Tumor Biology

, Volume 34, Issue 2, pp 1215–1224

CYP2E1 polymorphisms and colorectal cancer risk: a HuGE systematic review and meta-analysis

Authors

  • Ou Jiang
    • Department of Surgical OncologyThe Second People’s Hospital of Neijiang
  • Rongxing Zhou
    • Department of Biliary SurgeryWest China Hospital of Sichuan University
  • Daoquan Wu
    • Department of Surgical OncologyThe Second People’s Hospital of Neijiang
  • Yu Liu
    • Department of Surgical OncologyThe Second People’s Hospital of Neijiang
  • Wenjian Wu
    • Department of Surgical OncologyThe Second People’s Hospital of Neijiang
    • Department of Biliary SurgeryWest China Hospital of Sichuan University
Research Article

DOI: 10.1007/s13277-013-0664-8

Cite this article as:
Jiang, O., Zhou, R., Wu, D. et al. Tumor Biol. (2013) 34: 1215. doi:10.1007/s13277-013-0664-8

Abstract

Studies investigating the associations between Cytochrome P4502E1 (CYP2E1) polymorphisms and colorectal cancer (CRC) risk report conflicting results. We conducted a meta-analysis to assess the association between CYP2E1 gene Rsa I/Pst I, Dral T/A and 96-bp insertion polymorphisms and CRC susceptibility. Two investigators independently searched the Medline, Embase, CNKI, Wanfang, and Chinese Biomedicine Databases. Summary odds ratios (ORs) and 95 % confidence intervals (95 % CIs) for CYP2E1 polymorphisms and CRC were calculated in a fixed-effect model (the Mantel–Haenszel method) and a random-effects model (the DerSimonian and Laird method) when appropriate. Ultimately, 12, 5, and 4 studies were found to be eligible for meta-analyses of Rsa I/Pst I, Dral T/A, and 96-bp insertion polymorphisms, respectively. Our analysis suggested that the variant genotype of Rsa I/Pst I were associated with a significantly increased CRC risk (c2/c2 vs. c1/c1, OR = 1.36, 95 % CI = 1.04–1.77; recessive model, OR = 1.35, 95 % CI = 1.04–1.75). Moreover, similar results were observed between CYP2E1 96-bp insertion polymorphism and CRC risk (dominant model, OR = 1.25, 95 % CI = 1.07–1.45), while no association was observed between CYP2E1 Dral T/A polymorphism and CRC susceptibility in any genetic model. No publication bias was found in the present study. This meta-analysis shows that CYP2E1 Rsa I/Pst I and 96-bp insertion polymorphisms may be associated with CRC risk. The CYP2E1 Dral T/A polymorphism was not detected to be related to the risk for CRC.

Keywords

Colorectal cancerCYP2E1Gene polymorphismMeta-analysis

Abbreviations

CRC

Colorectal cancer

CYP2E1

Cytochrome P4502E1

OR

Odds ratio

CI

Confidence interval

PCR

Polymerase chain reaction

RFLP

Restriction fragment length polymorphism

SNP

Single nucleotide polymorphisms

HWE

Hardy–Weinberg equilibrium

Introduction

Colorectal cancer (CRC) is one of the most common forms of cancer and is the third leading cause of cancer-related death worldwide [1]. In Europe and the USA, CRC represents one of the primary causes of cancer deaths [1, 2]. In Asia, CRC is the fourth leading cause of mortality by cancer, and its incidence is increasing [3]. In recent years, the incidence of CRC is increasing in China, which accounts for about 6.5 % of total cancers in urban areas and 4.6 % in rural areas [4]. However, the mechanism of colorectal carcinogenesis is still not fully understood. As with other complex diseases, CRC is caused by both genetic and environmental factors [5].

Previous epidemiological studies have identified dietary factors, such as consumption of meat, especially red meat, and cigarette smoking as possible risk factors for the development of CRC [6, 7]. Obesity and low physical activity have also been associated with a higher incidence of this cancer [8]. Moreover, approximately 20 % of patients with colorectal neoplasia have a family history of CRC, implying a significant genetic contribution in this disease [9]. Because well-recognized genetic predisposition syndromes account for less than 3 % of CRC, low-penetrance genetic factors alone or in combination with environmental factors probably contribute to CRC development [10].

Cytochrome P450 2E1 (CYP2E1), ethanol-inducible enzyme, a member of the cytochrome P450 superfamily, is involved in the metabolic activation of many low molecular weight compounds, such as N-nitrosamines, aniline, vinyl chloride, and urethane [11, 12]. N-nitrosamines present in tobacco and diet are well recognized carcinogens involved in the development of tumors at various sites. The CYP2E1 gene is located on chromosome 10q26.3. It is 18,754 base pairs (bp) long, consists of nine exons and eight introns, and encodes a 493-amino acid protein. Among the known genetic polymorphisms in the CYP2E1 gene, two point mutations in the 5′-flanking region [Pst I (rs3813867); Rsa I (rs2031920)] which are in close linkage disequilibrium and the 96-bp insertion in the 5′-flanking region have drawn much interest because of their potential functionality [13]. According to the conventional nomenclature, the RsaI wild-type allele (commonly called c1 allele) and the variant c2 allele correspond to CYP2E1*5A and CYP2E1*5B, respectively. The insertion allele is named CYP2E1*1D, whereas the noninsertion allele is CYP2E1*1C. Another important polymorphism detectable with DraI in intron 6 is T7632A, a mutation of T to A (rs6413432), which is reported to enhance transcription of the CYP2E1 gene [14]. Polymorphisms in CYP2E1 are therefore believed to be risk factors for cancer.

Over the last two decades, a number of case–control studies were conducted to investigate the association between CYP2E1 polymorphisms and CRC risk in humans. However, the results of these studies are conflicting. In 2010, Zhou et al. [15] published a meta-analysis to assess the association between CYP2E1 Rsa I/Pst I polymorphism and CRC risk and found that the Rsa I/Pst I polymorphism may be associated with the increased risk of CRC in Caucasians. However, a study by Zhou et al. [15] had some limitations, such as relatively small sample size and some of the extracted data (revealed in Table 1 in Zhou et al. [15]), to our knowledge, were incorrect. In order to derive a more comprehensive estimation of the associations between CYP2E1 polymorphisms and CRC risk, we conducted a meta-analysis to assess the association between CYP2E1 Rsa I/Pst I, Dral (T/A), and 96-bp insertion polymorphisms and CRC susceptibility.
Table 1

Characteristics of studies included in this meta-analysis

Reference (by first author)

Year

Country

Ethnicity

Sample size (case/control)

Genotyping methods

Matching criteria

MAF in control

HWE

CYP2E1 Rsa I/Pst I polymorphism

 Butler [23]

2001

Australia

Caucasian

219/200

PCR-RFLP

Age, sex

Yes

 Le Marchand [6]

2002

Mixed

Mixed

521/639

PCR-RFLP

Age, sex, ethnicity

0.169

No

 Chen [24]

2005

China

Asian

139/338

PCR-RFLP

0.203

No

 Landi [25]

2005

Spain

Caucasian

341/299

APEX

0.028

Yes

 van der Logt [26]

2006

Netherlands

Caucasian

357/412

PCR-RFLP

0.030

No

 Kiss [27]

2007

Hungary

Caucasian

500/500

PCR-RFLP

Age, sex, smoking habits, and red meat consumption

0.046

Yes

 Küry [7]

2007

French

Caucasian

1013/1118

Taqman

Age, sex, geographic origins

0.040

Yes

 Gao [28]

2007

China

Asian

313/433

PCR-RFLP

Age, sex, ethnicity

0.208

Yes

 Cotterchio [29]

2008

Canada

Caucasian

832/1247

PCR-RFLP

Age, sex

0.034

Yes

 Morita [30]

2009

Japan

Asian

685/778

PCR-RFLP

Sex, smoking, red meat intake, residence area

0.236

Yes

 Sameer [32]

2011

India

Asian

86/160

PCR-RFLP

Age, sex

0.238

No

 Silva [33]

2012

Brazil

Caucasian

131/206

PCR-RFLP

0.051

Yes

CYP2E1 Dral T/A polymorphism

 Butler [23]

2001

Australia

Caucasian

219/200

PCR-RFLP

Age, sex

Yes

 van der Logt [26]

2006

Netherlands

Caucasian

365/410

PCR-RFLP

0.030

No

 Cotterchio [29]

2008

Canada

Caucasian

834/1248

PCR-RFLP

Age, sex

0.101

Yes

 Cleary [10]

2010

Canada

Caucasian

1165/1291

Taqman

Age, sex

0.105

Yes

 Darazy [31]

2011

Lebanon

Caucasian

70/70

PCR-RFLP

Age, sex

0.029

Yes

CYP2E1 96-bp insertion polymorphism

 Le Marchand [6]

2002

Mixed

Mixed

511/637

PCR-RFLP

Age, sex, ethnicity

0.158

No

 Morita [30]

2009

Japan

Asian

684/778

PCR-RFLP

Sex, smoking, red meat intake, residence area

0.217

Yes

 Sameer [32]

2011

India

Asian

86/160

PCR-RFLP

Age, sex

0.150

No

 Silva [33]

2012

Brazil

Caucasian

131/206

PCR-RFLP

0.046

Yes

PCR-RFLP polymerase chain reaction–restriction fragment length polymorphism, APEX an oligonucleotide microarray platform based on the arrayed primer extension technique, MAF minor allele frequency, HWE Hardy–Weinberg equilibrium

Material and methods

Literature search strategy

We searched the PubMed, Embase, China National Knowledge Infrastructure, Wanfang, and Chinese Biomedicine databases for all articles on the association between CYP2E1 polymorphisms and CRC risk (last search update 5 Dec 2012). The following key words were used: “Cytochrome P450 2E1” or “CYP2E1”, “colorectal” or “colo*”, “cancer” or “tumor” or “carcinoma”, and “polymorphism” or “variant” or “allele” or “genotype”. The search was without restriction to the language and on studies conducted on human subjects. The reference lists of reviews and retrieved articles were hand-searched at the same time. We did not consider abstracts or unpublished reports. If more than one article was published by the same author using the same case series, we selected the study where the most individuals were investigated.

Inclusion and exclusion criteria

We reviewed abstracts of all citations and retrieved studies. The following criteria were used to include published studies: (a) case–control studies were conducted to evaluate the association between at least one of these three polymorphisms (Rsa I/Pst I, Dral T/A, and 96-bp insertion) and CRC risk; (b) sufficient genotype data were presented to calculate the odds ratios (ORs) and 95 % confidence intervals (CIs); and (c) the paper should clearly describe CRC diagnoses and the sources of cases and controls. Major reasons for exclusion of studies were (a) review, or editorial, or comment; (b) duplicated studies; and (c) cell line studies.

Data extraction

Two investigators (Ou Jiang and Rongxing Zhou) extracted information from all eligible publications independently according to the inclusion criteria listed above. Disagreements were resolved by discussion between the two investigators. The following characteristics were collected from each study: the first author’s name, year of publication, the country of participants, ethnicity, source of control group (population- or hospital-based controls), number of cases and controls, genotypes, genotyping methods, minor allele frequency (MAF) in controls, and evidence of Hardy–Weinberg equilibrium (HWE) (Table 1).

Statistical analysis

We first assessed HWE in the controls for each study using goodness-of-fit test (chi-square or Fishers exact test), and a P < 0.05 was considered as statistically significant. The strength of the association between CRC and the CYP2E1 polymorphisms were estimated using ORs, with the corresponding 95 % CIs. In addition, Z test was also used, and the P value <0.05 indicated statistical significance for the association. The crude ORs and 95 % CIs were calculated by several comparisons. Taking CYP2E1 Rsa I/Pst I as an example as follows: codominant model (c2/c2 vs. c1/c1 and c1/c2 vs. c1/c1), dominant model (c1/2 + c2/c2 vs. c1/c1), and recessive model (c2/c2 vs. c1/c2 + c1/c1), respectively [16].

Both the Cochran’s Q statistic [17] to test for heterogeneity and the I2 statistic to quantify the proportion of the total variation due to heterogeneity [18] were calculated. A P value of more than the nominal level of 0.10 for the Q statistic indicated a lack of heterogeneity across studies, allowing for the use of a fixed-effects model (the Mantel–Haenszel method) [19];otherwise, the random-effects model(the DerSimonian and Laird method) was used [20]. Heterogeneity was explored using subgroup analysis with ethnicity, matched control (yes/no), HWE in controls (yes/no), and genotyping methods (polymerase chain reaction (PCR)-restriction fragment length polymorphism (RFLP)/other).

Sensitivity analyses were performed to assess the stability of the results, namely a single case–control study in this meta-analysis was omitted each time to reflect the influence of the individual data set to the pooled odds ratio (OR). Several methods were used to assess the potential publication bias. Visual inspection of funnel plot asymmetry was conducted. The Begg’s rank correlation method [21] and the Egger’s weighted regression method [22] were used to statistically assess publication bias (P < 0.05 was considered statistically significant). All analyses were done using STATA software, version 11.0 (STATA Corp., College Station, TX, USA). All the P values were two-sided.

Results

Literature search and study selection

Sixty-five papers were relevant to the search words. Through screening the title and reading the abstract and the entire article, 14 eligible articles [6, 7, 10, 2333] (13 [6, 7, 10, 23, 2533] in English and one [24] in Chinese) were included based on the search criteria (Fig. 1), for CRC susceptibility related to the CYP2E1 gene Rsa I/Pst I, Dral (T/A), and 96-bp insertion polymorphisms.
https://static-content.springer.com/image/art%3A10.1007%2Fs13277-013-0664-8/MediaObjects/13277_2013_664_Fig1_HTML.gif
Fig. 1

Flow chart of study selection based on the inclusion and exclusion criteria

Study characteristics were summarized in Table 1. There were four studies of subjects of Asian descent, nine studies of subjects of Caucasian descent, and one of subjects of mixed descent. Among these studies, five studies have investigated only CYP2E1 Rsa I/Pst I polymorphism, two studies have investigated only CYP2E1 Dral (T/A) polymorphism, whereas three studies included CYP2E1 Rsa I/Pst I and Dral (T/A) polymorphisms, and four studies included CYP2E1 Rsa I/Pst I and 96-bp insertion polymorphisms. Therefore, there were 12 case–control studies with 5,137 cases and 6,330 controls for Rsa I/Pst I polymorphism; five case–control studies with 2,653 cases and 3,219 controls for Dral (T/A) polymorphism; and four case–control studies with 1,412 cases and 1,781 controls for 96-bp insertion polymorphism. Studies had been carried out in Australia, China, Spain, Netherlands, Hungary, French, Canada, Japan, India, Brazil, Lebanon, and Hawaii. The controls were mainly from healthy population and blood donor and matched with age, sex, or other. All studies extracted DNA from peripheral blood and a classic PCR-RFLP assay was used in 11 out of 14 studies. Only 9/14 (64.3 %) studies described the use of positive controls and a different genotyping assay to confirm the data. The genotype distributions among the controls of all studies followed HWE except for four studies [6, 24, 26, 32] for the Rsa I/Pst I polymorphism, two studies [6, 32] for the 96-bp insertion polymorphism, and one study [26] for the Dral (T/A) polymorphism.

Overall meta-analysis

CYP2E1 Rsa I/Pst I polymorphism

Twelve case–control studies with 5,137 cases and 6,330 controls were included for association between CYP2E1 Rsa I/Pst I polymorphism and CRC risk. There was a wide variation in the CYP2E1 Rsa I/Pst I c2 allele frequency among different ethnicities, ranging from 2.8 % in a Caucasian population [25] to 23.8 % in an Asian population [32]. The genotype distributions among the controls of all studies were consistent with HWE except for four studies [6, 24, 26, 32] for the Rsa I/Pst I polymorphism.

The evaluations of the association of CYP2E1 Rsa I/Pst I polymorphism with CRC risk are shown in Table 2. The results of the combined analyses showed that the variant genotype of Rsa I/Pst I were associated with a significantly increased CRC risk (c2/c2 vs. c1/c1, OR = 1.36, 95 % confidence interval (CI) = 1.04–1.77, Pheterogeneity = 0.18; recessive model, OR = 1.35, 95 % CI = 1.04–1.75, Pheterogeneity = 0.27) (Fig. 2). In the subgroup analysis by ethnicity, matched controls, HWE in controls, and genotyping methods, the variant genotypes had no significant relationship with CRC in all of the subgroups except that a significantly increased CRC risk was observed among studies with matched controls (OR = 1.35, 95 % CI = 1.02–1.77, Pheterogeneity = 0.10) and among studies taking PCR-RFLP as genotyping method (OR = 1.34, 95 % CI = 1.03–1.75, Pheterogeneity = 0.14) in the recessive comparison model.
Table 2

Stratified analyses of the CYP2E1 Rsa I/ Pst I polymorphism on CRC risk

Variables

Na

c2/c2 vs. c1/c1

c1/c2 vs. c1/c1

Dominant model

Recessive model

OR (95 % CI)

Pb

OR (95 % CI)

Pb

OR (95 % CI)

Pb

OR (95 % CI)

Pb

Total

12

1.36 (1.04, 1.77)

0.18

1.00 (0.89,1.11)

0.10

1.09(0.92, 1.30)

0.02

1.35 (1.04, 1.75)

0.27

Ethnicity

 Asian

4

1.56 (0.89, 2.72)

0.05

1.02 (0.87, 1.20)

0.30

1.17 (0.90, 1.53)

0.06

1.50 (0.90, 2.50)

0.08

 Caucasian

7

2.30 (0.93, 5.69)

0.64

1.11 (0.84, 1.46)

0.08

1.11 (0.83, 1.49)

0.04

2.23 (0.90, 5.53)

0.66

 Others

1

0.94 (0.52, 1.71)

0.83 (0.63, 1.09)

0.84 (0.65, 1.09)

0.99 (0.55, 1.78)

Matched control

 Yes

8

1.52 (0.95, 2.43)

0.05

0.99 (0.82, 1.18)

0.06

1.05 (0.85, 1.29)

0.007

1.35 (1.02, 1.77)

0.10

 No

4

1.44 (0.59, 3.50)

0.60

1.25 (0.94, 1.65)

0.82

1.26 (0.96, 1.66)

0.67

1.36 (0.56, 3.30)

0.62

HWE in controls

 Yes

8

1.37 (0.97, 1.95)

0.12

0.99 (0.87, 1.13)

0.13

1.05 (0.86, 1.29)

0.05

1.38 (0.98, 1.96)

0.14

 No

4

1.33 (0.89, 1.99)

0.26

1.01 (0.83, 1.24)

0.12

1.21 (0.83, 1.76)

0.03

1.30 (0.88, 1.93)

0.42

Genotyping methods

 PCR-RFLP

10

1.49 (0.98, 2.26)

0.09

1.08 (0.91, 1.29)

0.07

1.14 (0.94, 1.39)

0.009

1.34 (1.03, 1.75)

0.14

 Others

2

1.49 (0.25, 8.88)

0.63

0.86 (0.64, 1.15)

0.65

0.87 (0.65, 1.17)

0.69

1.51 (0.25, 8.95)

0.62

HWE Hardy–Weinberg equilibrium

aNumber of comparisons

bP value of Q test for heterogeneity test. Random-effects model was used when P value for heterogeneity test <0.1; otherwise, fixed-effects model was used

https://static-content.springer.com/image/art%3A10.1007%2Fs13277-013-0664-8/MediaObjects/13277_2013_664_Fig2a_HTML.gifhttps://static-content.springer.com/image/art%3A10.1007%2Fs13277-013-0664-8/MediaObjects/13277_2013_664_Fig2b_HTML.gif
Fig. 2

Forest plots of ORs with 95 % CIs for CYP2E1 Rsa I/Pst I polymorphism and risk for CRC. The center of each square represents the OR, the area of the square is the number of sample and thus the weight used in the meta-analysis, and the horizontal line indicates the 95 % CI. a c2c2 vs. c1c1. b c1c2 vs. c1c1. c c2c2 + c1c2 vs. c1c1. d c2c2 vs. c1c2 + c1c1

CYP2E1 Dral T/A and 96-bp insertion polymorphisms

The evaluations of the association of CYP2E1 Dral T/A and 96-bp insertion polymorphisms with CRC risk are shown in Table 3. Five case–control studies with 2,653 cases and 3,219 controls for CYP2E1 Dral T/A, in which all subjects were Caucasian populations, were included eventually. There was a wide variation in the CYP2E1 Dral T/A A allele frequency among different countries, ranging from 2.9 % in a Caucasian population [31] to 10.5 % in another Caucasian population [10]. The results of the combined analyses showed that CYP2E1 Dral T/A was not associated with CRC risk for all genetic models (AA vs. TT: OR = 1.13, 95 % CI = 0.65–1.94, Pheterogeneity = 0.96; AT vs. TT: OR = 1.20, 95 % CI = 0.88–1.63, Pheterogeneity = 0.04; dominant model: OR = 1.13, 95 % CI = 0.86–1.49, Pheterogeneity = 0.06; and recessive model: OR = 1.11, 95 % CI = 0.64–1.91, Pheterogeneity = 0.96).
Table 3

The CYP2E1 Dral T/A and 96-bp insertion polymorphisms on CRC risk

Variables

Na

Subgroup

OR (95 % CI)

Pooled methods

Pb heterogeneity

P value

P Publication bias

Beggc

Eggerd

Dral T/A polymorphism

 AA vs. TT

5

All

1.13 (0.65, 1.94)

Fixed M-H method

0.96

0.67

1.00

0.99

 AT vs. TT

5

All

1.20 (0.88, 1.63)

Random D-L method

0.04

0.25

0.73

0.63

 Dominant

5

All

1.13 (0.86, 1.49)

Random D-L method

0.06

0.39

1.00

0.96

 Recessive

5

All

1.11 (0.64, 1.91)

Fixed M-H method

0.96

0.71

1.00

0.87

96-bp insertion/noninsertion polymorphism

 22 vs. 00

4

All

1.29 (0.93, 1.80)

Fixed M-H method

0.26

0.13

0.73

0.43

 11 vs. 00

4

All

1.37 (0.98, 1.92)

Random D-L method

0.04

0.06

0.31

0.34

 Dominant

4

All

1.25 (1.07, 1.45)

Fixed M-H method

0.15

0.005

0.31

0.30

 Recessive

4

All

1.20 (0.86, 1.66)

Fixed M-H method

0.26

0.28

0.73

0.48

aNumber of comparisons

bP value of Q test for heterogeneity test. Random-effects model was used when P value for heterogeneity test <0.1; otherwise, fixed-effects model was used

cP of Begg. P value of Begg rank correlation method for testing publication bias

dP of Egger. P value of Egger rank correlation method for testing publication bias

There were four case–control studies with 1,412 cases and 1,781 controls which had been performed to study the CYP2E1 96-bp insertion polymorphism and CRC risk. There was a wide variation in the CYP2E1 96-bp insertion allele frequency among different ethnicities, ranging from 4.6 % in a Caucasian population [33] to 21.7 % in an Asian population [30]. The results of the combined analyses showed that the variant genotype of CYP2E1 96-bp insertion was associated with a significantly increased CRC risk (dominant model, OR = 1.25, 95 % CI = 1.07–1.45, Pheterogeneity = 0.15).

Sensitivity analysis

The influence of a single study on the overall meta-analysis estimate was investigated by omitting one study at a time, and the omission of any study made no significant difference, indicating that our results were statistically reliable.

Publication bias

Begg’s Funnel plot and Egger’s test were performed to evaluate publication bias of the literature on CRC. Figure 3 displayed a funnel plot that examined the CYP2E1 Rsa I/Pst I polymorphism and overall CRC risk included in the meta-analysis. The shape of funnel plot did not reveal any evidence of funnel plot asymmetry. The statistical results still did not show publication bias (for CYP2E1 Dral T/A and 96-bp insertion polymorphisms were in Table 3; and for CYP2E1 Rsa I/Pst I polymorphism: c2/c2 vs. c1/c1: Begg’s test P = 1.00, Egger’s test P = 0.34; c1/c2 vs. c1/c1: Begg’s test P = 0.06, Egger’s test P = 0.09; dominant model: Begg’s test P = 0.30, Egger’s test P = 0.24; recessive model: Begg’s test P = 1.00, Egger’s test P = 0.35).
https://static-content.springer.com/image/art%3A10.1007%2Fs13277-013-0664-8/MediaObjects/13277_2013_664_Fig3_HTML.gif
Fig. 3

Funnel plot for publication bias test (c2c2 vs. c1c1). Each point represents a separate study for the indicated association

Discussion

The present meta-analysis, including 14 case–control studies, explored the association between the Rsa I/Pst I, Dral T/A, and 96-bp insertion polymorphisms of the CYP2E1 gene and CRC risk. We found that CYP2E1 Rsa I/Pst I polymorphism was associated with an increased CRC risk (5,137 cases and 6,330 controls). When subgroup analyses were performed by ethnicity, matched control, genotyping methods, and HWE in controls, significant association was observed among studies with matched controls and among studies taking PCR-RFLP as genotyping method. Similarly, we found that CYP2E1 96-bp insertion polymorphism was also associated with an increased CRC risk (1,412 cases and 1,781 controls). Nevertheless, we found that CYP2E1 Dral T/A polymorphism was not associated with CRC risk.

The CYP2E1 gene that encodes the CYP2E1 enzyme has been mapped to chromosome 10q24.3-qter. It is an important member of the cytochrome P-450 superfamily and is a critically important enzyme involved in the metabolic activation of nitroaromatic compounds, polycyclic aromatic hydrocarbons, arylamines, and numerous xenobiotics [34]. N-nitrosamines can be found in tobacco and diet, the active nitrosamines has been linked to the development of various cancers, including CRC, gastric cancer, and lung cancer. Genetic variants in CYP2E1 in the etiology of CRC have drawn increasing attention. Molecular biological studies showed that the rare allele of Rsa I/Pst I polymorphism in the CYP2E1 gene is associated with increased transcriptional activity [13]. Moreover, another important polymorphism detectable with DraI in intron 6 is T7632A, a mutation of T to A, which is reported to enhance transcription of the CYP2E1 gene [14]. Therefore, functional CYP2E1 gene polymorphisms might impact on susceptibility to CRC development.

As for the CYP2E1 PstI/RsaI polymorphism, a number of studies have focused on the role of RsaI polymorphism in the pathogenesis of cancers of the colon and rectum, and their results have been inconsistent. For example, Le Marchand et al. [6] found that individuals with the RsaI c2 allele tended to have decreased risks of colon and rectal cancers in a population-based study in Hawaii, whereas a statistically significant increase in CRC risk was reported for individuals with the RsaI c2 allele in Hungary [27] and for those homozygous for the c2 allele in China [28]. No association was detected between the RsaI polymorphism and the risk of CRC in Australia [23] and the Netherlands [26]. In 2010, Zhou et al. [15] published a meta-analysis to assess the association between CYP2E1 Rsa I/Pst I polymorphism and CRC risk and found that the Rsa I/Pst I polymorphism may be associated with the increased risk of CRC in Caucasians. However, the RsaI c2 allele is rather rare in Caucasians, and results from studies of Caucasians may have been subject to chance. This led us to undertake the present meta-analysis, which aims to derive a more comprehensive estimation of the associations between CYP2E1 polymorphisms and CRC risk. The main finding of this meta-analysis of 14 case–control studies is that individual carriers of CYP2E1 c2 homozygous do appear to have an increased risk of CRC, which further confirms the result reported by Zhou et al. [15]. This may be because of the increased transcriptional activation of the c2 variant of the CYP2E1 gene, with elevated expression levels of CYP2E1 mRNA and protein [35].

This meta-analysis, to the best of our knowledge, investigated the association between CYP2E1 Dral T/A and 96-bp insertion polymorphisms and risk of CRC for the first time. We found five studies that had examined the association between CYP2E1 Dral T/A polymorphisms and CRC. The pooled result showed that CYP2E1 Dral T/A was not associated with CRC risk for all genetic models. This null association was limited to studies from Western countries, and no studies were from Asian countries. Additional studies are therefore required to further validate the effect of DraI polymorphisms on CRC risk, especially in Asian populations. As for the CYP2E1 96-bp insertion polymorphism, we found that the variant genotype of 96-bp insertion was associated with a significantly increased CRC risk. As studies investigated the genotypes were not much enough, the results should be interpreted with caution, and more studies are needed to confirm our results.

One of the major concerns in a sound meta-analysis is publication bias due to selective publication of reports. We used funnel plots and two formal statistical methods (Egger’s weighted regression method and the rank correlation method of Begg and Mazumdar) to detect bias. Both the shape of funnel plots and statistical results did not show publication bias. Moreover, sensitivity analysis was performed to confirm the stability of the meta-analysis. Inappropriate selection of controls is a major source of bias in case–control studies. However, control groups for most of the studies used in this meta-analysis were selected from among healthy volunteers or blood donors, and CYP2E1 polymorphisms are unlikely to be associated with these conditions. Because CYP2E1′ locus is on chromosome 10 (and not sex chromosomes), the distribution of this polymorphism is not associated with sex. Therefore, in theory, matching for sex should not affect the results.

However, there are still some limitations in this meta-analysis. First, only published studies were included in the meta-analysis; therefore, publication bias may have occurred, even though the use of a statistical test did not show it. Secondly, our meta-analysis was based on unadjusted OR estimates because not all published studies presented adjusted ORs or when they did, the ORs were not adjusted by the same potential confounders, such as age, sex, ethnicity, and exposures. Lack of the information for the data analysis may cause serious confounding bias. Thirdly, we did not test for gene-to-environment interactions because of the issue of multiple testing and the lack of sufficient studies. It is possible for specific environmental and lifestyle factors to alter those associations between gene polymorphisms and CRC risk.

In conclusion, this meta-analysis suggests that CYP2E1 Rsa I/Pst I and 96-bp insertion polymorphisms may be associated with CRC risk. The CYP2E1 Dral T/A polymorphism was not detected to be related to the risk for CRC. Since limited studies were from Asian population, it is critical that larger and well-designed multicentric studies, especially Asian studies, should be performed to reevaluate the associations. Moreover, further studies estimating the effect of haplotypes and gene–environment interactions may eventually provide a better comprehensive understanding of the associations between the CYP2E1 polymorphisms and CRC risk.

Conflicts of interest

None.

Copyright information

© International Society of Oncology and BioMarkers (ISOBM) 2013