The FIP1L1-PDGFRA fusion gene results from a cytogenetically invisible interstitial chromosomal deletion on chromosome 4q12 and was recently identified as a recurrent molecular abnormality in patients with chronic eosinophilic leukemia (CEL) and systemic mastocytosis with eosinophilia (SME). The pathogenesis of FIP1L1-PDGFRA positive CEL/SME is similar to BCR-ABL positive chronic myeloid leukemia (CML) with constitutively increased tyrosine kinase activity of the fusion protein and excellent response to treatment with imatinib. The breakpoints within FIP1L1 are variable and a number of different exons are fused to a truncated PDGFRA exon 12. However, the numbers of sequenced fusion transcripts in single reports have been too small for a more detailed analysis of the anatomy and relative frequency of the different fusion transcripts. We therefore sought to collect data from FIP1L1-PDGFRA positive patients from several laboratories across Europe (France, Germany, Italy, UK) in a collaborative study within the workpackage "Minimal residual disease" of the European LeukemiaNet. A total number of 43 FIP1L1-PDGFRA positive cases were identified by RT-PCR. For yet unknown reasons a considerable number of cases were only found to be positive after nested RT-PCR despite adequate sample quality, high leukocyte counts and marked eosinophilia. Possible reasons might be a relatively low proportion of FIP1L1-PDGFRA positive cells, relatively low expression of the fusion transcript and/or relatively rapid fusion transcript degradation; although FIP1L1-PDGFRA was found to be expressed at a level comparable to the ABL control gene in RQ-PCR analysis of EOL-1 cell line. Sequence analysis revealed that all PDGFRA breakpoints fell exclusively within exon 12, thus retaining the entire kinase domain of PDGFRA in all cases. The truncated PDGFRA (p) exon 12 was fused to FIP1L1 (f) exons 9 to 13 (formerly described as 7a, 8, 8a, 9 and 10 -

Cools et al., NEJM. 2003;348: 1201–1214
): f9p12 (n=1; 2%), f10p12 (n = 10; 22%), f11p12 (n=15; 33%), f12p12 (n=7; 15%), f13p12 (n=10; 22%). An insertion of additional sequences of up to 107 bp was found in 24 patients (56%) leading to an open reading frame in all cases. These sequences were derived from introns of FIP1L1 in 14 cases, from FIP1L1 exon 13 in one case and from 4q33 in one case, the latter indicating a more complex rearrangement. Eight inserts of 2 – 6 bp could not be matched to known sequences because they were too short. An entirely identical fusion transcript resulting from an identical PDGFRA sequence fused to the same FIP1L1 exon without insert was found in 6 patients with a f12p12 fusion transcript and 4 patients with a f10p12 fusion transcript. We conclude that the combination of a great variability of breakpoints within FIP1L1 and PDGFRA plus the insertion of sequences which are variable in length and origin lead to unique FIP1L1-PDGFRA fusion sequences in the majority of patients. This carries important implications for strategies for molecular detection and development of RQ-PCR assays to determine response to imatinib or alternative tyrosine kinase inhibitors.

Topics:
eosinophilia, eosinophilic leukemia, chronic, fusion genes, heterogeneity, infantile severe myoclonic epilepsy, mastocytosis, systemic, platelet-derived growth factor alpha receptor, imatinib mesylate, polymerase chain reaction, reverse transcriptase polymerase chain reaction

Author notes

Corresponding author

2005, The American Society of Hematology
2004
Sign in via your Institution