Acute Non Lymphocytic Leukemias

AML patients with CEBPα mutations mostly retain identical mutant patterns but frequently change in allelic distribution at relapse: a comparative analysis on paired diagnosis and relapse samples

Article metrics


The roles of CEBPα mutations and its cooperating mutations in the relapse of acute myeloid leukemia (AML) are not clear. CEBPα mutations were analyzed on 149 patients with de novo AML at both diagnosis and relapse. Twenty-two patients (14.8%) had the mutations at diagnosis, two patients had N-terminal nonsense mutations alone, one had homozygous inframe duplication at the bZIP domain, and 19 patients had both N-terminal and bZIP mutations. Twenty patients relapsed with identical mutant patterns, two lost CEBPα mutations and none acquired the mutations at relapse. Cloning analysis showed that the N-terminal and C-terminal mutations occurred on separate cloned alleles and also on the same alleles in most of the diagnosis and relapse samples. Losing one of the two or more mutations on the same allele or acquiring the other mutation on the allele original carrying single mutation were observed not infrequently in the paired samples analyzed. Seven patients with CEBPα mutations had cooperating mutations with FLT3/ITD, FLT3/TKD or N-ras but not K-ras mutations. Our study showed that 91% of de novo AML harboring CEBPα mutations at diagnosis retained the identical mutant patterns but frequently changed in the allelic distribution at relapse.


Transcription factor CCAAT/enhancer binding protein α (C/EBPα) plays an important role in hematopoiesis and is essential for the differentiation of granulocytes.1, 2 CEBPα mutations have been described in 7–10% of adult patients with newly diagnosed acute myeloid leukemia (AML),3, 4, 5, 6, 7 and were associated with a favorable prognosis.5, 7 The role of CEBPα mutations in the progression of de novo AML is not clear. We have recently demonstrated that CEBPα mutations play a role in a subset of patients with myelodysplastic syndrome during the disease course; the patterns of CEBPα mutations in the progression to AML were heterogeneous, including retaining the identical mutant clones, loss of mutant clones or emergence of novel clones.8 Up to now, only one study of CEBPα mutations in relapsed AML had been reported.9 However, the patient number in that study was small, only two patients carried the mutations. Further study including a larger number of paired samples at both diagnosis and relapse from AML patients are required to define the role of CEBPα mutations in the relapse of AML.

In the present study, we analyzed CEBPα mutations in bone marrow (BM) samples collected at both diagnosis and relapse from 149 patients. We aimed to assess the role of the mutations in the relapse of AML and to determine whether the mutation pattern at relapse differed from those at initial diagnosis. In addition, the cooperating mutations with FLT3 and Ras genes in CEBPα mutation(+) patients were also analyzed. To the best of our knowledge, the present study is the largest series in which CEBPα mutations have been extensively analyzed in paired diagnostic and relapse samples.

Materials and methods


Between 1991 and 2004, BM samples from 132 adults and 17 children with de novo AML were examined at diagnosis and at first relapse. In addition, samples from 14 patients obtained at second relapse were also examined. Complete remission samples from patients who carried CEBPα mutations at diagnosis were also studied. Informed consent was obtained from each patient or the patient's guardians. This study was approved by the Human Research Committee of Chang Gung Memorial Hospital and Mackay Memorial Hospital. All relapse samples contained at least 20% BM blasts. The mononuclear cells from BM samples were enriched by Ficoll–Hypaque (1.077 g/ml, Amersham Pharmacia, Buckinghamshire, UK) density gradient centrifugation and cryopreserved in 10% dimethyl sulfoxide and 20% fetal bovine serum at −70°C or in liquid nitrogen until use.

Cytochemical study, immunophenotyping, cytogenetic analysis, Southern blot analysis of MLL rearrangement, and reverse transcriptase-polymerase chain reaction (PCR) assay for detection of common fusion transcripts were performed on the diagnosis samples as described previously.10 The morphologic subtypes were classified according to the French-American-British (FAB) criteria.11, 12 Cytogenetic risk category was classified according to the criteria adopted by the UK Medical Research Council (MRC) AML10 trial,13 with minor modification as follows: favorable risk group defined by the presence of t(8;21)/AML1-ETO, inv(16)/CBFβ-MYH11, or t(15;17)/PML-RARα, irrespective of the presence of additional cytogenetic abnormalities; unfavorable risk group defined by the presence of −5/del(5q), −7/del(7q), 3q abnormalities, or complex karyotype 3 unrelated abnormalities; the remaining group of patients composed of the intermediate risk group except for patients with MLL rearrangements that were segregated as an additional group. Before 1995, adult patients received induction chemotherapy with daunomycin and cytarabine (3+7) regimen followed by consolidation with (2+5) regimen for three cycles and then weekly maintenance (6-thioguanine and cytarabine) and intensification therapy (cytarabine plus daunomycin or etoposide) every 3 months for 2 years. After 1995, the postremission therapy consisted of high-dose cytarabine plus daunomycin alternating with etoposide for 4–6 courses. Since 1994, patients with AML-M3 received all-trans-retinoic acid in addition to chemotherapy. Seventeen patients underwent stem cell transplantation. Pediatric patients were treated with TPOG-AML-901 and TPOG-AML-97 as described previously.14

DNA PCR and direct sequencing for detection of CEBPα mutations

DNA PCR assay followed by direct sequencing for each PCR product was performed as described previously.8, 15 Briefly, the PCR reaction was carried out with two overlapping primer pairs PP1F and PP1R, and PP2F and PP2R which cover the entire coding region of human CEBPα.3 PCR products were purified (Qiagen, Hilden, Germany) and sequenced using BigDye Terminators and AmpliTaq FS (ABI 3730, Applied Biosystems, Foster City, CA, USA) in both directions. Samples with abnormal or ambiguous sequencing results were subjected to repeated PCR assays with an alternative primer pairs.8, 15 The PCR products were again purified and sequenced directly.

Expand long template PCR assay and cloning analysis

For samples carrying more than one mutant, expand long template PCR assay (Roche, Mannheim, Germany) using the primer pair of PP1F and PP2R was carried out to determine the allelic distribution of the multiple mutations on the same allele or on different alleles. The PCR product was subcloned into the pCRII-TOPO vector (Invitrogen, Carlsbad, CA, USA), and at least 10 clones were subsequently sequenced for each sample.

Detection of Ras and activating FLT3 mutations

The DNA PCR assay with direct sequencing for the detection of point mutations at codons 12, 13 and 61 of the N-ras gene, detection of internal tandem duplication of FLT3 (FLT3/ITD) and point mutations at tyrosine kinase domain of FLT3 (FLT3/TKD), GeneScan-based analysis of the FLT3/ITD mutant level, and sequencing of the duplicated fragments of FLT3/ITD were performed as previously described.10, 16 For detection of K-ras mutations, the DNA or cDNA PCR assays were performed as those described for N-ras mutations except with the different primers as follows: for cDNA PCR assay, using the forward primers of K-ras-cF: 5′-IndexTermCAT TTC GGA CTG GGA GCG AG-3′; and the reverse primer of K-ras-cR: 5′-IndexTermCTA TAA TGG TGA ATA TCT TCA AAT GAT TTA GT-3′ which amplifying a fragment of 387 bps covering codons 12, 13 and 61. For DNA-PCR assay, (exon1) K-ras-E1-F was 5′-IndexTermGGT GAG TTT GTA TTA AAA GGT ACT GGT G-3′ and K-ras-E1-R was 5′-IndexTermCCT CTA TTG TTG GAT CAT ATT CGT CC-3′; (exon2) K-ras-E2-F was 5′-IndexTermGGA TTC CTA CAG GAA GCA AGT AGT AA-3′ and same K-ras-cR as for cDNA-PCR assay.

Statistical analysis

Frequencies of CEBPα mutations at diagnosis and relapse were compared with Fisher's exact test. χ2 analysis was used to compare data among different subgroups. The clinicohematologic variables and the presence of CEBPα mutations were compared by t-test. Overall survival was estimated by the method of Kaplan and Meier, and were compared using the log-rank test. All P-values were calculated by using two-sided tests. P0.05 was considered as statistically significant. The statistic analysis was performed by using a software of SPSS 8.0 for Windows (SPSS Ins. Chicago, IL, USA).


CEBPα mutations at diagnosis and at relapse

The paired diagnosis and relapse BM samples from 132 adult patients (ages ranged from 15 to 74 years with a median of 43 years; 72 were male) and 17 children (ages ranged from 5 months to 15 years with a median of 7.4 years, 13 were boys) were analyzed for CEBPα mutations. The first relapse occurred at a median of 9.4 months (range, 1.1–59.7 months) after the initial diagnosis. One hundred and twenty-seven patients had wild-type CEBPα at both diagnosis and relapse. CEBPα mutations were detected in 22 patients at diagnosis, two of them lost mutations and none acquired the mutation at relapse. As shown in Table 1, 20 patients (nos. 1–20) had mutations located in the N-terminal part of the protein resulting in frameshift and truncation of the proteins. Another one patient (no. 21) had a termination of translation by introducing a nonsense codon (P109X) and two additional outframe deletion at TAD2 and bZIP domains inducing a frameshift and termination. Nineteen patients (nos. 1–18 and 21) with N-terminal mutations also had C-terminal mutations at the bZIP domain, which consisted of inframe duplication, insertion, and deletion. Two patients (nos. 19 and 20) had N-terminal mutations alone; both had heterozygous single allelic mutations at N-part, patient 19 relapsed with an identical mutation whereas patient 20 lost N-terminal CEBPα mutation at relapse. The only one patient (no. 22) without N-terminal mutation had a homozygous C-terminal mutation with a large fragment of 66 bps (Q305-K326) duplication. We did not find any point mutation at bZIP domain. Together, 20 of the 22 patients harboring CEBPα mutations relapsed with identical mutation patterns, three of them had a second relapse that also exhibited the same patterns as their initial diagnosis and first relapse. None of the second relapse samples from 11 patients who lacked CEBPα mutations at diagnosis and first relapse acquired the mutations. CEBPα mutations were not detected in all complete remission samples obtained from the 22 patients who had mutations at diagnosis.

Table 1 Clinicohematologic characteristics of acute myeloid leukemia patients carrying CEBPα mutations in paired diagnostic and relapse samples

Cloning analysis of samples carrying more than one mutations

Samples harbored both N-terminal and bZIP mutations at diagnosis and relapse were all available for the expand long template PCR system followed by cloning analysis. The allelic frequency and distribution of combined N-terminal and C-terminal mutations of CEBPα of the 18 paired diagnosis and relapse samples are shown in Table 2. At diagnosis, three patients (nos. 4, 9 and 10) had heterozygous biallelic mutations with N-terminal and bZIP mutations on the separate alleles, whereas the remaining 15 patients had at least one clone carrying combined mutations on the same allele apart from other clones harboring single mutations either at N-terminal or at bZIP domain. At relapse, three patients (nos. 1, 7 and 8) losing one of the two mutations on the same allele and two (nos. 9 and 10) acquiring the other mutation on the allele originally carrying single mutation. Patient 1 had combined mutations in most of the cloned alleles at diagnosis but relapsed with exclusively single mutation at bZIP domain in 15 of the 16 clones analyzed and the remaining one clone being wild-type. Patient 21 had three mutants at both diagnosis and relapse, cloning analysis showed a different allelic distribution between diagnosis and relapse samples (Table 3); four new cloned alleles were detected at relapse, with one clone carrying all three mutations on the same allele, and additional one each carrying N-terminal mutation plus mutation at TAD2 or bZIP domain, whereas the clone carrying 974_975 del alone was not observed in the 12 cloned alleles examined at relapse. Together, seven patients had changes in the alleleic distribution at relapse.

Table 2 Allelic frequency and distribution of combined mutations of CEBPα at diagnosis and at relapse in 18 patients
Table 3 Result of cloning analysis of multiple mutations of CEBPα for patient 21

Cooperating mutations

Of the 22 patients with CEBPα mutations, two had concomitant FLT3/ITD mutations at diagnosis, one of them lost FLT3/ITD at relapse. Another two patients acquired FLT3/TKD mutations at relapse. N-ras mutations were detected in three patients with CEBPα mutations at diagnosis and two of them retained the identical N-ras mutations at relapse. None had K-ras mutations either at diagnosis or at relapse. Taken together, seven patients with CEBPα mutations at diagnosis/relapse had coexistence of FLT3/ITD, FLT3/TKD or N-ras mutations. Of the 127 patients without CEBPα mutations, FLT3/ITD, FLT3/TKD, N-ras and K-ras mutations were found in 27, 9, 14 and 4 patients, respectively. There was no difference in the frequency of cooperating mutations with FLT3/ITD (P=0.249), FLT3/TKD (P=0.357), N-ras (P=0.719) or K-ras (P=1.000) between patients with and without CEBPα mutations.

Clinicohematologic characteristics of relapsed AML patients with CEBPα mutations

A comparison of clinicohematologic characteristics between patients with CEBPα mutations and those without the mutations was made. No significant differences in age, sex, hemoglobin level, platelet count, WBC count, or percentage of blasts in BM or peripheral blood were observed between the two groups. CEBPα mutations were strongly associated with AML-M1 or M2 as compared with CEBPα mutation(−) patients (P=0.022). CEBPα mutations were not present in our patients with AML-M0, M3 or M7. Cytogenetic analysis was performed in 128 patients. CEBPα mutations were detected in 15 of the 62 patients with intermediate cytogenetics. The mutations were not observed in the 40 patients with favorable cytogenetics including 24 patients with t(8;21)/AML1-ETO, seven with inv(16)/CBFβ-MYH11, and nine with t(15;17)/PML-RARα; or in the six patients with unfavorable cytogenetics. Twenty patients had MLL rearrangements, eight with partial tandem duplication of MLL gene (MLL-PTD) and 12 with other 11q23 translocations including three t(6;11), four t(9;11), two t(10;11), two t(11;19) and one with interstitial deletion of 11q23 resulting in MLL-CBL;17 of them, only one patient with MLL-PTD had CEBPα mutation. The difference in the prevalence of CEBPα mutations among the cytogenetic/molecular subgroups at diagnosis was statistically significant (P=0.002).

The treatment outcome was analyzed according to the CEBPα mutation status at diagnosis. Kaplan–Meier analysis revealed that the median remission duration was 11.6 months for CEBPα mutation(+) patients and 9.0 months for CEBPα mutation(−) patients (P=0.142). The median overall survival was 22.6 months for patients with CEBPα mutations compared with 16.6 months for patients without CEBPα mutations (P=0.064).


There has been only one small study including 26 patients with relapsed AML examined on CEBPα mutations,9 in which only two patients had CEBPα mutations at diagnosis and both retained the mutations at relapse, none lost or acquired a mutation at relapse. In the present study, the incidence of CEBPα mutations in the cohort of AML patients who later had a relapse was 14.8% at initial diagnosis, 9.1% (2/22) of these patients lost the mutations and we did not observe the emergence of a novel mutation at relapse. We have recently described a heterogeneous pattern of CEBPα mutations, that is retaining the identical clones, clonal change or clonal evolution in the progression of myelodysplastic syndrome and chronic myelomonocytic leukemia to AML.8 Unlike the heterogeneous mutation pattern in myelodysplastic syndrome, we found that the vast majority of CEBPα mutant clones or subclones detected at diagnosis persisted at relapse in de novo AML. BM samples obtained at complete remission from all our 22 patients carrying CEBPα mutations at diagnosis did not harbor detectable CEBPα mutations, indicating that CEBPα mutations were leukemia-associated and were essential for the leukemogenesis in those who had the mutations at both phases. The observation of CEBPα mutations being present only at diagnosis but not at relapse in our two patients suggested that chemotherapy might be able to eradicate the leukemic clones carrying the CEBPα mutations, and when the leukemia relapsed, the CEBPα mutant clones did not recur. For patients losing CEBPα mutations at relapse, it seems likely that other genetic alterations may be responsible for blocking CEBPα function.

In the present study, two patients with CEBPα mutations had N-terminal mutations alone, one patient had only C-terminal mutation. Taken together, 86% (19/22) of our patients had both N- and C-terminal mutations of CEBPα, which were higher than the reported frequencies varying from 11% (1/9) to 40% (6/15).3, 4, 5, 7 The discrepancy of the frequency of combined N- and C-terminal mutations between ours and previous studies might be attributed to the different patient population studied. We focused on relapsed AML patients and examined paired diagnosis and relapse samples whereas the previous studies focused only on diagnosis samples. All the CEBPα mutations detected in the present study were examined by repeated PCR with alternative primer pairs, which in conjunction with cloning analysis performed on all 19 paired samples carrying combined mutations; thus, at least three independent PCR assays reproducibly detected the presence of identical mutations in these patients. The very high frequency of combined N-terminal and bZIP domain mutations in our patients suggested that relapse was rarely detected in the subset of patients with only an N-terminal mutation. Leroy et al.18 recently reviewed the 137 reported nonsilent CEBPα mutations in 87 patients with newly diagnosed AML, they divided the protein into five regions (R1–R5) with R1 and R4 accounting for 76% of the mutations. The distribution of the mutations in our patients were R1 68.2%, R2 27.3% and R4 90.9%. All but one mutation in R1 and R2 were 5′ to the second ATG codon leading to the production of a truncated nonfunctional protein or overexpression of dominant-negative P30 isoform. Mutations at bZIP domain (R4) abrogate dimerization and DNA binding which in turn reducing the transactivation of CEBPα target genes.3, 4

Cloning analysis demonstrated that the majority of the cloned alleles carried single N-terminal or C-terminal mutations on different alleles; but at least one clone harboring more than one mutation on the same alleles were detected in 79% (15/19) of diagnosis samples and 78% (14/18) of relapse samples. The allelic distribution of multiple CEBPα mutations in the initial diagnostic samples were also examined on diagnostic samples by two other groups, one described exclusively on different alleles6 and the other observed a similar findings as ours with the presence of combined mutations on the same alleles or on different alleles.7 It was interesting to note that changes in the allelic frequency and distribution occurred not infrequently at relapse in patients carrying combined mutations at diagnosis. As we only screened 10–32 clones for each sample in the cloning analysis, it was possible that the allelic frequency of CEBPα mutations might be different if much more clones would have been analyzed. Nevertheless, changes in allelic distribution of CEBPα mutations at relapse was not infrequent as compared with that at diagnosis. Lack of stability of CEBPα mutations in some cases of AML argued in favor of the mutations being a second event in leukemogenesis.

Two-hit hypothesis for the pathogenesis of AML has recently been proposed; one class of mutations that drive cell proliferation and survival and the other class of mutations that block differentiation.19 Activating mutations of FLT3 or Ras genes lead to enhanced cellular proliferation and leukemic transformation,20, 21 and CEBPα mutations result in the disruption of granulocyte differentiation.22 Cooperating mutations with FLT3 and Ras genes were investigated on our patients with CEBPα mutations. Coexistence of FLT3/ITD, FLT3/TKD, or N-ras but not K-ras mutations were present in seven patients carrying CEBPα mutations at diagnosis and/or relapse. Our findings suggested that these genetic alterations played a role in the leukemogenesis through a collaborative manner in a subset of AML patients, which support the two-hit model of leukemogenesis. However, we failed to find a correlation between the frequency of cooperating mutations with FLT3/ITD, FLT3/TKD, N-ras or K-ras mutations and the mutation status of CEBPα.

We compared the clinicohematologic features of the population of relapsed AML according to the CEBPα mutation status. Most of CEBPα mutations were associated with AML-M1 or M2 morphology and intermediate risk cytogenetics as described previously in the newly diagnosed AML.3, 4, 5, 6, 7, 23 CEBPα mutations were not observed in AML-M6 in the previous report, we did find one patient with M6 carrying the mutation. 11q23 abnormalities/MLL rearrangements have been assigned to either the unfavorable cytogenetic risk category or intermediate risk category by different study groups,13, 24, 25 we thus segregated this genetic abnormalities to an additional subgroup. None of our patients with CEBPα mutations had favorable or unfavorable cytogenetic groups. CEBPα mutation was present in one patient with MLL-PTD and not detected in patients with other 11q23 translocations. As MLL-PTD is cytogenetically silent for 11q23 abnormality, there might have been cases with this genetic subtype in the reported cases classified as intermediate cytogenetics carrying CEBPα mutations. There was no difference in age, sex, blood counts, and percentage of blasts in BM or peripheral blood between patients with or without CEBPα mutations either at diagnosis or at relapse.

CEBPα mutations has recently been described to be associated with favorable outcome in adult AML with normal karyotypes.5, 6, 7 The prognostic impact of CEBPα mutations was also analyzed in this selected cohort of relapsing AML patients. Patients with CEBPα mutations had a longer remission duration and overall survival than those without the mutations, but the difference did not reach statistical significance. However, the present study specially focused on a selected AML patients who ultimately relapsed, the survival data did not represent the real prognostic significance of CEBPα mutations in AML patients, that should include those who did not have relapses.

The detection of minimal residual disease is of growing importance in AML therapy. Almost all our AML patients with CEBPα mutations lacked other specific molecular markers. In addition, approximately 90% of CEBPα mutation(+) patients retained the identical mutants and none acquired a new mutant clone at relapse, therefore, CEBPα mutations might serve as a useful marker for monitoring of minimal residual disease or for detecting early relapse in these subset of patients except for those who lost CEBPα mutations at relapse. Although the majority of the N-terminal mutations included 1 bp deletion/insertion, incorporating allele specific PCR using primers with sequences designed at mutation sites, into the real-time quantitative PCR assay would be able to measure the levels of CEBPα mutants for minimal residual disease monitoring.

Accession codes




  1. 1

    Zhang DE, Zhang P, Wang ND, Hetherington CJ, Darlington GJ, Tenen DG . Absence of granulocyte colony-stimulating factor signaling and neutrophil development in CCAAT enhancer binding protein alpha-deficient mice. Proc Natl Acad Sci USA 1997; 94: 569–574.

  2. 2

    Radomska HS, Huettner CS, Zhang P, Cheng T, Scadden DT, Tenen DG . CCAAT/enhancer binding protein alpha is a regulatory switch sufficient for induction of granulocytic development from bipotential myeloid progenitors. Mol Cell Biol 1998; 18: 4301–4314.

  3. 3

    Pabst T, Mueller BU, Zhang P, Radomska HS, Narravula S, Schnittger S et al. Dominant-negative mutations of CEBPA, encoding CCAAT/enhancer binding protein-alpha (C/EBP alpha), in acute myeloid leukemia. Nat Genet 2001; 27: 263–270.

  4. 4

    Gombart AF, Hofmann WK, Kawano S, Takeuchi S, Krug U, Kwok SH et al. Mutations in the gene encoding the transcription factor CCAAT/enhancer binding protein alpha in myelodysplastic syndromes and acute myeloid leukemias. Blood 2002; 99: 1332–1340.

  5. 5

    Preudhomme C, Sagot C, Boissel N, Cayuela JM, Tigaud I, de Botton S, et al., ALFA Group. Favorable prognostic significance of CEBPA mutations in patients with de novo acute myeloid leukemia: a study from the Acute Leukemia French Association (ALFA). Blood 2002; 100: 2717–2723.

  6. 6

    van Waalwijk van Doorn-Khosrovani SB, Erpelinck C, Meijer J, van Oosterhoud S, van Putten WL, Valk PJ et al. Biallelic mutations in the CEBPA gene and low CEBPA expression levels as prognostic markers in intermediate-risk AML. Hematol J 2003; 4: 31–40.

  7. 7

    Frohling S, Schlenk RF, Stolze I, Bihlmayr J, Benner A, Kreitmeier S et al. CEBPA mutations in younger adults with acute myeloid leukemia and normal cytogenetics: prognostic relevance and analysis of cooperating mutations. J Clin Oncol 2004; 22: 624–633.

  8. 8

    Shih LY, Huang CF, Lin TL, Wu JH, Wang PN, Dunn P et al. Heterogeneous patterns of CEBPα mutation status in the progression of myelodysplastic syndrome and chronic myelomonocytic leukemia to acute myeloid leukemia. Clin Cancer Res 2005; 11: 1821–1826.

  9. 9

    Tiesmeier J, Czwalinna A, Muller-Tidow C, Krauter J, Serve H, Heil G et al. Evidence for allelic evolution of C/EBP alpha mutations in acute myeloid leukaemia. Br J Haematol 2003; 123: 413–419.

  10. 10

    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.

  11. 11

    Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR et al. Proposed revised criteria for the classification of acute myeloid leukemia. A report of the French-American-British Cooperative Group. Ann Intern Med 1985; 103: 620–625.

  12. 12

    Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR et al. Proposal for the recognition of minimally differentiated acute myeloid leukaemia (AML-MO). Br J Haematol 1991; 78: 325–329.

  13. 13

    Grimwade D, Walker H, Oliver F, Wheatley K, Harrison C, Harrison G et al. The importance of diagnostic cytogenetics on outcome in AML: analysis of 1612 patients entered into the MRC AML 10 trial. The Medical Research Council Adult and Children's Leukaemia Working Parties. Blood 1998; 92: 2322–2333.

  14. 14

    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.

  15. 15

    Liang DC, Shih LY, Huang CF, Hung IJ, Yang CP, Liu HC et al. CEBP alpha mutations in childhood acute myeloid leukemia. Leukemia 2005; 19: 410–414.

  16. 16

    Shih LY, Huang CF, Wang PN, Wu JH, Lin TL, Dunn P et al. Acquisition of FLT3 or N-ras mutations is frequently associated with progression of myelodysplastic syndrome to acute myeloid leukemia. Leukemia 2004; 18: 466–475.

  17. 17

    Fu JF, Hsu JJ, Tang TC, Shih LY . Identification of CBL, a proto-oncogene at 11q23.3, as a novel MLL fusion partner in a patient with de novo acute myeloid leukemia. Genes Chromosomes Cancer 2003; 37: 214–219.

  18. 18

    Leroy H, Roumier C, Huyghe P, Biggio V, Fenaux P, Preudhomme C . CEBPA point mutations in hematological malignancies. Leukemia 2005; 19: 329–334.

  19. 19

    Kelly LM, Gilliland DG . Genetics of myeloid leukemias. Annu Rev Genomics Hum Genet 2002; 3: 179–198.

  20. 20

    Mizuki M, Fenski R, Halfter H, Matsumura I, Schmidt R, Muller C et al. Flt3 mutations from patients with acute myeloid leukemia induce transformation of 32D cells mediated by the Ras and STAT5 pathways. Blood 2000; 96: 3907–3914.

  21. 21

    Beaupre DM, Kurzrock R . RAS and leukemia: from basic mechanisms to gene-directed therapy. J Clin Oncol 1999; 17: 1071–1079.

  22. 22

    Tenen DG . Disruption of differentiation in human cancer: AML shows the way. Nat Rev Cancer 2003; 3: 89–101.

  23. 23

    Snaddon J, Smith ML, Neat M, Cambal-Parrales M, Dixon-McIver A, Arch R et al. Mutations of CEBPA in acute myeloid leukemia FAB types M1 and M2. Genes Chromosomes Cancer 2003; 37: 72–78.

  24. 24

    Slovak ML, Kopecky KJ, Cassileth PA, Harrington DH, Theil KS, Mohamed A et al. Karyotypic analysis predicts outcome of preremission and postremission therapy in adult acute myeloid leukemia: a Southwest Oncology Group/Eastern Cooperative Oncology Group Study. Blood 2000; 96: 4075–4083.

  25. 25

    Schoch C, Schnittger S, Klaus M, Kern W, Hiddemann W, Haferlach T . AML with 11q23/MLL abnormalities as defined by the WHO classification: incidence, partner chromosomes, FAB subtype, age distribution, and prognostic impact in an unselected series of 1897 cytogenetically analyzed AML cases. Blood 2003; 102: 2395–2402.

Download references


We thank Ms Meng-Chu Chou, Ms Yu-Shu Shih, Ms Huei-Ying Li and Mr Ching-Tai Lee for their excellent technical assistance, Ms Hsin-I Wang for the statistical analysis, and Ms Yu-Feng Wang for her secretarial assistance. This work was supported by Grants NSC92-2314-B-182-026, NSC93-2314-B-182-001 and NSC93-2314-B-195-008 from the National Science Council, Taiwan, Grant MMH-E-94009 from Mackay Memorial Hospital, and Grant NHRI-EX94-9434SI from the National Health Research Institute, Taiwan.

Author information

Correspondence to L-Y Shih.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Shih, L., Liang, D., Huang, C. et al. AML patients with CEBPα mutations mostly retain identical mutant patterns but frequently change in allelic distribution at relapse: a comparative analysis on paired diagnosis and relapse samples. Leukemia 20, 604–609 (2006) doi:10.1038/sj.leu.2404124

Download citation


  • CEBPα mutations
  • allelic distribution
  • mutation patterns
  • cooperating mutations
  • relapsed AML

Further reading