Etiology of Hypereosinophilia and Hypereosinophilic Syndromes in a Tropical Setting

Background: Hypereosinophilia (HE) often requires extensive workup and in some instances quickly enough to identify the cause, considering the nature of the end organ damage which is likely to occur if a proper treatment is not initiated on time. This includes primarily a search for various secondary causes. In their absence,an evaluation of a clonal process is required which is complicated by the absence of any definite morphological or immunophenotypic features differentiating clonal from non-clonal conditions. PDGFRA, PDGFRB and FGFR1 related clonal HE and lymphocytic variant of reactive HE are two distinct groups of hypereosinophilic disorders requiring specific management. There are several reports of these disorders in the recent past from different parts of the world. However, there is no comprehensive data about the prevalence or etiology of these disorders from a tropical country like India. While WHO 2008 classification is solely intended for classifying neoplastic hypereosinophilia, the system proposed by International Cooperative Working Group on Eosinophil Disorders (ICOG-EO) in 2011 appears to be more comprehensive covering familial, reactive, neoplastic, idiopathic and other HE related single and multisystem disorders.

Methods and materials: A prospective study was conducted from January 2014 to May 2015 and a retrospective analysis of cases diagnosed between January 2009 and December 2013 was performed using hospital records. A comprehensive clinico-pathological evaluation was performed in all prospective cases by extensive work up including flow cytometry for T cell abnormalities and FISH for PDGFRA, PDGFRB and FGFR1 abnormalities. An attempt was made to classify HE [>1.5 x 10⁹/L eosinophils in peripheral blood or >20% of all nucleated cells in the bone marrow (BM)] and hypereosinophilic syndromes (HES; HE with organ damage) according to the ICOG-EO classification.

Results: A diagnosis of HE was made in a total of 101 patients (2.6 cases per month) over a period of 77 months (1-72 years, male to female ratio 1.8:1). 71 patients had either a sufficient follow up or work-up required for final categorization. 26.7% of patients were children. A diagnosis of HES was made in 21 patients (29.6%). The reactive HE was the commonest cause (73.9% of HE or 47.9% of all patients) followed by clonal or neoplastic HE (21.7% of HE or 14.1% of all patients). Similar to HE, HES was most commonly secondary in nature (61.9% of HES or 18.3% of all patients), followed by Idiopathic HES (19% of HES or 5.6% of all) and neoplastic HES (14.2% of HES and 4.2% of all). Helminthic infections especially filariasis, toxocariasis and trichinellosis; and giardiasis were important causes of HE. The identification of exact etiological agent required testing for a wide spectrum of pathological agents. T-NHL was the most important cause of reactive HE secondary to neoplasms. HE preceded the diagnosis of ALL, AML, Hodgkin lymphoma and cutaneous T cell lymphoma in one patient each. In three patients (4.2% all patients or 14.3% of HES) L-HES was diagnosed based on immunophenotype abnormalities (CD3-CD4+ in one patients and CD4+CD8+ T cells in two patients), response to steroids and requirement of long-term steroid for control of HE. BCR-ABL1 and JAK2V617F positivity was detected in 4.2% and 1.4% patients. There were no patients with rearrangement of PDGFRA, PDGFRB or FGFR1 genes. However HE was responsive to imatinib (BCR-ABL1 negative) in three (4.2%) retrospectively recruited patients (PDGFRA, PDGFRB or FGFR1 not tested). A comparison of HE associated malignancies with HE associated with benign disorders showed that maximum TLC (p=0.03), presence of blasts in the peripheral blood (p=0.008), higher BM blast% (p=0.01), lower platelet counts (0.02) were associated with malignancy. Higher percentage of eosinophils in both peripheral blood (p=0.0004) and BM (p=0.04) was associated with benign disorders. Absolute eosinophil count did not differentiate malignancy from benign conditions. IgE levels were generally higher in patients with benign disorders.

Conclusions: Our study shows that HE is frequent in India and is most commonly secondary to infections. HE can precede, succeed or can occur concurrently with various malignancies and autoimmune diseases. There were no confirmed cases of PDGFRA, PDGFRB and FGFR1 associated HE. The exact prevalence of these cases needs to be examined in a larger cohort.