Effects of the NQO 1 609 C > T Polymorphism on Leukemia Susceptibility : Evidence from a Meta-analysis

NQO1 (NAD(P)H dehydrogenase, quinone 1) is a member of the NAD(P)H dehydrogenase family and encodes a cytoplasmic 2-electron reductase. This FADbinding protein forms homodimers and reduces quinones to hydroquinones (Snyder et al., 1996; North et al., 2011). In organism, NQO1 functions as a gatekeeper of the 20S proteasomes (Asher et al., 2005), it binds to a subset of short-lived proteins (such as p53, p73 and ornithine decarboxilase) and protects them from 20S proteasomal degradation. The NQO1 609C>T polymorphism is characterized by a single proline-to-serine amino acid substitution, that decreases the half time of NQO1 from 18 h (wild-type) to only 1.2 h via ubiquitination and proteasome pathways. Moreover, other research demonstrated that cell lines and tissues genotyped as homozygous for the NQO1 609C>T polymorphism are deficient in NQO1 activity (Siegel et al., 2001). NQO1 protein prevents one electron reduction of quinones that results in the production of radical species. Mutations in this gene have been associated with tardive dyskinesia (TD) (Pae et al., 2004; Pae, 2008), an increased risk of hematotoxicity after exposure to benzene (Iskander et al., 2005; Ross, 2005), increased risk of childhood asthma (David et al., 2003; Li et al., 2009), susceptibility to various forms of cancer (Sameer et al., 2010; Pandith et


Introduction
NQO1 (NAD(P)H dehydrogenase, quinone 1) is a member of the NAD(P)H dehydrogenase family and encodes a cytoplasmic 2-electron reductase.This FADbinding protein forms homodimers and reduces quinones to hydroquinones (Snyder et al., 1996;North et al., 2011).In organism, NQO1 functions as a gatekeeper of the 20S proteasomes (Asher et al., 2005), it binds to a subset of short-lived proteins (such as p53, p73 and ornithine decarboxilase) and protects them from 20S proteasomal degradation.The NQO1 609C>T polymorphism is characterized by a single proline-to-serine amino acid substitution, that decreases the half time of NQO1 from 18 h (wild-type) to only 1.2 h via ubiquitination and proteasome pathways.Moreover, other research demonstrated that cell lines and tissues genotyped as homozygous for the NQO1 609C>T polymorphism are deficient in NQO1 activity (Siegel et al., 2001).
NQO1 protein prevents one electron reduction of quinones that results in the production of radical species.Mutations in this gene have been associated with tardive dyskinesia (TD) (Pae et al., 2004;Pae, 2008), an increased risk of hematotoxicity after exposure to benzene (Iskander et al., 2005;Ross, 2005), increased risk of childhood asthma (David et al., 2003;Li et al., 2009), susceptibility to various forms of cancer (Sameer et al., 2010;Pandith et
The aim of this article is to review and evaluate the association between NQO1609C>T polymorphism and leukemia risk, mainly focusing on different ethnicity types.

Identification and eligibility of relevant studies
To identify all published articles that examined the association between NQO1 609C>T polymorphism and leukemia risk, we conducted a search in the PubMed database (before 2012-12-15).We identified 90 articles with the search terms (''NAD(P)H Dehydrogenase (Quinone)'' [MeSH] or ''NQO1'') and ''leukemia'' and limiting the search to studies in human populations.Articles with the following characteristics were excluded from the review: 1) non-English articles; 2) review articles; 3) non-epidemiological studies (e.g., studies on animals or cell culture); 4) treatment outcome studies (Figure 1); 5) studies with control that did not meet Hardy-Weinberg equilibrium (HWE).As of December 15, 2012, we had identified 22 published studies describing the association between NQO1 polymorphisms and leukemia included case-control analyses.

Data extraction and assessment of study quality
Two authors (Fei-fei Han and Chang-long Guo) extracted data and reached a consensus on all of the eligibility items, including author, journal and year of publication, location of study, selection and characteristics of cancer cases and controls, control source, age grades of patients, ethnicity, and leukemia types.

Meta-analysis
The risks (odds ratios, OR) of cancer associated with NQO1 609C>T polymorphism were estimated for each study independently.Also we estimated the risk for the NQO1 609C>T polymorphism and breast cancer, colorectal cancer separately.

Statistical analysis
The meta-analysis was performed in a fixed/random effect model.The OR and its 95% CI were estimated for each study.The chi-squared test-based Q-statistic was used to assess the between-study heterogeneity.Heterogeneity was significant for P < 0.10, and then the result of random effect model was selected.Otherwise, the result of fixed effect model was selected.Meanwhile, we measured the effect of heterogeneity by another measure, I 2 = 100% × (Q-df)/Q.The I 2 -statistic measures the degree of inconsistency in the studies by calculating what percentage of the total variation across studies is as a result of heterogeneity rather than by chance.
The effect of association was indicated as OR with the corresponding 95% confidence interval (CI).The combined OR was estimated using fixed effects (FE) models (Mantel-Haenszel) and random-effects (RE) models (DerSimonian and Laird) (Sarlauskas et al., 2004).We did the Q test to assess the heterogeneity between these studies, and it was considered statistically significant   Yuan et al., 2010).The heterogeneity was quantified by I 2 metric (I 2 = 100% × (Q-df)/Q), which is independent of the number of studies in the meta-analysis (I 2 < 25% no heterogeneity; I 2 = 25-50% moderate heterogeneity; I 2 > 50% extreme heterogeneity) and P value (P > 0.1 no heterogeneity).Publication bias was investigated by funnel plot and Egger's linear regression test (Egger et al., 1997).The significant of asymmetry was determined by t test and P < 0.05 was considered as a significant publication bias.Hardy-Weinberg equilibrium (HWE) was tested by the Chi-square test.Meta-analysis was performed using Review Manager 5.0 software.Sensitivity analysis was performed by sequential remove (statistics of study remove) of individual studies (Review Manager 5.0 software).

Eligible studies for meta-analysis
This study is focusing on NQO1 609C>T polymorphism and leukemia risk.After a careful evaluation of the published literature, only 22 studies met our inclusion criteria for this meta-analysis (Table 1).The retrieved papers were then read in entirety to assess their appropriateness for the inclusion in this study.The basic information, including leukemia type, ethnicity, the number of cases and controls of each study, are listed in Table Ⅰ.In all studies, the controls were free of leukemia.In the total 22 studies, 8 articles provided the data of AML (acute myeloid leukemia) patients and 10 articles provided the data of ALL (acute lymphoblastic leukemia) patients.All of the researches were then conducted in different ethnicity, mainly Asian and Caucasian: 13 studies provided Caucasian and 5 studies provided Asian data.

Leukemia susceptibility analysis
22 studies (3700 cases and 5027controls) examining the association between NQO1 609C>T polymorphism and leukemia were included.Significant heterogeneity was observed in dominant genetic model (Ser/Ser + Pro/Ser versus Pro/Pro) and the original data were combined by means of the random effect model.In this model there showed no association of NQO1 609C>T polymorphism with leukemia (OR=1.21,95%CI = 1.02-1.44,Heterogeneity<0.00001;I 2 =69%),), and there is an association of NQO1 609C>T polymorphism with leukemia in recessive genetic model (Ser/Ser versus Pro/Ser + Pro/Pro: OR = 1.25, 95%CI = 1.02-1.53,Heterogeneity =0.94;I 2 = 0%).The forest plot (Figure 3A.) showed that the distribution of the ORs from individual studies in relation to their respective standard deviation was symmetric in funnel plot.Similarly, the Egger's test provided no evidence of publication bias in reviewed studies for dominant genetic model (t = 1.22,P = 0.235) and for recessive genetic model (t = 1.01,P = 0.328) (Figure 3B).

Leukemia type analysis
Most of studies involved in this research provided the data of AML and ALL.7 studies (1111 cases and 1829 controls) examining the association between NQO1 609C>T polymorphism and AML were included.We found no association of NQO1 609C>T polymorphism with AML (Ser/Ser + Pro/Ser versus Pro/Pro OR = 0.96, 95% CI = 0.64-1.44,P = 0.85; Ser/Ser versus Pro/Ser + Pro/Pro OR = 0.93, 95% CI = 0.64-1.33,P = 0.68).10 studies (1248 cases and 1489 controls) examining the association between NQO1 609C>T polymorphism and ALL population were included.And we found an association of NQO1 609C>T polymorphism with ALL in recessive genetic model but not in dominant genetic model (Ser/Ser + Pro/Ser versus Pro/Pro OR = 1.18, 95% CI = 0.77-1.81,P = 0.45; Ser/Ser versus Pro/Ser + Pro/Pro OR = 1.33, 95% CI = 1.02-1.75,P = 0.04).The forest plot showed that the distribution of the ORs from individual studies in relation to their respective standard deviation was symmetric in funnel plot.Similarly, we performed an analysis for ethnicity in AML and ALL populations.The results were shown in Table 2.

Age phase analysis
Because of some studies applied the data of children or pediatric, we performed an analysis of NQO1 609C>T polymorphism and children leukemia.Nine studies (1248 cases and 1489 controls) examining the association between NQO1 609C>T polymorphism and children leukemia were included.We found there is no association between NQO1 609C>T polymorphism and children leukemia (Ser/Ser + Pro/Ser versus Pro/Pro OR = 1.13, 95% CI = 0.83-1.55,P = 0.44; T/T versus C/T+C/C OR = 1.28, 95% CI = 0.96-1.7,P = 0.1).The forest plot (Figure 3C) showed that the distribution of the ORs from individual studies in relation to their respective standard deviation was symmetric in funnel plot.

Discussion
The NQO1, which is generally involved in Xenobiotic-Metabolizing, has been studied extensively on its relationship with different types of cancer, such as breast cancer, colorectal cancer, leukemia and so on.Previous conclusions of numerous studies on association between NQO1 609C>T polymorphism and leukemia remain conflicting and contradictory, this was largely attributed to the small samples or the relatively low statistical power of published studies.Meta-analysis is a powerful method for resolving inconsistent findings with a relatively large number of subjects.So, this meta-analysis was applied to provide a quantitative approach for combining the different results.To the author's knowledge, this is the most comprehensive meta-analysis investigating the genetic susceptibility of NQO1 gene C609T polymorphism to leukemia.
In the present meta-analysis with 3700 cases and 5027controls, the variant TT homozygous genotype and the combined CT/TT genotype of the NQO1 609 C>T polymorphism was found to be associated with a increased risk of leukemia, especially in Caucasian populations.These findings suggestted that the NQO1 609C>T polymorphism may modify the risk of leukemia mainly in Caucasian populations but not in Asian populations.Publication bias was not observed in this study.In the subgroup analysis of age phase we found that NQO1 609C>T polymorphism was not associated with children leukemia.
Several limitations of this meta-analysis should be pointed out.First, although the Begg's test and Egger's test did not show any publication bias, selection bias could have occurred, because only studies published in English and Chinese were included in our meta-analysis.Second, this analysis was based on unadjusted published estimates, and hence, it was unable to adjust them by possible confounders such as sex, smoking status and living environment risk factors.Furthermore, due to a limited number of published studies available to be included, it was unable to perform further subgroup analyses for AML in Asian populations.
In summary, this meta-analysis provided robust evidence of the association between NQO1 609 C>T polymorphism and leukemia risk on Caucasian population, supporting the results of published paper that NQO1  doi.org/10.7314/APJCP.2013.14.9.5311Effects of the NQO1 609C>T Polymorphism on Leukemia Susceptibility: Evidence from a Meta-analysis 609 C>T polymorphism is a strong susceptibility marker of leukemia, especially in Caucasian population.Moreover, sophisticated gene-gene interaction should be considered in future analysis, which would lead a better, comprehensive understanding of the association between NQO1 609 C>T polymorphism and leukemia risk.

Figure 3 .
Figure 3. Begg's and Egger's Funnel Plot for Publication bias Test in the Recessive Genetic Model Analysis

Forest Plot of Leukemia Risk Associated with NQO1 609 C>T Polymorphism Analysis with
P < 0.10 (