Mutations in the epigenetic modifying gene DNA methyltransferase 3A (DNMT3A) are found in 14–35% of acute myeloid leukemia (AML) patients [1,2,3,4,5,6]. The majority of DNMT3A mutations cluster at codon R882 in exon 23 [1,2,3,4,5,6]. Most of the studies that have investigated the prognostic impact of these alterations reported inferior outcomes associated with the presence of DNMT3A mutations—particularly in codon R882 [1,2,3,4, 6,7,8,9]. However, the negative impact on outcomes may be dependent on the clinical and biological context, such as the presence of concurrent gene mutations (i.e. NPM1 mutations and/or presence of FLT3-ITD) and the DNMT3A mutation type (i.e. R882 vs. non-R882) [1,2,3,4,5,6, 9]. In older AML patients, DNMT3A mutations are more prevalent  and have been associated with poor outcomes [1, 6, 8].
DNMT3A mutations are often found to persist during remission (possibly due to a preexisting clonal hematopoiesis) and at AML relapse, and might contribute to chemotherapy resistance . This raises the issue of allogeneic hematopoietic stem cell transplantation (HSCT) as a therapeutic approach specifically for DNMT3A mutated AML. Studies that included HSCT treated patients—most of which focused on younger AML patients—observed no influence of an HSCT on the negative prognostic impact of DNMT3A mutations [3,4,5, 9, 11]. For example, Gaidzik et al. . described a shorter relapse-free survival in younger DNMT3A R882 mutated AML patients independent of the use of HSCT. Similarly, Ahn et al. . and Xu et al. . reported a negative impact of DNMT3A R882 mutations after HSCT in mainly younger patients with normal karyotype (NK) AML.
Since DNMT3A R882 mutations occur more frequently with age  and data on the influence of DNMT3A R882 mutations on outcome in older AML patients receiving HSCT are limited, we analyzed the prognostic impact of DNMT3A R882 mutations at diagnosis among mainly older (>60 years) AML patients who underwent non-myeloablative (NMA)-HSCT.
NMA-HSCT relies on an immunological donor leukocyte-mediated GvL effect and, because of the lower treatment-related toxicity, represents a feasible consolidating therapy for older AML patients and/or those with comorbidities [12, 13].
The patients studied here were treated with cytarabine-based chemotherapies followed by conditioning with fludarabine and 2 Gy TBI or 2 Gy TBI only before NMA-HSCT (for further details, see Supplementary Information) at the University Hospital Leipzig between February 2003 and December 2012. We retrospectively analyzed the presence of DNMT3A R882 mutations in pre-treatment bone marrow (n = 98) or peripheral blood (n = 10) samples of 108 AML patients (for further details, see Supplementary Information). All patients gave written informed consent in accordance with the declaration of Helsinki. Pre-treatment cytogenetics and the mutation status of NPM1, CEBPA, and the presence of FLT3-ITD were determined as previously described . Patients were grouped into four genetic risk groups according to the 2010 European LeukemiaNet (ELN) standardized reporting system [14, 15]. Associations of DNMT3A R882 mutations at diagnosis with baseline clinical and molecular features, as well as leukemia-free survival (LFS) and overall survival (OS), were calculated as described in the Supplementary Information.
Of the 108 AML patients analyzed, 18 (16.6%) harbored a DNMT3A mutation in exon 23 of which 17 affected codon R882. Amino acid changes of the DNMT3A R882 mutations were arginine to histidine (R > H; n = 11, 64.7%) and arginine to cysteine (R > C; n = 6, 35.3%). One patient had a missense point mutation at codon 904 that resulted in a proline to leucine (P > L) amino acid change. This patient was excluded from further analysis and associations with clinical and molecular parameters were calculated for the remaining patients (n = 107; Table 1). Our set of patients receiving NMA-HSCT was mainly composed of older (>60 years) AML patients (n = 78; 72.2%) with a median age at diagnosis of 63.9 (32.4–74.8) years (Table 1). DNMT3A R882 mutated patients were older by trend (P = .07) and significantly more likely to be NPM1 mutated (P = .004) at diagnosis. No further associations with other determined clinical, molecular, or cytogenetic parameters were observed (Table 1). When we restricted our analysis to NK-AML patients (n = 51), DNMT3A R882 mutations were present in 10 patients (19.6%) and significantly associated with the presence of NPM1 mutations (P = .02; Supplementary Information Table S1).
For outcome analysis we considered only those AML patients transplanted in CR (n = 75) or complete remission with incomplete peripheral recovery (CRi; n = 23; Supplementary Information Table S2). In this set of AML patients, the LFS and OS rates at 3 years were 39.4% and 46.3%, respectively. The presence of a DNMT3A R882 mutation at diagnosis did not associate with LFS (P = .37) or OS (P = .63) and the survival curves were superimposable (Fig. 1). Similarly, no impact of the DNMT3A R882 mutations status on outcome was observed when we restricted our analysis to NK-AML patients (Supplementary Information Figure S1) or older (>60 years) AML patients (Supplementary Information Figure S2 and Table S3). The median follow-up for patients alive after NMA-HSCT was 2.65 years.
The observed incidence of DNMT3A R882 mutations in our cohort, as well as the association with NPM1 mutations was in line with previous publications [1,2,3,4,5,6,7,8,9]. In concordance with others, we observed no associations of DNMT3A R882 mutations with CEPBA mutations (P = .70) [4, 5], nor did DNMT3A R882 mutations associate with the presence of FLT3-ITD, which was consistent with previous studies that focused on older AML patients [6, 8].
With the caveat of a limited sample size the results that DNMT3A R882 mutations do not significantly associate with outcome in our group of mainly older AML patients treated with NMA-HSCT diverges from previous findings [3, 5, 9, 11]. For instance, Ahn et al.  found a shorter OS with DNMT3A R882 mutation in 115 mainly younger NK-AML patients receiving HSCT after myeloablative and reduced-intensity conditioning. However, the negative impact of DNMT3A R882 mutations in this study was most evident in the presence of a FLT3-ITD, a mutation pattern that might represent a more aggressive leukemia phenotype . In our set, only four patients had both a DNMT3A R882 mutation and a FLT3-ITD mutation preventing further analysis of this subgroup. Xu et al.  also found a shorter OS and disease-free survival for DNMT3A-mutated patients in 308 NK-AML patients, of which 214 received HSCT. Interestingly in this study, when outcome was compared between DNMT3A-mutated patients (n = 55) receiving chemotherapy or HSCT as post-remission therapy, OS was significantly improved in the HSCT-treated group. A similar observation had been made earlier in a smaller cohort by Ley et al. , suggesting that DNMT3A-mutated patients might derive particular benefit from HSCT. Furthermore, Thol et al.  have recently reported that patients harboring DNMT3A mutations in T cells—which may indicate the presence of clonal hematopoiesis of indeterminate potential (CHIP) before the onset of leukemia—might also perform better with HSCT. Future studies should therefore characterize the influence of NMA-HSCT on treatment outcome in DNMT3A-mutated patients, since HSCT may be a promising treatment approach in these cases.
In our analysis, no impact on outcome was observed according to DNMT3A R882 mutation status in patients receiving NMA-HSCT. However, a major limitation of this study is the relatively small sample size. Thus results should be interpreted cautiously and within the specific context of the applied NMA-HSCT treatment regimen. However, to the best of our knowledge, our data represents the so far largest cohort analyzing DNMT3A R882 mutation status in mainly older, homogenously treated AML patients consolidated with NMA-HSCT. Some data in the literature suggests a survival benefit of DNMT3A-mutated patients receiving HSCT compared to chemotherapy-based consolidation [1, 11]. Older AML patients harboring a DNMT3A R882 mutation might derive particular benefit from NMA-HSCT. Ongoing randomized clinical trials testing chemotherapy-based versus HSCT-based consolidation in AML patients should help to address this important clinical question.
Presented in part at the 41st Annual Meeting of the European Society for Blood and Marrow Transplantation, Istanbul, Turkey, March 23rd, 2015 and published in abstract form.
Ley TJ, Ding L, Walter MJ, McLellan MD, Lamprecht T, Larson DE, et al. DNMT3A mutations in acute myeloid leukemia. N Engl J Med. 2010;363:2424–33.
Gale R, Lamb K, Allen C, El-Sharkarwi D, Stowe C, Jenkinson S, et al. Simpson’s paradox and the impact of different DNMT3A mutations on outcome in younger adults with acute myeloid leukemia. J Clin Oncol. 2015;33:2072–83.
Thol F, Damm F, Lüdeking A, Winschel C, Wagner K, Morgan M, et al. Incidence and prognostic influence of DNMT3Amutations in acute myeloid leukemia. J Clin Oncol. 2011;29:2889–96.
Ribeiro AF, Pratcorona M, Erpelinck-Verschueren C, Rockova V, Sanders M, Abbas S, et al. Mutant DNMT3A: a marker of poor prognosis in acute myeloid leukemia. Blood. 2012;119:5824–31.
Gaidzik VI, Schlenk RF, Paschka P, Stölzle A, Späth D, Kuendgen A, et al. Clinical impact of DNMT3A mutations in younger adult patients with acute myeloid leukemia: results of the AML Study Group (AMLSG). Blood. 2013;121:4769–77.
Marcucci G, Metzeler KH, Schwind S, Becker H, Maharry K, Mrózek 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. 2010;30:742–50.
Tsai CH, Hou HA, Tang JL, Liu CY, Lin CC, Chou WC, et al. Genetic alterations and their clinical implications in older patients with acute myeloid leukemia. Leukemia. 2016;30:1485–92.
Ostronoff F, Othus M, Ho PA, Kutny M, Geraghty DE, Petersdorf SH, et al. Mutations in the DNMT3A exon 23 independently predict poor outcome in older patients with acute myeloid leukemia: a SWOG report. Leukemia. 2013;27:238–41.
Ahn JS, Kim HJ, Kim YK, Lee SS, Jung SH, Yang DH, et al. DNMT3A R882 mutation with FLT3-ITD positivity is an extremely poor prognostic factor in patients with normal-karyotype acute myeloid leukemia after allogeneic hematopoietic cell transplantation. Biol Blood Marrow Transplant. 2016;22:61–70.
Thol F, Klesse S, Köhler L, Gabdoulline R, Kloos A, Liebich A, et al. Acute myeloid leukemia derived from lympho-myeloid clonal hematopoiesis. Leukemia. 2017. https://doi.org/10.1038/leu.2016.345
Xu Y, Sun Y, Shen H, Ding L, Yang Z, Qiu H, et al. Allogeneic hematopoietic stem cell transplantation could improve survival of cytogenetically normal adult acute myeloid leukemia patients with DNMT3A mutations. Am J Hematol. 2015;90:992–7.
McSweeney PA, Niederwieser D, Shizuru JA, Sandmaier BM, Molina AJ, Maloney DG, et al. Hematopoietic cell transplantation in older patients with hematologic malignancies: replacing high-dose cytotoxic therapy with graft-versus-tumor effects. Blood. 2001;97:3390–3400.
Hegenbart U, Niederwieser D, Sandmaier BM, Maris MB, Shizuru JA, Greinix H, et al. Treatment for acute myelogenous leukemia by low-dose, total-body, irradiation-based conditioning and hematopoietic cell transplantation from related and unrelated donors. J Clin Oncol. 2006;24:444–53.
Bill M, Jentzsch M, Grimm J, Schubert K, Lange T, Cross M, et al. Prognostic impact of the European LeukemiaNet standardized reporting system in older AML patients receiving stem cell transplantation after non-myeloablative conditioning. Bone Marrow Transplant. 2017. https://doi.org/10.1038/bmt.2017.42
Döhner H, Estey EH, Amadori S, Appelbaum FR, Büchner T, Burnett AK, et al. Diagnosis and management of acute myeloid leukemia in adults: recommendations from an international expert panel, on behalf of the European Leukemia-Net. Blood. 2010;115:453–74.
L.B. was supported by the foundation “Zusammen gegen den Krebs”. M.B. was supported by an intramural scholarship of the medical faculty of the University of Leipzig (#990101-089). The authors would like to thank Sabine Leiblein, Janet Bogardt, Annette Jilo, Dagmar Cron, Christel Müller, Evelin Hennig, Daniela Bretschneider, Martina Pleß, Ulrike Bergmann, Ines Kovacs, Scarlett Schwabe, Kathrin Wildenberger, and Christine Günther for assisting with and collecting patient data and material.
L.K.S. and L.B. designed and performed experiments. M.B. and K.S. performed experiments. L.K.S., L.B., S.S., J.G., and M.J. analyzed and interpreted data. M.C., V.V., T.L., W.P., G.N.F., and G.B. provided administrative and technical support. L.K.S., L.B., and S.S. wrote and all authors reviewed and approved the manuscript. S.S. and D.N. supervised the study.