Durable responses to imatinib in patients with PDGFRB fusion gene–positive and BCR-ABL–negative chronic myeloproliferative disorders

Marianna David, Nicholas C. P. Cross, Sonja Burgstaller, Andrew Chase, Claire Curtis, Raymond Dang, Martine Gardembas, John M. Goldman, Francis Grand, George Hughes, Francoise Huguet, Louise Lavender, Grant A. McArthur, Francois X. Mahon, Giorgio Massimini, Junia Melo, Philippe Rousselot, Robin J. Russell-Jones, John F. Seymour, Graeme Smith, Alastair Stark, Katherine Waghorn, Zariana Nikolova and Jane F. Apperley


Fusion genes derived from the platelet-derived growth factor receptor beta (PDGFRB) or alpha (PDGFRA) play an important role in the pathogenesis of BCR-ABL–negative chronic myeloproliferative disorders (CMPDs). These fusion genes encode constitutively activated receptor tyrosine kinases that can be inhibited by imatinib. Twelve patients with BCR-ABL–negative CMPDs and reciprocal translocations involving PDGFRB received imatinib for a median of 47 months (range, 0.1-60 months). Eleven had prompt responses with normalization of peripheral-blood cell counts and disappearance of eosinophilia; 10 had complete resolution of cytogenetic abnormalities and decrease or disappearance of fusion transcripts as measured by reverse transcriptase–polymerase chain reaction (RT-PCR). Updates were sought from 8 further patients previously described in the literature; prompt responses were described in 7 and persist in 6. Our data show that durable hematologic and cytogenetic responses are achieved with imatinib in patients with PDGFRB fusion–positive, BCR-ABL–negative CMPDs.


Despite the passage of more than 40 years since the discovery of the molecular hallmark of chronic myeloid leukemia (CML), t(9,22)(q34, q11), the reciprocal Philadelphia translocation, our knowledge of the causality of Philadelphia-negative chronic myeloproliferative disorders (CMPDs) remains inadequate.

These malignancies are uncommon and constitute a heterogeneous group of disorders, where the molecular pathogenesis remains unclear. Recently, fusion genes involving the platelet-derived growth factor receptor beta (PDGFRB) or platelet-derived growth factor receptor alpha (PDGFRA) genes have been associated with a subgroup of these disorders. The tyrosine kinase inhibitor imatinib (Glivec, Gleevec, STI571) interacts with the ATP-binding site of protein tyrosine kinases and has specific inhibitory activity for the ABL, ARG, PDGFRA, PDGFRB, and c-KIT kinases.1-2 Our previous publication described 4 patients with PDGFRB fusion genes (ETV6-PDGFRB) who responded rapidly to imatinib.3 In the current report, we extend our original observations on these 4 subjects and report an additional 8 patients treated and monitored by us and demonstrate that such responses are highly predictable and durable. In evaluable patients, some were able to attain clearance of disease as assessed by sensitive molecular techniques.

Patients, materials, and methods


We report the results of long-term imatinib treatment in 12 patients (6 from the United Kingdom, 4 from France, and 1 each from Austria and Australia) of median age 52 years (range, 19-80 years) with CMPD characterized by leukocytosis, peripheral-blood eosinophilia, and PDGFRB gene rearrangements. Five of the patients were treated within a Novartis-sponsored study of imatinib for patients with life-threatening diseases known to be associated with one or more imatinib-sensitive tyrosine kinases, approved by the local ethics review processes. The remaining patients were treated off-study on compassionate grounds. All patients consented to the use of their data, in accordance with the Declaration of Helsinki. All patients were negative for the BCR-ABL fusion as determined by reverse transcriptase–polymerase chain reaction (RT-PCR) analysis. Patient characteristics are summarized in Table 1 Five of these patients (nos. 1, 2, 3, 9, and 11) were reported previously with considerably shorter follow-up.3,4 We also update the outcome for an additional 8 patients previously published as case reports.5-12

Table 1

Characteristics of the patients included in the study


Cytogenetic analysis and a 2-color fluorescence in situ hybridization assay (FISH) to detect PDGFRB rearrangements.

Cytogenetic analysis of bone marrow cells was performed by conventional G-banding. We previously established a 2-color fluorescence in situ hybridization assay for PDGFRB rearrangements using 2 flanking cosmid probes for chromosomes 5, 9-4, and 4-1. As PDGFRB is the only gene that lies in the 53-kb interval between the 2 cosmids, separated interphase signals indicate disruption of this gene. For the split-apart FISH test, the 2 cosmids are differentially labeled and cohybridized to test interphase cells. Scoring of the results was performed on the basis of separation of the red and green signals. In an unaffected individual, cells carry 2 fused pairs of signals; in PDGFRB-rearranged cells, one fused pair and one separated pair are seen.13

RT-PCR for the ETV6-PDGFRB fusion transcript.

RNA was reverse transcribed with random hexamer primers and tested for the presence of ETV6-PDGFRB fusion sequence by single-step reverse transcriptase–polymerase chain reaction (RT-PCR) (limit of detection, 10−2) and nested RT-PCR (maximum limit of detection, 10−5). Single-step PCR was performed for 30 cycles at an annealing temperature of 60°C with the primers ETV6-3F (5′-CTGCTGACCAAAGAGGACTT-3′) and PD-C (5′-TGGCTTCTTCTGCCAAAGCA-3′). When the single-step RT-PCR was negative, samples were subjected to nested PCR. The products of each reaction were amplified with primers ETV6-J (5′-TTCACCATTCTTCCACCCTGGA-3′) and PD-J (5′-GGAGATGATGGTGAGCACCA-3′) for 30 cycles at an annealing temperature of 62°C. To confirm sample quality, we measured the number of ABL transcripts by real-time PCR.14 Samples in which the number of ABL transcripts was less than 104 in the same volume of cDNA as was used for minimal residual disease (MRD) analysis were considered failures.


Purified PCR products were sequenced using 10 ng of one primer and the BigDye Terminator kit (Applied Biosystems, Foster City, CA), cleaned up on Dye Ex columns (Quigen NV, Venlo, The Netherlands), dried, resuspended in dye, and loaded onto an ABI 377 sequencer (Applied Biosystems).

Results and discussion

The platelet-derived growth factors (PDGFs) represent a family of mitogens that includes 5 dimeric forms: PDGF-AA, -AB, -BB, -CC, and -DD. PDGF dimers activate 2 specific type III–receptor tyrosine kinases, the PDGFRA, which binds the A, B, and C chains, and the PDGFRB, which binds the B and D chains.15 Signaling through PDGFRB plays an important role in mitogenesis, cytoskeletal rearrangements, and chemotaxis.16,17

Disruption of PDGFRB, normally located on chromosome 5q33, was first described as the consequence of the t(5;12) in which the 5′ end of ETV6 (earlier known as TEL) is juxtaposed to the 3′ end of PDGFRB.18 Subsequently, several more translocations have been characterized that fuse other genes to PDGFRB, including HIP1, CCDC6 (H4), TRIP11 (CEV14), RAB5EP, NIN, PDE4DIP (myomegalin), TP53BP1, and KIAA1509.4,5,6,7-8,19 All the resulting fusion proteins are tyrosine kinases with constitutive enzymatic activity. CMPDs associated with t(5;12) abnormalities are infrequent, and their clinical features have recently been extensively reported.20 They appear to represent a unique and distinct clinical entity with Philadelphia-negative myeloproliferation, myelodysplasia, and eosinophilia. Most cases were male (as in our study). The 2-year survival of the 18 evaluable patients was only 55%. An additional group of patients with translocations involving chromosome 5q31-q35 and partner genes other than ETV6 also had peripheral-blood eosinophilia.

In the present study, we describe the treatment with imatinib and long-term follow-up of 12 patients with PDGFRB rearrangements. Eight patients had the ETV6-PDGFRB translocation. All 12 had prompt responses to imatinib, and all except 1 (no. 12) achieved normalization of the peripheral-blood cell counts with the disappearance of eosinophilia. Ten achieved complete cytogenetic remission (CCyR) and decrease or disappearance of fusion transcripts. In 7 patients, follow-up now exceeds 4 years. The initial dose of imatinib was 800-mg daily (1 patient), 400-mg daily (8 cases), 300 mg (1 patient), and 200 mg (1 patient). The variation in dosing reflects physician uncertainty regarding the optimal doses in these disorders and was not due to the occurrence of side effects. The median overall survival is 65 months since diagnosis (range, 25-234 months). Ten of the 12 patients are alive with ongoing durable responses (Tables 1-2). Hematologic and cytogenetic responses have been sustained in these patients for a median of 49 months (range, 19-60 months) and 47 months (range, 16-59 months), respectively. One patient (no. 11) died with relapsed disease 8 months after initiation of the imatinib treatment and has been reported in more detail previously. He was heavily pretreated prior to commencing imatinib.4 One further patient (no. 12) with a longstanding CMPD and a t(1;3;5) abnormality received 4 days of imatinib at 800-mg daily after entering a second blast crisis. She became pancytopenic without morphologic or chromosomal evidence of disease for 6 months before a further blast crisis ensued. She was treated with 20 days of imatinib at 200 mg and again became pancytopenic without evidence of leukemia until her death from fungal infection 3 months later. With the exceptions of patients 11 and 12, the spectrum of side effects was similar to that seen in CML and included mild nausea, fatigue, fluid retention, and myalgias not requiring cessation of therapy.

Table 2

Dose, duration, and response to imatinib and the results of the recent follow-up

All 8 patients with ETV6-PDGFRB rearrangements were serially studied using a nested RT-PCR technique with a maximum level of detection of 10−5. RT-PCR testing replaced or supplemented regular monitoring by karyotyping when CCyR was achieved. Negative results were recorded for 6 of the 8 patients at some point during their course, and 4 of the 8 were negative for ETV6-PDGFRB transcripts at the last examination: 5 are alive and well more than 4 years since initiation of imatinib. Residual disease was present in the remaining 4 patients at low level as indicated by single-step RT-PCR negativity and nested RT-PCR positivity (Table 3). This may be consistent with long-term control of their disease as fluctuating or persistent low-level positivity by sensitive RT-PCR methods for residual bcr-abl transcripts has also been described in long-term survivors of allogeneic stem cell transplantation without any evidence of disease progression.21

Eight additional patients, 6 male and 2 female, with CMPD characterized by peripheral eosinophilia and translocations involving PDGFRB have been published as case reports by other groups. All had partner genes other than ETV6. Most had received multiple prior cytotoxic therapy without achieving remission. Seven of the 8 responded promptly to treatment with 200 to 400 mg imatinib and entered complete cytogenetic and/or hematologic remission.5-12 More recent follow-up has been provided by the original authors and is shown in Table 4. This gives a total of 20 patients with Philadelphia-negative CMPD associated with translocations involving PDGFRB who have been treated with imatinib. Nineteen responded promptly, and in most cases durably, to imatinib. Screening for fusion genes that affect the PDGFRB genes is an important part of the diagnostic investigations for these poorly understood disorders, and management of fusion gene–positive patients with appropriate tyrosine kinase inhibitors should become first-line treatment in the future.

Table 3

RT-PCR results from molecular monitoring of patients with ETV6-PDGFRB rearrangements by time since initiation of treatment

Table 4

Updated results on previously published patients


We would like to thank the following individuals for providing further information on the patients described in references 5 to 12: Dr R. A. Aguiar, Dr J. Barrett, Dr S. Castaigne, Dr J. L. Garcia, Dr P. Giraldo, Dr J. M. Hernandez, Dr V. Pitini, Dr J. San Miguel, Dr M. A. Shipp, Dr R. Stone, Dr J. L. Vizmanos, and Dr M. Wadleigh.


  • Correspondence: J. F. Apperley, Department of Haematology, Imperial College at Hammersmith Hospital, Ducane Rd, London W12 0NN, United Kingdom; e-mail: j.apperley{at}
  • The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.

  • Conflict-of-interest disclosure: Two of the authors (G.M., Z.N.) are employed by a company (Novartis) whose product (imatinib) was studied in the present work.

  • Contribution: M.D. coordinated the collection and analysis of data, and wrote substantial parts of the manuscript. J.F.A., S.B., R.D., G.H., M.G., J.M.G., F.H., G.A.M., P.R., J.S., G.S., and A.S. cared for individual patients, instigated and monitored treatment, contributed data to the manuscript and are responsible for the integrity of their data. A.C., C.C., F.G., L.L., F.X.M., and K.W. performed the cytogenetic, FISH, and RT-PCR analyses for the patients on multiple occasions. N.C.P.C. designed the laboratory tests and supervised their execution. J.F.A., J.M.G., F.X.M., and R.J.R.-J. were responsible for the concept of treating this patient group with imatinib. G.M. and Z.N. were responsible for the design and conduct of the Novartis study of imatinib for patients with life-threatening diseases known to be associated with one or more imatinib sensitive tyrosine kinases (Novartis B2225) through which 5 patients were treated. They also provided critical appraisal and made valuable suggestions for the manuscript. J.V.M., A.C., F.H., G.M., and J.S. provided critical review and valuable suggestions for the manuscript. J.F.A. instigated the study, supervised M.D., and contributed substantially to the final manuscript.

  • Submitted January 2, 2006.
  • Accepted June 11, 2006.


View Abstract