Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Jun 5.
Published in final edited form as: Leukemia. 2012 Jun 22;27(1):238–241. doi: 10.1038/leu.2012.168

Mutations in the DNMT3A Exon 23 Independently Predict Poor Outcome in Older Patients with Acute Myeloid Leukemia: A SWOG report

Fabiana Ostronoff 1,2, Megan Othus 3, Phoenix A Ho 1,4, Matthew Kutny 5, Daniel E Geraghty 1, Stephen H Petersdorf 2, John E Godwin 6, Cheryl L Willman 7, Jerald P Radich 1,2, Frederick R Appelbaum 1,2, Derek L Stirewalt 1,2, Soheil Meshinchi 1,4
PMCID: PMC4457316  NIHMSID: NIHMS693932  PMID: 22722750

Acute myeloid leukemia (AML) represents a heterogeneous group of malignancies with great variability in clinical course and response to therapy. Currently, cytogenetics is the single most important prognostic factor in this disease. In recent years, an increasing number of molecular features with prognostic significance in AML have been identified [ref. 1-3].

DNMT3A encodes DNA cytosine-5-methyltransferase 3A, which is an enzyme that catalyzes the addition of a methyl group to the cytosine residue of CpG dinucleotides. By regulating the methylation of clusters of CpG, this enzyme mediates the down-regulation of downstream genes. The first evidence of somatic mutation of DNMT3A (DNMT3A+) gene in leukemia cells was provided by a Japanese study by large-scale array-based genomic re-sequencing of human leukemia [ref. 4]. Ley and colleagues then detected somatic mutations in this gene by massive parallel sequencing of an adult normal cytogenetic AML (CN-AML) genome. The investigators also studied the prevalence and prognostic significance of these mutations in AML patients. They found that DNMT3A+ were present in 62 out of the 282-screened patients (22%) and were independently associated with decreased survival in adult AML patients [ref. 5]. Subsequently, other studies confirmed the high prevalence and negative prognostic impact of DNMT3A+ in patients with AML [ref. 6-11]. As the previous studies included mostly young AML patients, currently there is a paucity of information regarding the prognostic value of these newly identified mutations in the older AML population. Thus, we studied the incidence and prognostic impact of these mutations in the context of other prognostic markers in a cohort of uniformly treated older patients with AML.

Pre-treatment samples were obtained from patients with AML enrolled in the SWOG clinical trials S-9031 and S-9333. All patients enrolled in these studies were >55 years old and had untreated AML. A total of 234 and 334 patients were registered for S-9031 and S-9333 trials, respectively. Details of the treatment protocols have been previously reported [ref. 12, 13]. In brief, in the S-9031 the patients were randomly assigned to a standard induction regimen (daunorubicin 45 mg/m2 per day for 3 days and Ara-C 200 mg/m2 per day for 7 days) plus either placebo or G-CSF (400 microg/m2 once daily). In the S-9333 the patients were randomized to receive mitoxantrone (10 mg/m2 per day for 5 days) and etoposide (100 mg/m2 per day for 5 days), or standard induction regimen. Patients with acute promyelocytic leukemia were excluded, as were SWOG-9333 patients randomized to induction chemotherapy with mitoxantrone and etoposide as this is not currently a standard therapy. At diagnosis, samples were analyzed in SWOG-approved laboratories for cytogenetic abnormalities using standard culturing and banding techniques. Cytogenetic abnormalities were grouped according to published criteria [ref. 14]. Written informed consent was obtained for all patients according to the Declaration of Helsinki, and the institutional review boards of member sites approved the studies.

As previous studies consistently showed that the majority of DNMT3A+ are located in exon 23, we amplified this exon via polymerase chain reaction and sequenced as previously described [ref. 15]. Other relevant genes were assessed for frequently occurring mutations as previously described (i.e., NPM1, FLT3 internal tandem duplication [FLT3-ITD], CEBPA and IDH1/2) [ref. 1-3,15]. Genomic DNA extracted from diagnostic specimens from 191 patients were available for evaluation. Eighty percent (153 out of 191) of the cases were de-novo AML and 20% (38 out of 191) were secondary AML. The source of the DNA was bone marrow in the case of 86.4% (165 out of 191) of the samples and peripheral blood in the case of 13.6% (26 out of 191).

Missense mutations of DNMT3A exon 23 were identified in 37 patients (19%). A total of 6 different mutations were identified, including R882H (n=20), R882C (n=13), R887I (n=1), L901R (n=1), P903L (n=1), and C911Y (n=1). Characteristics of patients with and without DNMT3A+ were compared (Table). There were no statistically significant differences in the median age, sex, white blood cell (WBC) count, blast percentage, hemoglobin, platelet count or French-American British (FAB) classification and ECOG performance scores at diagnosis between DNMT3A+ and DNMT3A wild-type (WT). There was no statistically significant difference in the incidence of DNMT3A+ among de novo and secondary AML (p=0.82).

Table. Comparison of Pretreatment Characteristics between Patients with AML and Mutated DNMT3A or Wild-Type DNMT3A.

Characteristic Mutated DNMT3A
N=37		 Wild-Type DNMT3A
N=154		 p-value
Median Age, years (range) 68 (57-81) 68 (56-89) 0.79
 Age ≥ 61 years 12 (32) 94 (61) 0.57
 Age < 61 years 25 (68) 60 (39)
Sex (%)
 Male 20 (54) 86 (56) 0.86
 Female 17 (46) 68 (44)
ECOG Performance Score
 0 7 (19) 34 (23) 0.47
 1 20 (54) 74 (49)
 2 8 (22) 23 (15)
 3 2 (5) 20 (13)
Cytogeneticξ
 Favorable 0 12 (11) <0.001
 Intermediate 24 (86) 54 (50)
 Unfavorable 1 (4) 32 (30)
 Unknown* 3 (11) 10 (9)
 No data 9 46
Normal cytogenetics (%) 21 (75) 45 (42) 0.003
BM blast (%), median 70 71 0.45
WBC ×109/L, median 37 28 0.46
Platelet count ×109/L, median 59 60 0.61
Hemoglobin (g/dL), median 9 9 0.18
FAB subtype (%)
 M0 2 (5) 3 (2) 0.10
 M1 9 (24) 35 (23)
 M2 6 (16) 55 (36)
 M3 0 1 (1)
 M4 10 (27) 38 (25)
 M5 7 (19) 13 (8)
 M6 0 2 (1)
 M7 0 5 (3)
FLT3-ITD present (%) 13 (35) 36 (24) 0.21
NPM1 mutated (%) 17 (65) 34 (26) <0.001
CEBPA mutated (%) 0 3 (2) 1
IDH1 mutated (%) 0 5 (3) 0.57
IDH2 mutated (%) 6 (16) 34 (22)
Molecular risk group ¥
 Low risk 12 (46) 18 (14) <0.001
 High risk 14 (54) 110 (86)
Open in a new tab

Abbreviations: AML, acute myeloid leukemia; ECOG, performance status of Eastern Cooperative Oncology Group; FAB, French-American British classification of acute myeloid leukemia; BM, bone marrow; WBC, white blood count.

¥

The high-risk molecular group is defined by the presence of FLT3-ITD Regardless of NPM1 status or the absence of FLT3-ITD and NPM1 wild-type. The low risk group is defined by the absence of FLT3-ITD and NMP1 mutated or CEPBA mutated.

*

Cytogenetic abnormalities of unknown prognostic significance.

ξ

The p-value is from a Fisher's exact test of the cytogenetics categories (favorable, intermediate, unfavorable, unknown) versus DNMT3A status (DNMT3A mutated versus wild-type). Patients with missing data were not included in the test.

NPM1 mutations (NPM1+) were more common among DNMT3A+ than in DNMT3A-WT patients (65% vs 26%, p<0.0001). There were no significant differences in the incidence of FLT3-ITD, CEBPA or IDH1/2 mutations among patients with DNMT3A+ as compared to DNMT3A-WT. Patients were then subdivided into high (presence of FLT3-ITD regardless of NPM1 status or the absence of FLT3-ITD and NPM1 wild-type) and low (absence of FLT3-ITD and NMP1 mutated or CEPBA mutated) molecular risk groups [ref. 16]. A greater proportion of DNMT3A+ patient were low-risk compared to DNMT3A-WT patients (46% vs 14%), while, comparably, a smaller proportion of DNMT3A+ patient were high-risk compared to DNMT3A-WT patients (54% vs 86%, p<0.001).

Clinical outcome data were examined for all patients with mutated and wild-type DNMT3A. The median follow-up time for patients who remained alive was 8.3 years. Patients with mutated or wild-type DNMT3A had similar CR rates (OR=1, 95% CI, 0.5, 2.1, p=0.98). The OS at 2 years from study entry for DNMT3A+ was 11% vs 21% for those with DNMT3A-WT (HR=1.42, 95% CI, 0.98-2.06; p=0.065, Figure 1A). For those with mutated or wild-type DNMT3A, 2-year EFS and RFS were 3% vs 16% (HR= 1.46, 95% CI, 1.01-2.1; p=0.043) and 8% vs 26% (HR=1.4, 95% CI, 0.81-2.44; p=0.23), respectively (Figure 1B and 1C). In multivariate Cox proportional hazards regression analyses after adjusting for other known prognostic factors such as age, ECOG performance status, WBC, blast count, cytogenetic risk groups, FLT3-ITD and NPM1 mutation status, the presence of DNMT3A+ was independently associated with worse OS (HR=2.69, 95% CI, 1.46-4.94; p=0.0015), EFS (HR=3.13, 95% CI, 1.7-5.75; p<0.001) and RFS (HR=2.6, 95% CI, 1.07-6.35; p=0.036).

Figure 1.

Figure 1

Open in a new tab

Prognostic impact of DNMT3A mutations in all investigated patients with acute myeloid leukemia (AML). (A) Overall survival in patients with AML having wild-type (WT) or mutated DNMT3A (DNMT3A+). (B) Event-free survival in patients with AML having WT or DNMT3A+. (C) Relapse-free survival in patients with AML having WT or DNMT3A+.

The prognostic impact of the DNMT3A mutations was evaluated in specific cytogenetic risk groups. CN-AML was more common among DNMT3A+ patients as compared to DNMT3A-WT patients (75% vs. 42%, p=0.003). DNMT3A+ were not observed in patients with core binding factor (CBF) AML and were rare in those with high-risk cytogenetics. In patients with CN-AML, DNMT3A+ was associated with worse OS (2-yr 10% vs 31%, HR=2.11, 95% CI, 1.2-3.7; p=0.0078), EFS (2-yr 0% vs 27%, HR=2.45, 95% CI, 1.39-4.31; p=0.0014) and RFS (2-yr 8% vs 39%, HR=2.82, 95% CI, 1.28-6.22; p=0.0073) at 2 years as compared to patients with DNMT3A-WT (Figure 2). In multivariate analysis among CN-AML, DNMT3A+ patients had worse OS (HR=2.51, 95% CI, 2.12-5.18; p=0.013), EFS (HR=3.76, 95% CI, 1.74-8.09; p<0.001) and RFS (HR=3.63, 95% CI, 1.25-10.56; p=0.018) as compared to those with DNMT3A-WT.

Figure 2.

Figure 2

Open in a new tab

Prognostic impact of DNMT3A mutations in patients with normal cytogenetics acute myeloid leukemia (CN-AML). (A) Overall survival in patients with CN-AML having wild-type (WT) or mutated DNMT3A (DNMT3A+). (B) Event-free survival in patients with CN-AML having WT or DNMT3A+. (C) Relapse-free survival in patients with CN-AML having WT or DNMT3A+.

The prognostic impact of the DNMT3A mutations was tested in the high and low molecular risk groups [ref. 16]. Among the high-risk subgroup, OS was significantly worse in the DNMT3A+ patients (2-yr 0% vs 22%, HR=2.02, 95% CI=1.13-3.61, p=0.02) as were the EFS (2-yr 0% vs 16%, HR=1.85, 95% CI= 1.04-3.28, p=0.033) and RFS (2-yr 0 vs 25%, HR=2.78, 95% CI=1.19-6.47; p=0.02). In the low-risk subgroup, there was no statistically significant difference between DNMT3A+ and wild-type in the OS (25% vs 30%, HR=1.54, 95% CI=0.69-3.45, p=0.29), EFS (8% vs 30%, HR=1.49, CI=0.71-3.13, p=0.29) or RFS (50% vs 31%, HR=0.38, CI=0.08-1.78, p=0.22) [Figure 3]. In this study we demonstrate that in older adults with AML, mutations in the exon 23 of DNMT3A gene are common, and their presence is highly associated with normal cytogenetics and with adverse clinical outcome. We further demonstrate that within the high-risk molecular cohort presence of DNMT3A+ provides further prognostic demarcation; where in the molecular high-risk cohort, those with DNMT3A+ have a dismal outcome, whereas outcome for those without DNMT3A+ is similar to those without high-risk molecular genotype. These findings are consistent with observations made in younger AML patients, highlighting the fact that DNMT3A mutations can be used to refine risk stratification in high-risk cohort of patients.

Figure 3.

Figure 3

Open in a new tab

Prognostic impact of DNMT3A mutations in patients with acute myeloid leukemia (AML) having wild-type (WT) or mutated DNMT3A (DNMT3A+) and low vs high-risk molecular risk groups. (A) Overall survival in patients with low and high-risk and WT vs DNMT3A+. (B) Event-free survival in patients with low and high-risk and WT vs DNMT3A+. (C) Relapse-free survival in patients with low and high-risk and WT vs DNMT3A+.

Note: The high-risk molecular group is defined by the presence of FLT3-ITD regardless of NPM1 status or the absence of FLT3-ITD and NPM1 wild-type. The low risk group is defined by the absence of FLT3-ITD and NMP1 mutated or CEPBA mutated.

Other investigators recently reported on the negative prognostic impact of DNMT3A+ in elderly AML patients [ref. 17 and 18]. Similar to ours and other studies [ref. 5-11], Marcucci et al. also found that mutation affecting arginine codon 882 (R882) were more common than those affecting other codons (non-R882). Interestingly, the investigators also observed that the prognostic significance of DNMT3A+ depended on the type of mutation. In younger patients, only non–R882-DNMT3A+ were associated with worse clinical outcome, whereas R882-DNMT3A+ had no prognostic significance. Conversely, in older patients, only R882-DNMT3A+, were independently associated with worse outcome. The reasons why the prognostic significance of different DNMT3A mutation types varies in younger and older patients are currently unknown [ref. 17].

Previous studies have shown that the majority of mutations in DNMT3A are located in the functional methyltransferase domain (exon 23) [ref. 5-7]. With currently available clinical assays, screening of entire DNMT3A is not feasible. It is important to note that by limiting mutation analysis to the functional “hot spot” of this gene, although a practical approach for the purpose of risk stratification, and to aid in clinical decision-making, approximately 15% to 20% of the mutations located outside the methyltransferase domain are missed.

As shown in younger adult AML patients, we also found in this cohort of older AML patients that DNMT3A+ can identify risk subgroups in the high-risk cohort, whereas in the low-risk group DNMT3+ have no prognostic effect [ref. 7]. Although this has been consistently demonstrated in previous studies the reason for these findings is unclear. It is conceivable that cooperating genetic and epigenetic alterations may lead to a distinct leukemic phenotype and enhanced drug resistance. Moreover, the higher prevalence of DNMT3A+ in CN-AML, the clustering of certain mutations (NPM1, FLT3-ITD and IDH1/2) and the strong selection against CBF abnormalities are likely not random and it would be important to determine if mutations in these genes have overlapping or cooperating functional consequences.

Our findings must be interpreted cautiously as they are based on a relatively small patient cohort and our mutation analysis was limited to the functional “hot spot” of DNMT3A. Nonetheless, the high prevalence of DNMT3A+ in exon 23 and its independent negative impact on prognosis identify DNMT3A as a potentially important molecular marker for older patients with AML, especially those with CN-AML. Within the high-risk molecular group, DNMT3A+ identifies a subset of patients with especially poor prognosis. These findings are likely to have a major impact on the clinical management of AML as these genetic alterations may not only represent a prognostic marker, but also may constitute targets for therapeutic intervention, which are especially needed in older AML patients. Mutation profiling of DNMT3A exon 23 can easily be incorporated into clinical practice to aid in risk based therapy allocation. In addition, this high-risk cohort may provide a target population for assessment of novel epigenetic modifying agents.

Acknowledgments

The authors thank the patients and families who consented to the use of biologic specimens in these trials and the AML Reference Laboratories of the COG and SWOG for providing diagnostic specimens. This investigation was supported in part by the following PHS Cooperative Agreement grant numbers awarded by the National Cancer Institute, Cancer Therapy Evaluation Program, DHHS: CA32102, CA38926, CA20319, CA12213.

Footnotes

Contribution: F.O. designed research, performed research, analyzed data, and wrote the manuscript; M.O. performed statistical analyses, and edited the manuscript; P.A.H., D.E.G., S.H.P., J.E.G., C.L.W., J.P.R., F.R.A. performed research and edited the manuscript; S.M. designed research, analyzed data, and edited the manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

References

  • 1.Thiede C, Steudel C, Mohr B, Schaich M, Schäkel U, Platzbecker U, et al. Analysis of FLT3-activating mutations in 979 patients with acute myelogenous leukemia: association with FAB subtypes and identification of subgroups with poor prognosis. Blood. 2002;99:4326–35. doi: 10.1182/blood.v99.12.4326. [DOI] [PubMed] [Google Scholar]
  • 2.Falini B, Mecucci C, Tiacci E, Alcalay M, Rosati R, Pasqualucci L, et al. GIMEMA Acute Leukemia Working Party. Cytoplasmic nucleophosmin in acute myelogenous leukemia with a normal karyotype. N Engl J Med. 2005;352:254–66. doi: 10.1056/NEJMoa041974. [DOI] [PubMed] [Google Scholar]
  • 3.Ho PA, Alonzo TA, Gerbing RB, Pollard J, Stirewalt DL, Hurwitz C, et al. Prevalence and prognostic implications of CEBPA mutations in pediatric acute myeloid leukemia (AML): a report from the Children's Oncology Group. Blood. 2009;113:6558–66. doi: 10.1182/blood-2008-10-184747. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Yamashita Y, Yuan J, Suetake I, Suzuki H, Ishikawa Y, Choi YL, et al. Array-based genomic resequencing of human leukemia. Oncogene. 2010;29:3723–31. doi: 10.1038/onc.2010.117. [DOI] [PubMed] [Google Scholar]
  • 5.Ley TJ, Ding L, Walter MJ, McLellan MD, Lamprecht T, Larson DE, et al. DNMT3A-Mut in acute myeloid leukemia. N Engl J Med. 2010;363:2424–33. doi: 10.1056/NEJMoa1005143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Yan XJ, Xu J, Gu ZH, Pan CM, Lu G, Shen Y, et al. Exome sequencing identifies somatic mutations of DNA methyltransferase gene DNMT3A in acute monocytic leukemia. Nat Genet. 2011;43:309–15. doi: 10.1038/ng.788. [DOI] [PubMed] [Google Scholar]
  • 7.Thol F, Damm F, Lüdeking A, Winschel C, Wagner K, Morgan M, et al. Incidence and Prognostic Influence of DNMT3A Mutations in Acute Myeloid Leukemia. J Clin Oncol. 2011;29:2889–96. doi: 10.1200/JCO.2011.35.4894. [DOI] [PubMed] [Google Scholar]
  • 8.Shen Y, Zhu YM, Fan X, Shi JY, Wang QR, Yan XJ, et al. Gene mutation patterns and their prognostic impact in a cohort of 1185 patients with acute myeloid leukemia. Blood. 2011;118:5593–603. doi: 10.1182/blood-2011-03-343988. [DOI] [PubMed] [Google Scholar]
  • 9.Marková J, Michková P, Burčková K, Březinová J, Michalová K, Dohnalová A, et al. Prognostic impact of DNMT3A mutations in patients with intermediate cytogenetic risk profile acute myeloid leukemia. Eur J Haematol. 2012;88:128–35. doi: 10.1111/j.1600-0609.2011.01716.x. [DOI] [PubMed] [Google Scholar]
  • 10.Lin J, Yao DM, Qian J, Chen Q, Qian W, Li Y, et al. Recurrent DNMT3A R882 mutations in Chinese patients with acute myeloid leukemia and myelodysplastic syndrome. PLoS One. 2011;6:e26906. doi: 10.1371/journal.pone.0026906. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Hou HA, Kuo YY, Liu CY, Chou WC, Lee MC, Chen CY, et al. DNMT3A mutations in acute myeloid leukemia: stability during disease evolution and clinical implications. Blood. 2012;119:559–68. doi: 10.1182/blood-2011-07-369934. [DOI] [PubMed] [Google Scholar]
  • 12.Anderson JE, Kopecky KJ, Willman CL, Head D, O'Donnell MR, Luthardt FW, et al. Outcome after induction chemotherapy for older patients with acute myeloid leukemia is not improved with mitoxantrone and etoposide compared to cytarabine and daunorubicin: a Southwest Oncology Group study. Blood. 2002;100:3869–76. doi: 10.1182/blood-2001-12-0354. [DOI] [PubMed] [Google Scholar]
  • 13.Godwin JE, Kopecky KJ, Head DR, Willman CL, Leith CP, Hynes HE, et al. A double-blind placebo-controlled trial of granulocyte colony-stimulating factor in elderly patients with previously untreated acute myeloid leukemia: a Southwest oncology group study (9031) Blood. 1998;91:3607–15. [PubMed] [Google Scholar]
  • 14.Slovak ML, Kopecky KJ, Cassileth PA, Harrington DH, Theil KS, Mohamad 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. [PubMed] [Google Scholar]
  • 15.Ho PA, Kutny MA, Alonzo TA, Gerbing RB, Joaquin J, Raimondi SC, Gamis AS, Meshinchi S. Leukemic mutations in the methylation-associated genes DNMT3A and IDH2 are rare events in pediatric AML: a report from the Children's Oncology Group. Pediatr Blood Cancer. 2011;57:204–9. doi: 10.1002/pbc.23179. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Döhner H, Estey EH, Amadori S, Appelbaum FR, Büchner T, Burnett AK, et al. European LeukemiaNet. Diagnosis and management of acute myeloid leukemia in adults: recommendations from an international expert panel, on behalf of the European LeukemiaNet. Blood. 2010;115:453–74. doi: 10.1182/blood-2009-07-235358. [DOI] [PubMed] [Google Scholar]
  • 17.Marcucci G, Metzeler KH, Schwind S, Becker H, Maharry K, et al. Age-related prognostic impact of different types of DNMT3A mutations in adults with primary cytogenetically normal acute myeloid leukemia. J Clin Oncol. 2012;30:742–50. doi: 10.1200/JCO.2011.39.2092. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Metzeler KH, Walker A, Geyer S, Garzon R, Klisovic RB, et al. DNMT3A mutations and response to the hypomethylating agent decitabine in acute myeloid leukemia. Leukemia. 2012;26:1106–1107. doi: 10.1038/leu.2011.342. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES