Society: AGA

INTRODUCTION:
Countries with a high prevalence of gastric cancers have well developed programs aimed at identifying and monitoring gastric cancer precursors. However, there is limited information about the performance of such programs in the United States. In 2015, we initiated a surveillance program to follow patients with gastric intestinal metaplasia (GIM) by upper endoscopy every three years. We aim to gauge the implementation and compliance of surveillance at 7-10 years of follow-up.

METHODS:
Between March 2010 to March 2014, 366 patients had GIM on index upper endoscopy at our large metropolitan safety-net hospital. These patients were followed as part of a prospective cohort study. In our gastroenterology and primary care clinics we implemented a practice to recommend surveillance endoscopy every three years beginning in January 2015, given demographic risk factors of our population. The primary outcome was the proportion of GIM patients who returned for follow-up endoscopy with gastric biopsies. Additional outcomes were predictors of compliance, progression and regression of lesions, and length of time between surveillance.

RESULTS:
Patients identified with GIM on index upper endoscopy were predominantly Hispanic (77%) and female (62.8%), with a mean age of 57.8 years. Of this group, 90 (24.6%) patients underwent surveillance endoscopy. The average time interval was 31.8 months following the index assessment and the number of follow-up endoscopies per patient ranged from 1 to 8.

Of the 34 patients with focal intestinal metaplasia on index endoscopy, 4 (12%) progressed to multifocal metaplasia while the remainder either maintained focal metaplasia or regressed to lower risk lesions. Of the 55 patients with non-focal metaplasia initially, 5 (9%) patients progressed to neoplasia (including neuroendocrine lesions), 20 (36%) maintained multifocal metaplasia, and the remainder (55%) regressed to focal metaplasia or less worrisome lesions. One patient with dysplasia on index endoscopy developed carcinoma and underwent gastrectomy.

Multivariable logistic regression analysis showed that patients with a high-risk indication for index endoscopy based on clinical outcomes research initiative (CORI) terms (weight loss, iron deficiency anemia, abnormal imaging, gastric ulcer, and epigastric abdominal pain) were more likely to undergo surveillance (OR 2.03, CI 1.20 – 3.45) (Table 1). Those found to be H. pylori positive during the index endoscopy were less likely to follow-up for surveillance (OR = 0.47, CI 0.26 – 0.87).

CONCLUSION:
Surveillance of GIM in a high-risk American population has a significant yield, particularly among those with multifocal involvement. However, in the underserved setting, surveillance compliance is suboptimal. Our findings underscore the need to develop strategies to improve participation and compliance in these important programs.

Background: The guidelines recommend gastric cancer surveillance with esophagogastroduodenoscopy (EGD) every 3 years in both extensive and incomplete type gastric intestinal metaplasia (GIM). To date, there has been no promising endoscopic tool to prevent gastric cancer development in these patients. Radiofrequency ablation (RFA) has been used to eradicate intestinal metaplasia in Barrett’s esophagus. This study aimed to evaluate the efficacy of RFA in diminishing and downgrading GIM in those patients with high risk GIM.
Methods: This randomized, self-control trial was conducted in a tertiary hospital in Thailand between June 2020 and October 2022. Patients with histology proven of extensive or incomplete GIM were enrolled in the study. All patients underwent surveillance EGD with white-light and narrow band imaging gastroscopy. The performing endoscopists took biopsy from either suspected GIM area or randomly (negative GIM read by NBI) to complete 5 areas as per the Sydney protocol at both left and right side of stomach (total 10 biopsies). Then, the included patients were randomized with 1:1 concealed allocation to receive either right or left side RFA (Barrx™ 90 RFA Focal Catheter, Medtronic, Thailand). The RFA procedures were repeated at the assigned right or left side of stomach every 2-3 months for 3 times by the same endoscopist. The primary outcome was the complete resolution of GIM (disappearance of GIM epithelium confirmed by histology after RFA at the previous complete or incomplete GIM) at 1 year after RFA. The gold standard for GIM diagnosis was the pathology read by two gastrointestinal pathologists.
Results: Forty-six patients with either extensive or incomplete GIM lesion were enrolled. Male and female was 1:1 with mean age of 66± 8 years. Majority of them (65%) had neither history of smoking nor alcohol drinking. One-third had a history previous H.pylori infection with successful eradication. Half had dyspeptic symptom. The incomplete and complete GIM lesions were not different in RFA vs. observation group (73 vs. 77, p=0.76 and 53 vs. 49, p=0.73, respectively). The complete resolution rate of incomplete GIM lesions after RFA was higher (22/73; 30%) than that of in the observation group (10/77; 13%) (p=0.05). In contrast, the complete resolution rate of complete GIM lesions was not different between RFA and observation groups (27/53; 51% vs. 24/49; 49%, respectively, p=1.00).When compare the downgrading (complete resolution or de-staging) of incomplete GIM lesions, there was no different between RFA and observational group (34/73 (46%) vs. 30/77 (39%), p=0.65) (Table)
Conclusion: This study showed that RFA could facilitate complete resolution of incomplete GIM lesions but not complete GIM. Instead of only diagnostic EGD during surveillance, RFA may be a viable therapy to diminish incomplete GIM.

Background & Aim: Aging is known to affect cancer initiation and progression in various organs including the stomach. On the other hand, the organoid culturing system has enabled us to culture patient-derived tissue samples and investigate the characteristics of these tissues in vitro. In DDW2022, we reported that organoids established from the stomachs of aged C57BL/6 mice exhibited enhanced proliferation due to suppressed cellular senescence as a result of augmented expression of transcription factor T-box 3 (Tbx3). In the present study, we aimed to establish organoids from human gastric intramucosal neoplasia and investigate the molecular mechanisms behind their carcinogenesis from the perspective of aging.
Methods & Results: We established gastric cancer organoids and normal gastric organoids from endoscopically obtained specimens of intramucosal neoplasia and the adjacent mucosae of six patients, respectively. Comprehensive gene expression analysis demonstrated that expression of WNT2B, TCF7, LEF1, TBX3, CCND2, and CDX2 was enhanced and that of DKK3 and SOX2 was diminished in all of the established gastric cancer organoids, suggesting enhanced Wnt/β-catenin signaling and intestinal differentiation in these organoids. Gene set enrichment analysis (GSEA) using array data of these organoids revealed that genes related to Wnt/β-catenin signaling and stem cell proliferation were enhanced in gastric cancer organoids. Indeed, augmented cellular proliferation in gastric cancer organoids was confirmed by 3D cell viability assay, and depletion of Wnt3a and R-spondin1 from the culture led to diminished proliferation, indicating that the proliferation of the organoids was dependent on Wnt/β-catenin signaling. In addition, the GSEA indicated suppressed cellular senescence in gastric cancer organoids. Mechanistically, gastric cancer organoids exhibited enhanced expression of Wnt target genes TBX3 and CCND2, in addition to pRB, and suppressed expression of p16, which led to repression of cellular senescence through G1/S transition. The enhanced Wnt/β-catenin signaling observed in gastric cancer organoids was due to decreased expression of Wnt inhibiting Dickkopf 3 (DKK3), which resulted from the methylation of the DKK3 gene confirmed by methylation-specific PCR. Finally, immunohistochemistry for 49 cases of intramucosal gastric neoplasia revealed a positive correlation between patients' age and the ratio of TBX3-positive cancer cells, which suggested the involvement of aging-related deterioration in gastric carcinogenesis.
Conclusion: Aberrant TBX3 expression in gastric cancer organoids resulted in evasion of cellular senescence, and the ratio of TBX3 positive cells in gastric cancers increased with patients' age. These findings suggested that TBX3 may contribute to aging-related carcinogenesis in human gastric cancers through the DKK3-WNT-TBX3 pathway.

Background: Helicobacter pylori (H. pylori) is a common gastric pathogen that infects approximately half of the world’s population. Chronic infection with these bacteria is considered the principal cause of gastric cancer. One of the bacterial virulence determinants closely linked to tumorigenesis is CagA protein. H. pylori injects CagA into gastric cells, where it acts as a scaffold protein, binding to multiple partners and affecting intracellular signaling, including activation of oncogenic pathways and inhibition of multiple tumor suppressors, among them is key tumor suppressor p53. Regulation of p53 is tightly related to the level of p14Arf (p19Arf in mice) protein, which inhibits E3 ubiquitin ligases MDM2 and ARF-BP1, preventing ubiquitination and proteasomal degradation of p53. Downregulation of p53 and p14Arf is very common for gastric cancers. Notably, the survival of gastric cancer patients is strongly correlated to the level of these tumor suppressors. Our previous studies shown that the downregulation of p14Arf and p53 occurs in strain-specific manner, i.e. H. pylori strains with high tumorigenic potential reveal stronger inhibition, with CagA playing a key role. CagA affects the activity of Arf-specific E3 ubiquitin ligases ULF and SIVA1 in the corpus and the antrum of the stomach to target p14Arf for degradation. In this study we investigated the mechanism of interaction between CagA and host E3 ubiquitin ligases SIVA1 and ULF, which results in downregulation of p14Arf and p53.
Methods: This study was conducted using gastric epithelial cells and organoids in vitro, mouse models and human biopsies in vivo. Samples, infected with H. pylori or transfected with CagA, SIVA1 and ULF expressing plasmids, were analyzed by immunochemical staining, immunoprecipitations and Western blotting. The set of CagA, SIVA1 and ULF mutants were used to determine motifs, responsible for interactions.
Results: Our studies demonstrated that H. pylori infection results in transient activation followed by fast degradation of SIVA1, strong induction of ULF, and subsequent ubiquitination and degradation of p14Arf and p53 tumor suppressors. CagA was found to reproducibly bind to E3 ubiquitin ligases SIVA1 and ULF in a sequential fashion, potentially functioning as an ubiquitin ligase switch that regulates p53 and p14Arf during H. pylori infection. We found that interactions of CagA with ULF and SIVA1 are dependent on tyrosine phosphorylation of CagA protein at the EPIYA motifs. WWE domain of ULF protein is important for CagA binding.
Conclusion: Our studies revealed novel mechanism of oncogenic stress response inhibition through direct binding of CagA to host E3 ubiquitin ligases SIVA1 and ULF, alteration of their activity, resulting in downregulation of p53 and p14Arf. This process is regulated by CagA protein phosphorylation at the EPIYA motifs.

Gastric cancer develops within a carcinogenic process of sequential steps from pre-cancerous metaplasia to dysplasia and adenocarcinoma. The dysplastic stage is considered the key transition state between precancerous and cancerous stages. Cancer cells rewire numerous metabolic pathways to meet their high energy demand for exponential growth and proliferation. However, it is not yet understood how metabolic pathways are involved in the progression of metaplasia to dysplasia. We have examined roles and patterns of metabolic reprogramming during gastric carcinogenesis using a novel mouse model which induces Kras activation in gastric chief cells.
Methods: We utilized a Gif-rtTA;TetO-Cre;Kras^G12D (GCK) mouse model that continuously induces active Kras expression only in chief cells following doxycycline treatment. The GCK mice were sacrificed at 2 to 14 weeks after the Kras induction and immunostaining was performed to examine mucosal changes in the stomachs. Imaging mass spectrometry was performed for spatial profiling of bioactive metabolites highly accumulated in the GCK mouse stomachs. Human and mouse gastric organoids were used for metabolic enzyme inhibitor treatment. The GCK mice were treated with A939572 to block the fatty acid desaturation pathway in dysplastic cells.
Results: The GCK mouse stomachs developed pyloric metaplasia and progressed to high-grade dysplasia within 4 months after the active Kras expression in chief cells. Glycolytic metabolism was first activated in the development of metaplasia and sequentially shifted toward fatty acid metabolism in the progression of metaplasia to dysplasia. In particular, the stomach with high-grade dysplasia showed highly accumulated patterns of monounsaturated fatty acid (MUFA). Stearoyl-CoA desaturase (SCD), a rate-limiting enzyme for the biosynthesis of MUFAs, was specifically expressed in the transitioning area where the cell lineage converts from metaplastic to dysplastic cells in the GCK mouse stomachs. Inhibition of fatty acid desaturation pathway using SCD inhibitor killed 100% of mouse dysplastic organoids within 3 days after treatment, while the dysplastic organoids did not display any significant changes after inhibition of glycolysis. Human gastric organoids exhibited a differential drug sensitivity to the SCD inhibitor depending on the level of SCD expression. Moreover, mouse dysplastic organoids mostly restored their viability by supplementation with the MUFA in the presence of the SCD inhibitor. Finally, the SCD inhibitor selectively killed dysplastic cells in the GCK mouse stomachs.
Conclusions: Our study indicates that active Kras expression only in gastric chief cells drives the full spectrum of gastric carcinogenesis and the metabolic switch from glycolysis to fatty acid metabolism is critical for metaplasia progression during gastric carcinogenesis.

Aim: Loss of parietal cells causes the development of spasmolytic polypeptide-expressing metaplasia (SPEM) through transdifferentiation of chief cells. However, whether initial chief cell loss cause SPEM needs to be further investigated. Acidic mammalian chitinase (CHIA) belongs to the 18 glycosidase family and is expressed in epithelial cells in various organ, which can degrade chitin-containing pathogens, participate in Th2-mediated inflammation. CHIA is highly expressed in the chief cells, and our previous study showed that loss of CHIA promotes gastric cancer cell proliferation and progression (Hu et al., KDDW 2021), indicating that it is a tumor suppressor in gastric carcinogenesis. However, the molecular and cellular basis of CHIA function remains to be fully understood.
Methods: Histopathological and immunohistochemical (IHC) analysis, as well as In Situ Hybridization (ISH) with specific markers were performed in CHIA wild type mice (WM) and knockout mice (KM) at different age. CHIA mRNA and protein expression were measured by qPCR and Western-blot in human normal gastric epithelium, SPEM and gastric cancer tissues.
Results: With advanced of age, CHIA deficient mice exhibited severe gastric pre neoplastic phenotype, including chief cell loss followed with chronic inflammation, SPEM, and high-grade intraepithelial neoplasia (HGIN). However, slight parietal cell loss was only found in the HGIN stage. Furthermore, deletion of CHIA in the chief cells resulted in upregulation of the specifical markers of SPEM, including TFF2 and MUC6. CHIA KM showed that progressively and significantly reduction of gastric stem cell (GSC) markers, including Lgr5 and Mist1 from 8 days to 15 months after birth. However, CHIA deletion in mice resulted in mild decrease of GSC maker for parietal cells (PCs) differentiation- Lrig1, as well as PCs labeling markers, including H⁺-K⁺-ATPase, Slc26a9 and SHH. Additionally, spontaneous inflammation was observed by upregulation of IL33, IL25, IL-6, IL-13 and IL17 in CHIA KM. Moreover, absence of CHIA caused epithelial cell proliferation, which was detected by Ki67, but suppressed apoptosis, including BCL-2, BCL-xl, CyclinD1 via activation of STAT3 signaling axis. Consistent with animal experiment, CHIA mRNA and protein expression gradually decreased in SPEM (n=78) and gastric cancer tissue (n=93) compared with human normal gastric epithelial (n=80), demonstrating that loss of CHIA expression in the chief cells is first step of normal gastric epithelial progressing to SPEM and malignant pathological process.
Conclusion: CHIA deficiency in the chief cells is the key event to induces SPEM and progress to gastric cancer, which is independent of parietal cell loss. CHIA functions as a protective role in safeguarding gastric epithelial cells against aberrant growth factor signaling and the resultant cellular plasticity and stemness.

Chunli Hu

Appearances