Exposure to Eosinophilic Esophagitis Limits Esophageal Carcinogenesis in a Murine Model

Abstract

IntroductionEpidemiological data suggest that eosinophilic esophagitis (EoE) patients do not develop esophageal cancer. Here, we paired murine models of EoE and esophageal squamous cell carcinoma (ESCC) to perform the first functional interrogation of the relationship between EoE and esophageal cancer.MethodsC57B6 mice were treated with MC903/Ovalbumin (OVA) to induce EoE and the esophageal carcinogen 4‐nitroquinoline 1‐oxide (4NQO) to induce ESCC according to the experimental scheme shown in Fig 1A. Esophagi of mice treated with vehicle controls (n=6), MC903/OVA alone (n=4), 4NQO alone (n=10), and MC903 in combination with 4NQO (n=5) were dissected. Tumor frequency was recorded and tumor load (tumor size x tumor number) was determined under a dissecting microscope. H&E stained tissues were evaluated for esophageal lesions. CD45‐positive immune cells were isolated from the spleens of mice with EoE or naïve controls then co‐cultured with the murine‐derived esophageal cell line AKR. After 24‐72 hours of co‐culture, live/dead dye was used to assess AKR viability.Results10/10 mice treated with 4NQO alone developed ESCC as expected in this robust tumor model. In mice co‐treated with MC903/OVA and 4NQO, ESCC was detected in 4/5 mice with a significant decrease in tumor load as compared to mice treated with 4NQO alone (Fig 1B, C). 6/6 mice treated with MC903/OVA alone and 4/4 mice treated with propylene glycol (vehicle for 4NQO) failed to display tumors (Fig 1C). We further evaluated esophageal epithelium of 4NQO‐treated animals to characterize distribution of ESCC lesions. While squamous dysplasia (pre‐malignant), early ESCC and late ESCC were present in all tumor‐bearing animals; there was a decrease in early ESCC and a lack of late ESCC lesions in 4NQO‐treated mice that were exposed to MC903/OVA‐induced EoE (Fig 1D). Moreover, the total percentage of esophageal epithelium occupied by neoplastic lesions was significantly diminished in mice co‐treated with 4NQO and MC903/OVA (Fig 1E). To investigate how EoE limits esophageal carcinogenesis, we exposed AKR esophageal tumor cells to CD45‐positive cells isolated from spleens of strain‐matched donor mice with or without EoE (Fig 2A). We found a robust induction of cell death in tumor cells exposed to immune cells from EoE donor mice compared to those exposed to immune cells from vehicle‐treated controls at 48 hours (n=3, p<0.05) (Fig 2B, C).ConclusionsThese data indicate that exposure to EoE activates an anti‐tumoral immune response that limits esophageal carcinogenesis. These studies have the potential for significant implications related to the clinical care of patients affected by EoE and ESCC.