Potential conflict of interest: Nothing to report.
Genetic polymorphisms in DNA repair genes may influence individual variations in DNA repair capacity, and this may be associated with the risk and outcome of hepatocellular carcinoma (HCC) related to aflatoxin B1 (AFB1) exposure. In this study, we focused on the polymorphism of xeroderma pigmentosum complementation group C (XPC) codon 939 (rs#2228001), which is involved in nucleotide excision repair. We conducted a case-control study including 1156 HCC cases and 1402 controls without any evidence of hepatic disease to evaluate the associations between this polymorphism and HCC risk and prognosis in the Guangxi population. AFB1 DNA adduct levels, XPC genotypes, and XPC protein levels were tested with a comparative enzyme-linked immunosorbent assay, TaqMan polymerase chain reaction for XPC genotypes, and immunohistochemistry, respectively. Higher AFB1 exposure was observed among HCC patients versus the control group [odds ratio (OR) = 9.88 for AFB1 exposure years and OR = 6.58 for AFB1 exposure levels]. The XPC codon 939 Gln alleles significantly increased HCC risk [OR = 1.25 (95% confidence interval = 1.03-1.52) for heterozygotes of the XPC codon 939 Lys and Gln alleles (XPC-LG) and OR = 1.81 (95% confidence interval = 1.36-2.40) for homozygotes of the XPC codon 939 Gln alleles (XPC-GG)]. Significant interactive effects between genotypes and AFB1 exposure status were also observed in the joint-effects analysis. This polymorphism, moreover, was correlated with XPC expression levels in cancerous tissues (r = −0.369, P < 0.001) and with the overall survival of HCC patients (the median survival times were 30, 25, and 19 months for patients with homozygotes of the XPC codon 939 Lys alleles, XPC-LG, and XPC-GG, respectively), especially under high AFB1 exposure conditions. Like AFB1 exposure, the XPC codon 939 polymorphism was an independent prognostic factor influencing the survival of HCC. Additionally, this polymorphism multiplicatively interacted with the xeroderma pigmentosum complementation group D codon 751 polymorphism with respect to HCC risk (ORinteraction = 1.71). Conclusion: These results suggest that the XPC codon 939 polymorphism may be associated with the risk and outcome of AFB1-related HCC in the Guangxi population and may interact with AFB1 exposure in the process of HCC induction by AFB1. (HEPATOLOGY 2010;)
Hepatocellular carcinoma (HCC) is the most common malignant tumor in the Guangxi Zhuang Autonomous Region of China.1 Epidemiological evidence has shown that exposure to aflatoxin B1 (AFB1), an important chemical carcinogen, is the most important cause of the high rate of HCC in this area.1 Many studies have suggested that AFB1 can induce the formation of AFB1 DNA adducts and cause DNA strand breakage, DNA base damage, and oxidative damage that may ultimately lead to cancer.2 AFB1-induced DNA damage can be repaired by base excision repair, strand break repair, and nucleotide excision repair (NER).3, 4
We have previously reported a relationship between HCC risk and polymorphisms in the X-ray repair cross-complementing group 1 (XRCC1), XRCC3, and xeroderma pigmentosum complementation group D (XPD) genes, which are involved in base excision repair, strand break repair, and NER.5-7 In this study, we focused on the xeroderma pigmentosum complementation group C (XPC) gene, which is one of the seven genetic complementation groups encoding for proteins involved in NER.8-10 The XPC codon 939 polymorphic locus (rs#2228001) has attracted particular interest in molecular epidemiology studies.11-13 Some studies have shown that this polymorphism may be associated with decreased DNA repair capacity and increased tumor risk.10-23 Therefore, we specifically conducted a hospital-based case-control study to examine whether this polymorphism modifies HCC risk and prognosis in a Guangxi population from an AFB1 exposure area.
AFB1, aflatoxin B1; CI, confidence interval; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; HR, hazard ratio; IRS, immunoreactive score; MST, median survival time; NER, nucleotide excision repair; OR, odds ratio; OReg, odds ratio for the presence of both high aflatoxin B1 exposure and risk genotypes of xeroderma pigmentosum complementation group C codon 939; ORe′g, odds ratio for the high-risk xeroderma pigmentosum complementation group C genotype in patients with low aflatoxin B1 exposure; OReg′, odds ratio for high aflatoxin B1 exposure for patients with the low-risk xeroderma pigmentosum complementation group C genotype; PCR, polymerase chain reaction; PP, percentage of positive cells; SI, staining intensity; XPC, xeroderma pigmentosum complementation group C; XPC-GG, homozygotes of the xeroderma pigmentosum complementation group C codon 939 Gln alleles; XPC-LG, heterozygotes of the xeroderma pigmentosum complementation group C codon 939 Lys and Gln alleles; XPC-LG/GG, genotypes of the xeroderma pigmentosum complementation group C codon 939 Gln alleles; XPC-LL, homozygotes of the xeroderma pigmentosum complementation group C codon 939 Lys alleles; XPD, xeroderma pigmentosum complementation group D; XPD-GG, homozygotes of the xeroderma pigmentosum complementation group D codon 751 Gln alleles; XPD-LG/GG, genotypes of the xeroderma pigmentosum complementation group D codon 751 Gln alleles; XPD-LL, homozygotes of the xeroderma pigmentosum complementation group D codon 751 Lys alleles; XRCC, X-ray repair cross-complementing group.
Patients and Methods
Study Population and Sample Collection.
To investigate gene-environment interactions based on large samples,24 this study consisted of 1156 HCC cases (including 635 previously studied subjects7, 25) and 1402 control individuals (including 712 previously studied subjects7, 25). All cases and controls were recruited from affiliated hospitals of the two main medical colleges in southwestern Guangxi (Guangxi Medical University and Youjiang Medical College for Nationalities) from January 2005 to November 2009. All cases and controls were residents of the Guangxi Zhuang Autonomous Region from AFB1 exposure areas and accepted enrollment in this study. The cases included in this study, representing a significant proportion (>90%) of HCC patients in the Guangxi population, were identified by histopathological diagnosis in 100% of the HCC cases. During the same period, controls without any evidence of liver disease were randomly selected from a pool of healthy volunteers who visited the general health check-up centers of the same hospitals for their routinely scheduled physical examinations supported by local governments.
To control the effects of confounders that were likely risk factors for Guangxi HCC patients, cases were individually matched (1:1 or 1:2) to controls with respect to age (±5 years), ethnicity (Han or minority), hepatitis B virus (HBV) infection status, and hepatitis C virus (HCV) infection status. Every potential control was first surveyed with a short questionnaire to elicit willingness to participate in the study and to provide preliminary demographic data for matching. With written, informed consent, the characteristic information for each subject, including age, gender, ethnicity, HBV infection status, and HCV infection status, was gathered with a standard interviewer-administered questionnaire and/or from medical records by a Youjiang Cancer Institution staff member; at the same time, 4 mL of peripheral blood was obtained for the extraction of genomic DNA. Additionally, we collected clinical pathological data (including the cirrhosis status, tumor size, and tumor stage) from case medical records for 834 HCC patients receiving the same surgical resection treatment for the evaluation of the severity of liver disease and surgically removed samples for the analysis of XPC expression levels.
Liver cirrhosis was diagnosed by pathological examination, and the tumor stages were confirmed according to the TNM system. Those who were hepatitis B surface antigen (HBsAg)–positive and anti-HCV–positive in their peripheral serum were defined as HBV-infected and HCV-infected. One hundred percent of those asked to take part in this study who did enroll agreed to participate in the investigative study. The protocol of the study was approved by the ethic committees of the involved hospitals.
AFB1 Exposure Years.
AFB1 exposure years were ascertained by our previously published methods.6, 7 Briefly, AFB1 exposure years were defined as the years that each subject lived in an AFB1 exposure region, and cumulative AFB1 exposure years were calculated with the following formula:
where migration years are the years that a person lived in a non–AFB1 exposure region. For analysis, AFB1 exposure by years was divided into three groups according to the number of AFB1 exposure years with two cutoff points of 40 and 48 years (the median numbers of exposure years for controls and cases, respectively): short (< 40 years), medium (40-48 years), and long (>48 years).
AFB1 Exposure Levels.
In this study, we evaluated the AFB1 exposure levels according to the AFB1 DNA adduct levels of DNA samples from all subjects' peripheral blood leukocytes; we used a comparative enzyme-linked immunosorbent assay, which is described in our previously published articles.7 For analysis, AFB1 DNA adduct levels were divided into three groups according to the values of the AFB1 DNA adduct levels with two cutoff points of 1.00 and 2.00 μmol/mol of DNA (the average adduct levels among controls and cases, respectively): low (≤1.00 μmol/mol of DNA), medium (1.01-2.00 μmol/mol of DNA), and high (≥2.01 μmol/mol of DNA).
The gene polymorphism analysis of XPC Lys939Gln was typed by TaqMan polymerase chain reaction (PCR) on an iCycler iQ real-time PCR detection system (iQ5, Bio-Rad). The primers (5′-AGCAGCTTCCCACCTGTTC-3′ and 5′-GTGGGT GCCCCTCTAGTG-3′) and the probes (5′-FAM-CACAGCTGCTCAAAT-MGB-3′ and 5′-Hex-CTCACAGCT TCTCAAAT-MGB-3′) were obtained from the Cancer Genome Anatomy Project SNP500 Cancer Database and were synthesized by Shanghai GeneCore BioTechnologies Co., Ltd. (Shanghai, China). PCR reactions were run in a 25-μL final volume containing 1× Premix Ex Taq (catalog number DRR039A, Takara), 0.2 μM of each probe, 0.2 μM of each primer, and 50 to 100 ng of genomic DNA. The cycling conditions were 95°C for 2 minutes and 45 cycles of 95°C for 10 seconds, 60°C for 1 minute, and 72°C for 10 seconds. Controls were included in each run, and repeated genotyping of a random 10% subset yielded 100% identical genotypes. Data analysis for allele discrimination was performed with the iCycler iQ software.
Immunohistochemistry for XPC.
The immunohistochemistry assay for XPC was performed according to the standard procedure (protocol 40441a, Maixin Biotechnology, Inc., Fuzhou, China). The corresponding anti-XPC polyclonal antibody (1:200 dilution; catalog number sc-30156) and the horseradish peroxidase–conjugated secondary antibody (catalog number KIT-9707) were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA), and Maixin Biotechnology, respectively. The quality control for immunohistochemistry assays was administered with negative and positive controls. The evaluation of the staining reaction was performed according to a previously published formula26:
where IRS is the immunoreactive score, SI is the staining intensity, and PP is the percentage of positive cells. In this study, XPC expression levels in cancerous tissues were divided into three groups: low expression (IRS < 3), medium expression (IRS = 3-6), and high expression (IRS > 6).
Follow-Up for HCC Patients.
The patients were followed for at least 0.5 years for medians and ranges. The last follow-up day was April 30, 2010, and the survival status was confirmed by patients or family contacts. In this study, the duration of survival was defined as the time from surgical resection to death or to the date on which the patient was last known to be alive.
Cases and controls were compared for any differences in demographic and etiological characteristics and XPC genotypes with the χ2 test. The odds ratio (OR) with the 95% confidence interval (CI) was calculated as an estimate of the relative risk (by conditional logistic regression with matching factors) to evaluate the association between the risk factors and HCC risk. Joint effects between genotypes and AFB1 exposure status on HCC risk were assessed with the full regression model, which included all possible confounders. The interactive effects were evaluated according to the following formula24:
where OReg is the odds ratio for the presence of both high AFB1 exposure (i.e., many AFB1 exposure years or high AFB1 exposure levels) and risk genotypes of XPC codon 939 (adjusted OR > 1), OReg′ is the odds ratio for high AFB1 exposure for patients with the low-risk XPC genotype, and ORe′g is the odds ratio for the high-risk XPC genotype in patients with low AFB1 exposure.
The Spearman r test was used to analyze the correlation between XPC genotypes and XPC expression levels. Kaplan-Meier survival analysis with the log-rank test was used to evaluate the relationship between this polymorphism and HCC prognosis. Risk factors for HCC prognosis were first selected with the Cox multivariate regression model (including all possible multiplicatively interactive variables) with stepwise forward selection based on the likelihood ratio test. Hazard ratios (HRs) and 95% CIs for risk factors were next calculated with the multivariate Cox regression model (including all risk factors, all possible multiplicatively interactive variables, and clinical variables known to be prognostic). A P value < 0.05 was considered statistically significant in this study. All statistical analyses were performed with SPSS version 18.0 (SPSS Institute, Chicago, IL).
Demographic and Etiological Characteristics of the Subjects.
There were no significant differences in sex, age, ethnicity, HBsAg status, or anti-HCV status (P > 0.05; Supporting Table 1); this suggests that the HCC patient data were comparable to the control data.
AFB1 Exposure Status and HCC Risk.
Table 1 summarizes the AFB1 exposure information for the entire study population. We found that the HCC cases (48 years) had more AFB1 exposure years than the controls (40 years), and the HCC risk gradually increased with an increasing number of exposure years (adjusted OR = 3.26-9.88, P < 0.01). We also found that the levels of AFB1 DNA adducts were associated with an increased risk for HCC (OR = 2.02 for medium-level adducts and OR = 6.58 for high-level adducts). These results are consistent with our previously published data.5, 7, 25, 27
The genotypic distribution of XPC Lys939Gln for both cases and controls is shown in Table 2. The genotypic distribution of this gene in controls was in Hardy-Weinberg equilibrium. The frequencies of the codon 939 Gln allele were higher in cases (0.40) versus controls (0.32). Logistic regression analyses showed that the adjusted OR for HCC for those individuals carrying the heterozygotes of the XPC codon 939 Lys and Gln alleles (XPC-LG) versus those exhibiting the homozygotes of the XPC codon 939 Lys alleles (XPC-LL) was 1.25 (95% CI = 1.03-1.52); the corresponding OR for those featuring the homozygotes of the XPC codon 939 Gln alleles (XPC-GG) was 1.81 (95% CI = 1.36-2.40). This showed that the HCC risk was associated with the number of codon 939 Gln alleles.
Conditional on the matched set adjusted for AFB1 exposure years and AFB1 DNA adduct levels (corresponding OR and 95% CI values can be found in Supporting Table 4).
Because more than 70% of the cases and controls were HBsAg-positive and this might have interacted with the XPC codon 939 polymorphism in the carcinogenesis of HCC, we investigated the modified effects of this confounder and XPC genotypes on HCC risk (Supporting Table 2 and Supporting Fig. 1). Similar risk values for HCC were observed among HBsAg-positive carriers (OR = 1.67) and among HBsAg-negative participants (OR = 1.82). Additionally, we analyzed possible modified effects of anti-HCV status and XPC genotypes on HCC risk and found similar risk values for HCC among anti-HCV–positive and anti-HCV–negative groups. Likelihood ratio tests for the interaction between the stratified variables, including HBsAg status (negative and positive) and anti-HCV status (negative and positive), and XPC genotypes showed that this was not statistically significant (Pinteraction > 0.05; Supporting Table 2).
Joint Effects of AFB1 Exposure and the XPC Codon 939 Polymorphism on HCC Risk.
To study the relationship between the polymorphism of XPC codon 939 and AFB1 exposure years in the risk for HCC, we analyzed the joint effects of AFB1 exposure years and genotypes on HCC risk (Table 3). In this analysis, we used as a reference the lowest risk group: those who had XPC-LL and short-term AFB1 exposure. The results showed that increasing the number of AFB1 exposure years consistently increased HCC risk; moreover, this effect was more pronounced among the XPC-LG and XPC-GG subjects. Additionally, we evaluated the multiplicatively interactive effects of genotypes and AFB1 exposure years according to the following formula (first shown in the Patients and Methods section)24:
Table 3. Joint Effects of AFB1 Exposure and the XPC Codon 939 Polymorphism on HCC Risk
OR (95% CI)
Conditional on the matched set adjusted for AFB1 DNA adduct levels (corresponding OR and 95% CI values can be found in Supporting Table 4).
Conditional on the matched set adjusted for AFB1 exposure years (corresponding OR and 95% CI values can be found in Supporting Table 4).
Interestingly, we found some evidence of interactive effects of genotypes and exposure years on HCC risk (22.33 > 8.69 × 1.88). A similar increased-risk trend was also found in the joint-effects analysis of XPC genotypes and AFB1 DNA adduct levels for the risk of HCC (18.38 > 4.62 × 1.11; Table 3).
XPC Codon 939 Polymorphism and XPC Expression Levels.
On the basis of a recent report showing that the dysregulation of XPC is highly related to HCC,28 using immunochemistry, we investigated whether XPC genotypes influenced its expression in the cancerous tissues of 834 HCC cases. Different expression levels were detected in tumor tissues from cases with different genotypes (r = −0.369; Table 4). Representative photographs exhibit the aforementioned correlation between the genotypes and expression levels (Fig. 1A-C).
Table 4. XPC Codon 939 Polymorphism and Expression Levels of XPC Protein
Spearman r test: r = −0.369 and P < 0.001.
XPC Codon 939 Polymorphism and HCC Prognosis.
An association analysis of the risk genotypes [i.e., genotypes of the XPC codon 939 Gln alleles (XPC-LG/GG)] or the nonrisk genotype (XPC-LL) and the clinical characteristics of HCC (including the etiology and severity of liver diseases) was first performed separately. Significant differences in the distributions of the genotypes were observed with respect to different AFB1 expression years or levels but not with respect to age, gender, minority status, HBsAg status, anti-HCV status, tumor size, cirrhosis status, or TNM stage (Supporting Table 3). Survival analysis next showed that HCC patients carrying XPC-LG/GG, compared to those with XPC-LL, had shorter overall survival [the median survival times (MSTs) were 25 and 19 months for patients with XPC-LG and XPC-GG, respectively; Fig. 2A] and a higher dying risk (adjusted HR = 1.37 for XPC-LG and adjusted HR = 1.51 for XPC-GG; Table 5), particularly under high AFB1 exposure conditions (Fig. 2B,C). Furthermore, some evidence of multiplicative interaction was found for XPC genotypes and AFB1 exposure years (Pinteraction = 0.012; Table 5).
Table 5. Cox Proportional Hazards Model Analysis for the Multivariate Analysis of Potential Predictor Factors for the Overall Survival of HCC Patients
HR (95% CI)
The likelihood ratio test was used to test the interaction of XPC genotypes and AFB1 exposure by years (Pinteraction = 0.012).
Joint Effects of the XPC Codon 939 Polymorphism and XPD Codon 751 Polymorphism on HCC Risk.
Because our previous study showed that the XPD codon 751 polymorphism, another DNA repair gene involved in NER, modifies HCC risk in the Guangxi population,7 we explored possible interactions between XPC and XPD in 618 cases and 712 controls who had been previously studied.7 Because of the small number of subjects with both the homozygotes of the XPD codon 751 Gln alleles (XPD-GG) and XPC-GG, the combination of these two genes was divided into four strata (Table 6). We found that individuals with both XPC-LG/GG and genotypes of the XPD codon 751 Gln alleles (XPD-LG/GG), in comparison with those with both XPC-LL and homozygotes of the XPD codon 751 Lys alleles (XPD-LL), might face a higher HCC risk (adjusted OR = 2.02, 95% CI = 1.42-2.87). A likelihood ratio test showed that there were multiplicatively interactive effects of XPC and XPD on the HCC risk (Pinteraction = 0.019).
Table 6. Joint Effects of the XPC Codon 939 Polymorphism and XPD Codon 751 Polymorphism on HCC Risk
The most common cause of HCC is AFB1 exposure via the consumption of corn and groundnuts, which are primary food types for the Guangxi population, and HCC is one of the major cancer types in the Guangxi Zhuang Autonomous Region of China.1 Our previous studies5-7, 25, 27 and this study also show that HCC patients have more AFB1 exposure years and higher AFB1 exposure levels; moreover, we found that HCC risk is associated with different AFB1 exposure statuses. AFB1 is an important chemical carcinogen that is mainly metabolized by cytochrome P450 into the genotoxic metabolic AFB1 epoxide; this can bind to DNA, cause the formation of AFB1 guanine adducts, and induce DNA damage, including base damage and oxidative DNA damage that can be repaired by the NER pathway.2-4
XPC is an important DNA damage recognition molecule involved in the detection of a variety of DNA adducts formed by exogenous carcinogens such as AFB1.8-10 Some recent studies have shown that defects in XPC are related to many types of malignant tumors.21, 29 For example, Takebayashi et al.21 analyzed the loss of heterozygosity of XPC in sporadic ovarian, colon, and lung carcinomas. Their results showed that a deficiency of XPC results in the carcinogenesis of human tumors. Transgenic mouse studies have also revealed a predisposition to many types of tumors in XPC gene knockout mice.30 Our data also imply that a deficiency of XPC function promotes the carcinogenesis of HCC induced by AFB1 exposure. These results suggest that the XPC gene plays an important role in the prevention of DNA damage–mediated malignant tumor occurrence.
More than 30 polymorphisms in the XPC gene have been identified, and more and more evidence has shown that the polymorphisms of this gene are associated with the DNA repair capacity, which might modify the risk for malignant tumors such as lung cancer, breast cancer, esophageal cancer, skin cancer, oral cavity cancer, gastric cancer, and head and neck cancer.11, 12 In this study, we analyzed only the XPC Lys939Gln polymorphism because this polymorphism locus localizes at conserved sites of the gene31 and changes the coded amino acids, which may be associated with a decreased DNA repair capacity,12, 13, 15-20, 22, 23, 32 an increased frequency of p53 mutations,33, 34 and increased tumor risk.11, 17 We found that this polymorphism not only increased HCC risk but also correlated with the levels of XPC expression. Supporting our results, recent studies have suggested that this polymorphism modifies the HBV infection–related HCC risk,35 and the dysregulation of XPC expression is highly related to HCC.28
In this study, we stratified the analysis of XPC codon 939 genotypes by AFB1 exposure status. This was done primarily because several previous studies have provided evidence showing that there might be interactive effects of this polymorphism and carcinogens on cancer risk.19, 23 For example, Mechanic et al.19 conducted a hospital-based case-control study (including 2311 cases and 2022 controls) to elucidate whether XPC codon 939 Gln alleles modify the risk for breast cancer associated with smoking. They found some multiplicatively interactive effects of the XPC polymorphism and smoking status on the risk of breast cancer. Zhou et al.23 also found a statistically significant interaction between this polymorphism and environmental carcinogen exposure with respect to esophageal cancer in another Chinese population, the Hebei population (OR = 2.05, 95% CI = 1.15-3.66). Our data not only support the aforementioned studies but also show positively modified effects of the XPC codon 939 Gln alleles on HCC carcinogenesis induced by AFB1 exposure. Interestingly, this polymorphism is associated with shorter survival times and a higher risk of dying from HCC, especially under the condition of high AFB1 exposure.
These results suggest that the XPC Lys939Gln polymorphism may alter the normal protein function and consequently may be associated with a reduction of the DNA repair capacity and the dysregulation of expression levels. The DNA damage induced by AFB1 cannot be repaired effectively and consequently may cause genic mutations (e.g., p53) and hepatocellular canceration. Thus, the XPC Lys939Gln polymorphism may play a role in the carcinogenetic pathway of AFB1 exposure–related HCC for Guangxi patients. In addition, we found some evidence of XPC-XPD interactive effects on HCC risk, possibly because this gene-gene interaction results in a more obvious decrease in the NER capacity and consequently correlates with a higher risk for HCC.
When we are investigating the relationship between genetic polymorphisms and AFB1-related HCC, it is important to establish effective methods for measuring the AFB1 exposure status, for avoiding the effects of confounders, and for collecting sufficiently large samples for gene-environment interaction analysis.24 In this study, we evaluated the AFB1 exposure levels according to AFB1 DNA adduct levels of DNA samples from all subjects' peripheral blood leukocytes for the following reasons: DNA samples of liver tissue were impossible to obtain from the controls, but according to our previous study,7 AFB1 DNA adduct levels of HCC cancerous tissue, although higher, are positively and linearly related to peripheral blood leukocyte adduct levels. Therefore, it was feasible to elucidate the AFB1 exposure status by means of an analysis of AFB1 DNA adduct levels of peripheral blood leukocytes in the case-control study.
In this study, the effects of possible confounders, such as HBV and HCV infection status, were controlled with an individually matched design. Actually, no significant interactive effects were found in the stratified analysis, and this implied that these factors should be effectually manipulated and not modify the correlation between the XPC codon 939 polymorphism and HCC related to AFB1 exposure.
To the best of our knowledge, this is the first report investigating an association between XPC codon 939 polymorphisms and the risk and prognosis of HCC in Guangxi patients. We have found evidence suggesting that the genotypes of XPC with codon 939 Gln alleles may be correlated with increased risk and poor prognosis for AFB1-related HCC, and the NER pathway may play an important role in the mechanism of action of this genotoxin. However, there were several limitations to our study. Potential selection bias might have occurred because the selection of control subjects in our study was hospital-based. There may have been a biased distribution of liver disease severity (e.g., the HBV DNA level). The increased risk with AFB1 exposure status seen in this study was probably underestimated because the liver disease itself may affect the metabolism of AFB1 and modify the levels of AFB1 DNA adducts. Despite the analysis of the XPC codon 939 polymorphism, we did not analyze other polymorphisms of this gene11, 12 possibly able to modify the risk of AFB1 for HCC. Therefore, more genes deserve further elucidation based on a large sample and the combination of genes and AFB1 exposure.
The authors thank Dr. Qiu-Xiang Liang, Dr. Yun Yi, and Dr. Min-Fa Wang for sample collection and management and Dr. Yong-Zhi Huang and Dr. Hua Huang for molecular biochemical techniques. They also thank all members of the Department of Medical Testing and Infection Control, Affiliated Hospital of Youjiang Medical College for Nationalities, for their help.