Relevance of XPD polymorphisms to neuroblastoma risk in Chinese children: a four-center case-control study

Neuroblastoma is a lethal tumor that commonly occurs in children. Polymorphisms in XPD reportedly influence risk for several types of cancer, though their roles in neuroblastoma remain unclear. Here we endeavored to determine the relevance of XPD gene polymorphisms and neuroblastoma susceptibility in Chinese children genotyping three XPD polymorphisms (rs3810366, rs13181 and rs238406) in 505 cases and 1070 controls and assessing their contributions to neuroblastoma risk. Overall, we detected no significant association between any single XPD genotype and neuroblastoma risk. When risk genotypes were combined, however, we found that patients with 2-3 risk genotypes were more likely to develop neuroblastoma (adjusted odds ratio =1.31; 95% confidence interval =1.06-1.62, P=0.013) than those with 0-1 risk genotypes. Stratification analysis of rs3810366 revealed significant relationships between the subgroups age ≤18 months and clinical stage I+II+4s and neuroblastoma risk. Moreover, the presence of 2-3 risk genotypes was significantly associated with increased neuroblastoma risk in the subgroups age ≤18 months, male, tumor originated from others, and clinical stage I+II+4s. Our findings provide novel insight into the genetic underpinnings of neuroblastoma and demonstrate that XPD polymorphisms may have a cumulative effect on neuroblastoma risk.


INTRODUCTION
Neuroblastoma, a solid tumor of the sympathetic nervous system, remains the most commonly occurring lethal cancer among children [1], accounting for 7%-10% of all cancers in children under 15 years of age [2,3]. The prevalence of neuroblastoma in the United States is about 1 in 7000 [4], but is only about 1 in AGING 13,000 in China [5]. Neuroblastomas have been classified into low-risk, intermediate-risk and high-risk groups [6]. Among them, the high-risk group accounts for 50% of all neuroblastoma patients. The 5-year survival rate among this high-risk group remains less than 35%, despite administration of intense multimodal therapy [7,8]. This low cure rate may be attributable to the presence of widespread tumor metastasis at the time of diagnosis [9].
The integrity and stability of the human genome is maintained primarily by DNA repair systems [23], among which the nucleotide excision repair (NER) pathway is essential in eliminating DNA damage caused by both exogenous and endogenous factors [24]. Dysfunction of the NER pathway could lead to failure to repair genomic defects, thereby increasing cancer risk [25][26][27]. Several critical genes (ERCC1, XPA, XPB, XPC, XPD, XPE, XPF and XPG) participate in the NER process and coordinately function to maintain genomic integrity [28]. XPD (xeroderma pigmentosum complementary group D), also known as ERCC2 (excision repair cross-complementation group 2), encodes an evolutionarily conserved ATP-dependent helicase [29]. This helicase forms a complex with the transcription factor TFIIH to function in basal transcription and nucleotide excision repair [30]. The importance of XPD to the NER pathway has prompted AGING several case-control studies to assess the effect of XPD polymorphisms on the risk of such cancers as nasopharyngeal carcinoma [31], renal cell carcinoma [32], prostate cancer [33], esophageal squamous cell carcinoma [34], and breast cancer [35]. However, the effect of XPD polymorphisms on neuroblastoma risk has not yet been studied. To address that issue, we conducted a four-center case-control study analyzing the relationship between three XPD polymorphisms and neuroblastoma risk in the Chinese population.

Correlation between XPD polymorphisms and neuroblastoma susceptibility
A total of 505 patients and 1070 healthy controls were successfully genotyped in our study. The demographic characteristics of the tested subjects can be found in our previously published articles [36][37][38][39] and in Supplemental Table 1. The genotype frequencies for the three selected XPD polymorphisms (rs3810366, rs13181, rs238406) in all subjects and in selected subject groups and their contributions to neuroblastoma risk are summarized in Table 1 and Supplemental Table  2. No single XPD polymorphism was significantly associated with neuroblastoma risk in any genetic model evaluated. On the other hand, participants harboring 2 or 3 risk genotypes were more likely to develop neuroblastoma (adjusted OR=1.31; 95% CI=1.06-1.62, P=0.013) than those with 0 or 1 risk genotype.

Stratification analysis of XPD polymorphisms and neuroblastoma susceptibility
To assess the correlations between XPD polymorphisms and neuroblastoma risk in particular subgroups of healthy individuals and neuroblastoma patients, we conducted analyses after stratifying based on age, gender, sites of origin, and clinical stages. We found a significant association between the rs3810366 GC/CC genotypes and neuroblastoma risk  (Table 3). However, no XPD haplotype was associated with neuroblastoma susceptibility when the most common haplotype (GTG) was used as the reference.

DISCUSSION
To identify XPD polymorphisms influencing neuroblastoma tumorigenesis, we performed a hospitalbased case-control study involving a total of 505 neuroblastoma patients and 1070 healthy control subjects. All participants were Chinese. To our knowledge, this study is the first investigation assessing the association between XPD polymorphisms and neuroblastoma risk in Chinese children.
XPD is located on chromosome 19p13.3 [40] and is composed of 23 exons encoding an evolutionarily conserved ATP-dependent helicase. This helicase is responsible for DNA unwinding and transcription initiation. More than 100 mutations have been mapped in XPD [41], which could potentially affect the helicase activity of the encoded protein, thereby impeding normal NER function and leading to increased cancer risk [29,42]. Several XPD polymorphisms are reportedly associated with cancer risk [43][44][45]. Among these, rs1799793 (Asp312Asn) and rs13181 (Lys751Gln) in the XPD coding region are the two most widely investigated. The rs13181 polymorphism at codon 751 in exon 23 is a non-synonymous A>C substitution, which results in an amino acid change from Lys to Gln. It has been demonstrated that this Lys751Gln polymorphism could decrease DNA repair capacity [46]. In a study conducted in Poland with 430 patients and 430 controls, Magdalena et al. [47] found that the Gln/Gln genotype of rs13181 is associated with an increased risk of ovarian cancer. In our earlier study investigation of the association between two XPD polymorphisms (rs238406 and rs13181) and esophageal squamous cell carcinomas risk, we found that rs238406, but not rs13181, was associated with elevated disease risk [34]. The results are somewhat inconclusive, however, since differing relationships between XPD polymorphisms and cancer risk have been reported. These discrepancies may reflect differences in the cancer types, sample sizes, population sources, selection criteria for subjects, and environmental exposures. It is therefore necessary to limit conclusions regarding the contributions of XPD polymorphisms to cancer risk to a particular population and cancer type.
Given the critical role of XPD polymorphisms in cancer risk and the lack of research on their contributions to neuroblastoma risk, we endeavored to assess the association between three XPD polymorphisms and neuroblastoma risk in Chinese children. Unexpectedly, www.aging-us.com 1992 AGING AGING we failed to detect any significant contribution by rs13181 or rs238406 to neuroblastoma risk in the overall analysis or in any of the selected subgroups after stratification. Several factors, including the relatively small sample size, low penetrance of a single polymorphism, and population bias may account for the null association. The etiology of neuroblastoma is complex and subject to heterogenetic influence by a variety of risk factors [15,48]. Although a single XPD polymorphism may have limited impact on neuroblastoma risk, it would be expected that the combined effects of several polymorphisms might bring about more significant findings. Indeed, in the present study, we observed that participants with two or more risk genotypes of these functional XPD polymorphisms were at significantly higher risk of neuroblastoma than those carrying one or fewer risk genotypes. This trend was also observed in earlier studies by ourselves and others [49,50]. These findings are biologically plausible, probably due to the joint effects of multiple functional polymorphisms.
While this study has its merits, it also has several limitations. First, inherent bias could not be excluded, as all the DNA samples were collected in hospitals. Second, although the sample size for the overall analysis was relatively large, after stratification some subgroups were less than 100, which inevitably diminished the statistical power. Third, the included subjects were restricted to unrelated Han Chinese; consequently, the results may not be applicable to other ethnicities. Fourth, only three XPD polymorphisms were selected and analyzed in this study. Other potentially functional XPD polymorphisms may also modify the activity of gene or the encoded helicase and thus should be involved in the ongoing study. Fifth, as neuroblastoma is a heterogeneous disease with a complex etiology, the genetic analysis in the present study only partially elucidated the etiology of neuroblastoma. Potentially important environmental factors such as diet, living environment, and parental exposures should be addressed in the future.
In summary, our findings provide insight into the potential role of XPD polymorphisms in neuroblastoma risk. Our results failed to detect a role for any single XPD polymorphism in neuroblastoma risk. It is anticipated that ongoing epidemiological studies with larger samples and more analysis of confounding factors will provide additional information on the contribution of XPD polymorphisms to neuroblastoma tumorigenesis.

Study subjects
A total of 505 cases and 1070 healthy controls were included in this study. Of those, 429 cases and 884 controls were described in our previous study [39]. The additional 76 cases and 186 controls were from the Second Affiliated Hospital of Xi'an Jiaotong University (Supplemental Table 1). The cases were individuals diagnosed with neuroblastoma, and the controls were recruited from the same hospitals between September 2009 and March 2018. The eligibility criteria for the included subjects were described previously [49,[51][52][53]. Written informed consent was provided by all subjects or their guardians. The study protocols were approved by the Institutional Review Board of each hospital.

Polymorphism selection and genotyping
In brief, we searched for potentially functional candidate SNPs located in the 5'-flanking region, 5' untranslated region, 3' untranslated region, and exon of XPD. The potentially functional XPD polymorphisms AGING were screened from the NCBI dbSNP database (http://www.ncbi.nlm.nih.gov/projects/SNP) and SNPinfo (http://snpinfo.niehs.nih.gov/snpfunc.htm) using previously described criteria [54]. Three polymorphisms (rs3810366 G>C, rs13181 T>G, and rs238406 G>T) in the XPD gene were ultimately selected: rs3810366 G>C, which is located within transcription factor binding sites; rs13181 T>G and rs238406 G>T, which may affect splicing regulation activity; and rs13181 T>G, which may also lead to Lys751Gln alteration. As shown in Supplementary  Figure 1, there was no significant LD (R 2 <0.8) between rs13181 and rs238406 (R 2 =0.026) or between rs13181 and rs3810366 (R 2 =0.001). However, there was a little LD between rs238406 and rs3810366 (R 2 =0.891).
For genotyping, DNA samples were mainly purified from venous blood using a TIANamp Blood DNA Kit (TianGen Biotech Co. Ltd., Beijing, China). Following standard methods, we used TaqMan real-time PCR to genotype the selected polymorphisms. Details of the genotyping protocol are provided elsewhere [54][55][56][57]. To control for result quality, approximately 10% of the samples were randomly selected to perform duplicate analyses. We obtained a concordance rate of 100% for all duplicate sets.

Statistical analysis
We first used a goodness-of-fit χ 2 test to assess whether the selected polymorphisms were in Hardy-Weinberg equilibrium in the controls. The demographic variables and allele frequencies were compared between the cases and controls using a two-sided χ 2 test. The association between XPD polymorphisms and neuroblastoma risk was estimated using logistic regression analysis providing odds ratios (ORs) and 95% confidence intervals (CIs). Values of P<0.05 were considered significant. All statistical analyses were performed using the SAS statistical package (version 9.1, SAS Institute, Cary, NC).

CONFLICTS OF INTEREST
No competing interests to declare.