Since the identification of the FIP1L1/PDGFRA fusion gene as a pathogenic cause of the hypereosinophilic syndrome (HES), the importance of the molecular classification of HES leading to the diagnosis of chronic eosinophilic leukemia (CEL) has been recognized. As a result, a new category, ‘myeloid and lymphoid neoplasm with eosinophilia and abnormalities in PDGFRA, PDGFRB or FGFR1’, has recently been added to the new WHO criteria for myeloid neoplasms. FIP1L1/PDGFRα-positive disorders are characterized by clonal hypereosinophilia, multiple organ dysfunctions due to eosinophil infiltration, systemic mastocytosis (SM) and a dramatic response to treatment with imatinib mesylate. A murine HES/CEL model by the introduction of FIP1L1/PDGFRα and IL-5 overexpression also shows SM, representing patients with FIP1L1/PDGFRα-positive HES/CEL/SM. The murine model and the in vitro development system of FIP1L1/PDGFRα-positive mast cells revealed the interaction between FIP1L1/PDGFRα, IL-5 and stem cell factor in the development of HES/CEL/SM. Current findings of FIP1L1/PDGFRα-positive HES/CEL are reviewed focusing on aberrant mast cell development leading to SM.
The importance of the molecular classification of the hypereosinophilic syndrome (HES) has been increasingly recognized. The fusion gene FIP1L1/PDGFRA was identified in a large number of patients initially diagnosed as having a myeloproliferative variant of HES or chronic eosinophilic leukemia (CEL) . Subsequently, other variant PDGFRA fusion genes as well as those involving PDGFRB or FGFR1 have also been described in myeloproliferative neoplasms with eosinophilia in the last years [2,3,4]. As a result, a new category of myeloid neoplasms, ‘myeloid and lymphoid neoplasm with eosinophilia and abnormalities of PDGFRA, PDGFRB or FGFR1’, has recently been added to the new WHO criteria . FIP1L1/PDGFRα fusion-positive disorders are characterized by clonal myeloproliferation resulting in hypereosinophilia, multiple organ dysfunctions due to eosinophil infiltration, a dramatic response to treatment with imatinib mesylate and systemic mastocytosis (SM) . Murine models of FIP1L1/PDGFRα-induced diseases have been reported recently [7,8]. Interestingly, these models demonstrated severe SM representing patients with FIP1L1/PDGFRα fusion-positive diseases . In this review, the clinical manifestation of FIP1L1/PDGFRα fusion-associated disorders are summarized, focusing on mastocytosis induced by FIP1L1/PDGFRα expression, and the mechanisms of mastocytosis in FIP1L1/ PDGFRα-positive HES/CEL are discussed.
Eosinophilia in SM Patients
Peripheral blood eosinophilia has been reported in 15–28% of SM patients [10,11,12]. This is no big surprise since crosstalk between eosinophils and mast cells is well known, especially in allergic inflammation. For instance, mast cell activation by major basic protein, an eosinophil granule protein, elicits the generation of lipid mediators and cytokines. Eosinophils also produce cytokines associated with mast cell activation such as stem cell factor (SCF), granulocyte/macrophage colony-stimulating factor and nerve growth factor . The D816V mutation in the KIT gene resulting in constitutive activation of the receptor tyrosine kinase has been shown in the majority of patients as a cause of SM . D816V-kit mutation-positive SM with eosinophilia has been clinically distinguished from that without eosinophilia. D816V-kit mutation-positive SM patients with eosinophilia present hepatosplenomegaly, lymphadenopathy, anemia and monocytosis more frequently as well as higher levels of circulating tryptase, whereas anaphylaxis is seen with a low frequency in these patients, in comparison to patients without eosinophilia .
Clinical Manifestations of Mastocytosis Associated with FIP1L1/PDGFRα
A recent report has shown that FIP1L1/PDGFRα-associated SM is a clinically distinguishable disease from D816V mutation SM with eosinophilia . FIP1L1/PDGFRα-associated SM shows lower tryptase levels in the circulation, less aggregation of bone marrow mast cells, more severe eosinophilia, higher serum vitamin B12 levels and more frequent pulmonary and cardiac involvement than D816V SM with associated eosinophilia. The clinical difference is important since it would justify the differential diagnosis between these two entities, based on the analysis of the expression of the FIP1L1/PDGFRA fusion gene. More recently, Klion  proposed a scoring system to approach FIP1L1/PDGFRα fusion-positive HES/CEL therapy appropriately.
SM in Murine Models of FIP1L1/PDGFRα-Positive Disorders
First, Cools et al.  reported that the introduction of the FIP1L1/PDGFRA fusion gene into bone marrow hematopoietic stem cells and progenitors (HSC/P) induces a murine model of a myeloproliferative disorder (MPD) similar to that found in p210-BCR/ABL-induced chronic myelogenous leukemia-like disease (F/P-MPD). Subsequently, an HES/CEL murine model was developed by the introduction of the FIP1L1/PDGFRA fusion gene into bone marrow HSC/Ps in the presence of T-cell overexpression of IL-5 (F/P-HES/CEL) . More recently, these two murine models were also shown to develop tissue mast cell infiltration and increased circulating mast cell protease 1 (MMCP-1) levels, which is a systemic assay of mast cell content and degranulation in the mouse resembling serum tryptase determination in SM patients . Similar to the patients with FIP1L1/PDGFRA fusion gene, tissue mast cell infiltration of hematopoietic organs, skin and intestine, where mast cell morphology is aberrant, is present in F/P-HES/CEL mice. Tissue mast cell shape is irregular with frequent cytoplasmic extensions reminiscent of the ‘spindle shape’ found in clinical SM. In addition, serum levels of MMCP-1 are extremely elevated in F/P-HES/CEL mice.
A possible interaction of IL-5 with the SM phenotype was analyzed in F/P-HES/CEL mice. F/P-HES/CEL and F/P-MPD mice showed significantly greater mast cell infiltration in their skin and intestine, and higher levels of MMCP-1 compared to both controls with and without IL-5 overexpression. Interestingly, intestinal mast cell infiltration and serum MMCP-1 levels in F/P-HES/CEL mice were significantly higher compared to F/P-MPD, suggesting that FIP1L1/PDGFRα in conjunction with IL-5 exacerbates mastocytosis in murine F/P-HES/CEL .
Mechanism of FIP1L1/PDGFRα-Promoted Mast Cell Development
Since the c-kit signaling pathway is pivotal for normal mast cell development and function, the question of whether FIP1L1/PDGFRα-associated SM is still c-kit dependent arises. In fact, following blockade of the the c-kit signaling pathway using anti-c-kit antibody, tissue mast cells and circulating levels of MMCP-1 were significantly decreased, suggesting that tissue infiltration of FIP1L1/PDGFRα fusion-positive mast cells are associated with SCF/c-kit signaling. The ex vivo cytokine-dependent mast cell differentiation of FIP1L1/PDGFRα-expressing HSC/Ps was largely dependent on the use of SCF in the culture conditions. Moreover, FIP1L1/PDGFRα fusion-positive mast cells showed prolonged survival and enhanced migration toward SCF. Specific synergistic stimulation of the Akt signaling pathway by FIP1L1/PDGFRα and SCF indicated collaboration of two tyrosine kinase activities in their downstream signaling pathways. Taken together, FIP1L1/PDGFRα synergizes with the SCF/c-kit pathway to promote mast cell development, activation and survival both in vivo and in vitro .
Pathogenesis of FIP1L1/PDGFRα-Positive HES/CEL/SM Associated with SCF and IL-5
Previously, we reported that the induction of murine HES/CEL by FIP1L1/PDGFRα requires a second event that is associated with IL-5 overexpression . In addition, the level of expression of IL-5Rα was exclusively up-regulated in FIP1L1/PDGFRα-positive splenocytes and FIP1L1/PDGFRα fusion protein shares the downstream JAK2/STAT5 pathway with IL-5 signaling (fig. 1) . Interestingly, polymorphisms of the human IL-5RA gene have been found linked to the constitutional IL-5RA genotype and the severity of FIP1L1/PDGFRα-positive CEL . These findings suggest that amplification of IL-5 signaling by FIP1L1/PDGFRα triggers a CEL-like disease. Interestingly, IL-5Rα is expressed on eosinophil and mast cell progenitors [18,19] as well as mature eosinophils  and mast cells , whereas c-kit expression is not only found on progenitors but also on mature eosinophils and mast cells [14,18,19]. Importantly, expression of the FIP1L1/PDGFRA fusion gene or deletion of the surrogate marker CHIC2 have been detected in non-eosinophilic cells, including neutrophils, monocytes, mast cells, lymphoid lineage cells and bone marrow CD34-positive cells in part of the patients, suggesting that the fusion of the FIP1L1/PDGFRA genes may occur in HSCs or early progenitors [22,23,24]. Taken together, these findings imply that FIP1L1/PDGFRα in collaboration with SCF may affect leukemic myeloproliferation and synergistically with IL-5 expand and activate mast cell and eosinophil lineages (fig. 2).
HES/CEL has attracted a lot of attention since the patients were successfully treated with imatinib mesylate, and subsequently the target, FIP1L1/PDGFRα, was discovered in a large number of patients initially diagnosed as myeloproliferative variant of HES. To our knowledge, there is little doubt that FIP1L1/PDGFRα preferentially affects eosinophil and mast cell proliferation, survival, differentiation and tissue infiltration, and leukemogenesis is induced combined with systemic or local extrinsic factors, as demonstrated by crucial roles of IL-5 and SCF in the pathogenesis of FIP1L1/PDGFRα-initiated HES/CEL/SM. This disease, an example of the crosstalk between oncogenesis and inflammation, represents an excellent model to study cellular integration of biochemical signals in cancer, being responsible for crucial aspects of cancer biology, e.g. cell proliferation, survival, tissue invasion and communication with the specific tissue microenvironment.
We are grateful to Drs. Marc E. Rothenberg (Cincinnati Children’s Hospital Medical Center) and David A. Williams (Children’s Hospital, Boston, Mass., USA) for their help in the development of the murine CEL/HES model. J.A.C. was supported by the National Institutes of Health (HL080759 and HL080759-01A2S1) and U.S. DoD (CM064050).
The authors declare that no financial or other conflict of interest exists in relation to the content of this article.