Genetic variations in LC risk assessment 2

CYP2C9
CYP2C9 is involved in the activation of PAH
and heterocyclic aromatic amines. The CYP2C9∗
2
(Arg144Cys) is themost intensively studied SNP, but
results have been inconsistent due to small study
sizes [77–79].
CYP2D6
CYP2D6 is responsible for the metabolism of NNK.
In a meta-analysis of 13 studies, no association was
noted with the variant genotype [80]. This was con-
firmed by studies in different ethnic groups [81,82].
CYP2E1
CYP2E1 metabolizes a wide variety of carcinogens
including NNK. There is an Rsa I polymorphism in
the 5
flanking region which has been shown to af-
fect gene transcription [83], and individuals in var-
ious populations with the variant allele exhibited a
significant reduction in LC risk [84–88]. Phenotypic
assays suggest that the variant allele is associated
with impaired host capacity to process chlorzoxa-
zone to its active metabolites [89]. However, a few
studies reported null results for this polymorphism
and LC risk [79]. Similar conflicting results were
also reported for the Dra I polymorphism located in
intron 6 [83,90–92].
MPO
Myeloperoxidase (MPO) transforms a broad range
of tobacco smoking-derived procarcinogens such as
B[a]P. A promoter SNP (–463G>A) has been asso-
ciated with reduced gene expression and enzymatic
activity, as well as reduced level of tobacco-derived
DNA adducts [93,94]. Ameta-analysis of 2686 cases
and 3325 controls [95] reported a nonsignificantly
reduced LC risk associated with the variant allele
with an OR of 0.86 (0.67–1.1). This finding is con-
sistent with another study [96].
GST family
GST is a family of soluble detoxification enzymes
that mainly catalyzes the conjugation of GSH with
intermediate cytotoxic compounds metabolized by
phase I enzymes. There are four classes of GST
enzymes, namely, GST alpha (GSTA), GST mu
(GSTM), GST pi (GSTP), and GST theta (GSTT).
Common variants in the GST family have been ex-
tensively investigated but the results have been in-
consistent. A recent meta-analysis compiling 119
previous studies with 19,729 cases and 25,921 con-
trols reported a 1.18-fold increased LC risk (1.14–
1.23) associated with the null genotype [97]. How-
ever, when the analysis was restricted to five large
studies (3436 cases and 3897 controls) with >500
cases in each to avoid potential publication bias,
only a slight increase in LC risk was identified with
an OR of 1.04 (0.95–1.14). For the whole-gene
deletion polymorphism of GSTT1, a meta-analysis
with 9632 cases and 12,322 controls from 44 previ-
ous studies reported a 1.09-fold increase in LC risk
(1.02–1.16) [97]; however, restricting the analysis
to four large studies with >500 cases eliminated this
effect with an OR of 0.99 (0.86–1.11). Publication
bias and study heterogeneity were not detected in
this GSTT1 meta-analysis, suggesting limited poten-
tial of this polymorphism alone as a risk factor for
LC. Two SNPs (Ile105Val and Ala114Val) have been
commonly studied for the GSTP1 gene. A reduction
in GSTP1 enzyme activity has been reported to be
associated with the variant allele of both SNPs [98].
However, ameta-analysis with 6221 cases and 7602
controls from 25 previous studies, and with 1251
cases and 1295 controls from4 studies did not reveal
any significant association with LC risk for either
variant [97]. GSTM3 has a 3bp-deletion polymor-
phism in intron 6, which has been investigated in
different populations but again with mixed results.
A meta-analysis of five studies with 1238 cases and
1179 controls did not note any significant associa-
tion with LC risk with an OR of 1.05 (0.89–1.23)
[97].
NAT family
Human N-acetyltransferases (NATs) are generally
classified into two categories, slow and fast acetyla-
tors, depending on the effect on protein enzymatic
activity by the polymorphisms. So far, the studies of
LC risk with NAT polymorphisms have yielded in-
consistent results. For NAT1, significant gene–dose
effects were observed between slow acetylators and
increased risk of LC in one study [99]. Conversely,
a German study [100] demonstrated that the NAT1
fast acetylators had an increased risk of lung
adenocarcinoma, but not squamous cell carcinoma,
which was consistent with the observation that fast
NAT1 acetylators exhibited a higher level of chro-
mosomal aberrations as well as increased LC risk
in smokers [101]. Similarly, discordant conclusions
have been noted for NAT2 polymorphisms with LC
risk.Arecentmeta-analysis of 16 studieswith a total
of 3865 cases and 6077 controls failed to detect any
statistically significant difference in NAT2 acetylator
status and LC risk [102].
UGT family
UDP-glucuronosyltransferase (UGT) superfamily
catalyzes the conjugation of a glucuronic acid moi-
ety to the nucleophilic substrate resulting from
the phase I bioactivation. A small Japanese study
showed that the UGT1A7∗
3 polymorphism was as-
sociated with a 4.02-fold (1.57–10.30) increased LC
risk compared with the wild-type UGT1A7∗
1 allele
[103].
SULT family
The sulfotransferase (SULT) family catalyzes the
sulfonation of a wide spectrum of exogenous and
endogenous compounds. SULT1A1 G638A is the
most studied SNP and the variant allele has been
associated with reduced sulfotransferase activity
[104]. Compared with the wild-type, the variant-
containing genotypes exhibited a significantly in-
creased LC risk with an OR of 1.85 (1.44–2.37) in a
Chinese population [105], which was in line with a
Caucasian study showing a 1.41-fold increased LC
risk (1.04–1.91) [106].
NQO1
NAD(P)H:quinono oxidoreductase 1 (NQO1) is a
cytosolic protein that metabolizes quinoid com-
pounds and derivatives. In vitro studies demon-
strated a significantly reduced enzyme activity asso-
ciated with the variant allele of a Pro187Ser SNP in
a dose-dependent fashion [107,108]. Nonetheless,
a meta-analysis of 19 studies with a total of 6980
cases and 8080 controls reported no significant LC
risk association [109].
EPHX1
There are two commonly studied SNPs (Try113His
and His139Arg) in the microsomal epoxide hydro-
lase 1 (EPHX1) gene that encodes a phase II pro-
tein involved in the hydrolysis of reactive aliphatic
and arene epoxides derived from PAH and aromatic
amines [110]. Mixed results were derived from a
number of small case–control studies evaluating
their associations with LC risk. A pooled analysis of
986 cases and 1633 controls from eight studies re-
ported a decreased LC risk associated with the vari-
ant allele of Try113His polymorphism with OR of
0.70 (0.51–0.96) [111],whichwas further validated
by an Austrian study [112]. In both studies, the rare
allele of His139Argwas associatedwith an increased
LC risk that did not reach statistical significance.
GPX1
Glutathione peroxidase I (GPX1) is a selenium-
dependent phase II antioxidant enzymewhich plays
a crucial role in detoxifying organic peroxides and
hydrogen peroxides [113]. A nsSNP (Pro198Leu)
has been widely assessed for an association with LC
risk, but the results appear to be conflicting in dif-
ferent populations [114–116].
DNA damage and repair
The capacity to repair DNA damage from endoge-
nous and exogenous sources is crucial to maintain-
ing genomic integrity. Nucleotide-excision repair
(NER), base-excision repair (BER), double-strand-
break repair (DSBR), and mismatch repair (MMR)
are four key DNA repair pathways that operate on
different types of DNA damage. Common genetic
variants in genes involved in these repair path-
ways have been reported in many LC susceptibility
studies [117].
Genetic polymorphisms in DNA repair genes
NER
NER is capable of removing a wide class of helix-
distorting lesions that interfere with base pair-
ing and generally disrupt transcription and normal
replication resulting from tobacco smoking such as
benzo[a]pyrene diol-epoxide (BPDE) [118]. NER is
themost versatile DNA repair pathway and includes
∼30 proteins involved in DNA damage recognition,