Eosinophils and Eosinophil Activation in the Pathogenesis of Human Disease

National Institute Of Allergy And Infectious Diseases

Investigators

Linked publications & trials

Abstract

Using a cohort of more than 580 subjects with a wide variety of eosinophilic disorders, ranging from benign eosinophilia to eosinophilic leukemia, we have continued to identify and characterize novel subgroups of patients with eosinophilia and to explore responses to targeted therapies as ways to increase our understanding of the underlying pathophysiologies in diverse groups of patients with hypereosinophilic syndromes (HES). A major focus of activity in the past year continues to be patients with HES and gastrointestinal involvement. An important clinical question is whether patients presenting with single organ involvement, such as eosinophilic gastrointestinal disease (EGID), are different from or at risk for progression to multisystem HES. To begin to address this, we compared patients with biopsy-proven EGID and hypereosinophilia (n=34) to patients with multisystem HES and biopsy evidence of gastrointestinal involvement (n=22) (Kuang J et al. Allergy Clin Immunol Pract 2020). Although the two groups were remarkably similar with respect to demographics, symptoms, segment involvement and comorbidities, patients with multisystem disease were more likely to receive glucocorticoid therapy (100% vs. 79%) and non-glucocorticoid therapies (77% vs 38%). Remarkably, one-third of patients with multisystem disease presented with isolated gastrointestinal symptoms before developing extraintestinal manifestations at a median of 1 year later. These data are currently being confirmed in a large multicenter retrospective study as they have implications for monitoring of patients with EGID and hypereosinophilia. In addition to these studies, we continue to participate in the rare disease consortium for eosinophilic gastrointestinal disorders (CEGIR) and were involved in a study demonstrating that lack of access to pediatric gastroenterologists in rural areas may explain observed differences in the prevalence of pediatric eosinophilic esophagitis in urban vs. rural settings (Sabet et al. J Allergy Clin Immunol Pract 2021), as well as a multicenter placebo-controlled, double-blind clinical trial of AK002 (an afucosylated antibody to Siglec-8) for the treatment of eosinophilic gastritis (Dellon et al. N Engl J Med 2020). Glucocorticoids (GC) are the mainstay of therapy for a variety of eosinophil-associated disorders, including hypereosinophilic syndromes (HES), although responses are not universal and GC are associated with significant toxicity. Prior to the first therapeutic use of GC, it was observed that the administration of exogenous adrenocorticotropic hormone or hydrocortisone led to a rapid, profound, and transient eosinopenia in humans. Based on data from a prior study suggesting a major role for CXCR4 in this process (Khoury et al. Allergy 2018), we developed an isolation technique to purify rhesus macaque eosinophils from peripheral blood and performed live tracking of zirconium-89-oxine labeled eosinophils by serial PET/CT imaging before and after glucocorticoid administration. Glucocorticoids induced rapid bone marrow homing of eosinophils with concomitant induction of CXCR4, and selective blockade of CXCR4 with plerixafor abrogated the early glucocorticoid-induced reduction in blood eosinophils supporting the hypothesis that glucocorticoid-induced eosinopenia results from CXCR4-dependent migration of eosinophils to the bone marrow (Hong et al. Blood 2020). We have recently completed a clinical trial demonstrating that the early response to a single dose of GC is a useful predictor of long-term GC response and are exploring the relevance of CXCR4 expression in this process. Targeted therapies not only provide new options for patients but provide insight into the effects of eosinophil depletion in humans and the mechanisms driving eosinophilia and eosinophil activation in HES. Ongoing studies include the use of tyrosine kinase inhibitors (imatinib and ruxolitinib) in the treatment of steroid-refractory HES, a recently initiated single site study of mepolizumab for the treatment of episodic angioedema and eosinophilia, a multicenter phase 3 trial of benralizumab for HES, as well as long-term followup studies of HES patients enrolled on investigator-initiated trials of benralizumab and dexpramipexole. Enrollment in several of these trials has been severely impacted by the COVID19 pandemic but is expected to improve in the coming year. Historically, eosinophils have been viewed as effector cells with primary roles in the response to helminth infection and allergens. Consistent with this, recent studies in our group have focused on the role of eosinophils and eosinophil activation in the clinical manifestations and post-treatment reactions in filariasis (Herrick et al. Clin Infect Dis 2020; Legrand et al. Clin Infect Dis 2017; Legrand et al. Clin Infect Dis 2020). Over the past decade, however, there has been increasing interest in the role of eosinophils in the modulation of responses to non-helminth pathogens. To address this issue, we initiated a collaboration with Dr. Katrin Mayer-Barber to examine the role of eosinophils in the host immune response to pulmonary infection with Mycobacterium tuberculosis (Mtb). After preliminary experiments demonstrated that purified eosinophils from healthy volunteers responded rapidly to both irradiated and live Mtb, secreting eosinophil granule proteins, the alarmin IL-1alpha, and the chemokines IL-8, CCL3 and CCL4, we used murine models to demonstrate that the early influx of neutrophils and NK cells into the lung post-Mtb infection is eosinophil-dependent, suggesting that eosinophils may be responsible for orchestrating the pulmonary innate immune response after Mtb infection. Consistent with the hypothesis that eosinophils are recruited to the lung in the setting of tuberculous inflammation, blood eosinophil numbers were decreased in patients with active tuberculosis and inversely correlated with sputum bacterial burden and tissue eosinophilia at sites of tubercular granulomatous inflammation. These findings were confirmed in a rhesus macaque model of Mtb and a zebrafish model of M. marinum. Finally, the absence of eosinophils in eosinophil-deficient mice infected with Mtb was associated with a dramatic increase in bacterial load and mortality, suggesting that eosinophils play a predominantly protective role in host resistance to Mtb infection.