
- •
Image-enhanced endoscopy (ie, chromoendoscopy), narrow-band imaging, and magnifying endoscopy increase the diagnostic yield for characterization of early gastric cancer.
- •
Endoscopic resection of intramucosal early gastric cancer with endoscopic mucosal resection or endoscopic submucosal dissection is currently performed in East Asian countries to prevent the development of advanced gastric cancer and to preserve patients’ quality of life after treatment.
Background
Gastric cancer is currently the fourth most common malignancy and the second most common cause of cancer deaths worldwide. Half the global total of gastric cancer occurs in East Asia. Age-standardized mortality rate is estimated as the highest (28.1 per 100,000 in men and 13.0 per 100,000 in women) in East Asia, whereas that in the United States is low (2.8 per 100,000 in men and 1.5 per 100,000 in women). Early detection and treatment are considered to be effective strategies in reducing mortality from gastric cancer as a secondary prevention. Thus, many attempts have been made in this direction, such as encouragement of mass screening or the development of accurate diagnostic procedures in East Asian countries.
Background
Gastric cancer is currently the fourth most common malignancy and the second most common cause of cancer deaths worldwide. Half the global total of gastric cancer occurs in East Asia. Age-standardized mortality rate is estimated as the highest (28.1 per 100,000 in men and 13.0 per 100,000 in women) in East Asia, whereas that in the United States is low (2.8 per 100,000 in men and 1.5 per 100,000 in women). Early detection and treatment are considered to be effective strategies in reducing mortality from gastric cancer as a secondary prevention. Thus, many attempts have been made in this direction, such as encouragement of mass screening or the development of accurate diagnostic procedures in East Asian countries.
Definition of early gastric cancer
Early gastric cancer (EGC) was first defined in 1962 by the Japanese Society of Gastroenterological Endoscopy as adenocarcinoma confined to the mucosa or submucosa irrespective of lymph node involvement. The need for such a definition was based on the observation that this type of gastric cancer has a favorable prognosis; 5-year survival rates are greater than 95%. The fact that lymph node or distant metastasis is uncommon explains the good prognosis for EGC. Lymph node invasion exists in 10% to 20% of cases; however, the metastatic lymph nodes of EGC are mostly restricted to a few regional nodes (N1). Therefore, gastrectomy with lymph node dissection shows an excellent outcome in patients with EGC. Moreover, the presence of nodal metastases is closely related to the depth of local invasion. When EGC is confined to the mucosa, lymph node involvement is much less common (≤3%). With the increase in the detection rate of EGC throughout the country, the Japanese national records show that the percentage of EGC among resected cases was 40% in 1985.
Many investigators attribute the high incidence of gastric carcinoma in East Asia to dietary and genetic factors, and to Helicobacter pylori infection. The high detection rate of EGC in Japan and Korea is explained by the availability of population-based screening programs. In addition, there are differences between the Japanese and Western criteria for the diagnosis of EGC that are considered relevant. In Western countries, gastric cancer is diagnosed when invasive growth of the neoplasm into the lamina propria of the mucosa or beyond is evident. By contrast, Japanese pathologists often use the term EGC for intramucosal lesions that Western pathologists classify as precursor lesions termed dysplasia or adenoma. In the authors’ opinion, this discrepancy does not express a biological difference in the tumor itself, but represents a difference of conception and terminology. Western pathologists diagnose high-grade dysplasia as a lesion that does not yet have malignant potential, but probably could develop it over time. Japanese pathologists diagnose EGC as a lesion that has malignant potential but has not yet expressed it. In practice, a biopsy diagnosis of high-grade dysplasia in the West or carcinoma in Japan would lead to consideration of therapeutic resection in both scenarios. Recently, a new system of categories classifying gastrointestinal neoplasia (ie, the Vienna classification) has been proposed ( Table 1 ) to bridge the East-West gap. Intramucosal EGC in the Japanese classification corresponds to mucosal high-grade neoplasia (Category 4) in this revised Vienna classification. The classification is important in transferring interpretations of epidemiologic, clinical, and pathologic studies from one arena to the other. This article regards EGC, including intramucosal EGC, as “intermediate lesions” for developing gastric cancer.
| Category | Diagnosis | Clinical Management |
|---|---|---|
| 1 | Negative for neoplasia | Optional follow-up |
| 2 | Indefinite for neoplasia | Follow-up |
| 3 | Mucosal low-grade neoplasia
| Endoscopic resection or follow-up a |
| 4 | Mucosal high-grade neoplasia
| Endoscopic or surgical local resection a |
| 5 | Submucosal invasion by carcinoma | Surgical resection a |
a Choice of treatment will depend on the overall size of the lesion; the depth of invasion as assessed endoscopically, radiologically, or ultrasonographically; and on general factors such as the patient’s age and comorbid conditions. For gastric, esophageal, and nonpolypoid colorectal well-differentiated and moderately differentiated carcinomas showing only minimal submucosal invasion (sm1) without lymphatic involvement, local resection is sufficient. Likewise, for polypoid colorectal carcinomas with deeper submucosal invasion in the stalk/base but without lymphatic or blood vessel invasion, complete local resection is considered adequate treatment.
Natural history of early gastric cancer
Although EGC has a long natural history, some cases do progress to advanced cancer ( Fig. 1 ). Rugge and colleagues followed up 118 consecutive patients with noninvasive neoplasia for an average of 52 (range 12–206) months, and found that 20 (17%) evolved into invasive gastric cancer. Among patients who underwent surgery for invasive cancer, 13 were pathologically staged as EGC while 2 were staged as advanced gastric cancer. Tsukuma and colleagues identified 56 patients with EGC (diagnosed as EGC with endoscopy and proved by biopsy) in whom neither endoscopic nor surgical resection was performed. Over a period of 6 to 137 months (mean 39 months), 20 remained in the early stage and 36 progressed to an advanced stage. The cumulative proportion of patients with advanced gastric cancer consistently increased with time, and the median time to develop advanced gastric cancer from EGC was 44 months.
Screening for the detection of early gastric cancer
At present, nationwide screening is undertaken in Japan and Korea, where gastric cancer is highly prevalent. Whether screening, especially that of the mass population, should be done remains controversial because the incidence of gastric cancer varies substantially among countries and within the same ethnic group. Even in a very high-risk area, there is only some evidence that mass screening reduces mortality from gastric cancer. Therefore, identification of high-risk populations to undergo screening is fundamental for the early detection of gastric cancer in countries with medium to low incidence.
High-Risk Populations for Gastric Cancer
In general, the incidence of gastric cancer increases after 40 years of age and is higher in men than in women. Furthermore, the risk of gastric cancer is increased at least 1.5 times in siblings or offspring of patients with gastric cancer. For the familial aggregation of gastric cancer, the clustering of H pylori in family members may be an important contributory factor, in addition to genetic factors. In multiethnic countries such as Malaysia and Singapore, gastric cancer is more common in Chinese people than in those of Malay and Indian origin. In a study from Singapore, screening for gastric cancer was cost-effective in high-risk groups such as Chinese men aged 50 to 70 years. Therefore, individuals with a high-risk ethnic background or those with a family history of gastric cancer may benefit most from screening and surveillance examination. Relatives with hereditary diffuse gastric cancer should be managed by specialists because genetic testing and prophylactic gastrectomy are sometimes advocated.
Despite the strong link between H pylori and gastric cancer, there are currently no data to suggest that a screen for gastric cancer should be limited to those infected with H pylori . Progression of intestinal metaplasia (IM) in the gastric mucosa can cause spontaneous eradication of H pylori or to an underestimation of H pylori infection ; therefore, patients with gastric cancer might no longer have detectable H pylori and, thus, the screening of only actively infected individuals is generally not effective.
Patients who have established precursor conditions such as mucosal atrophy or IM caused by chronic H pylori infection are at high risk for developing gastric cancer, especially the intestinal type. A Dutch nationwide cohort study indicated that the annual incidence of gastric cancer was 0.1% for patients with atrophic gastritis, 0.25% for IM, 0.6% for mild to moderate dysplasia, and 6% for severe dysplasia within 5 years after diagnosis. The potential benefits of endoscopic surveillance of gastric IM patients was suggested by a cancer incidence of 11% and improved survival in a retrospective study from the United Kingdom. At present, the diagnosis of atrophy and IM is based on the histology of the biopsy specimens. As a result of uneven distribution and topography of atrophy and IM in the stomach, multiple biopsies are recommended from certain anatomic locations of the gastric mucosa, as indicated by the updated Sydney System and the Operative Link on Gastritis Assessment (OLGA).
Although atrophy or IM are histologic entities, their presence can be predicted by endoscopic findings. Kimura and Takemoto suggested that the endoscopic findings of atrophic mucosa were the presence of a pale whitish mucosa and increased mucosal vessel visibility, indicating that it is related histologically to atrophy of the fundic gland ( Fig. 2 D). When the atrophic border remains on the lesser curvature of the corpus, the diagnosis is closed-type atrophic gastritis (antral predominant gastritis), whereas when the atrophic border no longer exists on the lesser curvature and extends along the anterior and posterior walls of the stomach, the diagnosis is open-type atrophic gastritis (pangastritis or corpus-predominant gastritis). These endoscopic diagnostic criteria are commonly accepted and are used in practice for the diagnosis of chronic atrophic gastritis in Japan and Korea. An association between the endoscopically diagnosed extent of atrophic gastritis and the risk for development of gastric cancer has been demonstrated in a large cohort study. Recently, newer endoscopic imaging technologies such as autofluorescence imaging (AFI; see Fig. 2 B, E), magnifying narrow-band imaging (NBI; see Fig. 2 C, F), and confocal laser endomicroscopy have been introduced, and these methods have been reported as useful for the diagnosis of atrophy or IM. In contrast to single-point evaluations of gastritis by biopsy, a possible advantage of these endoscopic imaging modalities is to enable evaluation of the prevalence, actual extent, and distribution of atrophy or IM in the gastric mucosa without taking multiple biopsies.
Screening Test Methods
Radiography using barium meal has been used in Japan for population-based screening for gastric cancer since 1960. When suspicious lesions are identified on barium examination, endoscopy is used to analyze the detected lesions. Barium radiography is the only recommended population-based screening method in the Japanese guidelines for gastric cancer screening, because several case-control studies have shown a 40% to 60% decrease in mortality from gastric cancer in those who have been screened with photofluorography. However, data from prospective cohort studies that defined death from gastric cancer as an end point do not give consistent results. The sensitivity of barium radiography for gastric cancer is reported to be 70% to 86%, but the sensitivity for advanced cancer is 92%, whereas for EGC it is only 32%. Because of the low sensitivity in detecting early lesions, barium radiography is becoming less common in clinical practice.
Serum pepsinogen (PG) is a popular serologic screening test for gastric cancer, particularly in Japan. Serum PG consists of 2 types, PG I and PG II. PG I concentrations decrease with loss of fundic-gland mucosa, whereas PG II remains constant. Therefore, a low PG I level or a low PG I/II ratio, or both, are good serum indicators of the presence of atrophic gastritis. In Japanese patients, a PG I/II ratio greater than 3.0 has a sensitivity of 93% and a specificity of 88% for the diagnosis of normal fundic glands. Serum testing for H pylori alone could miss high-risk gastric cancer populations, because the bacteria can no longer colonize the gastric mucosa with extensive atrophy or IM. Thus, the PG test is used in combination with serum H pylori antibody to assess the risk of gastric cancer. In a large Japanese cohort study, more than 9000 people were stratified according to H pylori antibody status (positive vs negative) and serum pepsinogen test (normal vs atrophic), and followed up endoscopically for a mean duration of 4.7 years ( Table 2 ). In patients with nonatrophic fundic mucosa (as defined by serum PG testing), risk of gastric cancer did not increase substantially, regardless of the presence of H pylori antibody. By contrast, in individuals with a low PG level that indicated atrophic gastritis, a significantly higher risk (6–8 times) of developing gastric cancer was found, compared with those with normal PG levels. Of note, among individuals with atrophic gastritis (as indicated with PG testing), those with negative H pylori serology had a higher risk than those with positive serology, presumably attributable to loss of H pylori in advanced gastric atrophy. Limitations of the serum PG test include suitability to detect mostly the intestinal type of gastric cancer, variation of cutoff values according to study or measurement kits, and normalization of test results after eradication of H pylori . Despite these limitations, in Japan the PG test is currently the most realistic and reliable marker for identifying high-risk populations that warrant endoscopy or more intensive surveillance.
| Group | H pylori Antibody | Serum Pepsinogen | Interpretation | Estimated Gastric Cancer Risk | |
|---|---|---|---|---|---|
| Annual Incidence of Gastric Cancer (%/y) | Adjusted Hazard Ratio a | ||||
| A | Negative | Normal | H pylori– naive individual | 0.04 | 1 |
| B | Positive | Normal | Active H pylori infection but no corpus atrophy | 0.06 | 1.1 |
| C | Positive | Atrophic | Active H pylori infection with corpus atrophy | 0.35 | 6.0 |
| D | Negative | Atrophic | Undetectable or previous H pylori infection with extensive corpus atrophy | 0.60 | 8.2 |
Endoscopy plays a pivotal role in screening for gastric cancer, because of its high lesion detection rate and the ability to remove biopsy specimens for histologic diagnosis. Endoscopy is particularly useful in identifying superficial lesions that may not be recognized by barium examination. A Japanese study comparing the diagnostic ability of endoscopy with barium meal examination in a mass screening population showed that the detection rate for gastric cancer was about 2.7 to 4.6 times higher with endoscopy than with barium studies or photofluorography. Despite these favorable data, the capacity of endoscopy is restricted in Japan because of the availability of gastroscopes and the number of experienced endoscopists. Moreover, endoscopy carries some risks, including perforation, cardiopulmonary events, aspiration pneumonia, and bleeding, that are not negligible in the general population. Therefore, mass screening with endoscopy alone is probably not feasible. Consequently, selection of high-risk candidates for endoscopic examination is important in efficient screening for gastric cancer.
Recently, various advanced endoscopic techniques such as chromoendoscopy, magnifying endoscopy, AFI, NBI, and confocal laser endomicroscopy have been introduced. However, currently available data about the utility of these new imaging techniques for the diagnosis of EGC is mainly concerned with characterization of superficial lesions detected by conventional endoscopy. Accordingly, white-light endoscopy is still the primary method used to detect EGC in routine screening practice.
Endoscopic screening procedure
The performance of endoscopy for detecting EGC depends heavily on the skill and knowledge of the endoscopist, so standardization and adequate training in endoscopic screening procedures are important.
Screening Procedure
To minimize the time and effort involved, methods have been developed to remove mucus and bubbles from the mucosal surface during the procedure to improve the detection of EGC. In Japan, a mixture of water with mucolytic and antifoaming agents (100 mL of water with 20,000 U pronase [Kaken Pharmaceutical, Tokyo, Japan], 1 g sodium bicarbonate, and 10 mL dimethylpolysiloxane [20 mg/mL; Horii Pharmaceutical, Osaka, Japan]) is administered before the procedure. An alternative mixture comprises 100 mL of water mixed with 2 mL of acetylcysteine (200 mg/mL Parvolex [Celltech, UK] or Mucomyst [Bristol-Myers Squibb, USA]), and 0.5 mL (40 mg/mL) activated dimethicone (Infacol; Forest Laboratories, Slough, Berkshire, UK) when pronase is not available. An anticholinergic agent, such as 10 to 20 mg scopolamine butylbromide (Buscopan, Nippon Boehringer Ingelheim Co, Ltd, Tokyo, Japan) or 1 mg glucagon (Glucagon G Novo; Eisai Co, Ltd, Tokyo, Japan), is given just before inserting the endoscope to inhibit peristalsis.
To avoid blind areas during gastroscopic observation, a standardized procedure to map the entire stomach is recommended. A basic technique for avoiding blind areas involves adequate air insufflation to extend the gastric wall to separate the folds, rinsing mucus and froth from the gastric mucosa through irrigation with defoaming agent solution, and mapping the entire stomach. The European Society of Gastrointestinal Endoscopy proposed a protocol for upper gastrointestinal endoscopy that includes taking 4 pictures in the stomach. However, 4 images are not enough to cover and record the whole stomach. The Japanese Society of Gastroenterological Cancer Screening has also published a standard protocol, however, with this protocol it is difficult to remember where to take pictures and how many pictures to take. Recently, Yao has proposed a minimum required standard, the “systematic screening protocol for the stomach (SSS),” as shown in Fig. 3 . With this method, images are arranged according to the order of the procedure, and pictures of 4 or 3 quadrant views are taken in either a clockwise or counterclockwise manner. Because the SSS is proposed as a minimum requirement, these images can be regarded as only checkpoints for the entire observation procedure.
