(2008) 22, 1999–2010 & 2008 Macmillan Publishers Limited All rights reserved 0887-6924/08 $32.00 www.nature.com/leu SPOTLIGHT REVIEW

Five years since the discovery of FIP1L1–PDGFRA: what we have learned about the fusion and other molecularly defined

J Gotlib1 and J Cools2,3

1Department of Medicine/Division of Hematology, Stanford Cancer Center, Stanford, CA, USA; 2VIB Department of Molecular and Developmental Genetics, VIB, Leuven, Belgium and 3Center for Human Genetics, K.U. Leuven, Leuven, Belgium

The year 2008 marks the fifth anniversary since the publication myeloid leukemia (AML) (for example, French-American-British which identified the FIP1L1–PDGFRA fusion in patients subtype M4Eo, inversion 16), myelodysplastic syndrome, with idiopathic hypereosinophilia. With the benefit of time, a more comprehensive picture has emerged regarding several systemic and so on). Although not formally characteristics of the fusion, including its incidence, biological adopted in the nomenclature of the World Health Organization features and the clinical profile of patients who carry the (WHO), the term ‘myeloproliferative variant of hypereosino- molecular rearrangement. A few prospective trials have now philic syndrome (M-HES)’ has been commonly used in the better defined the natural history of -treated FIP1L1– literature to refer to these marrow-derived eosinophilic MPNs PDGFRA-positive patients, from which some basic conclusions because they share one or more clinical or laboratory features can be drawn: the prognosis is outstanding, acquired resis- tance is exceedingly rare, but ongoing imatinib treatment is suggestive of CML or the classic MPDs: hepato/splenomegaly, likely required to prevent relapse. The emergence of genetically bone marrow hypercellularity or fibrosis, myeloid immaturity 2 assigned eosinophilias has led the World Health Organization and elevated serum B12 or serum tryptase levels. Lympho- in 2008 to adopt a semi-molecular classification scheme, with cyte-variant hypereosinophilia relates to the existence of one subcategory named ‘myeloid and lymphoid neoplasms clonal, pathogenetic T-cell subsets with an aberrant surface with and abnormalities of PDGFRA, PDGFRB or immunophenotype (for example, CD3ÀCD4 þ , CD4 þ CD7À, FGFR1.’ Molecular rearrangements involving other partner þ À À , such as ETV6 and JAK2, have also been associated CD3 CD4 CD8 ), which overproduce eosinophilopoietic with eosinophilic disorders, and will likely be assimilated into cytokines such as interleukin-5 (IL-5) and other Th2 cytokines such classifications over time. Despite the molecularly defined such as IL-4, IL-13 and granulocyte macrophage colony- eosinophilias comprising a small proportion of cases com- stimulating factor.3,4 In these cases, the eosinophilia is non- pared to the aggregate of other subtypes of hypereosinophilia, clonal. In rare instances, transformation of the clonal T cells to their recognition is critical because of the availability of highly T-cell has been reported, although lymphocyte- effective molecularly targeted therapy. Leukemia (2008) 22, 1999–2010; doi:10.1038/leu.2008.287; variant hypereosinophilia generally follows a benign course. published online 9 October 2008 The diagnosis of idiopathic hypereosinophilic syndrome (HES) Keywords: hypereosinophilic syndrome; chronic eosinophilic should be reserved for patients in whom no clonal marker has leukemia; FIP1L1–PDGFRA; PDGFRB; FGFR1; imatinib been identified, no increased blood or marrow blasts are present and for which no definite or suggestive clinical or laboratory features of a myeloproliferative or lymphocyte variant of hypereosinophilia can be found.1 According to the classic Terminology and classification definition of Chusid et al.,5 patients with idiopathic HES should also have an absolute count of 41500/mm3 and In contradistinction to ‘secondary’ or ‘reactive’ hypereosinophi- signs or symptoms of organ involvement; however, the lia, FIP1L1–PDGFRA-positive chronic eosinophilic leukemia requirement that the eosinophilia persist for 46 months has (CEL) and other molecularly defined myeloproliferative neo- generally fallen out of favor, in part because some patients may plasms (MPNs) are broadly categorized as either ‘primary’ or require more urgent treatment, and modern diagnostic evalua- ‘clonal’ eosinophilias, because they are acquired hematopoietic tions can proceed in a more timely manner. Table 1A and B SPOTLIGHT stem cell or progenitor cell marrow disorders for which a highlight the 2001 WHO classification of chronic eosinophilic specific genetic abnormality has been identified.1 CEL has also leukemia and HES and the revised 2008 WHO classification of been defined by an increase in peripheral blood (42%) or bone ‘myeloid and lymphoid neoplasms with eosinophilia and marrow (45%) blasts in the absence of a clonal marker by abnormalities of PDGFRA, PDGFRB,orFGFR1’ as well as conventional cytogenetic assays, fluorescence in situ hybridiza- ‘chronic eosinophilic leukemia, not otherwise specified tion (FISH) or other types of molecular studies (for example, (NOS).’1,6,7 X- inactivation analysis).1 In addition, eosinophilia- associated acute or chronic myeloid neoplasms must be excluded (for example, chronic myeloid leukemia (CML), acute Incidence of eosinophilic neoplasms associated with rearrangements of PDGFRA, PDGFRB and FGFR1 Correspondence: Dr J Gotlib, Stanford Cancer Center, Department of Medicine/Hematology, Stanford University School of Medicine, 875 The incidence rates for the molecularly defined eosinophilias Blake Wilbur Drive, Room 2324, Stanford, CA 94305-5821, USA. are not known, nor are there data regarding the proportion of E-mail: [email protected] patients with hypereosinophilia represented by these genetically or Dr J Cools, Herestraat 49, Box 602, Leuven B-3000, Belgium. E-mail: [email protected] defined cases. With these caveats, eosinophilic MPNs with Received 16 September 2008; accepted 16 September 2008; rearrangements of PDGFRA, PDGFRB and FGFR1 are consi- published online 9 October 2008 dered to be very rare entities (for example, incidence o1/100 000 h ra. PDGFRB /l must persist for at least 6 9 10 Â 1.5 4 or of ETV6–PDGFRB ) B ( d 6,7 PDGFRA . In addition, an eosinophil count of 5% but less than 19%), diagnose CEL. 4 ) and (2008) A ( FGR1 PDGFRB 1 or, demonstration of an c and FIP1L1–PDGFRA /l) J Gotlib and J Cools 9 2%) or marrow ( 10 PDGFRB , 4 Â fusion gene or other myeloproliferative neoplasms (PV, ET, PMF) or MDS/MPN (CMML or atypical b 1.5 4 PDGFRA BCR–ABL FGFR1 fusion gene or other rearrangement of fusion gene with eosinophilia associated with FIP1L1–PDGFRA a Five years since the discovery of FIP1L1–PDGFRA FIP1L1–PDGFRA WHO classification of eosinophilic disorders (2001) feature diagnostic of AML CML) bone marrow. diagnose HES or if blasts are present in the peripheral blood ( 3. There is no4. t(5;12)(q31–q35;p13) There or is other no 5. rearrangement There of is no6. rearrangement The of blast cell count in the peripheral7. blood There is and a bone clonal marrow cytogenetic is or less molecular than genetic 20% abnormality, or and blast there cells is are no more inv(16)(p13q22) than or 2% in t(16;16)(p13;q22) the or peripheral other blood or more than 5% in the 1. There is eosinophilia (eosinophil count 2. There is no Ph chromosome or Presence of a A myeloproliferative neoplasm withAND prominent eosinophilia A myeloproliferative neoplasm, oftenneutrophilia with or prominent monocytosis eosinophiliaAND and sometimes with Presence of t(5;12)(q31–q33;p12) or a variant translocation A myeloproliferative neoplasm withOR prominent eosinophilia andAcute sometimes myeloid with leukemia neutrophilia or orbone precursor monocytosis marrow T-cell eosinophilia) or precursorAND B-cell lymphoblastic leukemia/lymphomaPresence (usually of associated t(8;13)(p11;q12) with or peripheral a blood variant or translocation leading to FGFR1 rearrangement demonstrated in myeloid cells, lymphoblasts or both CML (Ph chromosome orAML BCR–ABL including positive) those withOther inv(16), myeloproliferative t(16;16)(p13;q22) diseases (PV,Myelodysplastic ET, PMF) syndromes Pulmonary diseases (hypersensitivity pneumonitis, Loeffler’s and so on) T-cell , including mycosisHodgkin’s fungoides, lymphoma Sezary syndrome Acute lymphoblastic leukemia/lymphoma Mastocytosis Allergy Parasitic disease Infectious disease ) ) Idiopathic hypereosinophilic syndrome is diagnosed when the following entities are excluded: reactive eosinophilia, lymphocyte-variant If appropriate molecular analysis is not available, this diagnosis should be suspected if there is a Ph-negative MPN with the hematological features Because t(5;12)(q31–q33;p12) does not always lead to an ETV6–PDGFRB fusion gene, molecular confirmation is highly desirable. If molecular Patients presenting with or lymphoblastic leukemia/lymphoma with eosinophilia and a FIP1L1–PDGFRA fusion gene are A B a 5q31–33 break point. analysis is not available, this diagnosis should be suspected if there is a Ph-negative MPN associated with eosinophilia and with a translocation wit also assigned to this category. essential thrombocythemia; HES, hypereosinophilic syndrome; MPN, myeloproliferative neoplasm; PMF, primary myelofibrosis; PV, polycythemia ve AML, acute myeloid leukemia; CEL, chronic eosinophilic leukemia; CML, chronic myeloid leukemia; CMML, chronic myelomonocytic leukemia; ET, d c a b Diagnostic criteria of MPN associated with ETV6–PDGFRB fusion gene or other rearrangement of PDGFRB Table 1 hypereosinophilia, chronic eosinophilicassociated leukemia, MPNs with NOS, rearrangements of clonal myeloid neoplasms-associated eosinophilia (criteria 2 above) and eosinophilia- of chronic eosinophilicincreased leukemia bone marrow associated mast with cells. splenomegaly, a marked elevation of serum vitamin B12, elevation of serum tryptase and Diagnostic criteria of chronic eosinophilic leukemia, not otherwise specified (NOS) Diagnostic criteria of MPN or acute leukemia associated with FGFR1 rearrangement ( months and tissue damage must be present. If there is no tissue damage, idiopathic hyerpeosinophilia is the preferred diagnosis. 5. If there is no demonstrable disease that could cause eosinophilia, no6. abnormal If T-cell population, requirements 1–4 and have no been evidence met, of and a if clonal the myeloid myeloid disorder, cells demonstrate a clonal cytogenetic abnormality or clonality is shown by other means, 4. Exclude T-cell population with aberrant phenotype and abnormal cytokine population 3. Exclude other neoplastic disorders in which eosinophilia is part of the neoplastic clone: 2. Exclude all neoplastic disorders with secondary, reactive eosinophilia: 1. Exclude all causes of reactive eosinophilia secondary to: ( Diagnostic criteria of an MPN Leukemia 2000 SPOTLIGHT Five years since the discovery of FIP1L1–PDGFRA J Gotlib and J Cools 2001 Table 2 Overview of selected molecular studies of HES/CEL that establish the frequency of the FIP1L1–PDGFRA fusion

Number of patients Number of FIP1L1–PDGFRA-positive Reference patients (frequency in %; no. of male/female)

16 9 (56; 8 M/1 F) Cools et al.8 72 27 (37; 27 M) Baccarani et al.9 376 40 (11; gender NA) Jovanovic et al.10 89 11 (14; 11 M) Pardanani et al.11 741 21 (3; gender NA) Pardanani et al.12 17 8 (47; 7 M/1 F) Vandenberghe et al.13 35 6 (17; 6 M) Roche-Lestienne et al.14 26 10 (38; 9 M/1 F) La Starza et al.15 Abbreviations: CEL, chronic eosinophilic leukemia; HES, hypereosinophilic syndrome; NA, not available.

Table 3 PDGFRA fusion genes identified in patients with MPNs and eosinophilia

Fusion gene Chromosomal aberration Number of Reference cases described

FIP1L1–PDGFRA del(4)(q12q12) 4100 Cools et al.8; Griffin et al.18; and others BCR–PDGFRA t(4;22)(q12;q11) 5 Trempat et al.16; Baxter et al.17; and others KIF5B–PDGFRA Complex karyotype involving 1 Score et al.19 3, 4 and 10 CDK5RAP2–PDGFRA ins(9;4)(q33;q12q25) 1 Walz et al.20 ETV6–PDGFRA t(4;12)(q12;p13) 1 Curtis et al.21 STRN–PDGFRA t(2;4)(p24;q12) 1 Curtis et al.21 Abbreviation: MPN, myeloproliferative neoplasm. persons). When the original case series of Cools et al.8 described FIP1L1–PDGFRA8,18 fusions were identified in 2002–2003 as the FIP1L1–PDGFRA fusion in 9 of 16 patients (56%) who were recurrent rearrangements, with the FIP1L1–PDGFRA fusion initially diagnosed as idiopathic HES or CEL, it was initially felt being the most common fusion. A few other variant PDGFRA that the majority of patients would have their idiopathic fusion genes have now also been described (Table 3).19–21 In hypereosinophilia explained by this cryptic molecular defect. 1994, the group of Golub and Gilliland22 described the ETV6– However, the study was biased due to its preselected study PDGFRB fusion as the first of these fusion genes in patients with population: patients were generally advanced cases for which chronic myelomonocytic leukemia with eosinophilia and other causes of hypereosinophilia had been thoroughly scruti- t(5;12). Since then, a large variety of fusion partners for PDGFRB nized, and they were being evaluated at tertiary referral centers have been described, most of which, however, are single case by expert hematologists who were familiar with the clinical reports (Table 4).23–36 Despite the rare frequency (o1%) of presentation of the myeloproliferative variant of HES with which PDGFRB rearrangements in cytogenetically defined cases of the fusion ultimately segregated. chronic myelomonocytic leukemia and other myeloid The median frequency of the FIP1L1–PDGFRA fusion in neoplasms (for example, atypical CML, juvenile myelomonocy- hypereosinophilia patients across eight published series enrol- tic leukemia, chronic basophilic leukemia, myelodysplastic ling more than 10 patients was 23% (range 3–56%) (Table 2). In syndrome/MPN overlap), their recognition is essential given the a prospective, multicenter Italian study of 169 patients with exquisite sensitivity of such cases to imatinib. eosinophilia, 72 were diagnosed with either primary eosino- Fusions involving the FGFR1 gene are similarly rare. The philia or HES. Twenty-seven of the 63 patients who provided association of t(8;13)(p11;q11) with lymphoblastic lymphoma consent for testing were found to carry the FIP1L1-PDGFRA with eosinophilia and myeloid hyperplasia (for example, SPOTLIGHT rearrangement (43%).9 However, the incidence of the FIP1L1– 8p11 myeloproliferative syndrome (EMS)) was initially de- PDGFRA fusion was only 16% of the 169 patients initially scribed in 1995, followed by the discovery of the ZNF198– enrolled with a diagnosis of eosinophilia. Similarly, the fusion FGFR1 fusion gene in 1998 by four groups.37–40 Additional was found in only 40 of 376 individuals (11%) in a European fusion partners for FGFR1, including BCR, have since been trial of patients with persistent, unexplained hypereosinophi- described (Table 5).41–47 The FGFR1 rearrangement can be lia.10 In a Mayo series, 11 of 89 patients (12%) with moderate- found in both myeloid and lymphoid cells, suggesting an origin to-severe eosinophilia were FIP1L1–PDGFRA positive.11 In a in a multipotent hematopoietic progenitor, and thus the basis for follow-up series of 714 unselected patients with eosinophilia, the disease’s alternate designation of ‘stem cell leukemia/ only 3% were fusion positive.12 Additional studies are listed in lymphoma syndrome.’ EMS manifests an aggressive course Table 2.13–15 Despite these studies having their own selection and therefore early allogeneic transplantation is often recom- biases, these data support a FIP1L1–PDGFRA fusion incidence mended. Small molecule inhibition of the constitutively of approximately 10–20% among patients presenting with activated FGFR1 may also hold promise, as idiopathic hypereosinophilia in developed countries. demonstrated in the case of a patient with a ZNF198–FGFR1 In addition to the FIP1L1–PDGFRA fusion gene, variant fusion who responded to PKC412.48 PDGFRA fusion genes, as well as different PDGFRB and FGFR1 Within all these different fusion genes, the FIP1L1–PDGFRA fusion genes have been described in MPNs with eosinophilia. In fusion is quite unique as it is generated by a cryptic the case of PDGFRA fusions, both the BCR–PDGFRA16,17 and chromosomal , rather than a translocation. All other

Leukemia Five years since the discovery of FIP1L1–PDGFRA J Gotlib and J Cools 2002 Table 4 PDGFRB fusion genes in addition to ETV6–PDGFRB identified in patients with MPNs and eosinophiliaa

Fusion gene Chromosomal abnormality Diagnosis Reference

WDR48–PDGFRB t(1;3;5)(p36;p21;q33) CEL Curtis et al.23 (Abstract) GPIAP1–PDGFRB der(1)t(1;5)(p34;q33), der(5)t(1;5)(p34;q15), CEL Walz et al.24 der(11)ins(11;5)(p12;q15q33) TPM3–PDGFRB t(1;5)(q21;q33) CEL Rosati et al.25 PDE4DIP–PDGFRB t(1;5)(q23;q33) MPN/MDS with Wilkinson et al.26 eosinophilia PRKG2–PDGFRB t(4;5;5)(q23;q31;q33) Chronic basophilic Walz et al.24 leukemia GOLGA4–PDGFRB t(3;5)(p21–25;q31–35) CEL Curtis et al.23 (Abstract) HIP1–PDGFRB t(5;7)(q33;q11.2) CMML with eosinophilia Ross et al.27 CCDC6–PDGFRB t(5;10)(q33;q21) aCML/MPN with Schwaller et al.28; Kulkarni et al.29 eosinophilia GIT2–PDGFRB t(5;12)(q31–33;q24) CEL Walz et al.24 NIN–PDGFRB t(5;14)(q33;q24) Ph-negative CML Vizmanos et al.30 KIAA1509–PDGFRB t(5;14)(q33;q32) CMML with eosinophilia Levine et al.31 CEV14–PDGFRB t(5;14)(q33;q32) AML with eosinophilia Abe et al.32 at relapseb 33

SPOTLIGHT TP53BP1–PDGFRB t(5;15)(q33;q22) Ph-negative CML Grand et al. with eosinophilia NDE1–PDGFRB t(5;16)(q33;p13) CMML La Starza et al.34 RABEP1–PDGFRB t(5;17)(q33;p13) CMML Magnusson et al.35 SPECC1–PDGFRB t(5;17)(q33;p11.2) JMML Morerio et al.36 Abbreviations: aCML, atypical chronic myeloid leukemia; CEL, chronic eosinophilic leukemia; CML, chronic myeloid leukemia; CMML, chronic myelomonocytic leukemia; JMML, juvenile myelomonocytic leukemia; MPN/MDS, myeloproliferative neoplasm/myelodysplastic syndrome; Ph: Philadelphia chromosome. aModified from the 2008 WHO Criteria.7 bNot a chronic MPN, but diagnosed in a case of AML at relapse.

Table 5 MPNs with eosinophilia and rearrangement of FGFR1a extramedullary hematopoiesis and bone marrow fibrosis.49 In 2005, the PCM1–JAK2 fusion was identified as a second Fusion gene Chromosomal Reference recurrent molecular abnormality, which results in dysregulation abnormality of JAK2 tyrosine kinase activity due to oligomerization mediated 50 37 by the coiled-coil domains of PCM1. The chimeric oncopro- ZNF198–FGFR1 t(8;13)(p11;q12) Xiao et al. ; tein results from the t(8;9)(p22;p24) chromosomal translocation Reiter et al.38; Popovici et al.39; and may have pleiotropic clinical presentations, including Smedlley et al.40 atypical CML, AML, acute B- and T-cell lymphoblastic CEP110–FGFR1 t(8;9)(p11;q33) Guasch et al.41 , often with peripheral eosinophilia.50–54 The clinical FGFR1OP1–FGFR1 t(6;8)(q27;p11–12) Popovici et al.42 course of the PCM1–JAK2 cases reported to date appears to be BCR–FGFR1 t(8;22)(p11;q11) Demiroglu et al.43 44 more aggressive than the JAK2 V617F-associated chronic MPDs. TRIM24–FGFR1 t(7;8)(q34;p11) Belloni et al. JAK2 inhibitors currently in phase I testing exhibit potential for MYO18A–FGFR1 t(8;17)(p11;q23) Walz et al.45 HERVK–FGFR1 t(8;19)(p12;q13.3) Guasch et al.46 treating these neoplasms characterized by constitutive JAK2 FGFR1OP2–FGFR1 ins(12;8)(p11;p11p22) Grand et al.47 activation. Abbreviation: MPN, myeloproliferative neoplasm. aModified from the 2008 WHO Criteria.7 In addition, FGFR1 rearrangement has been found in association with t(8;12)(p11;q15) Biology of the FIP1L1–PDGFRa fusion and t(8;17)(p11;q25) but the suspected involvement of FGFR1 in t(8;11)(p11;p15) was not confirmed. Mechanism of activation of the FIP1L1–PDGFRa tyrosine kinase The structure of the FIP1L1–PDGFRa fusion resembles PDGFRA, PDGFRB or FGFR1 fusions are generated by the structure of the ETV6–PDGFRb, ZNF198–FGFR1 and BCR– reciprocal translocations or by complex rearrangements, the ABL , for which homotypic oligomerization mediated by latter usually being identified in single cases. domains within ETV6, ZNF198 or BCR has been documen- Finally, in addition to rearrangements of PDGFRA, PDGFRB ted.55–57 Oligomerization of the corresponding fusion proteins and FGFR1, the PCM1–JAK2 fusion gene was recently dis- leads to activation of the tyrosine kinase domains, which in turn covered in various eosinophilia-associated leukemias. The activate downstream signaling pathways regulating cell proli- acquired JAK2 V617F is found in 495% of patients feration and survival. In contrast to this, we have been unable to with polycythemia vera, approximately 50% of patients with demonstrate oligomerization of the FIP1L1–PDGFRa fusion essential thrombocythemia or primary myelofibrosis and in a protein. However, we have observed that interruption of the small proportion of patients with atypical myeloproliferative juxtamembrane of PDGFRa is indispensable for kinase activa- disorders.49 The mutation results in constitutive activation of the tion in the context of FIP1L1–PDGFRa.58 Indeed, it was tyrosine kinase, and transplantation of JAK2 V617F-transduced previously shown that or duplications within the bone marrow to mice can recapitulate phenotypic aspects of juxtamembrane region of the PDGFR family of tyrosine kinases human myeloproliferative disease including erythrocytosis, can cause constitutive activation of their kinase activity.59,60

Leukemia Five years since the discovery of FIP1L1–PDGFRA J Gotlib and J Cools 2003 ligand (PDGF) dependent activation: lia, as mice expressing FIP1L1–PDGFRA in their bone marrow WW cells develop a general myeloproliferative disease without PDGFRα, PDGFRβ N C 65 TM JM kinase eosinophilia. Expression of FIP1L1–PDGFRA together with overexpression homodimerization dependent activation: of IL-5, however, mimics the disease much better in the mouse, with typical features of HES such as tissue infiltration of ETV6-PDGFRβ N C 67 DIMTM JM kinase . Similarly, a study of polymorphic variation at the IL-5 -a (IL5RA) gene revealed an association between a SNP in the 50 UTR of IL5RA and the eosinophil juxtamembrane interruption dependent activation: count/presence of tissue infiltration in FIP1L1–PDGFRA-positive 68 FIP1L1-PDGFRα N C HES patients. These data suggest that FIP1L1–PDGFRA alone kinase is not sufficient to explain the development of HES/CEL, and that partial JM additional factors such as IL-5 signaling may also be implicated Figure 1 The structure and mechanism of activation of PDGFRa or at least may influence the severity of the disease. and PDGFRb fusions. N: N-terminal site; C: C-terminal site; TM: transmembrane domain; JM: juxtamembrane domain; kinase: kinase domain; DIM: dimerization domain; W: tryptophan of the WW motif. Detection of the FIP1L1–PDGFRA fusion and variant PDGFRA fusions Also in cancer, this mechanism is well known from the internal tandem duplications in FLT3 and KIT in AML or gastrointestinal In many cases, patients expressing PDGFRA, PDGFRB,or 61,62 stromal tumors, respectively. Fusion of FIP1L1 to the FGFR1 fusion genes will have an abnormal karyotype indicating PDGFRa protein yields a constitutive active tyrosine kinase a rearrangement of 4q12 (PDGFRA), 5q31–33 (PDGFRB)or only if the juxtamembrane domain of PDGFRa is partially or 8p11–12 (FGFR1). Therefore, an important message to hemato- 58 completely removed. This is what happens in patients with the logists is not to ignore karyotyping in cases with eosinophilia in FIP1L1–PDGFRa fusion: there are very different break points order to rapidly identify patients with chromosomal rearrange- within FIP1L1, but the break points within the PDGFRA gene are ments who may benefit from targeted therapy with specific tightly clustered, invariably resulting in the removal of part of kinase inhibitors. In addition to karyotyping, FISH analysis with the juxtamembrane domain and activation of the kinase probes flanking the PDGFRA, PDGFRB and FGFR1 genes domain. A similar mechanism has now also been described in remains valid in cases with obvious chromosomal rearrange- cases of the PRKG2–PDGFRb fusion, in which the break points ments to confirm that the break points are indeed within these in the PDGFRB gene are also within the juxtamembrane genes, as well as in cases without these specific rearrangements 63 region. In contrast, in other PDGFRb fusions, the juxtamem- to check for possible cryptic rearrangements of these kinases. An brane is completely intact, and in these fusions the activation of important example of such cryptic rearrangement is the 4q12 PDGFRb kinase activity is obtained through oligomerization deletion that causes the FIP1L1–PDGFRA fusion. mediated by the fusion partner (Figure 1). The generation of the fusion between the 50 part of the FIP1L1 gene and the 30 part of the PDGFRA gene occurs through an uncommon mechanism by which the 800 kb genomic region Role of FIP1L1 between the two genes is deleted (Figure 2).8 This deletion On the basis of the results described above, the role of the begins within the FIP1L1 gene, with variable break points in the FIP1L1 part in the FIP1L1–PDGFRa fusion is less important than different patients, and ends in 12 of PDGFRA. Owing to the role of ETV6 in the ETV6–PDGFRb fusion. Nevertheless, it is the fact that the deletion is only 800 kb in size, this genomic still the case that the FIP1L1–PDGFRA fusion gene is under rearrangement remains undetectable by standard cytogenetics, control of the FIP1L1 promoter and translation start, and the but can be detected by FISH with specific probes. FISH probes FIP1L1 part in the fusion may determine the stability and that hybridize to the region between the FIP1L1 and PDGFRA subcellular localization of the fusion protein. Also, while FIP1L1 genes are now commonly used to detect the presence of the seems dispensable for transformation of Ba/F3 cells, Buitenhuis deletion. As the CHIC2 gene is located in this region, this FISH et al.64 documented the differences in in vitro colony formation test is sometimes referred to as ‘FISH to detect the CHIC2 between FIP1L1–PDGFRA transduced CD34 þ cells and cells deletion.’69 SPOTLIGHT transduced by a deletion variant lacking part of FIP1L1. A more sensitive way to detect the presence of the FIP1L1– PDGFRA fusion gene in the blood of eosinophilia patients is the use of (nested) reverse transcription (RT)-PCR. Despite the fact Insights in the mechanism of FIP1L1–PDGFRA-induced that the break points in the FIP1L1 gene can be very different eosinophilia from patient to patient, a single primer combination is sufficient FIP1L1–PDGFRa is required to stimulate proliferation and to detect the fusion transcript from most patients. In some mediate survival of the eosinophils in CEL patients, through patients, however, the fusion remains difficult to detect, which activation of several signaling pathways including phosphoino- may be due to low-level expression of the fusion gene, sitol 3-kinase, ERK 1/2 and STAT5.8,64 The exact mechanism, heterogeneity in the FIP1L1 break points and difficulties with however, by which FIP1L1–PDGFRa preferentially affects FISH in eosinophilia cases. Therefore, a combination of RT-PCR eosinophils remains unclear. The essential role of FIP1L1– and FISH provides the best chance of identifying FIP1L1– PDGFRA is clear from in vitro studies with the EOL-1 cell line, PDGFRA-positive cases, and several groups are working on from mouse models of FIP1L1–PDGFRA induced disease, and additional tests that could further limit the chances of false from the remarkable responses of FIP1L1–PDFGRA-positive CEL negative results. Despite some minor problems associated with patients to imatinib treatment.8,18,65,66 The mouse model, RT-PCR to detect the FIP1L1–PDGFRA transcript, nested however, has also suggested that expression of the FIP1L1– RT-PCR or quantitative RT-PCR remains the method of choice PDGFRA fusion is most likely not sufficient to cause eosinophi- to monitor the response of the disease to therapy (see below).

Leukemia Five years since the discovery of FIP1L1–PDGFRA J Gotlib and J Cools 2004 a - 4q12

centromeric telomeric 800 kb deleted region

FIP1L1 GSH2 PDGFRA

LNX CHIC2

120K16 3H20 238H24 24O10

b - del(4)(q12q12) cryptic splice site type I FIP1L1 PDGFRA SPOTLIGHT DNA exon x truncated exon 12 exon 13

mRNA ......

cryptic splice site type II FIP1L1 PDGFRA DNA exon x truncated exon 12 exon 13

mRNA ......

Figure 2 Schematic representation of the 4q12 and the consequences of the deletion causing the FIP1L1–PDGFRA fusion. (a) Shows the normal 4q12 region, indicating where the different genes are located and where the 800 kb deletion occurs in cases with the FIP1L1–PDGFRA fusion. The four green bars denote possible probes that can be used to detect the deletion by fluorescence in situ hybridization. (b) Shows the consequences of the deletion: FIP1L1 usually breaks within an intron, while PDGFRA always breaks within exon 12. The consequence of the deletion at the DNA level is that a part of an intron of FIP1L1 (various introns possible) is directly fused to a piece of exon 12 of PDGFRA. To obtain splicing between FIP1L1 and PDGFRA, cryptic splice sites need to be used, as the normal splice site at the beginning of exon 12 is removed by the deletion. Dependent on the break points within FIP1L1 and PDGFRA, this cryptic splice site is either located within exon 12 of PDGFRA (type I fusion) or within the intron of FIP1L1 (type 2 fusion). In all cases, this ‘abnormal’ splicing results in the generation of in-frame fusion transcripts encoding catalytically active fusion proteins.

As a variety of PDGFRA and PDGFRB fusion genes involving pathologically confirmed cases of systemic mastocytosis with partner genes other than FIP1L1 and ETV6 have also been eosinophilia (SM-eo).69 Histopathologically, the bone marrows detected in hypereosinophilia patients (Tables 3 and 4), of patients with FIP1L1–PDGFRA-positive SM-Eo exhibit less detection of these rare variants remains important, as these dense clusters of mast cells by tryptase immunostaining than are patients also benefit from imatinib treatment. These cases can be typically seen in SM, particularly cases with the common identified using specific primer sets for these fusions, or D816V KIT mutation.11 However, in some cases of CEL with alternatively, can be identified using quantitative RT-PCR to increased bone marrow mast cells, the mast cells may exhibit detect increased levels of PDGFRA expression.19 spindle-shaped morphology, form multifocal clusters and aberrant surface expression of CD25, major and minor criteria, which establish the basis for a WHO diagnosis of SM. Such Disease phenotypes associated with FIP1L1–PDGFRA cases may be considered a hybrid category of SM-CEL, wherein the CEL component is the associated hematologic non- FIP1L1–PDGFRA is a clonal marker associated with the lineage disease, pathogenetically driven by FIP1L1–PDGFRA. myeloproliferative variant of hypereosinophilia.2,8 These However, this may be an insufficient explanation as the FIP1L1– patients often present with organomegaly, hypercellular bone PDGFRA rearrangement has been found in a variety of myeloid marrows with increased mast cells and/or myelofibrosis, cell types (, , eosinophils), including mast increased serum tryptase levels, and historically carried a poor cells, consistent with a mutational origin in a multipotent prognosis before the successful therapeutic application of hematopoietic progenitor.70 When clinical and pathologic imatinib.2,8,13 Shortly, after its initial discovery in HES/CEL features of CEL and SM co-exist in the same patient, it is patients, the Mayo group linked the FIP1L1–PDGFRA fusion to certainly possible, if not likely, that both diseases originate from

Leukemia Five years since the discovery of FIP1L1–PDGFRA J Gotlib and J Cools 2005 the same clone. The relative ‘penetrance’ of eosinophil versus Clinical features mast cell symptoms and organ involvement may be modified by In the pre-fusion era, the cumulative frequencies of organ- host- or disease-related factors that have yet to be ascertained. specific manifestations of HES were previously described in The FIP1L1–PDGFRA fusion and D816V KIT appear to be three case series.73–75 In addition to universal involvement by mutually exclusive oncogenic mutations, as they have not been the bone marrow, the most common organ systems involved simultaneously reported in the same patient. Investigators from included cardiac (58%), dermatologic (56%), neurologic (54%), the NIH and Ann Arbor could reliably partition D816V KIT- pulmonary (49%), splenic (43%) and 20–30% involvement of positive SM-Eo from FIP1L1–PDGFRA-positive CEL into clini- the ocular and liver/gallbladder/GI systems.73–76 Although cally distinguishable entities based on several clinical and FIP1L1–PDGFRA-positive patients exhibit the previously al- laboratory features.71 In the D816V KIT-positive SM-Eo cohort, luded to myeloproliferative features, inconsistent reporting of gastrointestinal symptoms, urticaria pigmentosa, thrombocyto- their clinical features presentations makes comparisons to both sis, the median serum tryptase value and the presence of dense historically described HES and FIP1L1–PDGFRA-negative pa- mast cell aggregates in the bone marrow were statistically tients challenging. In one larger series, less frequent lung and significantly elevated or more frequently represented compared skin involvement, and more frequent splenomegaly character- to patients with FIP1L1–PDGFRA-positive CEL. Conversely, ized FIP1L1–PDGFRA-positive compared to FIP1L1–PDGFRA- male sex, cardiac and pulmonary symptoms, median peak negative cases of hypereosinophilia.9 It is possible that a absolute eosinophil count, the eosinophil to tryptase ratio and proportion of the fusion-negative patients may have represented serum B12 levels were significantly elevated or more frequently lymphocyte-variant hypereosinophilia (not tested in the study), represented in the FIP1L1–PDGFRA-positive CEL group. A as such individuals have a high rate of cutaneous manifestations scoring system incorporating these clinical findings and (pruritis, urticaria, angioedema, eczema, erythroderma). laboratory tests was generated that could reliably predict the molecular status (D816V KIT versus FIP1L1–PDGFRA)of patients with peripheral eosinophilia and increased marrow Therapy mast cell burden.71 More recently, the FIP1L1–PDGFRA fusion was also identified Imatinib therapy of FIP1L1–PDGFRA-positive CEL in five patients with AML (FAB subtypes M0, M2 and M4) and in The first report of imatinib treatment of HES was by Schaller two patients with lymphoblastic T-cell non-Hodgkin’s lympho- and Burkland77 in an online medical journal in 2001 (Table 6). ma.72 A search for the FIP1L1–PDGFRA rearrangement was Several case reports and small case series followed in 2001– prompted by the presence of eosinophilia either preceding or 2002, highlighting the dramatic hematological responses of contemporaneous with the diagnosis of AML or T-NHL, or patients with HES empirically treated with imatinib primarily in because eosinophilia persisted despite a complete hematologic the dose range of 100–400 mg daily.78–80 Complete and rapid remission after intensive . In the T-NHL cases, hematologic remissions, with normalization of eosinophilia, lymphoid involvement by FIP1L1–PDGFRA was confirmed by were observed in a high proportion of patients. the presence of the CHIC2 deletion by FISH in CD3 þ The presence of a normal karyotype in responding patients T lymphocytes. implicated a subtle mutation or cryptic rearrangement of a

Table 6 Different steps towards the discovery of the FIP1L1–PDGFRA fusion gene

2001: Imatinib is approved for the treatment of BCR–ABL-positive CML. 2002: A patient with HES is being treated with imatinib. There is no good reason to try imatinib therapy for this patient, but he suffers from the side effects of other treatments. It is reasoned that HES resembles CML and that maybe imatinib could also be efficacious to treat HES. The patient shows a ‘miraculous response’ to imatinib, with complete resolution of symptoms and normal eosinophil levels reached within a few weeks of therapy.77 2002: Five additional patients with HES are treated with imatinib: four of five patients respond to low doses of imatinib. The target of imatinib remains unknown, but BCR–ABL and KIT (D816V) are excluded.78 2002: The kinase domains of PDGFRA, PDGFRB, ABL, and KIT are being sequenced to identify activating mutations in any of these kinases in HES patients with response to imatinib. No mutations are identified. 2002: The study by the group of Gary Gilliland includes one HES/CEL patient with a t(1;4)(q44;q12) is included in the study. The 4q12 break point SPOTLIGHT points towards a possible role of the PDGFRA gene in this rearrangement, but at this time it is not believed to be very important since this is not a general finding in all HES patients with response to imatinib. The molecular analysis of this translocation is started by FISH analysis to determine if the 4q12 break point is indeed within the PDGFRA gene. May 2002: The FISH for the 4q12 region with FISH probes flanking the PDGFRA gene show puzzling results: it is clear that PDGFRA is translocated to chromosome 1, but the probe upstream of PDGFRA (the CHIC2 locus) seems to be deleted. It is concluded that this could be a relatively complex rearrangement involving a partial deletion of . These data indicate that PDGFRA is likely to be involved in the rearrangement, and RACE analysis to identify the possible fusion partner of PDGFRA is started. June 2002: Analysis of the sequencing results of the RACE using BLAST suggest that a part of a gene on chromosome 4 is fused to PDGFRA, while a gene on chromosome 1 was expected. Initially these data seem a bit strange and disappointing: could it be an artifact of the PCR protocol? Then, upon more detailed analysis it is found that the fusion with the novel gene on chromosome 4 is a nice in-frame fusion with PDGFRA, and it is noted that this gene is only 800 kb upstream of PDGFRA. Now everything becomes clear: the FISH data indicated a deletion upstream of PDGFRA, the fusion that is identified could be generated by an 800 kb deletion on chromosome 4 fusing the novel gene to PDGFRA, andFmost importantlyFsuch small deletion would not be visible by standard karyotyping, and thus could be present in other HES patients with response to imatinib. July 2002: Long distance inverse PCR is performed on DNA samples from five additional HES patients and confirms the presence of similar deletions upstream of PDGFRA in four of the five patients. It is now clear that this deletion, leading to the generation of the FIP1L1–PDGFRA fusion, can explain many cases of ‘idiopathic’ HES, and the response of these patients to imatinib. 2002: At around the same time, researchers from Theravance Inc. observe that the EOL-1 cell line is sensitive to imatinib, and identify the FIP1L1– PDGFRa fusion protein in these cells using mass spectrometry. They confirm that the same fusion is present in HES patients. March 2003 and June 2003: The data from these studies are published.8,18

Leukemia Five years since the discovery of FIP1L1–PDGFRA J Gotlib and J Cools 2006 tyrosine kinase as the therapeutic target of imatinib, which was was noted only several weeks after stoppage of imatinib in four ultimately identified as FIP1L1–PDGFRa.8,18 Of the 16 HES/CEL patients in the Mayo series.12 These data indicate that imatinib patients enrolled in the study of Cools et al.,8 11 were treated can suppress, but not eradicate the FIP1L1–PDGFRA clone, and with imatinib. Hematologic responses were observed in 10 of 11 that ongoing therapy is warranted. Although 100 mg daily may HES patients treated with imatinib doses of 100–400 mg daily. be sufficient to achieve a molecular remission in some patients, The median time to response was 4 weeks (range 1–12 weeks). others may require higher maintenance doses in the range of Nine of 10 patients demonstrated a durable hematologic 300–400 mg daily. However, in a recent series, maintenance response (lasting X3 months), with a median duration of 7 dosing of 100–200 mg weekly was sufficient to sustain a months at the time of publication. Now, more than 5 years later, molecular remission in five of six fusion-positive patients.83 the overwhelmingly majority of these patients remain in The ability of imatinib to produce a molecular remission may hematologic remission. reflect differences in drug metabolism/absorption between Numerous studies have since confirmed the hematologic individuals, disease burden and susceptibility of the various benefit of imatinib in FIP1L1–PDGFRA-positive CEL. Similar to FIP1L1–PDGFRA breakpoints to the drug; however, the poten- CML, sensitive real-time quantitative PCR-based assays are also tial contributions of these factors have not been systematically used to follow in-depth molecular responses. Molecular analyzed. remissions were first reported by the NIH group in five of six Finally, it must also be noted that FIP1L1–PDGFRA fusion FIP1L1–PDGFRA-positive patients after 1–12 months of imatinib negative HES patients may benefit from imatinib therapy. In this therapy.81 Several additional reports have since described group, however, hematologic responses tend to be partial, short- molecular remissions in imatinib-treated patients with FIP1L1– lived, and may reflect nonspecific drug-related myelosuppres- SPOTLIGHT PDGFRA-positive disease or after bone marrow transplantation. sion.8,9 Alternatively, some of the cases with complete The natural history of imatinib-treated FIP1L1–PDGFRA- responses may be patients in which the PDGFRA or PDGFRB positive CEL was recently reported by an Italian study which rearrangement remained undiscovered. It may thus be valid to prospectively followed 27 patients (all male) for a median try imatinib treatment in HES patients without detectable follow-up period of 25 months (range 15–60 months).9 Patients PDGFR rearrangements. were dose escalated from an initial dose of 100 mg daily to a final dose of 400 mg daily after the first month (median daily dose 339 mg). A complete hematologic remission was achieved Safety issues of imatinib in FIP1L1–PDGFRA-positive in all patients within 1 month, and all patients became RT-PCR disease negative for the FIP1L1–PDGFRA fusion after 1–10 months of The safety profile of imatinib-treated patients with FIP1L1– therapy (median 3 months). All 24 patients who continued PDGFRA-positive disease generally parallels that of CML. imatinib therapy remained PCR negative during a follow-up However, several cases of incipient cardiogenic shock have period of 6–56 þ months (median 19 months). been reported in several FIP1L1–PDGFRA-positive patients after Using real-time quantitative PCR, a European study prospec- initiation of imatinib therapy.84,85 Endomyocardial biopsy tively assessed the natural history of molecular responses to revealed myocyte injury, likely an acute inflammatory response imatinib (dose range100–400 mg daily) in 40 of 376 (11%) HES to imatinib resulting in degranulation of infiltrating eosinophils patients who were positive for the FIP1L1–PDGFRA fusion.10 exacerbated by imatinib. Early use of high-dose Fusion-positive patients exhibited higher absolute and % led to the improvement of left ventricular dysfunction and eosinophil counts compared to fusion-negative patients, but clinical recovery. Currently, prophylactic use of steroids during there was no correlation between the load of FIP1L1–PDGFRA the first 7–10 days of imatinib treatment is recommended for expression and variables such as the white blood cell count, patients with known cardiac disease and/or elevated serum absolute or % eosinophil count, or % cells with the CHIC2 troponin T levels.86 deletion by interphase FISH. A variability of up to 3 logs in the normalized FIP1L1–PDGFRA transcript load was found in patient samples before imatinib treatment. Among 11 patients Resistance to imatinib in FIP1L1–PDGFRA-positive with high pretreatment transcript levels, all achieved a 3-log disease reduction in transcript levels by 1 year of therapy, and 9 of 11 With more than 5 years of experience in the imatinib treatment patients (82%) achieved a molecular remission. of FIP1L1–PDGFRA positive disease, only four cases of acquired It has now become clear that despite the in-depth and durable resistance have been reported.8,18,86,87 We identified the first molecular responses with imatinib, discontinuation of the drug case of imatinib resistance in a patient with advanced AML can lead to relapse. In the Italian study, three patients who arising from CEL.8 He exhibited the FIP1L1–PDGFRA fusion in discontinued imatinib after 12, 14 and 15 months of therapy addition to a complex karyotype. Despite a complete hemato- experienced a rise in FIP1L1–PDGFRA transcript levels; upon logic remission, the patient relapsed after 5 months of therapy, restart of imatinib, fusion transcripts again became undetectable coinciding with the identification of a T674I mutation within the after 2–5 months of therapy.9 In the European trial, withdrawal ATP-binding domain of PDGFRa. In agreement with this, Ba/F3 of imatinib in two patients was followed by a rapid rise in cells transformed by the FIP1L1–PDGFRa T674I mutant were FIP1L1–PDGFRA fusion transcripts, with one of these patients 1000-fold more resistant to imatinib, compared to cells achieving a second molecular remission after reinstitution of transformed by the wild-type fusion.8 The observed acquired imatinib.10 In a dose de-escalation trial of imatinib in five resistance in this CEL patient also confirmed that the FIP1L1– patients who had achieved a stable hematologic and molecular PDGFRa fusion protein was indeed the therapeutic target of remission at 300–400 mg daily for at least 1 year, molecular imatinib. Additional cases of molecular resistance were similarly relapse was observed in all patients: in one patient after 5 due to the PDGFRa T674I mutation, one in a patient with CEL months of a reduced dose of 100 mg daily, and in four patients evolving to myeloid blast crisis (also after 5 months after 2–5 months after discontinuation of drug.82 Molecular remis- imatinib therapy), and one in a patient with Langerhans sions could be re-established with reinduction of imatinib in all histiocytosis with eosinophilia treated with multiagent chemo- cases at a dose range of 100–400 mg daily. Hematologic relapse therapy.86,87 Recently, we observed the development of

Leukemia Five years since the discovery of FIP1L1–PDGFRA J Gotlib and J Cools 2007 Table 7 Most important inhibitors for treatment of FIP1L1–PDGFRA-positive CEL

Inhibitor Main Activity against Activity against FIP1L1–PDGFRa Clinical use target FIP1L1–PDGFRa T674I

Imatinib (Gleevec, BCR–ABL Ba/F3 cells: IC50:3nM Ba/F3 cells: IC50: 45 mM Complete molecular response in Glivec, Novartis) EOL1 cells: IC50:1nM Resistance of this mutant to most patients with FIP1L1– Efficacy demonstrated in imatinib confirmed in mouse model PDGFRA-positive CEL; mouse model No activity in patients with FIP1L1–PDGFRA T674I PKC412 (Midostaurin, FLT3 Ba/F3 cells: IC50:B150 nM Ba/F3 cells: IC50: B100 nM Not tested in CEL Novartis) EOL-1 cells: IC50:B20 nM Efficacy demonstrated in mouse Efficacy demonstrated in mouse model model Sorafenib (Nexavar, BRAF, Ba/F3 cells: IC50:B5nM Ba/F3 cells: IC50: B50 nM One patient treated with Bayer) VEGFR EOL-1 cells: IC50:B1nM Efficacy tested in cellular models FIP1L1–PDGFRA T674I mutant Efficacy tested in cellular models

Abbreviation: IC50: 50% inhibitory concentration. resistance to imatinib in a fifth patient (Lierman E et al., benefit for patients affected by these MPNs. However, to reap unpublished data). This patient presented with FIP1L1– the benefits of this success, it is critical that clinicians test for the PDGFRA-positive AML with eosinophilia, and developed occult FIP1L1–PDGFRA fusion in the context of undiagnosed imatinib resistance again due to the T674I mutation. Taken hypereosinophilia and recognize hallmark translocations invol- together, these data suggest that the T674I mutation is the most ving 5q31Bq33, 4q12, 8p13 and 9p24, as these may represent common, if not the only, mutation that may cause clinical ‘druggable’ molecular rearrangements (for example, PDGFRB, resistance to imatinib in patients with FIP1L1–PDGFRA-positive PDGFRA, FGFR1 and JAK2, respectively). Ongoing work is acute leukemia. To date, resistance to imatinib has not been aimed at identifying the molecular basis for patients with reported in cases with the chronic phase of eosinophilic idiopathic hypereosinophilia and developing tyrosine kinase leukemia. Acquired resistance to imatinib in FIP1L1–PDGFRA- inhibitors with activity against MPNs with rearranged FGFR1 mediated disease is considerably rare compared to CML. It is and JAK2. Fortunately, 5 years after the discovery of FIP1L1– unknown whether this relates to the 100-fold sensitivity of the PDGFRA, acquired resistance to imatinib has been a rare FIP1L1–PDGFRA fusion to imatinib compared to the BCR-ABL problem. tyrosine kinase or other biological properties of the FIP1L1– PDGFRA containing clone. References Treatment of imatinib-resistant FIP1L1–PDGFRa T674I: insights from preclinical studies 1 Bain B, Pierre R, Imbert M, Vardiman JW, Brunning RD, Flandrin G. Chronic eosinophilic leukaemia and the hypereosinophilic a The FIP1L1–PDGFR T674I mutation is analogous to the T315I syndrome. In: Jaffe ES, Harris NL, Stein H, Vardiman JW (eds). BCR-ABL mutation in CML, which confers broad-spectrum World Health Organization of tumours: tumours of haematopoietic resistance to the tyrosine kinase inhibitors imatinib, dasatinib and lymphoid tissues. IARC Press: Lyon, France, 2001, pp 29–31. and nilotinib.88 We tested several known PDGFR inhibitors for 2 Klion AD, Noel P, Akin C, Law MA, Gilliland DG, Cools J et al. their activity against the imatinib-resistant T674I mutant form of Elevated serum tryptase levels identify a subset of patients with a FIP1L1–PDGFRa using a cellular screen in Ba/F3 cells (Table 7). myeloproliferative variant of idiopathic hypereosinophilic syn- drome associated with tissue fibrosis, poor prognosis, and imatinib PKC412, a potent FLT3 inhibitor that is in clinical development responsiveness. Blood 2003; 101: 4660–4666. for the treatment of AML, was the first inhibitor to be identified 3 Simon HU, Plotz SG, Dummer R, Blaser K. Abnormal clones of 65 with activity against the FIP1L1–PDGFRa T674I mutant. Using T cells producing interleukin-5 in idiopathic hypereosinophilia. both in vitro and in vivo mouse models, we demonstrated the N Eng J Med 1999; 341: 1112–1120. ability of PKC412 to induce in FIP1L1–PDGFRa 4 Roufosse F, Cogan E, Goldman M. Recent advances in patho- SPOTLIGHT T674I-transformed cells, and to significantly reduce leukocytosis genesis and management of hypereosinophilic syndromes. Allergy 2004; 59: 673–689. a 65 and splenomegaly in a FIP1L1–PDGFR T674I mouse model. 5 Chusid MJ, Dale DC, West BC, Wolff SM. The hypereosinophilic In a second study, we identified sorafenib, a BRAF and VEGFR syndrome. Analysis of fourteen cases with review of the literature. inhibitor approved for the treatment of renal cell carcinoma, as Medicine 1975; 54: 1–27. another potent inhibitor of both FIP1L1–PDGFRa and the T674I 6 Bain BJ, Gilliland DG, Horny H-P, Vardiman JW. Chronic mutant form.89 In addition, nilotinib was also shown to have eosinophilic leukaemia, not otherwise specified. In: Swerdlow S, some activity towards both FIP1L1–PDGFRa and the T674I Harris NL, Stein H, Jaffe ES, Theile J, Vardiman JW (eds). World 90,91 Health Organization Classification of Tumours. Pathology and mutant. These data show that several small molecule kinase Genetics of Tumours of Haematopoietic and Lymphoid Tissues. inhibitors are already available to treat FIP1L1–PDGFRA- IARC Press: Lyon, France, 2008, pp 51–53. positive patients who develop resistance to imatinib. 7 Bain BJ, Gilliland DG, Horny H-P, Vardiman JW. Myeloid and lymphoid neoplasms with eosinophilia and abnormalities of PDGFRA, PDGFRB,orFGFR1. In: Swerdlow S, Harris NL, Stein Conclusions H, Jaffe ES, Theile J, Vardiman JW (eds). World Health Organiza- tion Classification of Tumours. Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. IARC Press: Lyon, Increasing recognition of eosinophilia-associated cytogenetic France, 2008, pp 68–73. and molecular abnormalities, in conjunction with the advent of 8 Cools J, DeAngelo DJ, Gotlib J, Stover EH, Legare RD, Cortes J targeted small molecule inhibitors, has resulted in substantial et al. A tyrosine kinase created by fusion of the PDGFRA and

Leukemia Five years since the discovery of FIP1L1–PDGFRA J Gotlib and J Cools 2008 FIP1L1 genes as a therapeutic target of imatinib in idiopathic associated with eosinophilia: involvement of PDGFRB and hypereosinophilic syndrome. N Engl J Med 2003; 348: 1201–1214. response to imatinib. Blood 2003; 102: 4187–4190. 9 Baccarani M, Cilloni D, Rondoni M, Ottaviani E, Messa F, Merante 27 Ross TS, Bernard OA, Berger R, Gilliland DG. Fusion of S et al. The efficacy of imatinib mesylate in patients with FIP1L1- Huntington interacting protein 1 to platelet-derived growth factor PDGFRalpha-positive hypereosinophilic syndrome. Results of a beta receptor (PDGFbetaR) in chronic myelomonocytic leukemia multicenter prospective study. Haematologica 2007; 92: with t(5;7)(q33;q11.2). Blood 1998; 91: 4419–4426. 1173–1179. 28 Schwaller J, Anastasiadou E, Cain D, Kutok J, Wojiski S, Williams 10 Jovanovic JV, Score J, Waghorn K, Cilloni D, Gottardi E, IR et al. H4/D10S170, a gene frequently rearranged in papillary Metzgeroth G et al. Low-dose imatinib mesylate leads to rapid thyroid carcinoma, is fused to the platelet-derived growth factor induction of major molecular responses and achievement of receptor beta gene in atypical chronic myeloid leukemia with complete molecular remission in FIP1L1-PDGFRA-positive t(5;10)(q33;q22). Blood 2001; 97: 3910–3918. chronic eosinophilic leukemia. Blood 2007; 109: 4635–4640. 29 Kulkarni S, Heath C, Parker S, Chase A, Iqbal S, Pocock CF et al. 11 Pardanani A, Brockman SR, Paternoster SF, Flynn HC, Ketterling Fusion of H4/D10S170 to the platelet-derived growth factor RP, Lasho TL et al. FIP1L1-PDGFRA fusion: prevalence and receptor beta in BCR-ABL negative myeloproliferative disorders clinicopathologic correlates in 89 consecutive patients with with a t(5;10)(q33;q21). Cancer Res 2000; 60: 3592–3598. moderate to severe eosinophilia. Blood 2004; 104: 3038–3045. 30 Vizmanos JL, Novo FJ, Roman JP, Baxter EJ, Lahortiga I, Larrayoz 12 Pardanani A, Ketterling RP, Li CY, Patnaik MM, Wolanskyj AP, MJ et al. NIN, a gene encoding a CEP110-like centrosomal protein, Elliott MA et al. FIP1L1-PDGFRA in eosinophilic disorders: is fused to PDGFRB in a patient with a t(5;14)(q33;q24) and an prevalence in routine clinical practice, long-term experience with imatinib-responseive myeloproliferative disorder. Cancer Res imatinib therapy, and a critical review of the literature. Leuk Res 2004; 64: 2673–2676. 2006; 30: 965–970. 31 Levine RL, Wadleigh M, Sternberg DW, Wlodarska I, Galinsky I, 13 Vandenberghe P, Wlodarska I, Michaux L, Zachee P, Boogaerts M, Stone RM et al. KIAA1509 is a novel PDGFRB fusion partner in SPOTLIGHT Vanstraelen D et al. Clinical and molecular features of FIP1L1- imatinib-responsive myeloproliferative disease associated with a PDFGRA (+) chronic eosinophilic leukemias. Leukemia 2004; 18: t(5;14)(q33;q32). Leukemia 2005; 19: 27–30. 734–742. 32 Abe A, Emi N, Tanimoto M, Terasaki H, Marunouchi T, Saito H. 14 Roche-Lestienne C, Lepers S, Soenen-Cornu V, Kahn JE, Lai JL, Fusion of the platelet-derived growth factor beta receptor to a Hachulla E et al. Molecular characterization of the idiopathic novel gene CEV14 in acute myelogenous leukemia after clonal hypereosinophilic syndrome (HES) in 35 French patients with evolution. Blood 1997; 90: 4271–4277. normal conventional cytogenetics. Leukemia 2005; 19: 792–798. 33 Grand FH, Burgstaller S, Kuhr T, Baxter EJ, Webersinke G, Thaler J 15 La Starza R, Specchia G, Cuneo A, Beacci D, Nozzoli C, Luciano L et al. P53-binding protein 1 is fused to the platelet-derived growth et al. The hypereosinophilic syndrome: fluorescence in situ factor receptor beta in a patient with a t(5;15)(q33;q22) and an hybridization detects the del(4)(q12)-FIP1L1/PDGFRA but not imatinib-responsive eosinophilic myeloproliferative disorder. genomic rearrangements of other tyrosine kinases. Haematologica Cancer Res 2004; 64: 7216–7219. 2005; 90: 596–601. 34 La Starza R, Rosati R, Roti G, Gorello P, Bardi A, Crescenzi B. A 16 Trempat P, Villalva C, Laurent G, Armstrong F, Delsol G, Dastugue new NDE1/PDGFRB fusion transcript underlying chronic myelo- N et al. Chronic myeloproliferative disorders with rearrangement monocytic leukaemia in Noonan Syndrome. Leukemia 2007; 21: of the platelet-derived growth factor alpha receptor: a new clinical 830–833. target for STI571/Glivec. Oncogene 2003; 22: 5702–5706. 35 Magnusson MK, Meade KE, Brown KE, Arthur DC, Krueger LA, 17 Baxter EJ, Hochhaus A, Bolufer P, Reiter A, Fernandez JM, Senent L Barrett AJ et al. Rabaptin-5 is a novel fusion partner to platelet- et al. The t(4;22)(q12;q11) in atypical chronic myeloid leukaemia derived growth factor receptor beta in chronic myelomonocytic fuses BCR to PDGFRA. Hum Mol Genet 2002; 11: 1391–1397. leukemia. Blood 2001; 98: 2518–2525. 18 Griffin JH, Leung J, Bruner RJ, Caligiuri MA, Briesewitz R. 36 Morerio C, Acquila M, Rosanda C, Rapella A, Dufour C, Locatelli F Discovery of a fusion kinase in EOL-1 cells and idiopathic et al. HCMOGT-1 is a novel fusion partner to PDGFRB in juvenile hypereosinophilic syndrome. Proc Natl Acad Sci USA 2003; myelomonocytic leukemia with t(5;17)(q33;p11.2). Cancer Res 100: 7830–7835. 2004; 64: 2649–2651. 19 Score J, Curtis C, Waghorn K, Stalder M, Jotterand M, Grand FH 37 Xiao S, Nalabolu SR, Aster JC, Ma J, Abruzzo L, Jaffe ES et al. et al. Identification of a novel imatinib responsive KIF5B-PDGFRA FGFR1 is fused with a novel zinc-finger gene, ZNF198, in the fusion gene following screening for PDGFRA overexpression in t(8;13) leukaemia/lymphoma syndrome. Nat Genet 1998; 18: patients with hypereosinophilia. Leukemia 2006; 20: 827–832. 84–87. 20 Walz C, Curtis C, Schnittger S, Schultheis B, Metzgeroth G, Schoch 38 Reiter A, Sohal J, Kulkarni S, Chase A, Macdonald DH, Aguiar RC C et al. Transient response to imatinib in a chronic eosinophilic et al. Consistent fusion of ZNF198 to the fibroblast growth factor leukemia associated with ins(9;4)(q33;q12q25) and a CDK5RAP2- receptor-1 in the t(8;13)(p11;q12) myeloproliferative syndrome. PDGFRA fusion gene. Genes Chromosomes Cancer 2006; 45: Blood 1998; 92: 1735–1742. 950–956. 39 Popovici C, Adelaide J, Ollendorff V, Chaffanet M, Guasch G, 21 Curtis CE, Grand FH, Musto P, Clark A, Murphy J, Perla G et al. Jacrot M et al. Fibroblast growth factor receptor 1 is fused to FIM in Two novel imatinib-responsive PDGFRA fusion genes in chronic stem-cell myeloproliferative disorder with t(8;13). Proc Natl Acad eosinophilic leukaemia. Br J Haematol 2007; 138: 77–81. Sci USA 1998; 95: 5712–5717. 22 Golub TR, Barker GF, Lovett M, Gilliland DG. Fusion of PDGF 40 Smedley D, Hamoudi R, Clark J, Warren W, Abdul-Rauf M, receptor beta to a novel ets-like gene, tel, in chronic myelomo- Somers G et al. The t(8;13)(p11;q11–12) rearrangement associated nocytic leukemia with t(5;12) chromosomal translocation. Cell with an atypical myeloproliferative disorder fuses the fibroblast 1994; 77: 307–316. growth factor receptor 1 gene to a novel gene RAMP. Hum Mol 23 Curtis C, Apperley JF, Dang R, Jeng M, Gotlib J, Cross NCP et al. Genet 1998; 7: 637–642. The platelet-derived growth factor receptor beta fuses to two 41 Guasch G, Mack GJ, Popovici C, Dastague N, Birnbaum D, distinct loci at 3p21 in imatinib responsive chronic eosinophilic Rattner JB et al. FGFR1 is fused to the centrosome-associated leukemia. Blood 2005; 106: 909a [Abstract 3253]. protein CEP110 in the 8p12 stem cell myeloproliferative disorder 24 Walz C, Metzgeroth G, Haferlach C, Schmitt-Graeff A, Fabarius A, with t(8;9)(p12;q33). Blood 2000; 95: 1788–1796. Hagen V et al. Characterization of three new imatinib-responsive 42 Popovici C, Zhang B, Gregoire MJ, Gregoire MJ, Jonveaux P, fusion genes in chronic myeloproliferative disorders generated by Lafage-Pochitaloff M et al. The t(6;8)(q27;p11) translocation in a disruption of the platelet-derived growth factor receptor beta gene. stem cell myeloproliferative disorder fuses a novel gene, Haematologica 2007; 92: 163–169. FOP, to fibroblast growth factor receptor 1. Blood 1999; 93: 25 Rosati R, La Starza R, Luciano L, Gorello P, Matteucci C, Pierini V 1381–1389. et al. TPM3/PDGFRB fusion transcript and its reciprocal in chronic 43 Demiroglu A, Steer EJ, Heath C, Taylor K, Bentley M, Allen SL et al. eosinophilic leukemia. Leukemia 2006; 20: 1623–1624. The t(8;22) in chronic myeloid leukemia fuses BCR to FGFR1: 26 Wilkinson K, Velloso ER, Lopes LF, Lee C, Aster JC, Shipp MA et al. transforming activity and specific inhibition of FGFR1 fusion Cloning of the t(1;5)(q23;q33) in a myeloproliferative disorder proteins. Blood 2001; 98: 3778–3783.

Leukemia Five years since the discovery of FIP1L1–PDGFRA J Gotlib and J Cools 2009 44 Belloni E, Trubia M, Gasparini P, Micucci C, Tapinassi C, 65 Cools J, Stover EH, Boulton CL, Gotlib J, Legare RD, Amaral SM Confalonieri S et al. 8p11 myeloproliferative syndrome with a et al. PKC412 overcomes resistance to imatinib in a murine model novel t(7;8) translocation leading to fusion of the FGFR1 and TIF1 of FIP1L1-PDGFRalpha-induced myeloproliferative disease. genes. Genes Chromosomes Cancer 2005; 42: 320–325. Cancer Cell 2003; 3: 459–469. 45 Walz C, Chase A, Schoch C, Weisser A, Schlegel F, Hochhaus A 66 Cools J, Quentmeier H, Huntly BJ, Marynen P, Griffin JD, Drexler et al. The t(8;17)(p11;q25) in the 8p11 myeloproliferative HG et al. The EOL-1 cell line as an in vitro model for the study of syndrome fuses MYO18A to FGFR1. Leukemia 2005; 19: FIP1L1-PDGFRA-positive chronic eosinophilic leukemia. Blood 1005–1009. 2004; 103: 2802–2805. 46 Guasch G, Popovici C, Mugneret F, Chaffanet M, Pontarotti P, 67 Yamada Y, Rothenberg ME, Lee AW, Akei HS, Brandt EB, Williams Birnbaum D et al. Endogenous retroviral sequence is fused to DA et al. The FIP1L1-PDGFRA fusion gene cooperates with IL-5 to FGFR1 kinase in the 8p12 stem-cell myeloproliferative disorder induce murine hypereosinophilic syndrome (HES)/chronic eosino- with t(8;19)(p12;q13.3). Blood 2003; 101: 286–288. philic leukemia (CEL)-like disease. Blood 2006; 107: 4071–4079. 47 Grand EK, Grand FH, Chase AJ, Ross FM, Corcoran MM, Oscier 68 Burgstaller S, Kreil S, Waghorn K, Metzgeroth G, Preudhomme C, DG et al. Identification of a novel gene, FGFR1OP2, fused to Zoi K et al. The severity of FIP1L1-PDGFRA-positive chronic FGFR1 in 8p11 myeloproliferative syndrome. Genes Chromo- eosinophilic leukaemia is associated with polymorphic variation at somes Cancer 2004; 40: 78–83. the IL5RA locus. Leukemia 2007; 21: 2428–2432. 48 Chen J, DeAngelo DJ, Kutok JL, Williams IR, Lee BH, Wadleigh M 69 Pardanani A, Ketterling RP, Brockman SR, Flynn HC, Paternoster et al. PKC412 inhibits the zinc finger 198-fibroblast growth factor SF, Shearer BM et al. CHIC2 deletion, a surrogate for FIP1L1- receptor 1 fusion tyrosine kinase and is active in treatment of stem PDGFRA fusion, occurs in systemic mastocytosis associated with cell myeloproliferative disorder. Proc Natl Acad Sci USA 2004; eosinophilia and predicts response to imatinib mesylate therapy. 101: 14479–14484. Blood 2003; 102: 3093–3096. 49 Levine RL, Gilliland DG. Myeloproliferative disorders. Blood 70 Robyn J, Lemery S, McCoy JP, Kubofcik J, Kim YJ, Pack S et al. 2008; 112: 2190–2198. Multilineage involvement of the fusion gene in patients with 50 Reiter A, Walz C, Watmore A, Schoch C, Blau I, Schlegelberger B FIP1L1/PDGFRA-positive hypereosinophilic syndrome. Br J et al. The t(8;9)(p22;p24) is a recurrent abnormality in chronic and Haematol 2006; 132: 286–292. acute leukemia that fuses PCM1 to JAK2. Cancer Res 2005; 65: 71 Maric I, Robyn J, Metcalfe DD, Fay MP, Carter M, Wilson T et al. 2662–2667. KIT D816V-associated systemic mastocytosis with eosinophilia 51 Murati A, Gelsi-Boyer V, Ade´laı¨de J, Perot C, Talmant P, Giraudier and FIP1L1/PDGFRA-associated chronic eosinophilic leukemia are S et al. PCM1-JAK2 fusion in myeloproliferative disorders and distinct entities. J Allergy Clin Immunol 2007; 120: 680–687. acute erythroid leukemia with t(8;9) translocation. Leukemia 2005; 72 Metzgeroth G, Walz C, Score J, Siebert R, Schnittger S, Haferlach 19: 1692–1696. C et al. Recurrent finding of the FIP1L1-PDGFRA fusion gene in 52 Bousquet M, Quelen C, De Mas V, Duchayne E, Roquefeuil B, eosinophilia-associated acute myeloid leukemia and lympho- Delsol G et al. The t(8;9)(p22;p24) translocation in atypical blastic T-cell lymphoma. Leukemia 2007; 21: 1183–1188. chronic myeloid leukaemia yields a new PCM1-JAK2 fusion gene. 73 Fauci AS, Harley JB, Roberts WC, Ferrans VJ, Gralnick HR, Oncogene 2005; 24: 7248–7252. Bjornson BH. NIH Conference. The idiopathic hypereosinophilic 53 Ade´laı¨de J, Pe´rot C, Gelsi-Boyer V, Pautas C, Murati A, Copie- syndrome. Clinical, pathophysiologic, and therapeutic considera- Bergman C et al. A t(8;9) translocation with PCM1-JAK2 fusion in a tions. Ann Intern Med 1982; 97: 78–92. patient with T-cell lymphoma. Leukemia 2006; 20: 536–537. 74 Spry CJ, Davies J, Tai PC, Olsen EG, Oakley CM, Goodwin JF. 54 Huang KP, Chase AJ, Cross NC, Reiter A, Li TY, Wang TF et al. Clinical features of fifteen patients with the hypereosinophilic Evolutional change of karyotype with t(8;9)(p22;p24) and HLA-DR syndrome. Q J Med 1983; 52: 1–22. immunophenotype in relapsed acute myeloid leukemia. Int J 75 Lefebvre C, Bletry O, Degoulet P, Guillevin L, Bentana-Pessayre Hematol 2008; 88: 197–201. M, Le Thi Huiong D et al. Prognostic factors of hypereosinophilic 55 Golub TR, Goga A, Barker GF, Afar DE, McLaughlin J, Bohlander syndrome. Study of 40 cases. Ann Med Interne (Paris) 1989; 140: SK et al. Oligomerization of the ABL tyrosine kinase by the 253–257. Ets protein TEL in human leukemia. Mol Cell Biol 1996; 16: 76 Gotlib J, Cools J, Malone III JM, Schrier SL, Gilliland DG, Coutre 4107–4116. SE. The FIP1L1-PDGFRa fusion tyrosine kinase in hypereosino- 56 Xiao S, McCarthy JG, Aster JC, Fletcher JA. ZNF198-FGFR1 philic syndrome and chronic eosinophilic leukemia: implications transforming activity depends on a novel proline-rich ZNF198 for diagnosis, classification, and management. Blood 2004; 103: oligomerization domain. Blood 2000; 96: 699–704. 2879–2891. 57 McWhirter JR, Galasso DL, Wang JY. A coiled-coil oligomerization 77 Schaller JL, Burkland GA. Case report: rapid and complete control domain of Bcr is essential for the transforming function of Bcr-Abl of idiopathic hypereosinophilia with imatinib mesylate. oncoproteins. Mol Cell Biol 1993; 13: 7587–7595. MedGenMed 2001; 3:9. 58 Stover EH, Chen J, Folens C, Lee BH, Mentens N, Marynen P et al. 78 Gleich GJ, Leiferman KM, Pardanani A, Tefferi A, Butterfield JH. Activation of FIP1L1-PDGFRalpha requires disruption of the Treatment of hypereosinophilic syndrome with imatinib mesilate. juxtamembrane domain of PDGFRalpha and is FIP1L1-indepen- Lancet 2002; 359: 1577–1578. SPOTLIGHT dent. Proc Natl Acad Sci USA 2006; 103: 8078–8083. 79 Ault P, Cortes J, Koller C, Kaled ES, Kantarjian H. Response of 59 Chan PM, Ilangumaran S, La RJ, Chakrabartty A, Rottapel R. idiopathic hypereosinophilic syndrome to treatment with imatinib Autoinhibition of the kit by the cytosolic mesylate. Leuk Res 2002; 26: 881–884. juxtamembrane region. Mol Cell Biol 2003; 23: 3067–3078. 80 Gotlib J, Malone JM, DeAngelo DJ, Stone RM, Gilliland DG, Clark 60 Griffith J, Black J, Faerman C, Swenson L, Wynn M, Lu F et al. The J et al. Imatinib mesylate (GLEEVEC) induced rapid and complete structural basis for autoinhibition of FLT3 by the juxtamembrane hematologic remissions in patients with idiopathic hypereosino- domain. Mol Cell 2004; 13: 169–178. philic syndrome (HES) without evidence of BCR-ABL or activating 61 Gilliland DG, Griffin JD. The roles of FLT3 in hematopoiesis and mutations in c-KIT and platelet derived growth factor receptor- leukemia. Blood 2002; 100: 1532–1542. beta. Blood 2002; 100: 798a [Abstract 3152]. 62 Heinrich MC, Corless CL, Demetri GD, Blanke CD, von MM, 81 Klion AD, Robyn J, Akin C, Noel P, Brown M, Law M et al. Joensuu H et al. Kinase mutations and imatinib response in Molecular remission and reversal of myelofibrosis in response to patients with metastatic gastrointestinal stromal tumor. J Clin imatinib mesylate treatment in patients with the myeloproliferative Oncol 2003; 21: 4342–4349. variant of hypereosinophilic syndrome. Blood 2004; 103: 63 Lahortiga I, Akin C, Cools J, Wilson TM, Mentens N, Arthur DC 473–478. et al. Activity of imatinib in systemic mastocytosis with chronic 82 Klion AD, Robyn J, Maric I, Fu W, Schmid L, Lemery S et al. basophilic leukemia and a PRKG2-PDGFRB fusion. Haematologica Relapse following discontinuation of imatinib mesylate therapy for 2008; 93: 49–55. FIP1L1/PDGFRA-positive chronic eosinophilic leukemia: implica- 64 Buitenhuis M, Verhagen LP, Cools J, Coffer PJ. Molecular tions for optimal dosing. Blood 2007; 110: 3552–3556. mechanisms underlying FIP1L1-PDGFRA-mediated myeloproli- 83 Helbig G, Stella-Ho"owiecka B, Majewski M, Ca"becka M, feration. Cancer Res 2007; 67: 3759–3766. Gajkowska J, Klimkiewicz R et al. A single weekly dose of

Leukemia Five years since the discovery of FIP1L1–PDGFRA J Gotlib and J Cools 2010 imatinib is sufficient to induce and maintain remission of chronic disease with prominent eosinophilia. Leukemia 2005; 19: eosinophilic leukaemia in FIP1L1-PDGFRA-expressing patients. 286–287. Br J Haematol 2008; 141: 200–204. 88 Bradeen HA, Eide CA, O’Hare T, Johnson KJ, Willis SG, Lee FY 84 Pardanani A, Reeder T, Porrata LF, Li CY, Tazelaar HD, Baxter EJ et al. Comparison of imatinib mesylate, dasatinib (BMS-354825), et al. Imatinib therapy for hypereosinophilic syndrome and other and nilotinib (AMN107) in an N-ethyl-N-nitrosourea (ENU)-based eosinophilic disorders. Blood 2003; 101: 3391–3397. mutagenesis screen: high efficacy of drug combinations. Blood 85 Pitini V, Arrigo C, Azzarello D, La Gattuta G, Amata C, Righi M 2006; 108: 2332–2338. et al. Serum concentration of cardiac troponin T in patients with 89 Lierman E, Folens C, Stover EH, Mentens N, Van MH, Scheers W hypereosinophilic syndrome treated with imatinib is predictive of et al. Sorafenib is a potent inhibitor of FIP1L1-PDGFRalpha and the adverse outcomes. Blood 2003; 102: 3456–3457. imatinib-resistant FIP1L1-PDGFRalpha T674I mutant. Blood 2006; 86 Ohnishi H, Kandabashi K, Maeda Y, Kawamura M, Watanabe T. 108: 1374–1376. Chronic eosinophilic leukaemia with FIP1L1-PDGFRA fusion and 90 Stover EH, Chen J, Lee BH, Cools J, McDowell E, Adelsperger J T6741 mutation that evolved from Langerhans cell histiocytosis et al. The small molecule tyrosine kinase inhibitor AMN107 with eosinophilia after chemotherapy. Br J Haematol 2006; 134: inhibits TEL-PDGFRbeta and FIP1L1-PDGFRalpha in vitro and in 547–549. vivo. Blood 2005; 106: 3206–3213. 87 von Bubnoff N, Sandherr M, Schlimok G, Andreesen R, Peschel C, 91 von Bubnoff N, Gorantla SP, Thone S, Peschel C, Duyster J. The Duyster J. Myeloid blast crisis evolving during imatinib treatment FIP1L1-PDGFRA T674I mutation can be inhibited by the tyrosine of an FIP1L1-PDGFR alpha-positive chronic myeloproliferative kinase inhibitor AMN107 (nilotinib). Blood 2006; 107: 4970–4971. SPOTLIGHT

Leukemia