DOI:http://dx.doi.org/10.7314/APJCP.2012.13.11.

5817
Association Between GSTM1 Polymorphism and Nasopharyngeal Cancer Susceptibility

RESEARCH ARTICLE

Association Between GSTM1 Polymorphism and Nasopharyngeal

Cancer Susceptibility: a Meta-analysis
Zhen-Feng Sun, Jia Zhang, Hong-Ming Xu, Guo-Liang Wang, Pin Dong*
Abstract
Background/Aims: Glutathione S-transferase M1 (GSTM1) is a multifunctional enzyme that plays a critical
role in the detoxification of varieties of carcinogenic metabolites. Many studies have been conducted to investigate
the association between GSTM1 polymorphism and nasopharyngeal cancer (NPC) risk, but the findings among
those studies are inconsistent. To assess this relationship more precisely, we performed a meta-analysis of all
available studies on the subject. Methods: Case-control studies were identified by searching Pubmed, Embase,
ISI Web of Science, and Wanfang databases through September 6, 2012. We used the pooled odds ratio (OR)
with its corresponding 95% confidence interval (95%CI) to evaluate the association of GSTM1 polymorphism
with NPC susceptibility. Subgroup analyses by pathological types, sex and smoking status were performed to
further identify the association. Results: Overall, 11 published studies with 1,513 cases and 2,802 controls were
finally included into this meta-analysis according to the inclusion criteria. Meta-analysis of total studies showed
that the null genotype of GSTM1 was significantly associated with increased risk of NPC, when comparing with
the non-null genotype (OR=1.51, 95%CI=1.33-1.72, POR<0.001). The association was still statistically significant
in subgroup analysis of patients with nasopharyngeal squamous cell carcinoma (OR=1.73, 95%CI=1.24-2.42,
POR=0.001). Males with the null genotype of GSTM1 were more likely to subject to NPC than females. In
addition, the association between the null genotype of GSTM1 and NPC risk was strongest in individuals with
exposure to smoking. Sensitivity analysis by sequential omission of any individual studies one at a time further
demonstrated the significant association. Conclusions: The findings suggest that the null genotype of GSTM1 is
a risk factor for NPC, and there is a gene- smoking interaction in this association.
Keywords: Nasopharyngeal cancer - polymorphism - glutathione S-transferase M1 - meta-analysis

Asian Pacific J Cancer Prev, 13 (11), 5817-5821

Introduction individual’s susceptibility to carcinogens and toxins.

Previous evidence has demonstrated that the GSTM1
Nasopharyngeal cancer (NPC) is one of the most polymorphism is associated with susceptibility to a
common otolaryngological cancers, characterized by a number of malignant cancers, such as pancreatic cancer
high frequency of nodal and distant metastasis at diagnosis and renal cell carcinoma (Vrana et al., 2009; Ahmad et
(Chan, 2011). Although it is rare in many areas of the al., 2012). A recent meta-analysis indicated that the null
world, NPC still causes serious damage to public health genotype of GSTM1 was a risk factor for NPC, but the
(Chan, 2011). The exact pathogenesis of NPC has not yet sample size of this meta-analysis was not big enough to
been understood up till now. Apart from epstein-barr virus give a confirmed conclusion (Zhuo et al., 2009). Several
(EBV) infection and endogenous/exogenous carcinogens, previous studies (Guo et al., 2008; Jiang et al., 2011) on
genetic susceptibility seems to be a risk factor playing the association of GSTM1 polymorphism with NPC risk
an crucial role in the development of NPC (O’Neil et gave inconsistent results due to several factors, such as
al., 2008). Many published studies have revealed that environmental factors, family history and different genetic
polymorphisms of carcinogen-metabolizing genes backgrounds. Furthermore, a single study might not be
encoding detoxifying enzymes contribute to the variation powered sufficiently to detect a small effect of the genetic
of individual susceptibility to NPC (Chang et al., 2006; polymorphisms on NPC risk, particularly in studies with
Guo et al., 2008; Zhuo et al., 2009). small sample size. Meta-analysis by pooling data from
Glutathione S-transferase M1 (GSTM1) is a detoxifying all available studies takes the advantage of reducing
enzyme, which plays a critical role in the detoxification random error and obtaining a more precise estimate for
of varieties of carcinogenic metabolites (Moaven et the association between GSTM1 polymorphism and NPC
al., 2010). Genetic variation of GSTM1 results in loss susceptibility (Attia et al., 2003). Thus, we performed the
of its enzymatic activity and consequently affects an present meta-analysis of all eligible case-control studies to
Department of Otorhinolaryngology, Shanghai First People’s Hospital, Shanghai, China *For correspondence:
pindong2000@163.com
Asian Pacific Journal of Cancer Prevention, Vol 13, 2012 5817
Zhen-Feng Sun et al

clarify the effect of GSTM1 polymorphism on the risk of Heterogeneity analysis was assessed by the chi-square-
NPC and to shed some light on the contradictory findings. based Q statistic test (Cochran’s Q statistic) and the
I2 statistic (Cochran, 1950; Higgins et al., 2003). A P
Materials and Methods value larger than 0.05 indicated that there was lack of
heterogeneity among the included studies. The random-
Search strategy effects model was conducted using the DerSimonian and
All case-control studies assessing the association of Laird’s method (DerSimonian et al., 1986), while the
GSTM1 polymorphism with NPC risk published up to fixed-effects model was conducted using the Mantel-
September 6, 2012 were identified by searching Pubmed, Haenszel’s method (Mantel et al., 1959) according to the
Embase, ISI Web of Science, and Wanfang databases. results of heterogeneity analysis. Sensitivity analysis was
There was no language limitation. The following search performed by sequential omission of any individual studies
terms were used: Glutathione-S-Transferase M1, GSTM1, one at a time to validate the credibility of the outcomes
polymorphism, polymorphisms, mutation, variation, in this meta-analysis (Md et al., 1999). Publication bias
nasopharyngeal cancer, nasopharyngeal carcinoma and was investigated by funnel plot, Begg’s funnel plot and
NPC. For each article retrieved, manual search of the Egger’s linear regression test (Egger et al., 1997; Stuck
relevant references was simultaneously performed to et al., 1998). All analyses were performed using STATA
identify additional published articles. version 12.0 (StataCorp LP, College Station, Texas), and
the significance level was set at 0.05.
Inclusion criteria
Studies were included in this meta-analysis if they Results
satisfied with the inclusion criteria as follow: (1) Case-
control studies; (2) Studies assessing the association of Description of studies
GSTM1 polymorphism with NPC risk; (3) Providing A total of 34 potentially relevant publications up to
sufficient information for estimating Odds ratio (OR) September 6, 2012 were systematically identified by
with its 95% confidence interval (95%CI); (4) Providing searching Pubmed, Embase, ISI Web of Science, and
available data to acquire genotype frequency of GSTM1 Wanfang databases. According to the inclusion criteria, 11
polymorphism. If studies with overlapping cases or published case-control studies with 1,513 cases and 2,802
controls, the most recent and/or the largest study with controls were included into this meta-analysis (Nazar-
available data was included into the meta-analysis. Stewart et al., 1999; Da et al., 2002; Cheng et al., 2003; Liu
et al., 2003; Deng et al., 2004; Deng et al., 2005; Tiwawech
Data extraction et al., 2005; Bendjemana et al., 2006; Guo et al., 2008;
Essential data were carefully extracted from all Jiang et al., 2011; Wei et al., 2012), while the other 23
eligible studies independently by two investigators, ones were finally excluded because they did not examine
and discrepancies were finally resolved by consensus the relationship of GSTM1 polymorphism with NPC risk
between the two authors. The extracted data included or they were reviews. Two of the 11 publications (Deng
the first author’s name, the year of publication, ethnicity, et al., 2004; Deng et al., 2005) had the same first author,
countries, clinical status of controls, genotyping method, however, the both were considered as two separate study
source of controls, and the genotype distribution of cases because they were not based on the same participants with
and controls. NPC. There were 8 English literatures (Nazar-Stewart et
al., 1999; Cheng et al., 2003; Deng et al., 2005; Tiwawech
Statistical methods et al., 2005; Bendjemana et al., 2006; Guo et al., 2008;
We pooled the unadjusted OR with its 95% CI to Jiang et al., 2011; Wei et al., 2012) and 3 Chinese ones
assess the strength of the association between GSTM1 (Da et al., 2002; Liu et al., 2003; Deng et al., 2004).
polymorphism and NPC risk. Subgroup analyses by The characteristics of these 11 case-control studies
pathological typing of NPC, sex and exposure to smoking were briefly presented in Table 1. There were 7 studies
were performed to further identify the correlation. from China, one from Tunisie, one from Thailand, one

Table 1. Characteristics of Total 11 Available Studies in the Meta-analysis

First Author Publication Year Country Ethnicity Cases Controls
Null (%) Present (%) Null (%) Present (%)
Wei YP 2012 China Asians 78(61.9) 48(38.1) 305(47.6) 336(52.4)
Jiang Y 2011 China Asians 97(53.3) 85(46.7) 157(42.2) 215(57.8)
Guo X 2008 China Asians 204(59.8) 137(40.2) 328(55.6) 262(44.4)
Bendjemana K 2006 Tunisie Africans 23(51.1) 22(48.9) 33(33.0) 67(67.0)
Tiwawech D 2005 Thailand Asians 50(64.1) 28(35.9) 74(51.0) 71(49.0)
Deng ZL 2005 China Asians 78(61.4) 49(38.6) 95(45.9) 112(54.1)
Deng ZL 2004 China Asians 56(61.5) 35(38.5) 64(47.4) 71(52.6)
Liu ZG 2003 China Asians 28(60.9) 18(39.1) 18(34.0) 35(66.0)
Cheng YJ 2003 Taiwan Asians 173(55.1) 141(44.9) 169(50.1) 168(49.9)
Da SJ 2002 China Asians 48(60.0) 32(40.0) 36(45.0) 44(55.0)
Nazar-Stewart V 1999 USA Mixed 45(54.2) 38(45.8) 63(44.4) 79(55.6)

5818 Asian Pacific Journal of Cancer Prevention, Vol 13, 2012

 DOI:http://dx.doi.org/10.7314/APJCP.2012.13.11.5817
Association Between GSTM1 Polymorphism and Nasopharyngeal Cancer Susceptibility
Table 2. Summary Meta-analyses Results for Association of GSTM1 Polymorphism with Nasopharyngeal
Cancer Susceptibility
Null vs. Present* Studies ( Cases / Controls) Odds Ratio Model† Heterogeneity
OR[95%CI]* POR I (%)
2
‡PH
Total studies 11(1,513/2,802) 1.51[1.33-1.72] <0.001 Fixed 9.2 0.357
Subgroup analysis by pathological types 4(227/502) 1.73[1.24-2.42] 0.001 Fixed 0.0 0.958
Subgroup analysis by sex
Males 3(332/393) 1.36[1.01-1.83] 0.044 Fixed 0.0 0.505
Females 3(169/484) 1.20[0.84-1.71] 0.312 Fixed 37.1 0.204
Subgroup analysis by smoking or not
Smokers 2(111/124) 2.02[1.19-3.42] 0.009 Fixed 0.0 0.355
100.0
Nonsmokers 2(51/98) 1.06[0.54-2.10] 0.861 Fixed 0.0 0.625 6.

*OR=Odds Ratio; 95%CI=95% Confidence Interval; †Fixed=fixed-effects model; ‡PH, the P value of heterogeneity analysis
A 75.0
Study %
from Taiwan and still one from USA. The ethnicity among
these 11 studies was as follow: 9 publications from Asian,
ID OR (95% CI) Weight

56
one from Africa, and the last one from mixed populations.
Nazar-Stewart V 1999 1.48 (0.86, 2.56) 5.68

Da SJ 2002 1.83 (0.98, 3.43) 3.84

The genotype distributions were showed in great details50.0

Cheng YJ 2003 1.22 (0.90, 1.66) 19.54

Liu ZG 2003 3.02 (1.33, 6.87) 1.75

Deng ZL 2004 1.77 (1.03, 3.05) 5.29

in Table1. Four case-control studies of the 11 included
studies (Nazar-Stewart et al., 1999; Da et al., 2002; Deng
Tiwawech D 2005 1.71 (0.97, 3.02) 4.96

Deng ZL 2005 1.88 (1.20, 2.94) 7.44

et al., 2004; Tiwawech et al., 2005) made stratified analysis

Bendjemana K 2006 2.12 (1.04, 4.35) 2.67

25.0
Guo X 2008 1.19 (0.91, 1.56) 25.77

by pathological types. Three out of the total 11 studies

Jiang Y 2011 1.56 (1.09, 2.23) 12.86

Wei YP 2012 1.79 (1.21, 2.65) 10.19

Overall (I-squared = 9.2%, p = 0.357) 1.51 (1.33, 1.72) 100.00

(Nazar-Stewart et al., 1999; Tiwawech et al., 2005; Guo 31
et al., 2008) examined the association between the null
B genotype of GSTM1 and NPC risk in stratified analysis 0
.145 1 6.87
Study %

by gender (males and females). In addition, two of the

ID OR (95% CI) Weight

11 studies (Nazar-Stewart et al., 1999; Da et al., 2002)

explored the association in stratified analysis by smoking
Nazar-Stewart V 1999 1.48 (0.77, 2.86) 28.04

status (smokers and nonsmokers).

Da SJ 2002 1.87 (0.99, 3.54) 26.61

Tiwawech D 2005 1.92 (0.55, 6.66) 7.25

Meta-analysis of association between GSTM1

Deng ZL 2004 1.77 (1.03, 3.05) 38.10

polymorphism and NPC risk

Overall (I-squared = 0.0%, p = 0.958) 1.73 (1.24, 2.42) 100.00

Total included studies: The pooled OR of total 11 studies

.15 1 6.66
estimating the relationship of GSTM1 polymorphism with
C NPC susceptibility suggested that the null genotype of
 
Study %

GSTM1 was strongly associated with increased risk

ID OR (95% CI) Weight

of NPC, while comparing with the present genotype

Nazar-Stewart V 1999 1.16 (0.60, 2.25) 21.87

(OR=1.51, 95%CI=1.33-1.72, POR<0.001) (Table 2 and

Tiwawech D 2005 2.11 (0.92, 4.88) 10.16
Figure 1A). Sensitivity analysis by sequential omission
Guo X 2008 1.31 (0.91, 1.88) 67.97
of any individual studies further identified the significant
Overall (I-squared = 0.0%, p = 0.515) 1.36 (1.01, 1.83) 100.00 association (data were not shown). Heterogeneity was not
found in meta-analysis of total studies with an I2 equal to
9.2% (PH=0.357) (Table 2).
.205 1 4.88

D Subgroup analyses
 
Study %

Subgroup analysis by pathological types: In the

ID OR (95% CI) Weight

stratified analysis by pathological types, the risk for

Da SJ 2002 2.81 (1.16, 6.80) 30.21

nasopharyngeal squamous cell carcinoma was higher

Nazar-Stewart V 1999 1.67 (0.86, 3.25) 69.79
in individuals carrying the null genotype of GSTM1
Overall (I-squared = 0.0%, p = 0.355) 2.02 (1.19, 3.42) 100.00 (OR=1.73, 95%CI=1.24-2.42, POR=0.001, I2= 0.0%)
(Table 2 and Figure 1B). Sensitivity analysis indicated
that the result was stable (data were not shown). There
were lack of sufficient data reported on the associations
.147 1 6.8

Figure 1. Forest Plots of Pooled ORs with 95% CIs  

between GSTM1 null genotype and the other pathological
for Association Between GSTM1 Polymorphism and
types of NPC risk.
Nasopharyngeal Cancer Risk (A. Analysis of total studies;
Subgroup analysis by sex: A stratified analysis was
B. Subgroup analysis by pathological typing; C. Subgroup
analysis by sex; D. Subgroup analysis by smoking or not) conducted by sex. Interestingly, the pooled OR for three
(Results of individual and summary ORs, 95% CIs and weights studies with 332 cases and 393 controls was modestly
of each study were shown in the forest plots. Horizontal lines significant in the male population, indicated males with the
represented 95% CI and dotted vertical lines represent the value null genotype of GSTM1 were more likely to have NPC
of the summary ORs) than females (OR=1.36, 95%CI=1.01-1.83, POR=0.044,
Asian Pacific Journal of Cancer Prevention, Vol 13, 2012 5819
Zhen-Feng Sun et al

I2= 0.0%) (Table 2 and Figure 1C). results of sensitivity analyses by sequential omission of
Subgroup analyses by smoking or not: Two of the 11 individual studies one at a time suggested the significant
studies had explored the association of the null genotype association was highly unlikely due to chance.
of GSTM1 with NPC risk in stratified analysis by smoking Our meta-analysis firstly suggested that the null
status (smokers and nonsmokers) (Table 2). Significantly genotype of GSTM1 may increase susceptibility to
increased risk of NPC associated with the null genotype nasopharyngeal squamous cell carcinoma (OR=1.73,
of GSTM1 was observed for smokers (OR=2.02, 95%CI=1.24-2.42, POR=0.001) (Table 2 and Figure 1B).
95%CI=1.19-3.42, POR=0.009, I2= 0.0%), whereas no Furthermore, the results of subgroup analysis by sex
such association was observed for nonsmokers (OR=1.06, did show that males were at a higher risk for NPC than
95%CI=0.54-2.10, POR=0.861, I2= 0.0%) (Table 2 and females(Table 2 and Figure 1C), which were in agreement
Figure 1D). with the findings of a study by Tiwawech et al. in Thailand
population (Tiwawech et al., 2005). However, Nazar-
Publication bias Stewart et al. found that females with the null genotype
Begg’s funnel plot and Egger’s test were performed of GSTM1 were more susceptible to NPC than males
to identify the publication bias of the included studies (Nazar-Stewart et al., 1999), inversely. Worthy of note, our
on association of GSTM1 polymorphism with NPC meta-analysis had enlarged the sample size by combining
susceptibility. Funnel plots’ shape of all contrasts did data from all eligible studies, and thus had the advantage of
not reveal obvious evidence of asymmetry. The results obtaining a more precise estimate for the potential genetic
of Egger’s tests also suggested there was no publication association between the null genotype of GSTM1 and NPC
bias in this meta-analysis. risk in males. GSTM1 was involved in the metabolism of
tobacco and alcohol carcinogens (Ho et al., 1999). It has
Discussion been demonstrated that smoking and alcohol consumption
are considered to be risk factors for NPC (Ho et al., 1999;
To our knowledge, Glutathione S-Transferase (GST) is Nazar-Stewart et al., 1999). Our subgroup meta-analysis
one of the most common members of phase II isoenzymes by smoking or not revealed that an increasing risk of
playing crucial role in detoxifying kinds of electrophilic NPC associated with GSTM1 null genotype was observed
compounds, including carcinogens, chemotherapeutic in smokers (OR=2.02, 95%CI=1.19-3.42, POR=0.009),
drugs, and environmental toxins (Moaven et al., 2010). but not in nonsmokers (OR=1.06, 95%CI=0.54-2.10,
The absence of a homozygous allele of GSTM1 gene (the POR=0.861) (Table 2 and Figure 1D), which indicated
GSTM1 null genotype) yields a complete loss of enzyme that smoking was a risk factor for NPC in GSTM1 null
activity (Moaven et al., 2010). Previous studies have genotype carrying individuals.
demonstrated that the GSTM1 polymorphism is associated Some limitations of our meta-analysis should be
with susceptibility to a number of malignant cancers considered when interpreting the results. Firstly, the
(Vrana et al., 2009; Ahmad et al., 2012). The molecule results may be affected by additional confounding factors,
mechanism on how the null genotype of GSTM1 affects such as EBV infection status, tumor staging and age.
the development of cancers has not been defined till now. However, most studies did not estimate the relationship
Many case-control studies were published to assess the of GSTM1 polymorphism with NPC susceptibility on
association between the polymorphism of GSTM1, located these aspects, making it impossible to include them in
on chromosome 1p13.3, and NPC risk, but the existing the meta-analysis. Further studies with large sample size
evidence was still weak due to limited sample size, ethnic are needed to identify this relationship in different tumor
difference or disagreements among the published studies staging of NPC or provide the baseline data of EBV
(Tiwawech et al., 2005; Bendjemana et al., 2006; Guo infection status and age in details. Secondly, the sample
et al., 2008). Therefore, the present meta-analysis of all sizes of subgroup analyses by pathological typing, sex
available case-control studies was conducted to shed some and smoking status were still not large enough to give
light on those inconsistent results. a comprehensive analysis and a confirmed conclusion.
Similar to the meta-analytic results of Zhuo et al. Thus, more studies with large sample size are encouraged
(Zhuo et al., 2009), significant association of GSTM1 to evaluate the association of GSTM1 polymorphism with
polymorphism and NPC risk was demonstrated. In our NPC susceptibility in future. Finally, Guo et al. found
meta-analysis, 11 individual case-control studies with that the combined null genotype of GSTM1/GSTT1 was
1,513 cases and 2,802 controls were included (Nazar- associated with increased risk of NPC (Guo et al., 2008).
Stewart et al., 1999; Da et al., 2002; Cheng et al., 2003; It can be deduced that the gene-gene interactions should
Liu et al., 2003; Deng et al., 2004; Deng et al., 2005; be taken into account when exploring the association
Tiwawech et al., 2005; Bendjemana et al., 2006; Guo et al., between GSTM1 polymorphism and NPC risk. However,
2008; Jiang et al., 2011; Wei et al., 2012). Meta-analysis of the effect of GSTM1 polymorphism combined with other
total studies showed that the null genotype of GSTM1 was genes including GSTT1 on NPC susceptibility was not
strongly associated with increased risk of NPC without fully addressed in our meta-analysis due to insufficient
apparent heterogeneity (OR=1.51, 95%CI=1.33-1.72, data. Future studies are expected to further estimate the
POR<0.001, I2=9.2%) (Table 2 and Figure 1A). Subgroup possible association of combined genetic polymorphisms
analyses by pathological typing, gender and exposure to with NPC risk.
smoking further identified the association between the In conclusion, our meta-analysis shows that the null
null genotype of GSTM1 and susceptibility to NPC. The genotype of GSTM1 is a risk factor for NPC, and there
5820 Asian Pacific Journal of Cancer Prevention, Vol 13, 2012
 DOI:http://dx.doi.org/10.7314/APJCP.2012.13.11.5817
Association Between GSTM1 Polymorphism and Nasopharyngeal Cancer Susceptibility
is a gene-smoking interaction in this association. Future deletion and the relationship to hepatocellular caecinoma
studies may further explore the possible gene-gene from Guangxi. J Guangxi Med Univ, 20, 161-3 (in Chinese).
interactions in the association of GSTM1 polymorphism Mantel N ,Haenszel W (1959). Statistical aspects of the analysis
with NPC risk. of data from retrospective studies of disease. J Natl Cancer
Inst, 22, 719-48.
Md A, J W, A T, et al (1999). Facial artery and atherosclerosis.
Acknowledgements Folia Morphol (Warsz), 58, 199-206.
Moaven O, Raziee HR, Sima HR, et al (2010). Interactions
The author(s) declare that they have no competing between Glutathione-S-Transferase M1, T1 and P1
interests. polymorphisms and smoking, and increased susceptibility
to esophageal squamous cell carcinoma. Cancer Epidemiol,
34, 285-90.
References Nazar-Stewart V, Vaughan TL, Burt RD, et al (1999). Glutathione
S-transferase M1 and susceptibility to nasopharyngeal
Ahmad ST, Arjumand W, Seth A, et al (2012). Impact of
carcinoma. Cancer Epidemiol Biomarkers Prev, 8, 547-51.
glutathione transferase M1, T1, and P1 gene polymorphisms
O’Neil JD, Owen TJ, Wood VH, et al (2008). Epstein-Barr
in the genetic susceptibility of North Indian population to
virusencoded EBNA1 modulates the AP-1 transcription
renal cell carcinoma. DNA Cell Biol, 31, 636-43.
factor pathway in nasopharyngeal carcinoma cells and
Attia J, Thakkinstian A, D’Este C (2003). Meta-analyses of
enhances angiogenesis in vitro. J Gen Virol, 89, 2833-42.
molecular association studies: methodologic lessons for
Stuck AE, Rubenstein LZ ,Wieland D (1998). Bias in meta-
genetic epidemiology. J Clin Epidemiol, 56, 297-303.
analysis detected by a simple, graphical test. Asymmetry
Bendjemana K, Abdennebi M, Gara S, et al (2006). Genetic
detected in funnel plot was probably due to true heterogeneity.
polymorphism of gluthation-S transferases and N-acetyl
BMJ, 316, 469; author reply 70-1.
transferases 2 and nasopharyngeal carcinoma: the Tunisia
Tiwawech D, Srivatanakul P, Karalak A, et al (2005). Glutathione
experience. Bull Cancer, 93, 297-302.
S-transferase M1 gene polymorphism in Thai nasopharyngeal
Chan AT (2011). Current treatment of nasopharyngeal carcinoma.
carcinoma. Asian Pac J Cancer Prev, 6, 270-5.
Eur J Cancer, 47, S302-3.
Vrana D, Pikhart H, Mohelnikova-Duchonova B, et al (2009).
Chang ET, Adami HO (2006). The enigmatic epidemiology of
The association between glutathione S-transferase gene
nasopharyngeal carcinoma. Cancer Epidemiol Biomarkers
polymorphisms and pancreatic cancer in a central European
Prev, 15, 1765-77.
Slavonic population. Mutat Res, 680, 78-81.
Cheng YJ, Chien YC, Hildesheim A, et al (2003). No association
Wei YP, Long XD, Liu ZG, et al (2012). Genetic polymorphism
between genetic polymorphisms of CYP1A1, GSTM1,
of glutathione-S-transferase M1 and T1 in associated
GSTT1, GSTP1, NAT2, and nasopharyngeal carcinoma in
with carcinogenesis of hepatocellular carcinoma and
Taiwan. Cancer Epidemiol Biomarkers Prev, 12, 179-80.
nasopharyngeal carcinoma. Chin German J Clini Onc, 11,
Cochran WG (1950). The comparison of percentages in matched
138-41.
samples. Biometrika, 37, 256-66.
Zhuo XL, Cai L, Xiang ZL, et al (2009). GSTM1 and GSTT1
Da SJ, Liang B, Wu HL, et al (2002). Relationship between
polymorphisms and nasopharyngeal cancer risk: an
GSTM1 gene polymorphism and genetic susceptibility in
evidence-based meta-analysis. J Exp Clin Cancer Res, 28,
nasopharyngeal carcinoma. Prac J Cancer, 17, 617-26 (in
46.
Chinese).
Deng ZL, Wei YP, Luo W, et al (2005). Glutathione S-transferase
M1 and T1 gene deletion associated with increased
susceptibility to nasopharyngeal carcinoma. Chin German
J Clini Onc, 4, 276-8.
Deng ZL, Wei YP, Ma Y (2004). Frequent genetic deletion
of detoxifying enzyme GSTM1 and GSTT1 genes in
nasopharyngeal carcinoma patients in Guangxi Province,
China. Zhonghua Zhong Liu Za Zhi, 26, 598-600 (in
Chinese).
DerSimonian R, Laird N (1986). Meta-analysis in clinical trials.
Control Clin Trials, 7, 177-88.
Egger M, Davey Smith G, Schneider M, et al (1997). Bias in
meta-analysis detected by a simple, graphical test. BMJ,
315, 629-34.
Guo X, O’Brien SJ, Zeng Y, et al (2008). GSTM1 and GSTT1
gene deletions and the risk for nasopharyngeal carcinoma
in Han Chinese. Cancer Epidemiol Biomarkers Prev, 17,
1760-3.
Higgins JP, Thompson SG, Deeks JJ, et al (2003). Measuring
inconsistency in meta-analyses. BMJ, 327, 557-60.
Ho CK, Lo W CH, Huang PH (1999). Suspected nasopharyngeal
carcinoma in three workers with long term exposure to
sulphuric acid vapour. Occup Environ Med, 56, 426-8.
Jiang Y, Li N, Dong P, et al (2011). Polymorphisms in GSTM1,
GSTTI and GSTP1 and nasopharyngeal cancer in the east
of China: a case-control study. Asian Pac J Cancer Prev,
12, 3097-100.
Liu ZG, Wei YP, Ma Y, et al (2003). Population with GSTT1 gene

Asian Pacific Journal of Cancer Prevention, Vol 13, 2012 5821