Society: AGA

Background: Gut microbiota composition and diversity have been shown to influence the effectiveness of cancer immunotherapies. Recent evidence suggests that Heliobacter pylori infection may reduce the efficacy of immune checkpoint inhibitors (ICIs), but the validity of this association in large, diverse cohorts of gastric cancer patients is unknown. We aimed to determine the impact of H. pylori infection status on survival outcomes in individuals with gastric cancer undergoing ICI therapy.

Methods: This single-center, retrospective study included all individuals with primary gastric cancer undergoing treatment with ICIs and with documented H. pylori status at Memorial Sloan Kettering Cancer Center between July 2013 and October 2021. Positive H. pylori status was defined as a history of prior or active infection obtained via urea breath test, stool antigen test, and/or histopathology. Negative status was defined as explicitly negative infection testing; unknowns were excluded. The primary outcomes were progression-free survival (PFS; calculated as date of ICI initiation to date of first progression or death) and overall survival (OS; calculated as date of ICI initiation until death). Secondary outcomes included PFS and OS stratified by the usage of concurrent chemotherapy and demographic or histological differences between groups. Comprehensive reviews of medical charts, endoscopic procedures, and pathology reports were performed to identify H. pylori diagnoses and confirm outcomes.

Results: Of 215 ICI-treated gastric cancer patients meeting the inclusion criteria, 49 had a documented history of H. pylori infection prior to or upon cancer diagnosis and 166 had no documented infection. Compared to H. pylori negative patients, H. pylori positive individuals were more likely to be younger, non-white, and Hispanic while also demonstrating higher rates of non-cardia gastric cancers (73.5% vs. 41.6%, p<0.01) and intestinal-type histology (81.6% vs. 65.1%, p=0.08) (Table 1). The H. pylori positive group had significantly shorter PFS and OS than the negative group with an estimated median PFS of 3.2 months vs 6.8 months (HR 1.96, p<0.01) and an estimated median OS of 9.8 months vs 17.9 months (HR 1.54, p=0.02) (Figure 1). These findings persisted when excluding patients receiving concurrent chemotherapy. Multivariable analysis confirmed H. pylori status as an independent predictor of PFS (HR 1.89, p<0.01) and OS (HR 1.80, p<0.01).

Conclusions: In the largest study to date, H. pylori infection was found to be associated with shorter PFS and OS in gastric cancer patients undergoing ICI therapy. This suggests that H. pylori status may be a predictive marker of immune responsiveness. Future studies are needed to elucidate the specific mechanisms by which this immune regulation occurs and whether treatment of active infections would benefit immunotherapy outcomes.

Clinical, treatment, and outcome characteristics of gastric cancer (GC) patients treated with immune-checkpoint inhibitors (ICIs) by <i>Helicobacter pylori</i> infection status

Clinical, treatment, and outcome characteristics of gastric cancer (GC) patients treated with immune-checkpoint inhibitors (ICIs) by Helicobacter pylori infection status

<b>Figure 1.</b> Survival curves of progression-free survival (PFS) and overall survival (OS) in immune checkpoint inhibitor (ICI)-treated gastric cancer patients by <i>Helicobacter pylori </i>infection status: A, PFS in all patients receiving ICIs and concurrent chemotherapy; B, OS in all patients receiving ICIs and concurrent chemotherapy; C, PFS in patients only receiving ICIs; D, OS in patients only receiving ICIs

Figure 1. Survival curves of progression-free survival (PFS) and overall survival (OS) in immune checkpoint inhibitor (ICI)-treated gastric cancer patients by Helicobacter pylori infection status: A, PFS in all patients receiving ICIs and concurrent chemotherapy; B, OS in all patients receiving ICIs and concurrent chemotherapy; C, PFS in patients only receiving ICIs; D, OS in patients only receiving ICIs

Background. Gastric premalignant lesions, including atrophy, intestinal metaplasia, and dysplasia, are believed to develop into gastric cancer through multiple steps. Inflammation plays a vital role in the progression of lesions. This study aimed to explore the characteristics of macrophages and macrophage-related gene expression in premalignant gastric lesions to provide novel insights for the biomarkers and treatment targets.
Methods. We applied an algorithm of CIBERSORT to estimate the relative proportions of immune cell subsets in tissues based on the dataset (GSE55696), which was downloaded from the GEO database. Weighted gene co-expression network analysis (WGCNA) was applied to identify inflammation-related modules and hub genes. The datasets (GSE150290 and GSE134520) were used for single-cell analysis to identify pathology-related macrophage sub-clusters. Overlap genes in hub genes and macrophage sub-cluster were used for functional enrichment analysis. Gene set enrichment analysis(GSEA) analysis was carried out in mRNA-seq datasets to gain insight into the biological processes.
Results. A module (greenyellow) was found to be positively correlated with pathological changes and highly related to inflammation scores(figure 1A). The single-cell analysis identified the macrophage subpopulation and found that S100A8 and S100A9+ macrophages constituted a significantly larger proportion in EGC tissues (figure 1B-G). The functional enrichment analysis of the overlapping gene of greenyellow module and S100A8 and S100A9+ macrophages subpopulation found that S100A8 and S100A9+ macrophages might induce gastric tumorigenesis through the NFκB Signaling Pathway(figure 2A-F). GSEA analysis suggested that genes S100A8 and S100A9 could regulate the NUP98_HOXA9(FDR q-value=0.067, FWER p-value=0.047) and AML1_MTG8(FDR q-value=0.133, FWER p-value=0.047) protein fusion(figure 2G, H), which were known to affect the malignant proliferation of monocyte-macrophages.
Conclusions. The S100A8 and S100A9+ macrophages might promote gastric tumor progression through the NFκB signaling pathway. Monocyte-macrophage proliferation might play a crucial role in gastric cancer progression in gastric premalignant lesions.

Figure 1. WGCNA and single-cell RNA analysis of the gastric premalignant lesions. (A) Module-trait relationships of gastritis, LGIN, HGIN, EGC, and the absolute score of immune cells. (B) The tSNE plot of cell clusters. (C)& (D) AIF1 and CD14 expression distribution map. (E) The tSNE plot of macrophage cell clusters. (F) Percentages of cell clusters identified by pathology classification. (G) The relative expression heat map for macrophage cell cluster markers. Only the top five are shown.

Figure 1. WGCNA and single-cell RNA analysis of the gastric premalignant lesions. (A) Module-trait relationships of gastritis, LGIN, HGIN, EGC, and the absolute score of immune cells. (B) The tSNE plot of cell clusters. (C)& (D) AIF1 and CD14 expression distribution map. (E) The tSNE plot of macrophage cell clusters. (F) Percentages of cell clusters identified by pathology classification. (G) The relative expression heat map for macrophage cell cluster markers. Only the top five are shown.

Figure 2. (A) The tSNE plot of macrophage cell clusters. (B) & (C) S100A8 and S100A9 expression distribution map. (D) Venn plot of overlapped genes in the greenyellow module and S100A8 and subpopulation of S100A9+ macrophage. (E) GO analysis in overlapping genes. (F) KEGG pathways analysis in overlapping genes. (G) & (H) GSEA analysis of all genes in the greenyellow module.

Figure 2. (A) The tSNE plot of macrophage cell clusters. (B) & (C) S100A8 and S100A9 expression distribution map. (D) Venn plot of overlapped genes in the greenyellow module and S100A8 and subpopulation of S100A9+ macrophage. (E) GO analysis in overlapping genes. (F) KEGG pathways analysis in overlapping genes. (G) & (H) GSEA analysis of all genes in the greenyellow module.

Background and Aim: A subset of myeloid-derived suppressor cells (MDSCs) express Schlafen4 (Slfn4) in Helicobacter-infected stomachs coincident with the development of spasmolytic polypeptide-expressing metaplasia (SPEM). Aim: To define the identity of SLFN4⁺ cells and role of SLFN4 in these cells. Methods: We performed single-cell RNA sequencing on the CD45 pre-sorted immune cell population from PBMCs and stomachs with or without 6 mos of H. felis infection. Knockdown of Slfn4 by siRNA or PDE5/6 inhibition by sildenafil was performed in vitro using thioglycollate-elicited peritoneal myeloid cells. Intracellular ATP/GTP levels and GTPase enzyme activity of the immunoprecipitated SLFN4 complex was measured using the GTPase-Glo assay kit. The intracellular level of ROS was quantified using DCF-DA fluorescent staining, and apoptosis was determined by cleaved Caspase-3 and Annexin V expression. Gli1CreERT2 x Slfn4^fl/fl mice were generated and infected with H. felis. Sildenafil was administered twice over 2 weeks by gavaging H. felis infected mice ~4 months after inoculation once SPEM had developed. Gastric metaplasia was demonstrated by histological staining, western blots and qPCR for metaplastic markers as well SLFN4 and NOS2 levels. CD4/CD8 T cell numbers were quantified by flow cytometry. TNFα levels from gastric MDSCs was determined by ELISA. Results: Cell transcriptomes--13,076 from PBMCs and 8,394 from stomachs were analyzed. After H. felis infection, Slfn4 in gastric immune cells was highly expressed in both monocytic and granulocytic MDSCs. Both MDSC subsets exhibited a strong transcriptional signature for type I interferon responsive GTPases (e.g., GBP2,5) (Fig. 1). The increase in GTPase activity correlated with increased NOS2 activity and production of nitric oxide (NO), a cofactor for soluble guanylate cyclase. In vitro analysis of IFNα-induced MDSCs in culture showed that immunoprecipitation of SLFN4-containing complexes exhibit GTPase activity (Fig. 2), while knocking down Slfn4 or treatment with sildenafil reduced GTP and NOS2, yet enhanced generation of total cell ROS. Blocking PDE5/6 allowed cGMP to accumulate, which activates PKG and apoptosis since the PKG inhibitor KT5823 blocked both ROS and apoptosis. Similarly, increased ROS and apoptosis, induced by reducing Slfn4 (siSlfn4), was also mitigated by PKG inhibition, indicating that SLFN4 and sildenafil function through synergistic pathways. Accordingly, in vivo deletion of Slfn4 or pharmacologic inhibition of PDE5,6 with sildenafil also suppressed NOS2 induction, reversed T cell suppression and mitigated SPEM development. Conclusion: SLFN4 regulates the activity of the NO-GTP-cGMP pathway in MDSCs and precludes these cells from succumbing to massive ROS generation when they acquire MDSC function. Disrupting this pathway induces MDSC apoptosis and mitigates gastric metaplasia.

Figure 1. tSNE plot of stomach immune cells, color-coded by single cell expression of representative genes correlated with Slfn4<sup>+</sup> MDSCs.

Figure 1. tSNE plot of stomach immune cells, color-coded by single cell expression of representative genes correlated with Slfn4⁺ MDSCs.

Figure 2. Immunoprecipitated SLFN4 complexes exhibit GTPase activity.

Figure 2. Immunoprecipitated SLFN4 complexes exhibit GTPase activity.

Background: Cell cycle dysregulation is a hallmark of cancer, resulting in unregulated cell proliferation and, eventually, tumor development. Cyclin-dependent kinase 1 (CDK1) is a cell cycle regulatory protein that is involved in cell cycle maintenance. CDK1 has been discovered to be substantially elevated in a several tumors and is linked to poor overall and relapse-free survival. The aim of this study is to understand the regulation role of Helicobacter Pylori infection and inflammation on CDK1 expression in gastric cancer. Methods: Using TCGA data and our integrated comprehensive gene expression analysis, we found a significant overexpression of CDK1 in gastric cancer human and mouse tissues. We detected overexpression of CDK1 in human and mouse gastric glands in response to H. pylori infection. Our data demonstrated that H. pylori infection induced phosphorylation (S536) and activation of NF-kB in vitro and in vivo. Furthermore, H. pylori infection and TNF-α treatment increased the CDK1 mRNA and protein levels in gastric cancer cell lines. Using the ChIP assay, we detected direct biding of NF-κB on the CDK1 promoter regulating its transcription. CDK1 promoted activation of the β-catenin signaling pathway. Using the pTOP/pFOP luciferase reporter assays, as a measure of b-catenin/TCF transcription activity, we confirmed CDK1-dependent activation of β-catenin in response to H. pylori infection. Pharmacologic and genetic inhibition of CDK1 reversed these effects and decreased number and size of gastric tumors organoid from mouse and human, Conclusion: Our findings demonstrate NF-kB-mediated induction of CDK1 expression in response to H. pylori infection with subsequent activation of tumorigenic b-catenin signaling. This novel regulatory link between infection, CDK1, and b-catenin suggests the importance of considering CDK1 inhibitors in gastric cancer.

Gastric cancer is the fifth leading cause of cancer and the fourth leading cause of cancer deaths worldwide. Almost all cases of gastric cancer are associated with Helicobacter pylori infection, a pathogen that infects roughly half of the world’s population. Infection with this bacterium causes chronic , T cell-driven inflammation within the gastric mucosa that leadsleading to metaplasia development, a pre-neoplastic indicator of cancer development. Our lab has previously demonstrated that glucocorticoids are master regulators of gastric inflammation. In this study, we investigated how glucocorticoids regulate the T cell responsiveness response to Helicobacter infection and how theseT cells contributed to the development of gastric pathologies. We adrenalectomized (ADX) mice to remove endogenous glucocorticoids and infected them with H felis within one week of surgery. Two months post-infection, flow cytometry revealed that ADX-infected mice had significantly more T cell infiltration and a more significant proportion of activated T cells. increased leukocyte infiltration, and CD4+ and CD8+ T cells were the most predominant infiltrating immune cells. Analysis by quantitative RT-PCR revealed the increased expression of the cytokines Il1b, Tnf, and Ifn, suggesting an increased Th1-biased immune response in ADX mice. Long-term infection revealed that ADX mice infected for 12 months exhibited increased expression of the dysplasia marker TROP2, increased DNA damage depicted by H2AX phosphorylation, and carcinoma in situ. These data suggest that glucocorticoids are critical for suppressing pro-neoplastic gastric inflammation during Helicobacter infection and that disruption of glucocorticoids may promote gastric cancer development.Quantitative RT-PCR for characteristic Th1 genes (Tnf and Ifng) and Th2 gene, Il13, indicate that ADX-infected mice have a more Th1 skewed phenotype while intact-infected mice have a more Th2 skewed response. Th1 T cells are associated with increased gastric cancer risk in humans. At two months post infection, our results suggest that glucocorticoid signaling influences the T cell phenotype in response to H felis infection. After one year of H felis infection, ADX mice exhibit more severe pathologic changes within the gastric mucosa. Immunofluorescence for the recently established dysplasia marker, TROP2, indicates increased neoplastic changes. ADX-infected mice also exhibited increased staining for phosphorylated gH2AX, a marker of double-stranded DNA breaks. These data indicate that glucocorticoids may prevent the development of neoplastic lesions within the stomach.

Background: Among cancers globally, Gastric cancer (GC) still ranks fifth in incidence and fourth in mortality according to GLOBOCAN 2020 data. Patients with peritoneal metastasis of GC suffer from devastating complications, such as bowel obstruction, obstructive jaundice and massive ascites, resulting in a five-year overall survival rate of only 2% even if the tumor cells are still at the microscopic level. Immune checkpoint inhibitor (ICI) is effective in many cancer patients by releasing the brake in anti-tumor immunity, however it generally shows poor treatment effect on peritoneal metastasis of GC. We have established a chemically induced mouse gastric cancer cell line YTN16, which is transplantable in immune-competent C57BL/6 mice. We also have identified the neoantigens of YTN16 to monitor the dynamics of tumor-specific CD8+ T cells. The purpose of this study is to find a way to boost the effect of ICI treatment for peritoneal metastasis of GC.

Methods: The YTN16 peritoneal metastasis tumor model mice were treated with PD-1, CTLA-4, CD4 and CD8 antibodies alone or in combination. Mice were sacrificed on the 3rd or 5th week to compare the therapeutic effect. Peritoneum, peritoneal exudate cells and spleen were collected for subsequent analysis. Immunohistochemical staining, flow cytometry, bulk RNA-sequence analyses were performed to evaluate tumor microenvironment.

Results: Anti-PD-1 and anti-CTLA-4 combination therapy showed a good therapeutic effect in 3/5 mice. However, monotherapy with either ICI did not show a significant curative effect. Flow cytometry showed an increase in neoantigen-specific CD8+ T cells in peritoneum and peritoneal exudate cells of the combination treatment group. Immunohistochemical staining and mMCP-counter analysis also showed a significant increase in CD8+ T cells infiltrating within the tumors. The invasion of tumor was increased following depletion of CD8 + T cells. In mice resistant to dual ICI treatment, despite the infiltration of CD8+T cells, most of the T cells were exhausted, especially after the depletion of CD4+T cells. αSMA+ CAFs, CD68+ macrophages, Ly6G+MDSC and Foxp3+ Treg infiltration were observed in the tumor microenvironment, which indicated a state of immunosuppression. Gene set enrichment analysis (GSEA) revealed that some activation pathways, such as INTERFERON_GAMMA_RESPONSE, T_CELL_RECEPTOR_SIGNALING PATHWAY, and IL6_JAK_STAT3 SIGNALING were activated in mice with ICI resistant tumors compared to the untreated group.

Conclusion: The therapeutic effect of ICIs was different among individuals. Combination of anti-PD-1 and anti-CTLA-4 therapy may increase the efficacy by promoting tumor infiltration CD8+T cells in the peritoneal metastasis mouse model of GC. JAK inhibitor combined with dual ICI treatment should be further investigated to improve the response to ICI immunotherapy.

Hiroki Masuda

Appearances

DUAL IMMUNE CHECKPOINT INHIBITOR TREATMENT IMPROVES OUTCOMES IN GASTRIC CANCER PERITONEAL METASTASIS IN AN IMMUNE COMPETENT MOUSE MODEL

Presenter

Speakers