International Journal of Medical Laboratory

Acute leukemia includes acute lymphoblastic (ALL), and myeloid leukemia (AML) are heterogeneous malignancies in which undeveloped and dysfunctional hematopoietic progenitors grow and accumulate in the bone marrow [1]. Acute leukemia accounts for <3% of all cancers [2], and an estimated 18700 novel cases of acute leukemia are forecasted for 2008 in the United States alone [3]. An accounted for 6000 novel cases of acute lymphoblastic leukemia are recognized annually in the USA [4]. Almost three-fourths of childhood leukemia ages are ALL. AML accounts for approximately 25% of all leukemias in adults in the Western world [5]. AML is the most common form of acute leukemia in adults [6]. Globally speaking, the highest prevalence of AML is in the U.S., Australia, and western Europe [7].

The methylation of DNA is an epigenetic modification that plays a considerable role in controlling gene expression, cellular disinclination, X-inactivation, imprinting, silencing of transposable elements, and gene regulation [8]. DNA methyltransferase catalyzes the transfer of a methyl group from S-adenosyl-methionine at CpG sites in the genome. In various types of malignancies, abnormal methylation of CpG islands in the promoter region has been reported, resulting in their expression’s silencing [9]. Mutations in DNMT3A were first reported by Ley et al. [10]. Mutations in DNMT3A gene has been described in different hematologic malignancies, including AML in 14-34% of cases [11], 5-15% of myelodysplastic syndromes cases [12], 10% of chronic myelomonocytic leukemia (CMML) patients [13], 5.7% of primary myelofibrosis cases [14], 12% of cases with systemic mastocytosis [15], and about 18% of T cell acute lymphoblastic leukemia patients [16]. In previous studies reported the poor prognostic impact of DNMT3A mutations [17], and DNMT3A mutations were associated with reduced overall survival, increased risk of relapse [11], and more frequent in patients with normal cytogenetics in AML patients and poor outcome [10]. DNMT3A mutations are highly correlated with mutations in the NPM1 gene, FLT3 ITD and TKD gene, IDH1/2 gene and older age, and FAB M5 type [11]. Several studies reported that DNMT3A mutations are stable during the disease. So, those abnormalities could be a potential marker for minimal residual disease [18]. This study aimed to examine the distribution of DNMT3A mutations in Iranian de novo acute leukemia patients for better risk categorization and targeted therapy in the Iranian population.

Materials and Methods

Patients

Forty five acute leukemia patients were selected, including 23 cases of ALL and 22 cases of AML admitted to Ghaem Hospital of Mashhad University of Medical Sciences, Iran, May 2017 to February 2018. Sample size calculation was done based on a meta-analysis of DNMT3A mutation in AML patients [11]. According to the DNMT3A mutation prevalence of 34.2%, type I error of 0.05, and 85% of precision using the required sample size formula was 40. Mutation bone marrow and peripheral blood cells were gained from these patients at diagnosis. Laboratory data including French–American–British (FAB) classifications, bone marrow blast percentage, complete blood count, hemoglobin rate, and gene mutations were collected.

Polymerase chain reaction (PCR) amplification and direct sequencing

Mononuclear cells from bone marrow or peripheral blood at diagnosis were accumulated in ethylenediaminetetraacetic acid (EDTA)-containing tubes. Genomic DNA was extracted by DNA extraction kit (FABGK001) and stored at -70˚C in a refrigerator. The PCR amplification of exon 23 in the DNMT3A gene was done using forward primer 5’-TCCCAGTCCACTATACT GACGTCTC-3’and reverse primer 5'-TCTCTCCATCCTC ATGTTCTTGG-3’ [19]. PCR was carried out in a 50 µL mixture containing 2×PCR mixture (A290401) 25 µL, primers 10 pmol for each, and genomic DNA 100 ng. PCR cycles inclusive pre-denaturation at 95˚C for 5 min, denaturation at 95˚C for 30 s, annealing at 65˚C for 35s, extension at 72˚C for 30 s, final extension at 72˚C for 5 min and the product was kept at 4˚C. Direct sequencing was performed with forward and reverse primers after PCR amplification. They were sequenced on a Genetic Analyzer (3130xl, Applied Biosystems, USA). Gene Codes Sequencher v5.4.6 analyzed the sequencing results. The Ethical Committee approved this study at the Mashhad University of Medical Sciences.

Statistical analysis

In this study, the relationship between gene mutations and other variables was investigated. Fisher’s exact test was used to compare the differences between categorical variables. Mann-Whitney was performed to compare the differences between continuous variables. All data analyses were performed using SPSS version 16.0, and p<0.05 was considered statistically significant.

Results

In this study, a total of 45 patients were studied, of which 23 (51%) had ALL and 22 (49%) patients with AML. The mean age of all patients was 17.8 ± 15.2 years. The age also ranged from 0.09 to 20 and from 0.5 to 65 years for ALL and AML, respectively. Besides, among ALL patients, the frequency of males and among AML patients, females’ frequency was more. The PCR product is shown in Fig. 1.

The most and least frequent form of leukemia in patients with AML was M1 and M4, respectively. In patients with AML, NPM1 mutation has been studied in 15 patients, which was positive in 7 (46.7%) cases. CEBPA mutation has been evaluated in 12 patients, which was positive in 3 (25.2%) cases. Furthermore, FLT3ITD and FLT3KTD were also studied in 14 patients and were positive in 2 (14.3%) cases. In ALL patients, the blast percentage was 86.7±17.3 percent (ranged from 30 to 100 %), and in AML patients, on average, it was 11.6±83.1% (the range of variations was from 50 to 95%). 22.7% and 21.7% of patients with AML and ALL had polymorphism rs368516543, respectively (Fig. 2A). Among all 45 patients, DNMT3A mutation in exon 23 was found in 3 cases (6.6%) (Table 1). All three mutants were R882H mutation (c.2645G>A) (Fig. 2B).

 Overall, the frequency of mutation was not statistically different between ALL and AML (p=0.608). P.R882H mutation was identified in 2 (9.1%) of 22 AML patients (Table 2). The basic clinical data of these patients were presented in Table 1. No relationship was observed between p.R882H mutation and gender. The age of AML patients at diagnosis with DNMT3A p.R882H mutation was significantly more compared to those without mutations (Table 2). Even though the mean of white blood cell, red blood cell, hemoglobin, or platelet counts at diagnosis was higher in AML patients with p.R882H mutation than those without mutations, no statistical difference was seen. 1 (4.3%) heterozygous DNMT3A p.R882H mutation was identified in a girl aged 6 years old with ALL-L2. The difference in age and hematologic parameters was not seen confirmed between patients with and without mutations. However, this patient was recognized with normal karyotype and negative rs368516543 polymorphism (Table 3).

Table 1. The clinical and hematopoietic parameters of 3 patients with DNM3A mutations

WBC= White blood cell; RBC= Red blood cell

Table 2. Distribution of DNM3A R882H mutation in AML

^† Based on Mann -Whitney test; ^‡ Based on Fisher’s exact test.

WBC= White blood cell; RBC= Red blood cell

Table 3. Distribution of DNM3A p.R882H mutation in ALL

^† Based on the Mann -Whitney test;  ^‡ Based on Fisher’s exact test.

WBC= White blood cell; RBC= Red blood cell

Discussion

Gene mutations and epigenetic changes play an essential role in the pathogenesis and prognosis of leukemia. DNMT3A is a crucial enzyme in the de novo DNA methylation at CpG sites, and DNMT3A mutations have been reported in a diversity of malignancies and are one of the significant molecular aberrations associated with AML [11]. In the present study, a group of 45 de novo acute leukemia patients was investigated with a Direct Sequencing approach of 23 exons in the DNMT3A gene to identify the frequency of DNMT3A mutations as well as their associations with other molecular abnormalities. The mutation taster tools were used to evaluate DNA sequence variants for their disease-causing potential [20]. Generally, the DNMT3A mutation rate of 6.6% and 22.2% polymorphism rs368516543 was observed in the total cohort. In this study, a missense mutation p.R882H (4.34%) was found in ALL patients. In previous studies, the frequency of DNMT3A gene mutations in ALL patients was reported to be between 0% to 18% [21].In this article, two missense mutation P.R882H (9.09%) were found in AML patients, and in previous studies, the frequency of DNMT3A gene mutations in AML patients was reported to be between 4% to 30%. The difference in the prevalence of DNMT3A mutations in various studies may be due to differences in patient populations, methods used, and ethnic history [22]. No significant correlation was found between platelet count and hemoglobin levels with DNMT3A mutations, which were matched with the study’s results done by Ibrahem et al. in 2015. While in other studies, DNMT3A mutations have been reported with higher platelet counts [23] and lower hemoglobin levels [24].

In the present study, no significant correlation was found between sex and white blood cell count with DNMT3A mutations, which were agreed with the results of Ibrahem et al.’s study in 2015 [23], while in Lin et al.’s study in 2011 [25] and other studies, DNMT3A mutations have been reported with higher white blood cell counts [26] and in the study of Liu et al., in 2015, DNMT3A mutations were reported with lower white blood cell counting [19]. In AML, patients with DNMT3A mutations were older than the patients without DNMT3A mutations, which were in line with the results of Lin et al.’s study in 2011 [25] and other studies. While in ALL patients, no significant difference was found between mutant patients and without mutation patients. It was in contrast with Liu’s results et al.’s in 2015 [19].

In this study, a significant relationship was not found between the blast percent of bone marrow and DNMT3A mutations, which were matched with the results of Ibrahem et al.’s study in 2015 [23] and Paganin et al. in 2011 [27]. Thol et al. in 2011 reported DNMT3A mutation strongly related to NPM1 mutations [28]. Although positive relationships were not observed between DNMT3A and NPM1 mutations, both AML patients mutated had NPM1 mutations.

There was no significant relationship between DNMT3A and CEBPA mutations. It was according to the results of Marcucci et al.’s study in 2012 [29]. In this study, a significant association between DNMT3A and FLT3mutations was not found. While in previous studies, a strong association between DNMT3A and FLT3 mutations was reported [30].

Although a significant correlation between FAB classification and DNMT3A mutations was not discovered in this study. However, similar to previous studies, DNMT3A mutations in the AML population were observed in the monocyte category [10].

No significant relationship was found between cytogenetic abnormalities and DNMT3A mutations in AML patients, but both mutant patients had normal cytogenetics. As regards, normal cytogenetics tends to be an intermediate prognosis [31]. Therefore, patients with DNMT3A mutations with normal cytogenetics may also be prone to intermedia prognosis in this study, which was in line with previous results [24, 28]. The frequency of DNMT3A mutations in AML patients was higher than ALL patients, but further studies with larger groups are needed to compare these two groups.

22.7% and 21.7% of patients with AML and ALL had polymorphism rs368516543, respectively. In previous studies, the frequency of polymorphism rs368516543 has not been reported. In mutation, tester has been reported polymorphism rs368516543 as the cause of the disease. Consequently, further studies of the relationship between polymorphism rs368516543 with DNMT3A mutations and prognosis in acute leukemia patients are recommended.

Conclusion

This study showed that DNMT3A mutation occurred in Iranian acute leukemia patients. DNMT3A mutations are probably new biomarkers in the early examination and treatment of acute leukemia, although further studies involving larger groups are needed to assess prognosis and compare AML and ALL patients.

Conflict of Interest

There is no conflict of interest to declare.

Acknowledgments

This work was supported by a grant from Mashhad University of Medical Sciences. The authors
would like to acknowledge the Cancer Molecular Pathology Research Center professors and staff at Ghaem Hospital of Mashhad.

References