The Underappreciated Role of Eosinophils in Esophageal Cancer

Esophageal cancer is among the deadliest forms of human malignancy, exerting a substantial global health burden. Esophageal squamous cell carcinoma (ESCC) is the predominant subtype of esophageal cancer, comprising approximately 90% of cases worldwide.¹ Because ESCC is highly resistant to chemotherapy and radiation treatment, immunotherapy has emerged as a viable strategy for improving ESCC patients’ outcomes. Although traditional approaches to cancer immunotherapy focus on effector lymphoid cells, emerging evidence suggests that eosinophils may exert anticancer activities via direct eosinophil-mediated cytotoxicity or by influencing T-cell–mediated cytotoxicity.2, 3, 4, 5 Eosinophils are particularly interesting with regard to esophageal cancer because epidemiologic data suggest that patients with eosinophilic esophagitis, a chronic inflammatory disease characterized by eosinophilic infiltration into esophageal mucosa, have a reduced risk of developing esophageal malignancy.⁶ Consistent with these data, it recently was shown that exposure to eosinophilic esophagitis–like inflammation in mice limits ESCC carcinogenesis.⁵ Furthermore, high levels of eosinophils and mast cells each correlate with favorable outcomes in esophageal cancer patients.7, 8, 9, 10 Notably, however, the direct role of eosinophils in ESCC has remained elusive.

Now, Jacobse et al¹¹ address this critical knowledge gap, identifying potent antitumor activity of eosinophils in ESCC by coupling in vivo murine models with in vitro 3-dimensional co-culturing of eosinophils and ESCC cells. In this elegant study, the investigators examined the presence of eosinophilic infiltration in tumor and tumor-adjacent tissue from ESCC patients, finding that eosinophils are active in human ESCC and localized to the tumor microenvironment. They further report that eosinophil infiltration was more prominent in early ESCC compared with late-stage ESCC. The investigators then used the carcinogen 4-nitroquinoline-1-oxide (4-NQO) to induce esophageal tumorigenesis in mice, administering 4-NQO for 8 weeks to generate precancer or 16 weeks to induce carcinoma. In agreement with their findings in ESCC patients, eosinophilic infiltration was more pronounced in precancerous esophageal lesions compared with carcinoma in mice. Three mouse models were used to examine the functional consequences of eosinophil depletion on ESCC carcinogenesis in vivo: eosinophil chemoattractant eotaxin-1 (Ccl11-/-)-deficient mice, eosinophil-deficient (ΔdblGATA) mice, and interleukin (IL)5 monoclonal antibody–treated mice. All 3 models showed enhanced tumorigenesis when treated with 4-NQO. In complement, mice treated with recombinant IL5 and 4-NQO displayed increased esophageal eosinophilia concomitant with diminished precancerous and cancerous lesions. To gain mechanistic insight into the antitumor activity of eosinophils, the investigators first performed RNA sequencing in esophagi from ΔdblGATA and wild-type controls with precancerous esophageal lesions, identifying gene expression patterns consistent with eosinophil-mediated induction of oxidative stress in esophageal epithelium. Co-culture experiments with degranulating eosinophils and ESCC cells supported this finding, showing tumor cell apoptosis that is reversible in the presence of reactive oxygen species scavenger N-acetylcysteine. Cytokine profiling revealed enrichment of IL17 and associated protumorigenic pathways in ΔdblGATA mice with precancerous ESCC lesions compared with wild-type controls. Overall, these findings show that eosinophils exert direct cytotoxic effects on ESCC tumor cells that are associated with reactive oxygen species and may contribute to the development of a local microenvironment that further impairs ESCC carcinogenesis.

The finding that eosinophils are common in precancerous ESCC lesions but less so in frank ESCC suggests that augmenting eosinophils may be beneficial in patients with ESCC. Such an approach represents a novel ideation of cancer immunotherapy that has the potential for significant impact in ESCC, particularly given that the disease often is diagnosed at late stages. Of course, although the current study shows that eosinophils promote apoptosis in ESCC cells in vitro, it will be imperative to determine whether eosinophils are capable of inducing ESCC regression in vivo. Because eosinophilia also has been detected in subsets of patients with Barrett’s esophagus or esophageal adenocarcinoma,^10,12 it will be of further interest to determine if eosinophils exert antitumor activity in this context as well. The technical feasibility of harnessing the antitumor activity of eosinophils in esophageal cancer lesions also represents an important consideration. Although eosinophils are not normally found in the esophagus, their presence in the esophageal mucosa of subjects affected by widely prevalent esophageal diseases indicates that local delivery of eosinophils or factors that promote their migration (eg, IL5) to esophageal cancer lesions may indeed be feasible. It is possible, however, that inflammatory cells or stimuli beyond eosinophils will be required to drive the antitumor activity of these cells. Moreover, given the increasing appreciation of heterogeneity among eosinophils,¹³ it is possible that a particular subset mediates their antitumor activities. The work of Jacobse et al¹¹ has laid a solid technical and conceptual foundation for addressing these critical questions. For immunotherapy techniques in clinical cancer treatment to continue to advance, it remains critical to understand the functional and mechanistic role of each player in the tumor microenvironment.