Advances in endoscopic diagnosis of early gastric cancer

Reviewed by An Yongping and Liu Zhibing Yan Wenming (Reviewer) Department of Radiotherapy, Affiliated Hospital of Inner Mongolia Medical University Yan Wenming
                   Chinese Journal of Practical Medicine, Vol. 6, No. 24, 2006
Gastric cancer is one of the most common malignant tumors in clinical practice. Early detection and early treatment are the keys to improve the efficacy, and gastroscopy is the most direct, accurate and reliable diagnostic method. The survival rate of early gastric cancer is >90% in 10 years after surgery and nearly 100% for micro gastric cancer, while the survival rate of middle and late gastric cancer is only about 30% in 3 years despite active surgery, chemotherapy and radiotherapy. It takes 2-7 years for early gastric cancer to develop into progressive gastric cancer, with an average of about 3 years [1]. At present, there are many types of gastroscopes used for clinical examination, each with its own characteristics. General electronic gastroscopy has been widely used in clinical practice, while magnifying endoscopy, pigmented endoscopy, fluorescent endoscopy, ultrasound endoscopy and near-infrared electronic endoscopy have their unique values and roles in the diagnosis of early gastric cancer. They should be selected in clinical practice to improve the diagnosis rate of early gastric cancer.
1 Definition of Earlygastriccancer (EGC)
  Early gastric cancer (EGC) is defined as gastric cancer with vertical infiltration not exceeding the submucosal layer with or without metastasis, i.e. early gastric cancer = mucosal cancer (M cancer) + submucosal layer cancer (SM cancer) [2]. Endoscopically, early gastric cancer can be divided into (naked eye classification) type I (elevated type), type II (flat type), and type III (sunken type), where type II can be further subdivided into type IIa (flat elevated type), type IIb (flat type), and type IIc (flat sunken type). Generally, EGC is 1-4 cm in diameter, and those less than 0.5 cm are called microscopic gastric cancer [3].
The prognosis of EGC is good, with a 5-year survival rate of more than 90%, and that of microscopic gastric cancer is almost 100% [4], while the 5-year survival rate of progressive gastric cancer is only 30%-40% despite prompt surgery, radiotherapy, and/or chemotherapy after diagnosis and other aggressive treatments. It takes about 2-7 years to progress from early stage to progressive stage, with an average of about 3 years. Therefore, timely detection and treatment of early gastric cancer is of great importance to improve the treatment effect of gastric cancer [3].
2 Endoscopic diagnosis of early gastric cancer
2.1 General gastroscopy At present, electronic endoscopy has been widely used in clinical practice, and its imaging is mainly done by the micro image sensor (CCD) installed at the apex of the endoscope, and then the image is displayed on the TV monitor after being processed by the video processor. Electronic endoscopy has a clear image, no blind area, and can be displayed on a color screen for multiple physicians to view, so that the small, controversial lesions can be discussed while observing, so the chance of missing early lesions is smaller, which can improve the diagnosis rate of EGC. However, the endoscopic manifestations of EGC are not characteristic, so it is mainly dependent on the endoscopist to make a comprehensive and careful observation, especially to pay attention to small elevated or depressed lesions and superficial erosions, and to take biopsies for pathological examination in suspicious areas, and if necessary, to perform large gastric mucosal biopsies, which is an important part of the detection of EGC [5].
With general gastroscopy, morphological changes in the stomach can be directly observed, and lesions can be biopsied, and endoscopic multiple biopsies can increase the positive diagnostic rate. The detection rate of early gastric cancer under plain gastroscopy in China is only 15%-20% [6] EGC alertness. Endoscopy can be routinely performed in patients over 40 years old with unexplained upper abdominal symptoms, and gastroscopy should be reviewed regularly in patients with chronic gastric disease. Patients with moderate to severe atypical hyperplasia on gastroscopic biopsy should repeat gastroscopy and biopsy several times to avoid delaying the diagnosis.
When performing endoscopy, attention should be paid to: 1) elevated or depressed mucosal changes; 2) ulcerative or erosive changes; 3) abnormal hue; 4) abnormal mucosal folds; 5) bleeding susceptibility; 6) changes in vascular translucency; and 7) sclerosis and deformation of the gastric wall. However, the final diagnosis is based on the pathological tissue findings, and when the above changes are solitary or limited, aggressive biopsy of the suspicious lesions is extremely important [7].
Table 1 Classification of gastric cancer by depth of infiltration
Early stage gastric cancer
M carcinoma: Carcinoma confined to the mucosa; 
SM carcinoma: carcinoma infiltrating into the submucosa
Progressive gastric cancer
MP carcinoma: carcinoma infiltrating into the lamina propria;
SS carcinoma: Infiltrates into the subplasma layer;
S carcinoma: carcinoma infiltrating into the plasma membrane;
2.2 Ultrasound endoscopy (EUS) Early gastric cancer has different sonograms depending on the type. The accuracy of EUS in distinguishing EGC from progressive gastric cancer can be 70% to 80% [4], and EUS can still clearly determine EGC that is not easily identified by the naked eye, i.e., the cancer is confined to the submucosa without obvious morphological and color changes on the mucosal surface. Liu Jun [8] examined 119 cases of early gastric cancer with EUS: 64 cases of mucosal infiltrating carcinoma and 55 cases of submucosal infiltrating carcinoma, and the compliance rate was 93.4% when compared with pathological specimens. The average size of the nests was 24.7×21.3 mm, which was not significantly different from the pathological specimens (P>0.05). 37 cases were operated, 25 cases had positive lymph node metastasis and 12 cases had negative lymph node metastasis in the preoperative EUS examination, and the compliance rate with mucosal resection and postoperative pathological examination was 95.2%. It can be seen that EUS can accurately determine the depth of infiltration and lymph node metastasis of early gastric cancer, and its diagnosis of the extent of infiltration of cancer nests is better than that of ordinary endoscopy, and it has a high compliance rate with pathological specimens. In addition, a recent study showed that three-dimensional endoscopic ultrasound (3D-EUS) can image early gastric cancer well and thus assess the depth of tumor infiltration [3].
The ultrasound endoscope is a miniature high-frequency ultrasound probe placed on the tip of the endoscope, which allows direct observation of the lumen morphology through the endoscope and real-time ultrasound scanning to further obtain ultrasound images of the layers of the gastric wall and surrounding adjacent organs; it can determine the site and extent of the lesion, as well as the depth of infiltration of the lesion, the presence of invasion of adjacent organs and the presence of surrounding lymph node enlargement. In a retrospective survey conducted by Mancino et al [9], the preoperative ultrasound gastroscopy of 79 patients was compared with the histopathological analysis after surgery, and the compliance rate was 87.3% for early-stage gastric cancer and 91.1% for progressive gastric cancer. The compliance rate was 91.1% for progressive gastric cancer. However, it is difficult to distinguish hypoechoic lesions from tumors or inflammation and fibrosis by ultrasound gastroscopy, and relying on fine-needle aspiration biopsy under ultrasound endoscopy is one of the effective means to solve this problem [5].
In addition, if the infiltrated site forms an ulcer or is an ulcerated carcinoma, the depth of infiltration is often overestimated due to the inflammatory response and fibrotic lesions deep in the ulcer, and it is possible to diagnose early gastric cancer as a progressive stage. According to a Japanese report, the overall accuracy of EUS in identifying EGC and progressive gastric cancer was more than 80%, and the accuracy rates of three types of cancer foci, namely, confined to the mucosa, involving the submucosa, and invading the plasma membrane, were 80.4%, 78.8%, and 60%, respectively. It can be seen that EUS can accurately determine the depth of infiltration and lymph node metastasis of early gastric cancer, and its diagnosis of the extent of infiltration of cancer nests is better than that of general endoscopy, and it has a high rate of agreement with pathological specimens. In addition, a recent study has shown that three-dimensional endoscopic ultrasound (3D-EUS) can provide good imaging of early gastric cancer to assess the depth of tumor infiltration [10].
2.3 Pigmented endoscopy The accuracy of early gastric cancer diagnosis can be improved by staining the gastric mucosa with drugs sprayed on the mucosa and observing the color change of the mucosa [6]. The advantages of pigmented gastroscopy include: different staining of benign and malignant lesions, which makes it easy to make differential diagnosis; more accurate judgment of the cancerous area, which can increase the positive biopsy rate of gastric cancer; the ability to observe the size, shape and arrangement of gastric cells; and the ability to show the small bumpy changes on the mucosal surface. Early gastric cancer is not easily detected under normal gastroscopy and is easily missed, so staining can improve its diagnosis rate [3].
The use of certain dyes to stain the mucosal tissues under endoscopy can more clearly highlight the original face of the lesion. In particular, methylene blue solution staining shows light staining for enterosis and atypical hyperplasia, and dark blue for cancerous mucosa, while inflammatory lesions such as normal mucosa, ulcers and edges of erosions are not stained, so that the site of EGC can be effectively detected and its extent can be more accurately determined, which is useful for biopsy [11].
The pigmented gastroscopic methods commonly used for the diagnosis of early gastric cancer include fluoroscopic gastroscopy, exogenous fluorescent substance injection (e.g., sodium fluorescein staining, hematoporphyrin derivative fluorescence detection), autofluorescence diagnosis, fluorescein-labeled CEA monoclonal antibody for the diagnosis of gastric cancer, brilliant blue staining, combined Congo red 2 US blue staining, indigo carmine intraperitoneal arterial injection, indigo carmine oral and spray staining, US blue Oral staining method, toluidine blue staining method, methylene blue 2 indigo carmine staining method, Congo red 2 Evans blue staining method, etc. Among the various staining methods mentioned above, brilliant blue staining method and fluorescence method are more effective [12]. Brilliant blue staining of benign and malignant lesions has contrasting hues, and endoscopically benign lesions are blue and malignant lesions are red, making them easy to identify. The fluoroscopic method is complementary to conventional gastroscopy with the latest technology in fluorescence-based gastroscopic imaging and gastrointestinal spectroscopy, with exogenous fluorescent substance injection and autofluorescence diagnosis, which provides reliable detection results and allows monitoring of disease progression. The use of pigmented gastroscopy has greatly improved the detection rate of early gastric cancer and is widely used. The ideal method is to label monoclonal antibodies to gastric cancer with pigment or fluorescence, which bind specifically to gastric cancer cells and can be easily observed under gastroscopy. In-depth study of pigments with high sensitivity and specificity will make pigmented gastroscopy more promising [3].
2.4 Magnification endoscopy Attempts to determine the nature of lesions from subtle changes in the gastric mucosa using magnification endoscopy have been made long ago. The magnifying endoscope, equipped with a zoom lens, allows the endoscopist to observe changes in the microstructure of the gastrointestinal mucosa in order to determine the benignity and malignancy of the lesion, to distinguish the histological type, and to determine the depth and extent of the lesion. In the case of the stomach, the observation mainly includes 2 aspects: pit pattern and microvascular pattern. Since there is no perfect and unified standard for the typing of gastric dimples, reports on the characteristic manifestations of dimples in EGC vary. Tajiri et al [13] reported that under magnification endoscopy, the dimples of bulging EGC were irregularly shaped, the mucosal microvessels of tubular adenocarcinoma were irregular, thicker and shorter than those of hyperplastic polyps or adenomas, and the microvessels of papillary adenocarcinoma were long and tortuous, whereas in In depressed EGC, the dimples were mostly irregular and tubular in shape, and the dimples were smaller in size compared with the surrounding mucosa, and the microvasculature was finely reticulated and spiral in hypofractionated EGC. (2) irregular shape; (3) disappearing and indistinct microscopic morphology, mostly found in poorly differentiated tumors. Niwa et al [15] reported that EGC is characterized by small concave size, variable morphology, irregular dendritic shape and irregularly shaped tumor vessels. Nakayoshi et al [17] reported that the local microvascular morphology in early gastric cancer was mainly of three types: (1) fine mesh; (2) spiral; and (3) irregular. Although the above findings are not completely consistent, we can see that the small concavity, irregular shape and the disappearance of normal collecting veins and true capillary network in the mucosa, as well as the appearance of tumor neovascularization with irregular diameter and shape are the more characteristic manifestations of EGC endoscopically.
This may be due to the fact that in the early stage when there are few new tumor vessels, the diffuse infiltration of cancer cells from the poorly differentiated tumors damages the normal blood vessels, while the damage to the normal vessels from the early highly differentiated cancers is not obvious [18,19]. To determine the depth of tumor infiltration, it is generally believed that the degree of submucosal cancer affecting the pattern of collecting veins and true capillaries is related to the density of cancer cells in the lamina propria, and obviously the effect of intramucosal cancer on collecting veins and true capillaries is relatively smaller. Of course, these can only be used as an auxiliary basis for judgment. Although many reports have confirmed the advantages of magnification endoscopy in the diagnosis of EGC compared with ordinary endoscopy, due to the lack of unified diagnostic criteria and the fact that magnification endoscopy is easily influenced by gastrointestinal motility and respiratory movements, and the different magnification of zoom also affects the analysis of images, magnification endoscopy is still in the early stage of research and further experience is needed for diagnosis [11].
While conventional electronic gastroscopy has some magnification, the magnification of magnifying endoscopes can range from several times to a maximum of 170 times. Nowadays, it has been developed from fiberoptic magnification endoscopy to electronic magnification endoscopy, which can clearly distinguish the shape of the opening of the mucosal ducts of the digestive tract and the microscopic changes of the lesions, and combined with mucosal pigment staining, it can more accurately reflect the pathological background of the lesions, distinguish proliferative, adenomatous and cancerous lesions, and improve the detection rate of flat and indurated early cancers [20]. It can improve the purpose of biopsy and avoid unnecessary biopsy trauma. The microstructure of the gastric mucosa has been divided into dotted type A, short linear type B, banded type C and reticulated type D. The irregular type D structure is the characteristic change of gastric cancer [21 ]. The sensitivity of magnification endoscopy for the diagnosis of small gastric cancer and EGC was reported to be 96.0% with a specificity of 95.5%, and the fine mucosal structures and microvascular features observed by it had high correlation with histopathological diagnosis [22-24].
Magnified endoscopy of gastric cancer requires observation of fine structures and microvascular patterns on the mucosal surface. Small depressions, depressions of different sizes, irregular branching depressions and irregular vessels characterize the surface structures of early gastric cancer. Small depressions were common in differentiated early gastric cancer (88%), while undifferentiated type was relatively rare (50%). Analysis of microvascular patterns and fine surface structures was important for magnified endoscopic observation using 0.1% indocyanine. The use of magnified gastroscopy on the basis of gastric mucosal staining can improve the detection rate of EGC even more [10].
2.5 Fluorescence endoscopy Biological tissues can produce fluorescence under light excitation, and the fluorescence spectra of tumor tissues and normal tissues are different. Xiao Shudong et al [25] studied the fluorescence spectra of gastric cancer tissues in vivo under endoscopy and found that the autofluorescence spectra of gastric cancer and some patients with chronic atrophic gastritis were different, and the sensitivity of gastric cancer diagnosis was 82.9%, the specificity was 91.5%, and the accuracy was 87.5% based on the ratio of primary peak 510 nm to secondary peak 710 nm fluorescence intensity.
The endogenous fluorescence groups in tumor tissues and the specific fluorescence signal induced by laser were used to determine the nature of the tissues. It has the advantages of rapid and easy detection of early gastrointestinal tumors and precancerous diseases, real-time detection of lesions, and help to guide biopsies. Endoscopic fluorescence imaging has the following advantages: it can theoretically detect suspicious areas in all the tissues observed by endoscopy. Therefore, endoscopic fluoroscopy is less dependent on the experience of the endoscopist [26]. A total of 54 lesions (33 EGC and 21 benign lesions) in 52 patients with benign and malignant gastric disease have been detected by fluoroscopy with a positive rate of 85% (28/33) and a diagnostic sensitivity and specificity of 94% and 86%, respectively. The endoscopic intrinsic fluorescence biopsy technique currently used can automatically identify and diagnose tissues based on their intrinsic fluorescence spectral characteristics, which can immediately indicate whether the tested tissues are normal, benign lesions or early gastric cancer tissues, with a high positive compliance rate, which can help to correctly assess the extent of early gastric cancer lesions, and this deserves further in-depth study [5].
2.6 Capsule endoscopy
The capsule endoscope, also known as wireless capsule endoscopy, was produced by GIVEN imaging company in Israel in April 2000 and formally launched, its trade name is GIVEN diagnostic imaging system, which was approved by the U.S. Food and Drug Administration (FDA) in August 2001 and approved for clinical use, and was approved by the Chinese State Food and Drug Administration (SDA) in April 2002. In April 2002, the GIVEN diagnostic imaging system was approved by the State Food and Drug Administration (SDA) of China for clinical use in China. Capsule endoscopy is mainly used for occult gastrointestinal blood loss and other small bowel diseases [27], but also for patients with suspected gastrointestinal diseases who cannot complete, tolerate, or cooperate with conventional endoscopy and other examinations for other reasons [28]. Because of the large gastric cavity and many folds, capsule endoscopy can only observe part of the mucosal folds, with more blind areas, and the focal length of the camera of capsule endoscopy is only a few millimeters, so the diagnosis of intra-gastric lesions is limited.
2.7 Near-infrared electron endoscopy
Near-infrared light penetrates deeply into the tissue, whereas the light from conventional endoscopy does not [29]. In vivo spectrophotometry showed that infrared light penetrates the abdomen and stomach wall at 620-820 nm