Cooperating mutations of receptor tyrosine kinases and Ras genes in ...

Leukemia (2008) 22, 303–307 & 2008 Nature Publishing Group All rights reserved 0887-6924/08 $30.00 www.nature.com/leu

ORIGINAL ARTICLE Cooperating mutations of receptor tyrosine kinases and Ras genes in childhood core-binding factor acute myeloid leukemia and a comparative analysis on paired diagnosis and relapse samples L-Y Shih1,2, D-C Liang3, C-F Huang1, Y-T Chang3, C-L Lai2, T-H Lin2, C-P Yang4, I-J Hung4, H-C Liu3, T-H Jaing4, L-Y Wang3 and T-C Yeh3 1 Division of Hematology-Oncology, Chang Gung Memorial Hospital, Taipei, Taiwan; 2School of Medicine, Chang Gung University, Taoyuan, Taiwan; 3Division of Pediatric Hematology-Oncology, Mackay Memorial Hospital, Taipei, Taiwan and 4Chang Gung Children’s Hospital, Linkou, Taiwan

c-KIT mutations have been described in core-binding factor (CBF) acute myeloid leukemia (AML) at diagnosis. The role of c-KIT mutations in the relapse of CBF–AML is not clear. The role of CSF1R mutation in the pathogenesis of AML remains to be determined. We analyzed receptor tyrosine kinases (RTKs) and Ras mutations on 154 children with AML. Also, we examined the paired diagnosis and relapse samples in CBF–AML. CBF–AML accounted for 27% (41/154). c-KIT mutations were detected in 41.5% of CBF–AML at diagnosis (6 in exon 8, 10 in exon 17 and 1 in both exons 8 and 17) , FLT3–TKD 2.7%, N-Ras mutations 7.3% and K-Ras mutations 4.9%. FLT3–LM and CSF1R mutations were not found in CBF–AML. The mutations of RTKs and Ras were mutually exclusive except for one patient who had both c-KIT and N-Ras mutations. Eight of the 41 CBF–AML patients relapsed; four patients retained the identical c-KIT mutation patterns as those at diagnosis, the remaining four without c-KIT mutations at diagnosis did not acquire c-KIT mutations at relapse. Our study showed that 54% of childhood CBF–AML had RTKs and/or Ras mutations; c-KIT but not CSF1R mutations play a role in the leukemogenesis of childhood CBF–AML. Leukemia (2008) 22, 303–307; doi:10.1038/sj.leu.2404995; published online 25 October 2007 Keywords: c-KIT mutation; CSF1R mutation; core-binding factor; childhood acute myeloid leukemia; receptor tyrosine kinase; Ras mutation

to RTKs induces receptor dimerization and autophosphorylation. Activating mutations of RTKs result in ligand-independent signals for cell growth.3 Among the class III RTKs, FLT3 mutations have been intensively studied in childhood AML.4–8 The frequencies of c-KIT mutations in childhood CBF–AML varied considerably.9–12 The numbers of patients in most of the published series were relatively small, especially, in children with inv(16)/CBFb-MYH11 AML, when compared with those of adult patients.13–15 Since the frequencies of genetic mutations might differ between pediatric and adult AML, study on a larger number of pediatric CBF–AML is warranted to define the frequencies and patterns of c-KIT mutations. Moreover, the occurrence of other mutations of RTKs and Ras pathways, especially the CSF1R gene, has not been systematically studied in childhood CBF–AML. We previously reported that CBF–AML accounted for 24% of childhood AML in Taiwan.16 In the present study, we sought to delineate the cooperating mutations of CSF1R, c-KIT, FLT3–LM (length mutation of FLT3), FLT3–TKD (tyrosine kinase domain of FLT3), N- and K-Ras genes in CBF– AML. Furthermore, the roles of cooperating mutations in relapse of childhood CBF–AML were not clear. We also aimed to make a comparative analysis on paired diagnosis and relapse samples to elucidate the possible roles of class III RTKs and Ras mutations in the relapse of childhood CBF–AML.

Introduction

Materials and methods

Core-binding factor (CBF) acute myeloid leukemia (AML) includes AML with t(8;21)/RUNX1 (AML1)-RUNX1T1 (MTG8, ETO) and AML with inv(16)/CBFb-MYH11, which are two of the four well-characterized subgroups of recurrent genetic abnormalities defined by the World Health Organization (WHO) classification.1 A ‘two-hit’ model of leukemogenesis of AML, that is, cooperation of class I mutations which confer a proliferative advantage to leukemic cells and class II mutations which lead to differentiation block in leukemic cells, has been proposed.2 Activating mutations of FLT3, c-KIT, CSF1R (c-FMS) and Ras genes are of class I mutations, whereas RUNX1RUNX1T1 and CBFb-MYH11 are of class II mutations. FLT3, c-KIT and CSF1R belong to class III receptor tyrosine kinases (RTKs) and play a crucial role in hematopoiesis.3 Ligand binding

Patients and Materials

Correspondence: Professor D-C Liang, Division of Pediatric Hematology-Oncology, Mackay Memorial Hospital, 92, Sec. 2, Chung-San North Road, Taipei 10449, Taiwan. E-mail: [email protected] Received 9 July 2007; revised 26 September 2007; accepted 27 September 2007; published online 25 October 2007

Bone marrow (BM) samples obtained from 154 consecutive patients (age 17 years and younger) with de novo AML diagnosed at Mackay Memorial Hospital and Chang Gung Children’s Hospital, Linkou were examined. BM samples were collected at diagnosis, at achieving complete remission and at BM relapse. Informed consent was obtained from the parents or guardians of the patients. The study was approved by the Institutional Review Board. Leukemic cells of BM samples were enriched by Ficoll-Hypaque (1.077 g ml1; Amersham Bioscience, Uppsala Sweden) density-gradient centrifugation and cryopreserved in 10% dimethyl sulfoxide and 20% fetal bovine serum at 70 1C or in liquid nitrogen until test. At diagnosis, BM samples underwent Romanowsky system staining, cytochemical staining, immunophenotyping and cytogenetic analysis. The morphologic subtypes of AML were classified according to the French–American–British Cooperative Group.17–19 Reverse transcriptase (RT)–PCR assay for the detection of fusion transcripts of PML-RARa, RUNX1-RUNX1T1 and CBFb-MYH11 were performed as described previously.16

Cooperating mutations in childhood CBF-AML L-Y Shih et al

304 Detection of MLL rearrangement and characterization of fusion partner genes of MLL rearrangement were carried out as described before.20 The treatments were given according to the TPOG-AML protocols as described previously.21

Methods

DNA and RNA extraction Genomic DNA (gDNA) was extracted from frozen BM cells by using a DNA extraction kit (Puregene Gentra System, Minneapolis, MN, USA) according to the manufacturer’s instructions. RNA was extracted and reversely transcribed to complementary DNA (cDNA) as described previously.22

PCR buffer, Expand Long Polymerase (Roche, Mannheim, Germany) and 1 mM primers on a DNA thermal cycler (ABI 9700) using a touch-down program consisting of five cycles at 95 1C for 30 s, 68–64 1C for 30 s each and 68 1C for 90 s, and then 35 cycles at 95 1C for 30 s, 63 1C for 30 s and 68 1C for 90 s, using primers 245F and 413R to amplify exons 5–9. For amplification of exons 9–13 and exons 13–22, five cycles of 95 1C for 30 s, 65–61 1C for 60 s each, 68 1C for 150 s followed by 35 cycles of 95 1C for 30 s, 60 1C for 1 min and 68 1C for 150 s using primer pairs of both KIF/KIR and K2F/K2R, respectively. All PCR products were then directly sequenced in both directions as described above.

Detection of FLT3 and Ras mutations Detection of c-KIT mutations Mutation analysis of c-KIT was performed by direct sequencing for all RT–PCR products amplified with five overlapping primer pairs which cover the entire coding region of c-KIT gene from exon 1 through exon 21 (Supplementary Table S1). The RT–PCR reaction was carried out in a mixture containing cDNA, 0.2 mM dNTP, 2 mM MgCl2, 1.6 M Betaine, PCR buffer, Taq polymerase (Invitrogen, Carsbad, CA, USA), and 1 mM primers on a DNA thermal cycler (Applied Biosystems 9700) using a touch-down program consisting of eight cycles at 95 1C for 45 s, 65–58 1C for 45 s and 72 1C for 1 min and then 35 cycles at 95 1C for 45 s, 57 1C for 45 s and 72 1C for 90 s, with an initial preheating at 94 1C for 5 min and a final extension for 10 min at 72 1C. All PCR products were directly sequenced using BigDye Terminators and AmpliTaq FS (Applied Biosystems 3730) in both directions. Samples with abnormal sequencing results for c-KIT mutations were subjected to a repeated PCR using gDNA with selfdesigned intronic primers for amplification of the fragment of exon 8, c-KIT-8F 50 -GGAGTGAAGTGAATGTTGCTGAG-30 and c-KIT-8R 50 -CAAGTGAATTGCAGTCCTTCCC-30 , and using the primers to amplify the fragment of exon 17 according to the published primer sequences of c-KIT-17F 50 -GTTTTCACTCT T-TACAAGT-30 and c-KIT-17R 50 -TTACATTATGAAAGTCA CAGGAAAC-30 .23 The gDNA PCR reaction was carried out in a mixture containing gDNA, 0.2 mM dNTP, 2 mM MgCl2, 1 M Betaine, PCR buffer (pH8.5), Taq polymerase (Invitrogen) and 0.5 mM primers. The program for the gDNA-PCR reaction was initially denatured at 94 1C for 5 min and DNA amplification was achieved by 35 cycles of denaturation (95 1C for 45 s), annealing (59 1C, 45 s for exon 8; and 48 1C, 45 s for exon 17), primers extension (72 1C for 45 s) and a final extension at 72 1C for 10 min.

Cloning analysis For samples carrying two mutations, cloning analysis was carried out to clarify whether the two mutations were on the same allele or on different alleles. The PCR product was run on a gel, cut, purified and subcloned into the pCRII-TOPO vector (Invitrogen). At least 20 clones were subsequently sequenced for each sample.

Detection of CSF1R mutations Analysis of CSF1R mutations was performed by RT–PCR assay. The primers were designed to amplify the coding sequences from exon 5 through exon 22 of CSF1R gene (Supplementary Table S2). The RT–PCR reaction was carried out in a mixture containing cDNA, 0.5 mM dNTP, 0.75 mM MgCl2, 1.6 M Betaine, Leukemia

gDNA PCR assay was performed to identify FLT3–LM followed by Genescan analysis to determine the allelic distribution when abnormal PCR products were present.22 The detection of FLT3– TKD mutations was performed as described previously.24 DNA PCR or RT–PCR followed by direct sequencing was performed to detect mutations at codons 12, 13 and 61 of the N- and K-Ras genes.25

Statistical analysis The Fisher exact test (two sided) was used to compare the frequencies of mutations and BM relapse rates between two groups. Estimates of survival were calculated according to the Kaplan–Meier method. Comparisons of estimated survival curves were analyzed by the log-rank test. For all analyses, the P-valueso0.05 were considered statistically significant.

Results

Receptor tyrosine kinases and Ras mutations in CBF– AML at diagnosis A total of 27% of pediatric AML (41/154) had CBF–AML. In patients with t(8;21)/RUNX1-RUNX1T1, 42.9% (12/28) had cKIT mutations. In patients with inv(16)/CBFb-MYH11, 38.5% (5/ 13) had c-KIT mutations. Of the 17 patients with c-KIT mutations, five had single mutations in exon 8: one each with Y418_D419delinsG, T417_D419delinsF and T417_R420delinsRGK and two with D419del. Nine patients had single mutations in exon 17: D816Y in four, N822K in three, D816H and D816V in one each. Three patients had two c-KIT mutations. The two independent PCR assays reproducibly gave the identical results. Cloning analysis indicated that Y418N and Y418_D419insFF were on the same allele in one patient, Y418_D419del and D816H were on different alleles in the other patient. Another patient harboring both D816Y and N822K, the two mutations were found either on the same allele (one clone) or on different alleles (seven D816Y clones and eight N822K clones, respectively) by cloning analysis. CSF1R mutation was not detected in the whole cohort of AML. Of the 41 CBF–AML patients, one had FLT3–TKD (D835V) mutation; three had N-Ras mutations: one at codon 12 (Gly12Asp) and two at codon 13 (Gly13Arg and Gly13Cys); two had K-Ras mutations, one at codon 12 (Gly12Asp) and the other at codon 13 (Gly13Asp). No FLT3–LM mutations were detected in children with CBF–AML. Together, 54% of children with CBF–AML had collaboration with RTKs and/or Ras mutations at diagnosis, only one patient had two mutations (c-KIT and N-Ras).


Comparison of RTKs and Ras mutations between CBF– AML and non-CBF AML The frequencies of cooperating mutations of RTKs and Ras in CBF–AML and non-CBF AML at diagnosis are shown in Table 1. In contrast to the high frequency of c-KIT in CBF–AML, c-KIT mutations were not detected in children with non-CBF AML. Patients with non-CBF AML were associated with higher frequencies of FLT3–LM as compared with CBF–AML group. CSF1R mutations were not detected in both CBF–AML and nonCBF AML. In 62 patients with normal karyotype, 19 had FLT3–LM mutations, 9 had N-Ras mutations, 3 had K-Ras mutations and 6 patients had FLT3–TKD mutations. In seven patients with complex chromosomal abnormalities, none had RTKs or Ras mutations. Similarly, in three patients with monosomy 7, no any mutations were detected.

Remission marrows in AML with c-KIT mutations A total of 14 of the 17 patients harboring c-KIT mutations at diagnosis had complete hematologic remission marrow samples available for mutational analysis, no c-KIT mutations were detected in all the remission samples.

Mutational analysis on bone marrow relapse samples in CBF–AML A total of 8 of the 41 CBF–AML, including 7 of 28 patients with t(8;21)/RUNX1-RUNX1T1 and 1 out of 13 patients with inv(16)/ CBFb-MYH11, had BM relapses. The only one patient with inv(16)/CBFb-MYH11 who had a relapse did not harbor c-KIT mutation at initial diagnosis. All the eight CBF–AML patients with relapse had relapse marrow samples available for mutational analyses. Four patients with t(8;21)/RUNX1-RUNX1T1 and c-KIT mutations at diagnosis relapsed with the identical patterns of c-KIT mutations. One had double c-KIT mutations: Y418N; Y418_D419insFF and three had single mutations in exon 17: one each with D816H, D816V and D816Y. The remaining four children with CBF–AML did not have c-KIT mutations at both diagnosis and relapses. No mutations of CSF1R, FLT3–LM, FLT3–TKD, N-Ras or K-Ras genes were detected at diagnosis or at relapse in all of the eight CBF–AML patients with relapse.

The impact of c-KIT mutations on outcome of CBF–AML The BM relapse rate (8/41) in CBF–AML was lower than that of non-CBF AML (42/113) (P ¼ 0.051). The BM relapse rate in those with t(8;21)/RUNX1-RUNX1T1 (7/28) did not significantly differ from that in inv(16)/CBFb-MYH11 (1/13) (P ¼ 0.40). In patients with t(8;21)/RUNX1-RUNX1T1, the BM relapse rates did not significantly differ between those with and without c-KIT mutations (P ¼ 0.42).

Table 1

305 There were no statistically significant differences in overall survival (OS) between c-KIT mutations ( þ ) and c-KIT mutations() groups in the whole cohort of CBF–AML (63%±12.1 vs 72%±10.3, P ¼ 0.45), in RUNX1-RUNX1T1 subgroup (57%±14.6 vs 74%±11.2, P ¼ 0.40) or in CBFbMYH11 subgroup (80%±17.9 vs 70%±18.2, P ¼ 0.99). Also, no statistically significant differences were found in event-free survival (EFS) between c-KIT mutations( þ ) and c-KIT mutations() groups in the whole group of CBF–AML (71%±12.1 vs 80%±8.8, P ¼ 0.47), in RUNX1-RUNX1T1 subgroup (60%±15.5 vs 77%±11.5, P ¼ 0.30) or in CBFbMYH11 subgroup (100% vs 86%±13.2, P ¼ 0.45).

Discussion We have analyzed the c-KIT mutations in a large cohort of childhood CBF–AML. The frequency of c-KIT mutations in our series of AML was 11% compared with the reported frequencies ranging from 3.3 to 12.2%.9–11,26 The frequency of c-KIT mutations in our childhood CBF–AML was 41.5% compared with the reported frequencies of 0–37%.9–11,26 We found that there was no difference in the frequencies of c-KIT mutations between AML with RUNX1-RUNX1T1 and CBFb-MYH11 (42.9% vs 38.5%). The types of c-KIT mutations in our patients included in-frame deletion/insertion of exon 8, or point mutations in the activating loop of exon 17 and one patient had a combined exons 8 and 17 mutations. Double c-KIT mutations in the same individuals have been described in adults,15 but not in children. Kohl et al.27 demonstrated that c-KIT exon 8 mutants represent gain-of-function mutations that are sensitive to tyrosine kinase inhibitors and provide as a molecular target for therapy. All the mutations at exon 17 of c-KIT involved residues in the activating loop, D816 and/or N822, that exhibit increased tyrosine kinase activity.28,29 The reported patterns of pediatric c-KIT mutations varied. Meshinchi et al.26 only examined mutations on exon 17 of c-KIT gene and found none of the 14 children with CBF–AML had a c-KIT mutation. Screening exons 8, 9, 11, 17 and 18 of c-KIT, Shimada et al.12 found that 17.4% of patients with t(8;21) had c-KIT mutations that were exclusively located in exon 17. Two other groups reported that c-KIT mutations occurred in exon 8 or at D816 of exon 17 but not at N822.9,11 Whether the differences in the frequencies and mutations patterns of c-KIT among the published series and ours were attributed to the different ethinic composition, small sample size or different methodology used for detection of mutations remain to be determined. It is important to characterize the types of c-KIT mutations since their responses to tyrosine kinase inhibitors are different, for example, imatinib could inhibit N822 mutant but not D816 mutant.29 Data from others and the present study suggested that only one mutation of RTKs and Ras pathways would be required for leukemogenesis.10 Wang et al.29 have demonstrated that a

Comparison of the frequencies of mutations of RTKs and Ras between childhood CBF–AML and non-CBF AML at diagnosis

Groups CBF–AML Non-CBF AML P-value

c-KIT

CSF1R

FLT3–LM

FLT3–TKD

N-Ras

K-Ras

Total

17/41 (41.5%) 0/109 o0.0001

0/38 0/110 F

0/41 25/111 (22.5%) 0.001

1/40 (2.5%) 8/111 (7.2%) 0.446

3/41 (7.3%) 12/112 (10.7%) 0.760

2/41 (4.9%) 8/110 (7.7%) 0.729

22/41 (53.7%) 50/112 (44.6%) 0.363

Abbreviations: AML, acute myeloid leukemia; CBF, core-binding factor; LM, length mutation; RTKs, receptor tyrosine kinases; TKD, tyrosine kinase domain. Leukemia


306 stepwise leukemogenesis in RUNX1-RUNX1T1 leukemia by inducing RUNX1-RUNX1T1 expression into U937-A/E cells resulted in significantly upregulated mRNA and protein levels of c-KIT. Their findings suggested that c-KIT pathway may be needed for certain signaling events for efficient leukemogenesis in CBF–AML. In contrast, as hypothesized by other investigators, FLT3 mutations seem to be much efficient in driving leukemic proliferation in non-CBF AML.10,30 The human CSF1R functions as the receptor of monocytecolony stimulating factor.31 Early studies suggested a role of CSF1R mutation in leukemogenesis.3 By using allele-specific oligonucleotide (ASO) hybridization, activating mutations at codons 301 and 969 have been found in two small series of adult AML,32,33 whereas others did not.34–36 To the best of our knowledge, there was only one report on mutational analysis of CSF1R in pediatric AML,26 however, which only focused on codons 301 and 969 and no mutation was detected. Recently, Abu-Duhier et al.36 screened the entire CSF1R gene in 60 adult AML patients by using direct sequencing, they failed to find codons 301 or 969 mutations but identified two novel mutations at codons 245 and 413 in three patients, one of them carrying inv(16). We used RT–PCR assay followed by direct sequencing to examine the coding sequences from exons 5 through exon 22 of CSF1R, we were unable to detect any mutations at codons 245 (exon 6), 301 (exon 7), 413 (exon 9) or 969 (exon 22) of CSF1R gene. It is of note that the previously reported mutations at codons 301 or 969 were all detected with the relatively insensitive and less specific ASO hybridization technique,32,33 by which codon 969 mutant was even found in one of the 51 normal subjects.33 In the present study, the lack of CSF1R mutations in a systematic analysis on a large cohort of childhood AML indicated that CSF1R mutations is of no pathogenetic relevance either in CBF or non-CBF AML. There have been very rare studies on the molecular alterations associated with the relapse of CBF–AML. We performed a comparative analysis on paired diagnosis and relapse marrow samples. We were unable to find acquisition or loss of any of the mutations involving c-KIT, CSF1R, FLT3–LM, FLT3–TKD, N-Ras and K-Ras. Of patients harboring the c-KIT mutations at diagnosis, all retained the identical mutations at relapse, suggesting that c-KIT mutations play a crucial role in the leukemogenesis in a substantial proportion of patients with CBF–AML. Recently, c-KIT mutations have been found to adversely affect outcome in adult AML patients with t(8;21)/RUNX1RUNX1T1,11,14,15,37 and inv(16)/CBFb-MYH11.15 c-KIT mutations were also described to be associated with adverse outcome in childhood AML with t(8;21)/RUNX1-RUNX1T1 in one series,12 but not in other series.10 In the present study, there seemed no adverse impact of c-KIT mutations on BM relapse rate, EFS and OS of childhood CBF–AML. Further study including a larger patient population with CBF–AML is required to clarify this. Our results showed that 54% of childhood de novo CBF–AML had RTKs and/or Ras mutations, with c-KIT being the most common. The frequency of c-KIT mutations in CBF–AML was significantly higher than that in non-CBF AML, whereas the frequency of FLT3–LM in non-CBF AML was significantly higher than that in CBF–AML. There were no significant differences in the frequencies of CSF1R, FLT3–TKD, N-Ras and K-Ras mutations between CBF and non-CBF AML.

Acknowledgements This work was supported by grants MMH-E-94009 and MMH-E-95009 from the Mackay Memorial Hospital, grant NSC Leukemia

94-2314-B-195-003 from the National Science Council and grant NHRI-EX96-9434SI from the National Health Research Institute, Taiwan.

References 1 Vardiman JW, Harris NL, Brunning RD. The World Health Organization (WHO) classification of the myeloid neoplasms. Blood 2002; 100: 2292–2302. 2 Kelly LM, Gilliland DG. Genetics of myeloid leukemias. Annu Rev Genomics Hum Genet 2002; 3: 179–198. 3 Reilly JT. Class III receptor tyrosine kinases: role in leukaemogenesis. Br J Haematol 2002; 116: 744–757. 4 Kondo M, Horibe K, Takahashi Y, Matsumoto K, Fukuda M, Inaba J et al. Prognostic value of internal tandem duplication of the FLT3 gene in childhood acute myelogenous leukemia. Med Pediatr Oncol 1999; 33: 525–529. 5 Iwai T, Yokota S, Nakao M, Okamoto T, Taniwaki M, Onodera N et al. Internal tandem duplication of the FLT3 gene and clinical evaluation in childhood acute myeloid leukemia. The Children’s Cancer and Leukemia Study Group, Japan. Leukemia 1999; 13: 38–43. 6 Xu F, Taki T, Yang HW, Hanada R, Hongo T, Ohnishi H et al. Tandem duplication of the FLT3 gene is found in acute lymphoblastic leukaemia as well as acute myeloid leukaemia but not in myelodysplastic syndrome or juvenile chronic myelogenous leukaemia in children. Br J Haematol 199; 105: 155–162. 7 Meshinchi S, Woods WG, Stirewalt DL, Sweetser DA, Buckley JD, Tjoa TK et al. Prevalence and prognostic significance of Flt3 internal tandem duplication in pediatric acute myeloid leukemia. Blood 2001; 97: 89–94. 8 Liang DC, Shih LY, Hung IJ, Yang CP, Chen SH, Jaing TH et al. Clinical relevance of internal tandem duplication of the FLT3 gene in childhood acute myeloid leukemia. Cancer 2002; 94: 3292– 3298. 9 Beghini A, Ripamonti CB, Cairoli R, Cazzaniga G, Colapietro P, Elice F et al. KIT activating mutations: incidence in adult and pediatric acute myeloid leukemia, and identification of an internal tandem duplication. Haematologica 2004; 89: 920–925. 10 Goemans BF, Zwaan CM, Miller M, Zimmermann M, Harlow A, Meshinchi S et al. Mutations in KIT and RAS are frequent events in pediatric core-binding factor acute myeloid leukemia. Leukemia 2005; 19: 1536–1542. 11 Boissel N, Leroy H, Brethon B, Philippe N, de Botton S, Auvrignon A et al. Incidence and prognostic impact of c-KIT, FLT3, and Ras gene mutations in core binding factor acute myeloid leukemia (CBF-AML). Leukemia 2006; 20: 965–970. 12 Shimada A, Taki T, Tabuchi K, Tawa A, Horibe K, Tsuchida M et al. KIT mutations, and not FLT3 internal tandem duplication, are strongly associated with a poor prognosis in pediatric acute myeloid leukemia with t(8;21): a study of the Japanese Childhood AML Cooperative Study Group. Blood 2006; 107: 1806–1809. 13 Care RS, Valk PJ, Goodeve AC, Abu-Duhier FM, GeertsmaKleinekoort WM, Wilson GA et al. Incidence and prognosis of cKIT and FLT3 mutations in core binding factor (CBF) acute myeloid leukaemias. Br J Haematol 2003; 121: 775–777. 14 Cairoli R, Beghini A, Grillo G, Nadali G, Elice F, Ripamonti CB et al. Prognostic impact of c-KIT mutations in core binding factor leukemias: an Italian retrospective study. Blood 2006; 107: 3463– 3468. 15 Paschka P, Marcucci G, Ruppert AS, Mrozek K, Chen H, Kittles RA et al. Adverse prognostic significance of KIT mutations in adult acute myeloid leukemia with inv(16) and t(8;21): a Cancer and Leukemia Group B Study. J Clin Oncol 2006; 24: 3904–3911. 16 Liang DC, Shih LY, Yang CP, Hung IJ, Chen SH, Liu HC. Molecular analysis of fusion transcripts in childhood acute myeloid leukemia in Taiwan. Med Pediatr Oncol 2001; 37: 555–556. 17 Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR et al. Criteria for the diagnosis of acute leukemia of megakaryocyte lineage (M7). A report of the French-AmericanBritish Cooperative Group. Ann Intern Med 1985; 103: 460–462. 18 Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR et al. Proposed revised criteria for the classification of


307 19 20

21 22

23

24 25

26

27

acute myeloid leukemia. A report of the French-American-British Cooperative Group. Ann Intern Med 1985; 103: 620–625. Keenan FM, Barnett D, Reilly JT. Clinico-pathological features of minimally differentiated acute myeloid leukaemia (AML-M0). Br J Haematol 1992; 81: 458–459. Shih LY, Liang DC, Fu JF, Wu JH, Wang PN, Lin TL et al. Characterization of fusion partner genes in 114 patients with de novo acute myeloid leukemia and MLL rearrangement. Leukemia 2006; 20: 218–223. Liang DC, Chan TT, Lin KH, Lin DT, Lu MY, Chen SH et al. Improved treatment results for childhood acute myeloid leukemia in Taiwan. Leukemia 2006; 20: 136–141. Shih LY, Huang CF, Wu JH, Lin TL, Dunn P, Wang PN et al. Internal tandem duplication of FLT3 in relapsed acute myeloid leukemia: a comparative analysis of bone marrow samples from 108 adult patients at diagnosis and relapse. Blood 2002; 100: 2387–2392. Spritz RA, Giebel LB, Holmes SA. Dominant negative and loss of function mutations of the c-KIT (mast/stem cell growth factor receptor) proto-oncogene in human piebaldism. Am J Hum Genet 1992; 50: 261–269. Liang DC, Shih LY, Hung IJ, Yang CP, Chen SH, Jaing TH et al. FLT3-TKD mutation in childhood acute myeloid leukemia. Leukemia 2003; 17: 883–886. Liang DC, Shih LY, Fu JF, Li HY, Wang HI, Hung IJ et al. K-Ras mutations and N-Ras mutations in childhood acute leukemias with or without mixed-lineage leukemia gene rearrangements. Cancer 2006; 106: 950–956. Meshinchi S, Stirewalt DL, Alonzo TA, Zhang Q, Sweetser DA, Woods WG et al. Activating mutations of RTK/ras signal transduction pathway in pediatric acute myeloid leukemia. Blood 2003; 102: 1474–1479. Kohl TM, Schnittger S, Ellwart JW, Hiddemann W, Spiekermann K. KIT exon 8 mutations associated with core-binding factor (CBF)acute myeloid leukemia (AML) cause hyperactivation of the

28

29

30 31 32

33 34

35 36 37

receptor in response to stem cell factor. Blood 2005; 105: 3319–3321. Kitayama H, Tsujimura T, Matsumura I, Oritani K, Ikeda H, Ishikawa J et al. Neoplastic transformation of normal hematopoietic cells by constitutively activating mutations of c-KIT receptor tyrosine kinase. Blood 1996; 88: 995–1004. Wang YY, Zhou GB, Yin T, Chen B, Shi JY, Liang WX et al. AML1ETO and C-KIT mutation/overexpression in t(8;21) leukemia: implication in stepwise leukemogenesis and response to Gleevec. Proc Natl Acad Sci USA 2005; 102: 1104–1109. Gilliland DG, Griffin JD. Role of FLT3 in leukemia. Curr Opin Hematol 2002; 9: 274–281. Sherr CJ. Colony-stimulating factor-1 receptor. Blood 1990; 75: 1–12. Tobal K, Pagliuca A, Bhatt B, Bailey N, Layton DM, Mufti GJ. Mutation of the human FMS gene (M-CSF receptor) in myelodysplastic syndromes and acute myeloid leukemia. Leukemia 1990; 4: 486–489. Ridge SA, Worwood M, Oscier D, Jacobs A, Padua RA. FMS mutations in myelodysplastic, leukemic, and normal subjects. Proc Natl Acad Sci USA 1990; 87: 1377–1380. Natazuka T, Matsui T, Ito M, Nakata H, Nakagawa T, Nakamura H et al. Rare point mutation at codon 301 and 969 of FMS/M-CSF receptor in acute myelomonocytic and monocytic leukemia. Leuk Res 1992; 16: 541–543. Springall F, O’mara S, Shounan Y, Todd A, Ford D, Iland H. CSF1R point mutations in acute myeloid leukemia: fact or fiction? Leukemia 1993; 7: 978–985. Abu-Duhier FM, Goodeve AC, Wilson GA, Peake IR, Reilly JT. CSF1R mutational analysis in acute myeloid leukaemia. Br J Haematol 2003; 123: 749–750. Schnittger S, Kohl TM, Haferlach T, Kern W, Hiddemann W, Spiekermann K et al. KIT-D816 mutations in AML1-ETO-positive AML are associated with impaired event-free and overall survival. Blood 2006; 107: 1791–1799.

Supplementary Information accompanies the paper on the Leukemia website (http://www.nature.com/leu)

Leukemia