Letters to the Editor 2439 Figure 2. Immune recovery and tracking of donor NK cells after infusion of CD3/19–depleted, IL-15-stimulated boosts. (a) Immune reconstitution after infusion of IL-15-stimulated boosts (mean values). A fast recovery of white blood cells and lymphocyte subsets (CD56 þ NK cells, CD3 þ /CD4 þ and CD3 þ /CD8 þ T cells) was observed in all patients (n ¼ 8), with comparable or even higher levels after 15–30 days compared with levels before . (b) Tracking of infused IL-15-stimulated NK cells by activation markers (mean values). IL-15 stimulation induced strong expression of CD69 on donor NK cells. Activated NK cells (CD56 þ /CD69 þ ) were detectable in all patients analyzed with CD69 (n ¼ 7) in the first week after infusion of ex vivo-stimulated boosts. (c) Tracking of infused IL-15-stimulated NK cells by HLA flow cytometry. Top: patient 3 received NK cells from a haploidentical donor after matched unrelated donor transplantation. Infused donor NK cells were detectable by additional staining with an anti-HLA-A2 mAb (autologous patient NK cells HLA-A2 þ , donor NK cells HLA-A2 À ). Bottom left: infused donor NK cells expanded in the patient, with a maximum at day 10. Afterwards, NK cells were rejected and disappeared completely after 30 days. Bottom right: autologous NK cells from the patient paralleled the course of the haploidentical NK cells, indicating a stimulating effect of the graft infusion also on the autologous cells. a.c., after chemotherapy; b.c., before chemotherapy. isolation of NK cells from additional unstimulated leukapheresis AUTHOR CONTRIBUTIONS products. In contrast, the aim of our approach was not to obtain a The design of this scientific work was done by MMP, MS and PL. MMP did the pure population of NK cells but to combine the donation of both experimental work; IM, RH and PL took care of the patients and provided stem cells and stimulated effector cells. As bone marrow selected samples. RH directed the transplantation program and RH, MS and insufficiency occurred in all patients after induction chemo- PL supervised the research. Data analysis was done by MMP and PL. The therapy, the stem cell component of our IL-15-stimulated grafts manuscript was written by MMP and PL and critically reviewed by all authors. was of great clinical importance: a fast hematopoietic recovery (including T cells) could be observed and enabled further MM Pfeiffer1, M Schumm1, Ingo Mu¨ller2, maintenance therapy. Thus, with our approach no additional R Handgretinger1 and P Lang1 leukapheresis was necessary and depletion of unwanted cells was 1Department of Pediatric Hematology/Oncology, University done in one step. IL-15 was more potent in short-time stimulation Children’s Hospital, Eberhard Karl’s University, compared with IL-2, and administration of those cell products was Tuebingen, Germany and well tolerated without any acute and late side effects, especially 2Department of Pediatric Hematology/Oncology, without induction of GvHD. 5 University Medical Center Hamburg-Eppendorf, Miller et al. used haploidentical CD3-depleted leukapheresis Hamburg, Germany products after overnight stimulation with IL-2 in a non-transplant E-mail: [email protected] setting, and demonstrated that NK cells persisted for around 28 days in most patients. In our patients we observed a comparable duration of donor NK cell expansion, which might be further REFERENCES increased by in vivo IL-2 administration. 1 Ruggeri L, Capanni M, Urbani E, Perruccio K, Shlomchik WD, Tosti A et al. Effec- In summary, the major goals of this trial were to demonstrate that tiveness of donor natural killer cell alloreactivity in mismatched hematopoietic infusions of IL-15-stimulated grafts are safe and feasible even in the transplants. Science 2002; 295: 2097–2100. haploidentical setting and lead to a fast hematopoietic recovery, 2 Leung W, Iyengar R, Turner V, Lang P, Bader P, Conn P et al. Determinants of with detectable and activated CD69 þ NK cells. Combinations with antileukemia effects of allogeneic NK cells. J Immunol 2004; 172: 644–650. different were possible in the vulnerable post- 3 Miller JS, Cooley S, Parham P, Farag SS, Verneris MR, McQueen KL et al. Missing KIR transplant period and contributed to prolonged survival. However, ligands are associated with less relapse and increased graft-versus-host disease no firm conclusions can be drawn about the GvL or graft versus (GVHD) following unrelated donor allogeneic HCT. Blood 2007; 109: 5058–5061. tumor effect of the cell infusions themselves to date. 4 Schumm M, Handgretinger R, Pfeiffer M, Feuchtinger T, Kuci S, Faul C et al. Determination of residual T- and B-cell content after immunomagnetic depletion: proposal for flow cytometric analysis and results from 103 separations. Cytotherapy 2006; 8: 465–472. CONFLICT OF INTEREST 5 Miller JS, Soignier Y, Panoskaltsis-Mortari A, McNearney SA, Yun GH, Fautsch SK et al. Successful adoptive transfer and in vivo expansion of human haploidentical The authors declare no conflict of interest. NK cells in patients with cancer. Blood 2005; 105: 3051–3057. 6 Rubnitz JE, Inaba H, Ribeiro RC, Pounds S, Rooney B, Bell T et al. NKAML: a pilot study to determine the safety and feasibility of haploidentical natural killer cell trans- ACKNOWLEDGEMENTS plantation in childhood acute myeloid . JClinOncol2010; 28: 955–959. We thank Anni Babarin-Dorner, Olga Bartuli, Ulrike Junker, Claudia Treuner and Gabi 7 Lang P, Pfeiffer M, Handgretinger R, Schumm M, Demirdelen B, Stanojevic S et al. Hochwelker for their excellent technical assistance and Barbara Lang for reviewing Clinical scale isolation of -depleted CD56 þ donor lymphocytes in children. this work. This work was supported by grants (SFB 685 and TranSaRNet) from the Bone Marrow Transplant 2002; 29: 497–502. Deutsche Forschungsgemeinschaft (DFG) and the Bundesministerium fu¨r Bildung 8 Koehl U, Sorensen J, Esser R, Zimmermann S, Gru¨ttner HP, Tonn T et al. IL-2 und Forschung (BMBF) Tuebingen to PL and from the Jung-Stiftung fu¨r Wissenschaft activated NK cell immunotherapy of three children after haploidentical stem cell und Forschung and the Jose Carreras Leukemia Foundation to MMP. transplantation. Blood Cells Mol Dis 2004; 33: 261–266.

Long-term follow-up of FIP1L1-PDGFRA-mutated patients with : survival and clinical outcome

Leukemia (2012) 26, 2439–2441; doi:10.1038/leu.2012.162 patients with eosinophilia.2–4 As per the World Health Organization (WHO), patients harboring this molecular Identification of the karyotypically occult fusion oncogene FIP1L1- abnormality are classified as ‘Myeloid and lymphoid neoplasms PDGFRA by Cools et al.1 in 2003 provided the molecular basis associated with eosinophilia and abnormalities of PDGFRA, for the dramatic response to therapy in a subset of PDGFRB or FGFR1’.

Accepted article preview online 18 June 2012; advance online publication, 10 July 2012

& 2012 Macmillan Publishers Limited Leukemia (2012) 2414 – 2444 Letters to the Editor 2440 Ever since the original report was published, several studies One patient remains alive 213 months after diagnosis. He was have described the clinico-pathological features of this disease treated with high-dose and underwent emergent entity and confirmed the achievement of rapid and complete cardiac surgery (biventricular thrombectomy/endocardial resection clinical and hematological responses with imatinib at doses with tricuspid valve replacement). Subsequent steroid-sparing treat- ranging from 100 to 400 mg/day.5–11 The majority of patients also ment strategies included interferon-a (IFN-a)and/orhydroxyurea achieve molecular remission with imatinib therapy, with clinical (HU); to our knowledge, this patient has not received imatinib responses being durably maintained on imatinib doses as low treatment. The other two patients were treated with corticosteroids/ as 50–200 mg/week.9,12 Discontinuation of imatinib therapy is IFN-a and corticosteroids/IFN-a/HU/cyclosporine-A/2-chlorodeoxyade- associated with molecular relapse within weeks to months, which nosine, respectively. may be reversible on treatment resumption;8,10,11,13 secondary Of the 19 contemporaneous patients, 18 were treated with resistance to imatinib is relatively rare and occurs in the context of imatinib; the starting dose was 100 mg/day in 14 patients, and accelerated or blast phase disease and acquisition of PDGFRA 400 mg/day, 200 mg/day, 100 mg thrice weekly and 100 mg twice T674I and/or D842V .14 weekly for one patient each. All imatinib-treated patients, except The objective of the current paper is to describe the long-term one, achieved complete hematological remission (CHR). The natural history of eosinophilia patients with the FIP1L1-PDGFRA sole patient without CHR was intolerant to imatinib, leading to seen at our institution. treatment discontinuation; currently he is asymptomatic and The current study was approved by the Mayo Clinic institutional remains off treatment with an AEC of 1–1.5 Â 109/l. Another review board. Consecutive patients with mutated FIP1L1-PDGFRA patient discontinued imatinib after 1 year of his own volition; by seen at our institution were identified by querying the institutional report, he has not had recurrent eosinophilia after 5 years of cytogenetics database. A review of clinical, laboratory and follow-up, although confirmation is lacking as he has not returned histological material confirmed the diagnosis of ‘Myeloid and to our institution. One patient was lost to follow-up. All 10 patients lymphoid neoplasms associated with eosinophilia and abnormal- who were studied post imatinib treatment achieved complete ities of PDGFRA, PDGFRB or FGFR1’. Clinical and laboratory molecular remission (CMR) by FISH. The most recent recorded parameters were obtained at the time of first referral. FIP1L1- imatinib maintenance dose in the remaining patients is 100 mg/ PDGFRA was identified by fluorescence in situ hybridization (FISH). day (n ¼ 4), 100 mg/week (n ¼ 3), 100 mg thrice weekly (n ¼ 3), One of two FISH strategies were used depending upon the time 100 mg twice weekly (n ¼ 1), 100 mg every other day (n ¼ 1) frame during which the patient was seen: either the early- and 50 mg every other day (n ¼ 1). Two deaths have occurred to generation consecutive 1- and 2-color,7 or the more recent 3-color date. The imatinib-naive patient was treated with corticosteroids; FISH assay(s) were employed.15 he died of complications related to pancreatitis before imatinib Patients’ information was updated in March 2012 through a treatment could be initiated. The other death was due to review of patient histories and correspondence, social security transformation to (AML); in this patient, death index or a telephone call to the patient or their the baseline bone marrow (8 months prior) showed many local physician. The primary analysis was to examine the overall clusters of very immature precursors that were survival (OS); patients were stratified on the basis of imatinib use. CD34 þ by immunohistochemistry, thereby suggesting acceler- OS curves were prepared by the Kaplan–Meier method and ated disease at presentation. PDGFRA T674I or D842V mutation compared by using the log-rank test. P-values o0.05 were status was not available for this patient either at presentation or at considered significant. The Statview (SAS Institute, Cary, NC, relapse. USA) statistical package was used for all calculations. Clinical data have been previously reported for 11 of the current 22 patients.5,7,12 Imatinib response criteria have been previously 7 published. Table 1. Clinical and laboratory characteristics at presentation and A total of 22 patients with mutated FIP1L1-PDGFRA were clinical outcome studied; their relevant clinical and laboratory characteristics are presented in Table 1. In all, 19 patients either presented with or Total number of patients 22 were seen in follow-up for a primary eosinophilic disorder at our Male (%) 22 (100%) Age at diagnosis (years): median (range) 45.5 (26–72) institution after the discovery of FIP1L1-PDGFRA (circa 2002); of Hemoglobin (g/dl): median (range) 12.8 these, 18 patients were treated with imatinib. Three other patients (6.6–15.1) were retrospectively identified as harboring FIP1L1-PDGFRA upon White blood cell count ( Â 109/l): median (range) 14 (2.8–126.3) analysis of archived samples (collected from 1994 to 1997), as part Absolute eosinophil count ( Â 109/l): median (range) 5.3 (1–109.9) of the development/validation process for internal FISH assays; Platelet count ( Â 109/l): median (range) 144.5 none of these patients received imatinib treatment. (11–315) Not surprisingly, all patients were male and had eosinophilia at Organ involvement (%) presentation; the median absolute eosinophil count (AEC) was Constitutional (fever, night sweats, weight loss) 10 (45%) 5.3 Â 109/l (range 1–110). A significant proportion presented with Fatigue/bone pain 15 (68%) Skin (pruritus, rash) 6 (27%) fatigue, bone pain and/or other constitutional symptoms. The Cardiac (CHF, LV thrombus, characteristic ECHO findings) 7 (32%) proportion of patients with involvement of skin, lungs, heart, gut Pulmonary (cough, dyspnea, lung infiltrate, rhinosinusitis) 6 (27%) and nervous system was roughly similar. There was a wide Gastrointestinal (diarrhea, abdominal pain) 6 (27%) variation in the percentage of bone marrow nuclei harboring Neurologic (stroke, peripheral neuropathy) 5 (23%) FIP1L1-PDGFRA by FISH (8.5–95.5%); however, the confounding % of nuclei harboring FIP1L1-PDGFRA by FISH (n ¼ 21): 48 (8.5–95.5) effect of recent or concurrent treatment cannot be median (range) Abnormal karyotype (n ¼ 17) (%)a 1 (6%) discounted. Median follow-up for the imatinib-treated patients Number of patients treated with imatinib 18 (82%) (n ¼ 18) was 72.9 months (range 4.9–205.4); one patient started Follow-up (months): median (range) 72.5 imatinib treatment prior to the discovery of FIP1L1-PDGFRA.2 (4.3–213.1) Number of patients alive (%) 18 (82%) All three patients who were retrospectively identified had Transformation to AML 2 (9%) cardiac involvement; in addition, one patient had hepatic infiltration with /abnormal mast cells. Two of the three Abbreviations: AML, acute myeloid leukemia; CHF, congestive ; ECHO, echocardiographic; FISH, fluorescence in situ hybridization; LV, left patients have died; the cause of death is known for one case, a ventricular; n, number. Abnormal karyotype: þ 8(n ¼ 1). namely, leukemic transformation 24 months following diagnosis.

Leukemia (2012) 2414 – 2444 & 2012 Macmillan Publishers Limited Letters to the Editor 2441 1 and analyzed the FISH assays for FIP1L1-PDGFRA. CAH reviewed the bone marrow histology. All authors contributed towards writing the paper and 0.8 approved the submitted draft.

0.6 A Pardanani1, A D’Souza1, RA Knudson2, CA Hanson3, RP Ketterling2 and A Tefferi1 0.4 1Division of Hematology, Department of Medicine, Cum. Survival Mayo Clinic, Rochester, MN, USA; 0.2 2Division of Cytogenetics, Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN, USA and 3 0 Division of Hematopathology, Department of Laboratory Medicine 0 25 50 75 100 125 150 175 200 225 and Pathology, Mayo Clinic, Rochester, MN, USA E-mail: [email protected] Time Figure 1. Kaplan–Meier survival curves for patients treated with imatinib (blue curve, median survival not reached) versus other treatments (green curve, median survival 10.3 months). The color REFERENCES reproduction of this figure is available at the Leukemia journal 1 Cools J, DeAngelo DJ, Gotlib J, Stover EH, Legare RD, Cortes J et al. A tyrosine online. kinase created by fusion of the PDGFRA and FIP1L1 as a therapeutic target of imatinib in idiopathic hypereosinophilic syndrome. N Engl J Med 2003; 348: 1201–1214. 2 Schaller JL, Burkland GA. Case report: rapid and complete control of idiopathic Survival of imatinib-treated patients (n ¼ 18; median survival hypereosinophilia with imatinib mesylate. MedGenMed 2001; 3:9. not reached) is illustrated in Figure 1. The four patients 3 Gleich GJ, Leiferman KM, Pardanani A, Tefferi A, Butterfield JH. Treatment of who received non-imatinib therapy had a median survival of hypereosinophilic syndrome with imatinib mesilate. Lancet 2002; 359: 1577–1578. 4 Pardanani A, Reeder T, Porrata LF, Li CY, Tazelaar HD, Baxter EJ et al. Imatinib 10.3 months. therapy for hypereosinophilic syndrome and other eosinophilic disorders. Notable aspects of the current study are the long-term follow- Blood 2003; 101: 3391–3397. up after diagnosis and description of clinical outcome in a subset 5 Pardanani A, Ketterling RP, Brockman SR, Flynn HC, Paternoster SF, Shearer BM et al. of FIP1L1-PDGFRA-mutated patients who were imatinib-naive. CHIC2 , a surrogate for FIP1L1-PDGFRA fusion, occurs in systemic masto- Given the retrospective nature of the study, the imatinib-treated cytosis associated with eosinophilia and predicts response to imatinib mesylate and -untreated groups cannot be directly compared. Regardless, therapy. Blood 2003; 102: 3093–3096. the analysis confirms the excellent survival outcome of FIP1L1- 6 Klion AD, Noel P, Akin C, Law MA, Gilliland DG, Cools J et al. Elevated serum PDGFRA-mutated patients who receive imatinib therapy. The risk tryptase levels identify a subset of patients with a myeloproliferative variant of leukemic transformation in the current study was not trivial of idiopathic hypereosinophilic syndrome associated with tissue fibrosis, poor prognosis, and imatinib responsiveness. Blood 2003; 101: 4660–4666. (9%), although a more definitive estimate in this regard will 7 Pardanani A, Brockman SR, Paternoster SF, Flynn HC, Ketterling RP, Lasho TL et al. require study of more patients; the sole imatinib-treated patient FIP1L1-PDGFRA fusion: prevalence and clinicopathologic correlates in 89 who developed AML in the current cohort likely had evolving consecutive patients with moderate to severe eosinophilia. Blood 2004; 104: leukemia at presentation. Overall, imatinib treatment results 3038–3045. in excellent disease control, with remissions being maintained 8 Metzgeroth G, Walz C, Erben P, Popp H, Schmitt-Graeff A, Haferlach C et al. Safety at very low doses, confirming the significantly increased sensitivity and efficacy of imatinib in chronic eosinophilic leukaemia and hypereosinophilic of FIP1L1-PDGFRA to inhibition by imatinib, as compared syndrome: a phase-II study. Br J Haematol 2008; 143: 707–715. with BCR-ABL. These data also confirm the low likelihood of 9 Helbig G, Stella-Holowiecka B, Majewski M, Calbecka M, Gajkowska J, developing secondary resistance to imatinib in this setting, as Klimkiewicz R et al. A single weekly dose of imatinib is sufficient to induce and maintain remission of chronic eosinophilic leukaemia in FIP1L1-PDGFRA-expres- compared with chronic myeloid leukemia. For these reasons, sing patients. Br J Haematol 2008; 141: 200–204. it is unlikely to be clinically useful to monitor the depth of 10 Baccarani M, Cilloni D, Rondoni M, Ottaviani E, Messa F, Merante S et al. The molecular response with imatinib therapy in FIP1L1-PDGFRA- efficacy of imatinib mesylate in patients with FIP1L1-PDGFRalpha-positive mutated patients, with rare exception. A significant proportion hypereosinophilic syndrome. Results of a multicenter prospective study. of patients in this cohort had cardiac involvement that resulted in Haematologica 2007; 92: 1173–1179. significant morbidity and mortality in the imatinib-naive patients. 11 Jovanovic JV, Score J, Waghorn K, Cilloni D, Gottardi E, Metzgeroth G et al. Finally, no conclusion is possible regarding the lack of overt Low-dose imatinib mesylate leads to rapid induction of major molecular disease relapse in the single patient who discontinued imatinib responses and achievement of complete molecular remission in FIP1L1-PDGFRA therapy after achieving CMR. Our current view is that FIP1L1- positive chronic eosinophilic leukemia. Blood 2007; 109: 4635–4640. 12 Pardanani A, Ketterling RP, Li CY, Patnaik MM, Wolanskyj AP, Elliott MA et al. PDGFRA-mutated patients should continue imatinib therapy FIP1L1-PDGFRA in eosinophilic disorders: prevalence in routine clinical practice, indefinitely to maintain response. long-term experience with imatinib therapy, and a critical review of the literature. Leuk Res 2006; 30: 965–970. 13 Klion AD, Robyn J, Maric I, Fu W, Schmid L, Lemery S et al. Relapse following dis- continuation of imatinib mesylate therapy for FIP1L1/PDGFRA-positive chronic eosi- CONFLICT OF INTEREST nophilic leukemia: implications for optimal dosing. Blood 2007; 110: 3552–3556. 14 Metzgeroth G, Erben P, Martin H, Mousset S, Teichmann M, Walz C et al. Limited The authors declare no conflict of interest. clinical activity of nilotinib and sorafenib in FIP1L1-PDGFRA positive chronic eosino- philic leukemia with imatinib-resistant T674I mutation. Leukemia 2011; 26: 162–164. AUTHOR CONTRIBUTIONS 15 Fink SR, Belongie KJ, Paternoster SF, Smoley SA, Pardanani AD, Tefferi A et al. Validation of a new three-color fluorescence in situ hybridization (FISH) method AP and AT contributed patients. AP, AD and AT reviewed, summarized and to detect CHIC2 deletion, FIP1L1/PDGFRA fusion and PDGFRA translocations. analyzed the clinical and laboratory data. RAK and RPK designed, performed Leuk Res 2009; 33: 843–846.

& 2012 Macmillan Publishers Limited Leukemia (2012) 2414 – 2444