Keywords

9.1 Introduction

Colorectal cancer (CRC) is most common malignancy worldwide. The incidence of CRS varies over tenfold in different geographical regions. Developed countries such as Australia, Europe, and North America have higher incidence rates compared to the developing countries like Africa and South-Central Asia (Fitzmaurice et al. 2017). Further, due to early detection of CRC, polypectomy and introduction of effective primary and adjuvant treatments, the death rates from CRC was declined in western countries (Siegel et al. 2019). However in countries lacks strong healthcare infrastructure and limited resources a continuous increment in mortality rates was documented (Center et al. 2009). Although majority of CRCs are sporadic, a considerable inherited susceptibility has been observed in the CRC patients. Hence the likelihood of CRC development is the net results of environmental and genetic factors (Chan and Giovannucci 2010). Modern western lifestyles and clinical environmental factors are often associated with the increased risk of CRC. Several lines of evidences have demonstrated the long-term consumption of processed foods and foods cooked at high temperatures are implicated in CRC risk (Joshi et al. 2015). Cooking meats at high temperatures produce some compound such as polycyclic aromatic hydrocarbons (PAHs) and heterocyclic amines (HCAs) which has carcinogenic and mutagenic properties (Adeyeye 2018).

Microsomal epoxide hydrolase (mEH) (EPHX1; EC 3.3.2.3) is a phase II biotransformation enzyme that detoxifies epoxides, including PAHs and carcinogens (Okat 2018). The mEH provide protection against the toxicities of reactive epoxide intermediates by converting them as less reactive and less toxic intermediates (Oesch et al. 2004). The EPHX1 gene encoding mEH is positioned at chromosome 1q42.1 and possesses two functional polymorphisms (Hartsfield et al. 1998). The coding region of the EPHX1 gene has two genetic variants (Tyr113His and His139Arg) that alter enzyme activity. In vitro expression studies revealed that the Tyr113His polymorphism decreased mEH enzymatic activity by 40%, while His139Arg polymorphism increased mEH activity by 25% (Hassett et al. 1994). Both polymorphisms exhibit differences in alleles and genotypes among different ethnic populations (Bhaskar et al. 2013; Lakkakula et al. 2013). As mEH involved in detoxification of epoxides together with carcinogens such as PAHs and HCAs present in cigarette smoke also in cooked meats, the functional polymorphisms of EPHX modulate the rate of PAHs metabolism and subsequently modulate CRC risk. A number of studies have analysed the association between EPHX1 gene polymorphisms and the risk of various cancers, but the results are inconclusive. As the results from the previous studies investigating the correlation between colorectal cancer and EPHX1 polymorphism were not similarly conclusive (Harrison et al. 1999; Ikeda et al. 2008; Kiss et al. 2007; Mitrou et al. 2007), we performed a meta-analysis of all available data to investigate the role of EPHX1 Tyr113His and His139Arg polymorphisms with respect to the colorectal cancer risk.

9.2 Materials and Methods

9.2.1 Data Extraction

Studies related to association between EPHX1 polymorphisms and colorectal cancer risk were collected by searching PubMed, Embase, and Google Scholar. To harvest more comprehensive information published till May 2019, search terms such as “EPHX1 or mEH”, “polymorphism or mutation” and “colorectal cancer or carcinoma,” were used without any language restrictions. To facilitate the proper elucidation of results, potentially relevant studies were selected based the following criteria: (i) evaluation of the EPHX1 Tyr113His or His139Arg and risk of CRC, (ii) case-control study, and (iv) availability of genotypes. The studies matching with the above mentioned basic criteria were included in this meta-analysis. After assessing the methodological quality of individual papers, first author’s name, publication year, country of origin, and genotype frequencies were collected independently by two authors.

9.2.2 Statistical Analysis

To find the departure of Hardy–Weinberg equilibrium (HWE) in the control groups of EPHX polymorphisms, Chi-Square goodness of fit test was performed. To measure the strength of the association between the EPHX1 polymorphisms and CRC risk of cancer, odds ratios (ORs) and their corresponding 95% confidence intervals (CIs) were calculated for each study. The pooled OR with 95% CI was calculated in allelic and dominant genetic models. Between study heterogeneity in these genetic models were calculated using Chi-square test and I2 test. Random effects model (REM) or Fixed effects model (FEM) was selected respectively in the presence or absence of the heterogeneity. Subgroup analysis was also conducted according to ethnicity. To assess the influence of the individual studies to the pooled results, sensitivity analysis was conducted by omitting one study at a time. To test the publication bias, Begg’s and Egger’s tests were used. MetaGenyo, a web tool was used to calculate results of the meta-analysis in this study (Martorell-Marugan et al. 2017).

9.3 Results

9.3.1 Characteristics of Studies

A total 20 publications dealing with EPHX1 polymorphisms and CRC risk were included in the meta-analysis. The workflow of study identification is illustrated in Fig. 9.1. The characteristics of each study were summarized in Table 9.1. For Tyr113His, 20 publications and for His139Arg, 14 publications from several countries involving Caucasian and Asian subjects were investigated. All genotype distribution in controls was in accordance with HWE with the exception of 4 studies for Tyr113His (Kiss et al. 2007; Sachse et al. 2002; Sahin et al. 2012; Tranah et al. 2005).

Fig. 9.1
figure 1

Flow chart of the study selection process identifying studies comparing EPHX1 Tyr113His and His139Arg polymorphisms with the colorectal cancer

Full size image
Table 9.1 Characteristics of the studies included in the meta-analysis and their genotype distributions of the EPHX1 Tyr113His and His139Arg gene polymorphisms
Full size table

9.3.2 Meta-Analysis of EPHX1 Tyr113His Polymorphism with CRC Risk

In this meta-analysis, a total of 20 studies (Cleary et al. 2010; Cotterchio et al. 2008; Fernandes et al. 2016; Harrison et al. 1999; Hlavata et al. 2010; Huang et al. 2005; Ikeda et al. 2008; Kiss et al. 2007; Kury et al. 2008; Landi et al. 2005; Mitrou et al. 2007; Nisa et al. 2013; Northwood et al. 2010; Pande et al. 2008; Sachse et al. 2002; Sahin et al. 2012; Skjelbred et al. 2007; Tranah et al. 2005; van der Logt et al. 2006; Wang et al. 2012) involving 9770 CRC patients and 11,634 controls were included to investigate the associations between EPHX1 Tyr113His and the risk of CRC (Table 9.1). Pooled data showed that EPHX1 Tyr113His polymorphism was not significantly associated with an increased risk of CRC in both allelic and dominant genetic models (allelic C versus T: OR = 0.95, 95% CI: 0.89–1.02 and dominant CC + CT versus TT: OR = 0.94, 95% CI: 0.85–1.03) (Table 9.2; Fig. 9.2). In addition, a further subgroup analysis by ethnicity in Caucasians, Asians and mixed populations indicated that no association between EPHX1 Tyr113His polymorphism and CRC was observed for both allelic and dominant genetic models.

Table 9.2 Meta-analysis of the relationships of the EPHX1 Tyr113His and His139Arg polymorphisms with the colorectal cancer
Full size table
Fig. 9.2
figure 2

Forest plot of the studies assessing the association between colorectal cancer and EPHX1 gene polymorphisms. (a) EPHX1 Tyr113His and colorectal cancer; (b) His139Arg and colorectal cancer

Full size image

9.3.3 Meta-Analysis Between EPHX1 His139Arg Polymorphism and CRC Risk

A meta-analysis of the association between EPHX1 His139Arg polymorphism and CRC risk included 14 independent studies (Fernandes et al. 2016; Harrison et al. 1999; Hlavata et al. 2010; Huang et al. 2005; Kiss et al. 2007; Landi et al. 2005; Nisa et al. 2013; Northwood et al. 2010; Pande et al. 2008; Sachse et al. 2002; Sahin et al. 2012; Skjelbred et al. 2007; Tranah et al. 2005; van der Logt et al. 2006) with a total of 5532 CRC cases and 6529 controls (Table 9.1). Pooled data revealed that EPHX1 His139Arg polymorphism was correlated with the risk of CRC in the both allelic and dominant model (allelic G versus A: OR = 0.95, 95% CI: 0.89–1.01; dominant GG + GA versus AA: OR = 0.92, 95% CI: 0.85–0.99). In addition, a further subgroup analysis by ethnicity in Caucasians, Asians and mixed populations indicated that no association between EPHX1 His139Arg polymorphism and CRC was observed for both allelic and dominant genetic models.

9.3.4 Heterogeneity and Sensitivity Analysis

The heterogeneities that observed within the EPHX1 Tyr113His or His139Arg studies and within each subgroup of studies are shown in Table 9.2. The heterogeneity test showed a moderate heterogeneity between EPHX1 Tyr113His studies (Dominant model Pheterogeneity = 0.003, I-squared = 59%). However, no heterogeneity was detected in EPHX1 His139Arg studies (Dominant model Pheterogeneity = 0.505, I-squared = 0%). Hence in our meta-analysis, to calculate the summary ORs we applied REM and FEM respectively for Tyr113His and His139Arg polymorphisms. Further, to evaluate the sensitivity of this meta-analysis, we conducted pooled analyses by omitting one study each time (leave-one-out method). The results of the leave-one-out method for Tyr113His and His139Arg polymorphisms was shown in Fig. 9.3a and b. For both polymorphisms, there is no change in statistical significance of the results when any single study was omitted indicating the stability and reliability of the results.

Fig. 9.3
figure 3

Sensitivity analysis diagram by omitting a single study. (a) Sensitivity analysis of the OR coefficients for the association between EPHX1 Tyr113His and colorectal cancer; (b) Sensitivity analysis of the OR coefficients for the association between His139Arg and colorectal cancer

Full size image

9.3.5 Publication Bias

The results of publication bias for EPHX1 Tyr113His and His139Arg studies were shown in Table 9.2 and Fig. 9.4a and b. The shape of Begg’s funnel plot did not reveal any obvious asymmetry for both EPHX1 Tyr113His and His139Arg studies (Fig. 9.4a and b). Further, Egger’s linear regression tests did not reveal publication bias for both EPHX1 Tyr113His and His139Arg studies in all genetic models tested (Table 9.2).

Fig. 9.4
figure 4

Funnel plots analyses to assess the publication bias between colorectal cancer and EPHX1 gene polymorphisms. (a) Funnel plot based on EPHX1 Tyr113His and colorectal cancer; (b) Funnel plot based on His139Arg and colorectal cancer. The dotted vertical line indicates the triangular region within which 95% of studies are expected to lie in the absence of bias and heterogeneity. The solid vertical line corresponds to no intervention effect

Full size image

9.4 Discussion

Indeed cancer initiation and progression has been linked to activation of the immune system and oxidative stress. Throughout the life, the colon is exposed to microbiota and free radicals that respectively cause immune responses and oxidative damage. Although the etiology of CRC is not fully known, several lines of evidences indicated that the red meat intake of one of the major risk factors for the CRC (Bernstein et al. 2015; Demeyer et al. 2016; Domingo and Nadal 2017; Zhao et al. 2017a). In contrast to this some studies showed an inverse association of fiber intake with risk of CRC (Lee et al. 2017; Song et al. 2015). The meat cooking processes produce several carcinogens such as PAHs, HCAs and dioxin-like compounds. Dietary consumption is one of the highest sources of these environmental carcinogens (Zhang et al. 2013). To become carcinogenic, these PAHs and aromatic amines have to undergo phase I and phase II biotransformation reactions (Turesky 2004).

Microsomal epoxide hydrolase is one of the biotransformation enzymes that involved either detoxification or bio-activation of a wide range of substrates. Hence it is hypothesized that the polymorphisms of EPHX1 are crucial for the susceptibility of colon cancer. The Tyr113His variation is linked with low enzyme activity and His139Arg variation increases enzyme activity. Two nested case-control studies from the cohorts Nurses’ Health Study and Health Professionals’ Follow-up Study, did not reveal significant association EPHX1 gene polymorphisms and CRC risk (Tranah et al. 2004). However, individuals with ≥25 pack-year smoking history showed increased CRC risk (Tranah et al. 2004). The Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial demonstrated that the EPHX1 polymorphisms increased risk of CRC in non-Hispanic current and recent smokers (Huang et al. 2005). Subsequent studies showed that the EPHX1 polymorphisms are not associated with CRC risk (Landi et al. 2005; Mitrou et al. 2007; van der Logt et al. 2006). Meta analysis of interaction of smoking with the EPHX1 Tyr113His polymorphisms showed that the carriers of low metabolizer allele had slightly lower risk of CRC compared to its high metabolizer (Raimondi et al. 2009). Analysis of interactions among the charred meat consumption, smoking, EPHX1 polymorphisms and CRC, did not reveal significant association between EPHX1 genotype and colorectal polyps (Burnett-Hartman et al. 2011). However, meta-analysis of observational studies demonstrated that the increased intake of red and processed meat is associated with significantly increased risk of CRC (Xu et al. 2013). Further, meta-analysis also indicated that the red and processed meat intake was associated with an increased CRC incidence but not for CRC recurrence (Zhao et al. 2017b).

This meta-analysis included 20 independent case-control studies of CRC to investigate its correlation with EPHX1 variants. The results of our meta-analysis showed that EPHX1 gene Tyr113His and His139Arg polymorphisms were not associated with the risk of CRC in allelic and dominant genetic comparison models. Further in subgroup analysis by ethnicity, these polymorphisms were not associated with the risk of CRC in both Asian and Caucasians. There was no evidence of publication bias.

9.5 Conclusion

In summary, the results of this meta-analysis demonstrated that no evidence supporting the relationship between EPHX1 polymorphism and CRC risk was detected. As the pathogenesis of CRC is complex and involving interactions of gene with gene and gene with dietary factors some limitations should be taken into consideration when interpreting results. A major limitation of current meta-analysis is non-availability of the data on meat consumption, alcohol intake and smoking, which limited the evaluation of the potential interactions between these risk factors and EPHX1 polymorphisms. Well-designed studies with large sample size are warranted to establish the role of EPHX1 polymorphisms in CRC, especially for Tyr113His and His139Arg.