ASA Main Site  |  Past & Future Meetings
American Surgical Association

Back to 2022 Abstracts


The Microbiome of Gastric Cancer: A Novel Target for Cancer Therapy?
Miseker Abate1, Santosh Vardhana2, Chad Vanderbilt3, Vivian Strong1
1Surgical Oncology, Memorial Sloan Kettering Cancer Center, New York, New York, United States, 2Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York, United States, 3Diagnostic Molecular Pathology, Memorial Sloan Kettering Cancer Center, New York, New York, United States

Introduction: The microbiome is the collective genome of microbes that live in and on the human body. Altered microbiome promotes carcinogenesis through DNA damage, metabolite production leading to oncogenesis or tumor suppression, and immunomodulation. However, there is a significant lack in understanding of the bacteria and viruses involved in carcinogenesis. If identified, they could potentially provide novel targets for cancer therapy. In this study, we characterize the full spectrum of the microbes within gastric cancer (GC) and assess differences across GC subtypes.

Methods: Clinically relevant cancer-related genomic mutations were queried from GC tumors with institutional Integration Mutation Profiling of Actional Cancer Targets (IMPACT) assays. Bacterial genomes were separated from human genomes on a binary alignment map (BAM) file. A basic local alignment search tool (BLAST) algorithm was used to specify and quantify bacteria on a sub-species level using matching sequences of taxonomy IDís found on the National Center for Biotechnology and Information database. The microbial differences between GC and normal tissue, and among GC molecular subtypes: 1) Diffuse 2) Intestinal 3) microsatellite instability-high (MSI-H) 4) Epstein-Barr Virus (EBV) were analyzed. Shannon alpha diversity plot and enrichment analysis by odds ratio were used to compare cohorts. Statistically significant enrichment analysis in IMPACT cohort were validated with The Cancer Genome Atlas (TCGA) data for which the microbiome was extracted through the same pipeline.

Results: The microbiome of 676 GC tumors, 237 IMPACT and 439 TCGA, were compared with 137 normal tissue samples. Compared to normal tissue, GC tissue had decreased alpha diversity, with Azotobacter (OR:5.78, 95%CI:1.37, 24.31)), Bordetella (OR:4.38, 95%CI:1.03,18.58)), Campylobacter (OR:4.03, 95%CI:1.23,13.26), Allycycliphius (OR:2.89, 95%CI:1.12, 7.43)) and Escherichia (OR:1.67, 95%CI:1.09, 2.55) present in significantly higher likelihood (p < 0.05) (Figure 1). Among GC molecular subtypes, Lactobacillus (OR: 3.42, 95% CI:1.14, 10.20), Selenomonas (OR: 4.37, 95%CI:1.11, 17.29), and Prevotella (OR: 2.55, 95%CI: 1.04, 6.26) were significantly more likely to be present in MSI-H GC tumors compared to other subtypes (Table 1). There were no statistically significant differences in enrichment among other GC subtypes.

Discussion: Distinct and significant patterns of microbial dysbiosis and associated species are identified in GC compared to normal gastric tissue in the largest GC cohort to date. Furthermore, we identified microbial signatures unique to various molecular subtypes of GC, such as MSI-H tumors, suggesting that the milieu of various tumor subtypes may be altered differentially. By identifying unique microbial biomarkers, we hope to elucidate predictors that can be further studied for their interactions with the host inflammatory and immune response to identify possible targets for therapy.


Back to 2022 Abstracts