“Clinical and Molecular Heterogeneity of Eosinophilia-Associated Myeloproliferative Neoplasms with a Translocation t(5;12)”.

Abstract 4983

Chromosomal or molecular aberrations in eosinophilia-associated myeloproliferative neoplasms (Eos-MPN) often involve receptor (e.g. PDGFRA at 4q12, PDGFRB at 5q31, FGFR1 at 8p11) or intracellular (e.g. ABL1 at 9q34, JAK2 at 9p24) tyrosine kinases (TK). Fusion genes with rearrangement of PDGFRA or PDGFRB are known to be associated with rapid and durable complete hematological and clinical remission on treatment with imatinib; many of these cases also achieve complete molecular remission (undetectable fusion gene transcripts by nested reverse transcriptase polymerase chain reaction, RT-PCR). The most common fusion genes in Eos-MPN are FIP1L1-PDGFRA caused by a cytogenetically invisible interstitial deletion on 4q12, and ETV6-PDGFRB as consequence of a reciprocal translocation t(5;12)(q31-33;p13). Due to restricted availability of molecular diagnostic tools, a substantial proportion of Eos-MPN patients with t(5;12) are treated with imatinib based on karyotype but without confirmation of the underlying fusion gene by fluorescence-in-situ-hybridization (FISH) or RT-PCR. We report here on three patients (male, n=2; female, n=1) (median age 58 years, range 38-71) with Eos-MPN (median absolute eosinophil numbers in peripheral blood 8,100/μL, range 2,300-62,000) in chronic (n=2) or secondary acute myeloid leukemia (AML)/blast phase (n=1). Cytogenetic analysis revealed a t(5;12) (n=2) or a complex karyotype with involvement of chromosome arms 5q and 12p (n=1). Despite the fact that the clinical and hematological phenotype was resembling ETV6-PDGFRB positive disease (e.g. leukocytosis, eosinophilia, hypercellular marrow, splenomegaly), all three cases tested negative for ETV6-PDGFRB. A fourth patient with t(5;12)(q31;p13) but negative for ETV6-PDGFRB had identical clinical characteristics except basophilia of 7% without eosinophilia. Further molecular analyses revealed an ETV6-ACSL6 fusion gene in all four patients. Two patients received imatinib (400 mg/d) without knowledge of the molecular status and two patients were treated with sorafenib (400-800 mg/d) due to its multitargeted activity towards signal transduction molecules. No responses were observed to imatinib or sorafenib. Two male patients received an allogeneic stem cell transplantation (SCT) from a related or a matched unrelated donor, respectively. The first patient died on day +67 due to relapse of secondary AML/blast phase while the second patient died on day +64 due to transplant related complications. After failure to imatinib, the female patient with secondary AML/blast crisis only received supportive care because of comorbidity and died 7 months after diagnosis due to cytopenia-related complications. The remaining chronic phase male patient is alive 5 months after diagnosis. In addition to patients known from the literature (n=6), primary AML or rapid progression to secondary AML/blast phase has been observed in 7 of overall 10 patients with t(5;12) and an ETV6-ACSL6 fusion gene indicating a potentially aggressive clinical course. Consistent with the lack of involvement of a TK, the disease is primarily resistant to currently available TK-inhibitors and allogeneic SCT should be considered in eligible patients. In conclusion, treatment with TK-inhibitors in patients with myeloproliferative neoplasms and a t(5;12) should only be initiated if involvement of PDGFRB is confirmed by FISH or PCR analysis.