Association between NAT2 polymorphisms and acute leukemia risk

Supplemental Digital Content is available in the text

Abstract

Background:

N-acetyl-transferase 2 (NAT2) polymorphisms have been demonstrated to be associated with acute leukemia (AL); however, the results remain controversial. The present meta-analysis was performed to provide more precise results.

Methods:

Pubmed, Embase, Cochrane Library, China National Knowledge Infrastructure, and Wanfang databases were used to identify eligible studies. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated to evaluate the strength of the association between NAT2 polymorphisms and AL risk.

Results:

Increased risk was found under both heterozygous (OR 1.24, 95% CI 1.02–1.51) and recessive model (OR 1.28, 95% CI 1.06–1.55) for rs1801280. The slow acetylator phenotype (OR 1.22, 95% CI 1.07–1.40) also increased AL risk. Subgroup analysis demonstrated that rs1801280 increased AL risk under the recessive model (OR 1.14, 95% CI 0.93–1.41) in Caucasian population and the co-dominant (OR 1.77, 95% CI 1.40–2.23), homozygous (OR 3.06, 95% CI 1.88–4.99), dominant (OR 2.22, 95% CI 1.56–3.17), recessive model (OR 2.06, 95% CI 1.35–3.16) in the Mixed populations. Association between rs1799929 and decreased AL risk was found in the co-dominant (OR 0.82, 95% CI 0.70–0.97), homozygous (OR 0.65, 95% CI 0.46–0.93), heterozygous (OR 0.71, 95% CI 0.51–1.00), and the recessive model (OR 0.68, 95% CI 0.49–0.94) in the Caucasian group. As for rs1799931, the same effects were found in the co-dominant (OR 0.68, 95% CI 0.49–0.94) and the dominant model (OR 0.68, 95% CI 0.48–0.97) in the mixed group.

Conclusion:

rs1801280 and the slow acetylator phenotype are risk factors for AL.

1. Introduction

Acute leukemia (AL) is a hematopoietic system malignant tumor, characterized by deregulated clonal expansion of abnormal white blood cells in the bone marrow and impaired production of normal blood cells.^[1] According to the cell type from which it arises, AL can be subdivided into 2 clinical forms: acute myeloid leukemia (AML) and acute lymphocytic leukemia (ALL).^[2] It has been reported that the occurrence of AL is related to genetic disorders, physical and chemical exposure, and chemotherapy^[3,4]; however, the biologic mechanisms and etiology of AL remain unclear.^[5] Studies have demonstrated that major enzymes associated with metabolism of carcinogens are associated with AL susceptibility.^[6] Different polymorphisms of those enzymes cause the differences among individuals in their ability to metabolize environmental carcinogens via different levels of enzyme activity, and this may contribute to the risk of AL.^[7] Therefore, identification of the role of the polymorphisms of these crucial enzymes may further elucidate the pathogenesis and provide potential novel biomarkers for demonstrating AL risk.

N-acetyltransferase 2 (NAT2) is a phase II enzyme, encoded by NAT2 which is located on chromosome 8. NAT2 is responsible for the metabolism of a number of aromatic and heterocyclic amine carcinogens.^[8–10]NAT2 is highly polymorphic, and its most common variants are rs1799930 (G > A), rs1799929 (C > T), rs1801280 (T > C), and rs1799931 (G > A).^[11] All mentioned sequence variants of NAT2 can lead to a major decrease in NAT2 acetylation activity or protein stability.^[6,12]NAT2 polymorphisms have been associated with drug-induced toxicities and incidence of cancers including colorectal cancer, lung cancer, breast cancer, and AL.^[11,13–16] According to acetylation activity, the phenotype of NAT2 polymorphisms can be classified into slow, intermediate, and rapid acetylator.^[17,18] It was reported that the NAT2 acetylation phenotype was also involved in the pathogenesis of cancers such as bladder cancer and AL.^[19–22] Several studies have explored the association between AL risk and either the NAT2 polymorphisms or the phenotypes of NAT2, but no unanimous conclusion has been reached.^[6,23–25] Therefore, the present meta-analysis included 12 studies comprising 2629 AL cases and 4078 healthy controls was conducted to further assess the association between the four mutations (rs1801280, rs1799929, rs1799930, and rs1799931) of NAT2, and also NAT2 phenotypes and AL risk.

2. Material and methods

2.1. Literature search

The study was performed strictly abiding by the standards of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses.^[26] We carried out a search in the Pubmed, Embase, Cochrane Library, China National Knowledge Infrastructure, and Wanfang databases for relevant studies published before September 1, 2018. We used the following search terms to identify all potential relevant studies: (‘N-acetyltransferase 2’ or ‘NAT2’ or ‘acetylator polymorphism’) and (‘acute leukemia[s]’) and (‘genetic polymorphism[s]’ or ‘genetic variant[s]’ or ‘genetic mutation[s]’ or ‘single-nucleotide polymorphism’). In addition, the following Mesh terms were used: ‘polymorphism, genetic.’ The references in the searched studies were examined by manual retrieval to identify the studies that may not be included in these databases.

2.2. Inclusion and exclusion criteria

The first review of the literature search results was conducted to exclude letter, review, not human study, and irrelevant studies. The remaining papers were further reviewed with the following inclusion and exclusion criteria.

Inclusion criteria were as follows: the design of original article was a cohort study or a case-control study; studies that evaluated the association between the NAT2 polymorphisms and/or NAT2 phenotypes with ALL and/or AML; the genotype or phenotype frequencies in cases and controls in each study must be given in the original study or the genetic distribution can help infer the needed results.

Exclusion criteria were as follows: duplication of previous publications; comment, review, meta-analysis, or editorial; study without detailed genotype or phenotype data; study without healthy control population. The selection of these studies was achieved by 2 investigators, according to the inclusion and exclusion criteria by screening the title, abstract, and full-text. Any dispute was solved by discussion.

2.3. Data extraction

The data extraction was conducted by 2 investigators from all the eligible publications. The following information from all identified studies according to a standardized data collection form: the first author name, publication year, study country, leukemia subtype, ethnicity, genotype method, the number of cases and controls for each NAT2 SNP when available, genotype frequencies in cases and controls, and the P value of Hardy-Weinberg equilibrium (HWE) in controls. The frequencies of the rapid and slow acetylators were calculated to assess the possible association of NAT2 phenotypes with AL susceptibility. Ethnicities were classified as Caucasian, mixed, and Asian. Divergences were resolved through discussion.

2.4. Quality assessment of included studies

A quality assessment was independently performed for all of the included studies by 2 authors using the Newcastle-Ottawa Quality Assessment Scale (NOS),^[27] and any disagreement was resolved by discussion and consensus. The NOS comprised the following 3 parameters of quality: selection, comparability, and exposure. The range of the score was from 0 to 9 and studies with scores range of 6 to 9 points were considered to be high quality.

2.5. Statistics analysis

We evaluated HWE by goodness-of-fit chi-square test in control groups, and P < .05 was considered as a significant departure from HWE. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated to evaluate the strength of the association between NAT2 polymorphisms and AL risk. Meanwhile, we also evaluated the association of NAT2 phenotypes with the risk for AL using the frequencies of the rapid and slow acetylators by ORs and 95% CIs. Pooled ORs were calculated for the homozygous, heterogeneous, dominant, recessive, and co-dominant models based on the genotype frequencies in cases and controls. Heterogeneity belief associated with the selected studies was investigated by the chi-square based Q test and I² test.^[28] While calculating the pooled ORs, the random-effects model was used when I² more than 70%; otherwise, the fixed-effects model was selected.^[29,30] The significance of pooled ORs was determined by Z test at the P < .05 level of significance. To evaluate the effect of 1 single study on the overall risk of AL, sensitivity analyses were performed by sequential omission of individual studies and the robustness of the pooled estimate was tested. Potential publication bias was explored by funnel plot and Egger linear regression test.^[31,32] Apart from this, subgroup analyses were stratified by ethnicity (Caucasian, mixed, and Asian), and clinical types (AML and ALL). All analyses were performed by STATA software (Version 12.0; State Corporation, College Station, TX). P values of 2-sided were considered statistically significant if it was less than .05.

3. Results

3.1. Studies selection results

The detailed process of study selection is summarized in Fig. 1. Based on the search strategy in the “Materials and methods” section, 155 records were retrieved. After screening the titles and abstracts, 139 studies were excluded, and 16 records remained on the association between NAT2 polymorphisms or NAT2 phenotypes and AL risk. After screening the full articles, a total of 12 studies published between 1999 and 2015 were included in the final meta-analyses.^{[3,15,21–25,33–37]}

Figure 1.

Flow diagram of the study identification and selection process.

3.2. Study characteristics

The characteristics of all included studies on the relationship between NAT2 polymorphisms (rs1799930, rs1799929, rs1801280, and rs1799931) and AL risk are presented in Table 1. A total of 7 studies with 1664 AL cases and 2193 controls were included.^{[3,15,21,24,33–35]} For rs1799930, 6 studies with 8 datasets involving 1180 cases and 1752 controls were included. For rs1799929, 5 studies with 7 datasets involving 1070 cases and 1608 controls were included. Four studies with 7 datasets involving 1211 cases and 1546 controls were included for rs1801280; 5 studies with 6 datasets involving 777 cases and 1262 controls were included for rs1799931. The NAT2 polymorphisms (rs1799930, rs1799929, rs1801280, and rs1799931) genotypic frequencies in all the subjects of control groups were consistent with HWE except 1 study. Study quality was assessed by NOS, and the scores ranged from 6 to 8, so the studies were considered to be high quality (Table 1). Apart from this, 8 studies with 12 datasets involving 1522 AL patients and 2688 controls of the rapid and slow acetylators phenotypes were included to assess the association between NAT2 phenotypes and AL risk^{[15,21–23,25,34,36,37]} and study quality was also assessed by NOS, and the scores ranged from 6 to 8, which are summarized in Table 2.

Table 1.

Summary of major characteristics of the studies on N-acetyl-transferase 2 polymorphisms included in the present meta-analysis.

Table 2.

Summary of major characteristics of the studies on the acetylator phenotypes of N-acetyl-transferase 2 included in the present meta-analysis.

3.3. Association between NAT2 SNPs and AL risk

The main results of ORs and 95% CIs estimated in the present analysis on the association between individual NAT2 SNPs and AL susceptibility are shown in Table 3. The fixed-effects model was adopted to calculate the pooled ORs for each individual polymorphism with low heterogeneity; otherwise, the random-effects model was selected. Overall, significant associations of NAT2 rs1801280 polymorphism with AL risk were shown under the heterozygous model (OR 1.24, 95% CI 1.02–1.51, P = .032) and recessive model (OR 1.28, 95% CI 1.06–1.55, P = .009) (Fig. 2A and B), indicating that individuals carrying the variant allele at this site may have an increased AL risk compared with those bearing the wild-type allele.

Table 3.

Pooled odds ratios and 95% confidence intervals of the relationship between four polymorphisms in N-acetyl-transferase 2 gene and acute leukemia risk.

Figure 2.

Forest plot for the heterozygous and recessive model of NAT2 rs1801280 and acetylator phenotypes of NAT2. (A) Heterozygous genetic model analysis. (B) Recessive genetic model analysis. (C) Slow versus rapid acetylator phenotypes analysis. NAT2 = N-acetyl-transferase 2.

Considering the potential impact of the confounding factors on the overall results, subgroup analyses were stratified by ethnicity (Caucasian, mixed, and Asian) and clinical types (AML and ALL) for 4 individual NAT2 allelic SNPs (Table 3). In subgroup analysis according to ethnicity, as for rs1799929, decreased AL risk was observed in the 4 genetic models among the Caucasian group (co-dominant model: OR 0.82, 95% CI 0.70–0.97, P = .017; homozygous model: OR 0.65, 95% CI 0.46–0.93, P = .017; heterozygous model: OR 0.71, 95% CI 0.51–1.00, P = .05; recessive model: OR 0.68, 95% CI 0.49–0.94, P = .021). As for rs1799931, the same effects were found in the co-dominant model (OR 0.68, 95% CI 0.49–0.94, P = .021) and the dominant model (OR 0.68, 95% CI 0.48–0.97, P = .034) in the mixed group. For rs1801280, an increasing AL risk was found in the 4 models (co-dominant model: OR 1.77, 95% CI 1.40–2.23, P < .001; homozygous model: OR 3.06, 95% CI 1.88–4.99, P < .001; dominant model: OR 2.22, 95% CI 1.56–3.17, P < .001; recessive model: OR 2.06, 95% CI 1.35–3.16, P = .001) in the mixed group, and the recessive model (OR 1.14, 95% CI 0.93–1.41, P = .024) in the Caucasian group. However, when stratified by clinical types, no significant association was found.

The data above indicated that SNPs rs1799929, rs1799931, and rs1801280 of NAT2 were significantly associated with AL. Among them, rs1799929 and rs1799931 might decrease AL risk, whereas rs1801280 might increase AL risk and those effects were associated with ethnicity.

3.4. Association between acetylator phenotypes and AL risk

The results of the present analysis on the association between acetylator phenotypes and AL susceptibility are shown in Table 4. The meta-analysis results showed a significant association between slow acetylators and AL risk (OR 1.22, 95% CI 1.07–1.40, P = .004) (Fig. 2C). Subgroup analyses were also stratified by ethnicity (Caucasian, mixed, and Asian) and clinical types (AML and ALL) for acetylator phenotypes (Table 4). However, the results showed that no significant association between acetylator phenotypes and AL was found for either ethnicity or clinical type analyses. The data indicated that slow acetylators of NAT2 might also increase the AL risk.

Table 4.

Main results of the pooled data on the association between the acetylator phenotypes of N-acetyl-transferase 2 and acute leukemia risks.

3.5. Sensitivity analysis

Sensitivity analysis was conducted to detect the influence of each individual study on the pooled ORs by sequentially removing 1 single study each time. The results demonstrated that when each study was removed, the pooled ORs were stable under any of the models mentioned above of 4 SNPs and there was also no heterogeneity observed in the meta-analysis of phenotype of NAT2 (Fig. 3A–C).

Figure 3.

Sensitivity analysis for heterozygous model and recessive model of NAT2 rs1801280 and acetylator phenotypes of NAT2. (A) Heterozygous genetic model analysis. (B) Recessive genetic model analysis. (C) Slow versus rapid acetylator phenotypes analysis. NAT2 = N-acetyl-transferase 2.

3.6. Publication bias

Publication bias was assessed using funnel plots and Egger linear regression tests. The funnel plots seemed symmetrical for both the models of NAT2 SNPs and the phenotypes of NAT2, indicating there was no publication bias (Supplementary Fig. 1). Additionally, the Egger test also did not find significant publication bias under any models of the four SNPs and phenotype of NAT2 (Table 5).

Table 5.

The result of publication bias tests for the overall data.

4. Discussion

The NAT2 gene encodes the drug phase II metabolic enzyme NAT2, which is mainly expressed in liver and gut. The function of NAT2 is to metabolize aromatic and heterocyclic amines comprising drugs, xenobiotics, and carcinogens.^[38] There are some sites of mutation resulting in NAT2 polymorphisms. Among these, SNP rs1801280, SNP rs1799929, SNP rs1799930, and SNP rs1799931 were high-profile and contribute to a change in enzyme activity, leading to the dysfunction in its ability to modify drug toxicity and an increase in the incidence of cancer.^[6,16,39,40] According to enzyme activity, the phenotypes of NAT2 can be divided into slow, intermediate, and rapid acetylators.^[17] It has been demonstrated that NAT2 polymorphisms and acetylator phenotypes were associated with the risk of a variety of tumors such as breast cancer, lung cancer, and AL.^{[11,13,15,35,41]} There has been a recent focus on the relationship of NAT2 polymorphisms, acetylator phenotypes, and AL risk. However, no consistent opinion has been reached. Some studies had reported that NAT2 polymorphisms and acetylator phenotypes were markedly associated with AL risk.^[15,33,34] While other studies had demonstrated that there was no association between them.^[21,22,36] Therefore, to further assess the relationship between NAT2 polymorphisms, acetylator phenotypes, and susceptibility to AL, we performed the present meta-analysis combining the data from different studies.

We found that rs1801280 of NAT2 increased the risk of AL, which indicates that it may serve as a potential prediction, diagnosis biomarker for AL. Stratified by ethnicity (Caucasian, mixed, Asian), we found that rs1801280 polymorphism contributes to AL risk, particularly in the mixed population in 4 models (co-dominant, homozygous, dominant, and recessive models). As for rs1799929 of NAT2, a significantly decreased AL risk was found in co-dominant, homozygous, heterozygous, and recessive models in the Caucasian population. Additionally, rs1799931 of NAT2 was also found to be associated with decreased AL risk in both co-dominant and dominant models in the mixed population. Stratified by clinical classification (AML and ALL), no association was found. The results indicated that different SNPs of NAT2 might have different association with AL susceptibility in different ethnicities; a possible explanation of this phenomenon is the linkage disequilibrium patterns in alleles between different ethnic populations. Moreover, we also extracted rapid and slow acetylator phenotype frequencies to evaluate the association between NAT2 enzyme activity and the risk of AL. These findings indicated that the slow acetylator phenotype of NAT2 might contribute to the occurrence of AL. Previous studies have demonstrated that rs1801280—1 SNP of NAT2 associated with the slow acetylator phenotype—conferred an increased risk association with AL.^[3,33] Some articles also suggested that the slow acetylator phenotype was a significant risk determinant of AL.^[15,25] Our results are consistent with those previous studies and further confirmed the relationship between NAT2 and AL. As stratified by ethnicity and clinical type, no significant association was found in either the rapid or slow acetylators phenotypes. The clinical type of AL is very complicated. AL can be divided into AML and ALL, and AML can be further classified into M0 to M7 subtypes, whereas AL can be classified into L1 to L3 subgroups according to the French-American-British criterion.^[42] However, because of insufficient data in the included studies, we could not perform the accurate subgroups stratification analysis. Additionally, the genotype distribution of the controls in 1 article was deviated from HWE generally, and this may affect the result of the analysis. Regardless, publication bias did not exist according to funnel plot and Egger linear regression test, and the NOS score also indicated that the included studies were credible.

There were several limitations in the present study. First, significant heterogeneity between studies was detected in this meta-analysis that might have an effect on the accuracy of the results, despite the fact that we used the random-effects model to calculate pool ORs. Second, because the data were insufficient, we were unable to conduct relative research on other factors such as gene–gene or gene–environment interactions that, along with NAT2 polymorphisms, might influence AL susceptibility, and we could not proceed with further subgroup stratification analysis by age, sexual distinction and so on. Third, the polymorphisms of NAT2 were detected by different methods such as PCR-RFLP and multiplex PCR, which might influence the accuracy of the result. Moreover, even though all geographical information could be obtained from the included studies, the information of the ethnic origin of AL patients could not be acquired from some included studies.

5. Conclusions

To the best of our knowledge, the present study was the first to systematically analyze the association between NAT2 polymorphism, NAT2 acetylator phenotypes, and AL risk. The results revealed that rs1801280 polymorphism and the slow acetylator phenotype are closely correlated with the occurrence of AL. As for the rs1801280 polymorphism, this effect is more pronounced in the mixed population. Meanwhile, our meta-analysis suggested that in the mixed population, rs1799931 polymorphism is related to decreased AL risk, and this phenomenon was also found between rs1799929 polymorphism and AL risk in the Caucasian population. These findings will shed light on the role of gene polymorphism in the risk of AL and provide potential biomarkers for the prediction and prognosis of AL, or even contribute to the exploration novel targets for AL therapy.

Author contributions

Data curation: Yanbing Liu, Guangwu Chen, Qiang Guo.

Formal analysis: Ran Wei, Xunqiang Yin, Yunhong Zhang.

Funding acquisition: Xia Li.

Investigation: Bin Wang, Xia Li.

Methodology: Guangwu Chen, Qiang Guo.

Resources: Xiaoxiao Zhu, Yanbing Liu.

Software: Ran Wei, Xunqiang Yin, Yunhong Zhang.

Supervision: Zhen Zhang, Lin Zhao, Bin Wang.

Validation: Zhen Zhang, Lin Zhao.

Writing – original draft: Xiaoxiao Zhu.

Writing – review & editing: Xia Li.