Reviews focusing on the clinical aspects of GA-FG remain limited despite the increasing number of recorded cases. According to the 2019 WHO Classification of Gastric Tumors, the typical onset age for GA-FG is 60–70 years(6). In Japan, the reported average age of GA-FG is 67.7 years (range 42–82 years)(20). In our study, OGA patients had an average age of 50.7 years, while GA-FG patients averaged 56.3 years, and 55.7 years for other patients. Additionally, a gender bias is present in our cases, where a global male-to-female ratio of 2.2 was observed, whereas in Japanese GA-FG cases, the ratio is 1.4 (20). In contrast, the male-to-female ratio for GA-FG is 1:2 in our study, possibly limited by the small-sample-size and single-center nature of this study.
GA-FG initially presents in Hp-negative and non-atrophic mucosa and is classified as an Hp-negative gastric cancer. Numerous studies have reported GA-FG as Hp infection irrelevant. However, some studies propose a positive correlation between fundic gland polyps and fundic adenocarcinoma, even in patients without a history of continuous proton pump inhibitor (PPI) use or Hp infection/eradication(21). Their rationale is that the prevalence of Hp infections has risen concurrently with the increase in these cases. Differentiated adenocarcinomas are believed to arise from intestinal metaplasia, which is often associated with Hp infection. Most FG-GA originates in the non-atrophic fundic gland area, they found that in 15 out of 20 patients, lesions predominantly occurred in the non-atrophic gastric mucosa, regardless of Hp positivity or eradication status(6). Currently, it remains unclear how Hp influences GA-FG progression and severity. Mucosal atrophy and Hp infection status may not be primary factors in GA-FG development(18). In this study, none of the GA-FG patients had an Hp infection, aligning with previous research findings.
In terms of GA-FG tumor size, most reported tumors have a diameter of less than 1 cm, though some reach up to 8.5 cm(18). The risk of GA-FG transplantation and invasion increases with tumor diameter due to greater submucosal invasion. Sarcomeric GA-FG cancers are most commonly 0-IIa lesions, followed by 0-IIb, with 0-IIc and 0-I lesions being less frequent. Most lesions showed submucosal (SM) infiltration, predominantly SM1 infiltration. Fewer lesions exhibited SM2 or deeper infiltration, with intramucosal lesions being less common. The disease's clinical presentation often includes frequent SM infiltration, even in lesions of a few millimeters in size(6). Some studies note that GA-FG can be identified through various methods(19). As the tumor evolves, the fundic glandular type may transform into variants like the undifferentiated or tiny concave types, which possess a greater malignancy potential than GA-FG's predominantly cellular differentiation. In our study, the average size of GA-FG was 0.73 cm, and OGA was 0.53 cm. This is consistent with previous reports. It is uncommon to find multiple GA-FG lesions, as most of them are solitary. Although our study focused on single lesions, we did notice reports of patients with two simultaneous lesions in the non-atrophic zones of Hp active atrophic gastritis(21).
Typical endoscopic diagnosis signals of GA-FG include discolored tones, tumor surface vasodilation, and an SMT-like appearance. In a previous Janapnese report(22), discolored lesions, as the hallmark of GA-FG, were noted in 52 out of 67 (77.6%) under endoscopy. Additionally, vascular dilation of the tumor surface was observed in 49 out of 67 cases (73.1%) in the same report. It is important to differentiate between vasodilation caused by Stomach oxyntic gland neoplasms and that caused by MALT lymphomas(23), neuroendocrine tumors, and carcinomas. Stomach oxyntic gland neoplasms and neuroendocrine tumors share similar morphological characteristics, making it difficult to distinguish between the two. Stomach oxyntic gland neoplasms are typically found in Hp-negative environments and originate in the deep mucosal layer of tumor bulging type lesions with SMT-like changes, while neuroendocrine tumors originate in the deep mucosal layer and extend to the submucosal layer, also with SMT-like changes. Both ultrasound endoscopy and white light endoscopy can only show as deep as submucosal hypoechoic occupations, which makes ultrasound endoscopy difficult to distinguish between the two, and white light endoscopy is even worse. Gastric fundic gland polyp is originated in the epithelial layer, with a clear boundary and visible elevation, and the surface microstructure is the regular structure of the gastric fundus gland. In contrast, GA-FG has poorly defined borders, uneven microstructural expansion, extended epithelium at the crypt margins, and crypt middle region enlargement. As a result, biopsy or excision for pathologic distinction is the most effective method.
Consequently, vasodilation on the tumor surface is an indication of lesions in the deep mucosal/submucosal layers, which is common to fundic glandular gastric cancers proliferating in the submucosa, but not exclusive to GA-FG. Lesions, primarily located in the mid to deep mucosa and covered by normal superficial mucosa, are similar to SMT. SMTs are appropriately described as subepithelial tumors, as they are named after their predominant location in the submucosal layer. NBI endoscopic magnification reveals features of GA-FG with indistinct boundaries, expanded glandular apertures, increased interfollicular areas, and the absence of asymmetric microvessels. In our study, all solitary tumors were located in the fundus and body of the stomach. Magnified endoscopy on 14 patients of OGA and GA-FG showed negative boundaries, positive microstructure, and negative microvessels, attributed to surface discoloration, filamentous erythematous changes, and clear capillary dilation. Subepithelial symptoms were observed in all GA-FG patients and in seven cases of OGA. White light endoscopy revealed that most GA-FG lesions exhibited an elevated morphology, similar to an SMT. In our cases, OGA and GA-FG lesions were predominantly yellow, discolored, and reddish, consistent with high-grade lesions (Fig. 3).
GA-FG cells largely express MUC6, while MUC5AC-positive cells, indicative of indurated epithelium, are rare, showing scant P53 expression and a low Ki-67 index (average 3.6%)(24). Consequently, GA-FG cells exhibit minimal invasiveness and low proliferative activity(5). In our study, Fourteen OGA specimens were stained with MUC6, and out of which, 11 were diffusely positive. In the GA-FG group, all specimens were MUC6 positive. Ki-67 labeling index was expressed equally in both groups with 0.05. 4 of the 10 OGA specimens were positive for Syn, but all GA-FG specimens were positive for Syn. And 12 of the 14 OGA specimens were positive for CgA, but for GA-FG there were no specimen showed positive expression of CgA. The immunohistochemical staining results agreed with positive and negative MUC2 reports in past literature. Consequently, it is challenging to examine changes in the stomach's deepest layers, where GA-FG and OGA tumor components reside. Therefore, magnification endoscopy is not a reliable diagnostic tool, but rather a supplementary evaluation method. Conversely, magnified NBI identified some GA-FG lesions, along with typical mucosal vascular pattern (MVP) and mucosal surface pattern (MSP). Abnormal MSP and MVP observed under magnification staining may indicate GA-FG(14). Chiba et al. monitored ten GA-FG lesions and found no morphologic changes after 16 months(6).
GA-FG tumors typically exhibit slow growth, low malignancy potential, and minimal invasiveness. Although the submucosa is frequently involved, most GA-FG tumors are benign. The GCTG recommends local excision or gastrectomy with lymph node dissection for typical gastric cancers(24). Treatment decisions consider tumor size, histologic type, and estimated depth of invasion. In recent years, most GA-FG cases have been treated with surgical resection via ESD(24).
Overall, this study reports 24 cases of stomach oxyntic gland neoplasms. While most clinical variables align with previous reports, a few of them, such as gender bias, were observed among the three groups in this study. Other key features, such as endoscopy, the hall marker stating, and treatment methods, were characterized throughout this study. We believe our findings provide valuable insights into the clinical and pathological characteristics of stomach oxyntic gland neoplasms, contributing to the understanding and management of this rare gastric cancer subtype.