XPA gene polymorphisms and risk of neuroblastoma in Chinese children: a two-center case-control study

Abstract

Neuroblastoma is a malignant tumor arising from the developing sympathetic nervous system, which mainly affects children. Variations in XPA gene have been shown to confer cancer susceptibility. However, no investigation has been reported regarding the association between XPA polymorphisms and neuroblastoma risk. This study was conducted to measure the association of XPA polymorphisms with neuroblastoma susceptibility in Chinese children. In this hospital-based case-control study with 393 cases and 812 controls, we genotyped two polymorphisms (rs1800975 T>C, and rs3176752 G>T) in XPA gene to access their contributions to neuroblastoma risk by TaqMan methods. The strength of the association with neuroblastoma risk was estimated by odds ratios (ORs) and 95% confidence intervals (CIs). No single polymorphism was found to predispose to neuroblastoma susceptibility. When risk genotypes were combined, we found that carriers of 1-2 risk genotypes had significantly increased neuroblastoma risk (adjusted OR=1.28; 95% CI=1.001-1.64, P=0.049), when compared to non-carriers. Stratification analysis by age, gender, sites of origin and clinical stages failed to show any significant association. Our study provides cues that XPA gene polymorphisms may exert a weak effect in neuroblastoma risk. This finding needs further validations by larger sample size studies.

Introduction

Neuroblastoma, a solid cancer originating from neural crest progenitor cells, is one of the most common tumors in children ¹. The median age at diagnosis of neuroblastoma is about 17 months. Neuroblastoma accounts for approximately 10% of all pediatric cancers, yet its proportion of all pediatric cancer mortality accounts for up to 15% ². The incidence rate of neuroblastoma in China is about 1 out of 13,000 ³. In contrast, America witnesses is much higher, nearly 1 out of 7000 ⁴. Neuroblastoma is characterized by high heterogeneity in its clinical phenotypes and symptoms ⁵. Based on the clinical, pathologic, and prognostic factors, neuroblastoma cases are generally classified into low-, intermediate-, high- and ultra-high-risk groups ⁶. Among them, high- and ultra-high-risk groups constitute around 50% of all cases. The outcome of these high- and ultra-high-risk groups are poor, with less than 40% cases achieving long-term survival ¹. Poor prognosis was mainly attributed to the widespread metastasis of tumor at the time of diagnosis ^{7, 8}.

Approximately 1-2% of all cases of neuroblastoma are familial ⁵. Using large-scale genetic linkage studies, several candidate predisposition chromosomal regions (2p, 12p, and 16p) have been found to associate with familial neuroblastoma risk ^{9, 10}. These findings further led to the discovery of certain gene mutations. Most of the familial cases have highly penetrant mutations in PHOX2B ^{11, 12} and ALK gene ¹³. However, the underlying etiology of sporadic neuroblastoma, the most common form, remains to be elucidated. Although environmental factors such as the conception and pregnancy exposures are potential risk of sporadic neuroblastoma, direct link between has not yet been established ^{14, 15}.

Expanding body of evidence has suggested that genetic factors and genetic-environmental interactions could contribute to the susceptibility of neuroblastoma. Recent genome-wide association studies (GWASs) have identified several neuroblastoma susceptibility loci, including single nucleotide polymorphisms (SNPs) in HACE1, LIN28B, BARD1, CASC15, TP53, and LMO1 ^16-21. The association of neuroblastoma risk with LMO1, BARD1, LIN28B, and TP53 polymorphisms have been confirmed in many replication case-control studies ^22-26. Another powerful way to investigate the role of gene polymorphisms in cancer risk is to analyze the genetic linkage between potential functional SNPs in core genes and cancer susceptibility. This method could overcome the limitation of GWAS, as most of the GWAS-identified SNPs only have modest risk effects due to the adoption of restricted P-value (1 × 10⁵) ²⁷. Therefore, additional neuroblastoma susceptibility loci or gene polymorphisms are warranted to be identified.

The stability of cellular functions and genomic integrity is primarily maintained by DNA repair systems ²⁸. Among the various DNA repair systems, the nucleotide excision repair (NER) pathway is responsible for excising bulky DNA lesions ²⁹. A myriad of evidences have proven that defects in NER pathway might weaken the DNA repair capacity and thus predispose to cancer risks ³⁰. The NER process mainly consists of four steps: damage recognition, unwinding of the DNA, removal of the damage, and ligation ^{31, 32}. Eight critical genes (XPA-XPG and ERCC1) participate in the repair process and coordinately maintain genomic integrity ³³. XPA (xeroderma pigmentosum complementary group A) gene encodes a zinc finger DNA binding protein. This DNA binding-protein, XPA, is essential in organizing the DNA damage site and then functioning as a scaffold to excise lesions ^{34, 35}. Owing to the important role of XPA protein in NER pathway, the association between XPA polymorphisms and cancer risk remains a hot spot of research. By now, several epidemiological studies have been performed to determine of the association with the risk of cancer, including colorectal cancer ³⁶, laryngeal cancer ³⁷, lung cancer ³⁸, and B-cell lymphoma ³⁹.

Because of the universal importance of XPA in cancer, we conducted a case-control study to determine whether the SNPs in XPA gene could predispose to neuroblastoma in Chinese population.

Materials and methods

Study subjects

Totally, 393 cases with neuroblastoma and 812 healthy controls were recruited in this hospital-based epidemiological study ^40-42. To be specific, 275 cases and 531 controls were from Guangzhou Women and Children's Medical Center and the rest were from The First Affiliated Hospital of Zhengzhou University ^{25, 43, 44}. All the participants were unrelated Chinese Han children. Written informed consent was provided before the study by all participants or their guardians. The study protocols obtained approval from the Institutional Review Board of Guangzhou Women and Children's Medical Center and the Institutional Review Board of The First Affiliated Hospital of Zhengzhou University.

SNP selection and genotyping

We chose potentially functional polymorphisms of XPA gene from dbSNP database (http://www.ncbi.nlm.nih.gov/) and an online tool, SNPinfo (http:// snpinfo.niehs.nih.gov/) using the selection criteria described in our published studies ⁴⁵. Two SNPs (rs1800975 T>C, and rs3176752 G>T) in the XPA gene were selected in the final analysis. In them, rs1800975 T>C might affect transcription activity, while rs3176752 G>T might affect the microRNA binding site activity. Genomic DNA was extracted from peripheral blood donated by all the included subjects using a TIANamp Blood DNA Kit (TianGen Biotech Co. Ltd., Beijing, China). TaqMan real-time PCR was performed to genotype the selected SNPs. The details of genotyping protocol could be obtained in previous studies ^46-49. As a quality control, 10% of the random samples were carried out twice. A concordance rate of 100% was got for all duplicate sets.

Statistical analysis

First, deviation from Hardy-Weinberg equilibrium (HWE) of the selected SNPs among controls was assessed by goodness-of-fit χ² test. Two-sided chi-squared test was used to identify the difference of the demographic variables and allele frequencies between the cases and controls. Logistic regression analysis, served to provide odds ratios (ORs) and 95% confidence intervals (CIs), was adopted to estimate the association between XPA SNPs and neuroblastoma risk. Significance was achieved when P values were less than 0.05. All statistical analyses were performed using the version 9.4 SAS statistical package (SAS Institute, Cary, NC).

Results

Correlation of XPA gene polymorphisms with neuroblastoma susceptibility

The detailed characteristics of all the cases and controls could be found in our previously published articles ^40-42. In brief, no statistically significant differences were found between cases and controls regarding age (P>0.05) and gender (P>0.05). According to the INSS criteria, 69 (17.56%), 93 (23.66%), 68 (17.30%), 143 (36.39%), and 11 (2.80%) patients were diagnosed with clinical stages I, II, III, IV, and 4s disease, respectively. In terms of tumor sites, 153 (38.93%) neuroblastomas occurred in the adrenal glands, 109 (27.74%) in the mediastinum, 87 (22.14%) in retroperitoneal regions, and 36 (9.16%) in other regions. To evaluate the relationship between these two SNPs as well as the combined risk genotypes of XPA gene (rs1800975 T>C, rs3176752 G>T) and risk of neuroblastoma, crude and adjusted ORs and their 95% CIs were calculated. The main results of the above calculations were listed in Table 1. In general, we failed to detect any statistically significant differences in the genotype frequencies distribution of rs1800975 T>C or rs3176752 G>T polymorphism. Moreover, none of the two polymorphisms were significantly associated with neuroblastoma risk. The allele frequencies for controls in both rs1800975 and rs3176752 genotypes were in Hardy-Weinberg equilibrium, with P values equal to 0.068 and 0.416, respectively. Further analysis of the combined effect of risk genotypes revealed that 1-2 combined risk genotypes was markedly associated with an increased risk of neuroblastoma, compared with those without any risk genotypes (adjusted OR=1.28; 95% CI=1.001-1.64, P=0.049).

Table 1.

Logistic regression analysis for the association between XPA gene polymorphisms and neuroblastoma susceptibility

Genotype	Cases (N=393)	Controls (N=812)	P a	Crude OR (95% CI)	P	Adjusted OR (95% CI) b	P b
rs1800975 (HWE=0.068)
TT	111 (28.24)	191 (23.52)		1.00		1.00
TC	197 (50.13)	432 (53.20)		0.79 (0.59-1.05)	0.099	0.79 (0.59-1.05)	0.101
CC	85 (21.63)	189 (23.28)		0.77 (0.55-1.10)	0.147	0.78 (0.55-1.10)	0.150
Additive			0.207	0.88 (0.74-1.04)	0.134	0.88 (0.74-1.04)	0.137
Dominant	282 (71.76)	621 (76.48)	0.076	0.78 (0.60-1.03)	0.077	0.78 (0.60-1.03)	0.079
Recessive	308 (78.37)	623 (76.72)	0.522	0.91 (0.68-1.22)	0.523	0.91 (0.68-1.22)	0.527
rs3176752 (HWE=0.416)
GG	301 (76.59)	639 (78.69)		1.00		1.00
GT	83 (21.12)	160 (19.70)		1.10 (0.82-1.48)	0.526	1.10 (0.82-1.49)	0.516
TT	9 (2.29)	13 (1.60)		1.47 (0.62-3.48)	0.381	1.47 (0.62-3.48)	0.382
Additive			0.576	1.13 (0.88-1.46)	0.331	1.14 (0.88-1.47)	0.325
Dominant	92 (23.41)	173 (21.31)	0.408	1.13 (0.85-1.51)	0.409	1.13 (0.85-1.51)	0.400
Recessive	384 (97.71)	799 (98.40)	0.402	1.44 (0.61-3.40)	0.405	1.44 (0.61-3.40)	0.407
Combine risk genotypes c
0 d	232 (59.03)	527 (64.90)	0.132	1.00		1.00
1 d	119 (30.28)	206 (25.37)		1.31 (1.00-1.73)	0.052	1.31 (1.00-1.72)	0.054
2 d	42 (10.69)	79 (9.73)		1.21 (0.81-1.81)	0.361	1.21 (0.81-1.82)	0.354
0	232 (59.03)	527 (64.90)		1.00		1.00
1-2	161 (40.97)	285 (35.10)	0.048	1.28 (1.002-1.64)	0.048	1.28 (1.001-1.64)	0.049

^a χ² test for genotype distributions between neuroblastoma cases and controls.

^b Adjusted for age and gender.

^c Risk genotypes were rs1800975 TT, and rs3176752 GT/TT.

^d 0 was no risk genotype, 1 was rs1800975 TT or rs3176752 GT/TT, 2 was rs1800975 TT and rs3176752 GT/TT.

Stratification analysis of XPA gene polymorphisms with neuroblastoma susceptibility

We further conducted subgroup analyses by age, gender, tumor sites of origin and clinical stages to evaluate the contributions of the selected polymorphisms and their combined risk genotypes to the risk of neuroblastoma (Table 2). However, the calculated results indicated that there lacks significant difference association between the selected polymorphisms as well as their combined risk genotypes and neuroblastoma risk under any of the evaluated subgroups.

Table 2.

Stratification analysis of XPA gene polymorphisms with neuroblastoma susceptibility

Variables	rs1800975 (cases/controls)		Adjusted OR a	P a	rs3176752 (cases/controls)		Adjusted OR a	P a	Risk genotypes (cases/controls)		Adjusted OR a	P a
	TT	TC/CC	(95% CI)		GG	GT/TT	(95% CI)		0-1	2	(95% CI)
Age, month
≤18	31/64	95/241	0.81 (0.50-1.33)	0.411	93/240	33/65	1.31 (0.81-2.12)	0.273	78/203	48/102	1.23 (0.80-1.89)	0.357
>18	80/127	187/380	0.78 (0.56-1.08)	0.133	208/399	59/108	1.05 (0.73-1.51)	0.782	154/324	113/183	1.31 (0.96-1.77)	0.085
Gender
Females	43/74	125/268	0.81 (0.52-1.24)	0.324	133/268	35/74	0.94 (0.60-1.48)	0.798	104/226	64/116	1.19 (0.81-1.75)	0.364
Males	68/117	157/353	0.78 (0.55-1.11)	0.169	168/371	57/99	1.27 (0.87-1.84)	0.213	128/301	97/169	1.33 (0.96-1.84)	0.086
Sites of origin
Adrenal gland	43/191	110/621	0.80 (0.54-1.19)	0.269	119/639	34/173	1.06 (0.70-1.61)	0.777	88/527	65/285	1.35 (0.95-1.92)	0.096
Retroperitoneal	24/191	63/621	0.80 (0.49-1.33)	0.392	68/639	19/173	1.02 (0.60-1.75)	0.931	53/527	34/285	1.19 (0.75-1.87)	0.461
Mediastinum	32/191	77/621	0.73 (0.47-1.13)	0.158	80/639	29/173	1.33 (0.84-2.10)	0.226	63/527	46/285	1.36 (0.91-2.05)	0.138
Others	10/191	26/621	0.79 (0.37-1.66)	0.528	27/639	9/173	1.22 (0.56-2.65)	0.613	22/527	14/285	1.19 (0.60-2.36)	0.624
Clinical stages
I+II+4s	44/191	118/621	0.81 (0.55-1.19)	0.297	119/639	43/173	1.33 (0.90-1.96)	0.150	95/527	67/285	1.32 (0.93-1.86)	0.115
III+IV	64/191	147/621	0.72 (0.51-1.01)	0.055	167/639	44/173	0.98 (0.68-1.43)	0.929	123/527	88/285	1.31 (0.96-1.79)	0.089

^a Adjusted for age and gender without the corresponding stratification factor.

Haplotype analysis

As shown in Table 3, we did not find any haplotype can influence neuroblastoma susceptibility significantly.

Table 3.

Association between inferred haplotypes of XPA gene and neuroblastoma susceptibility

Haplotypes a	Cases (N=786)	Controls (N=1624)	Crude OR (95% CI)	P	Adjusted OR b (95% CI)	P b
CG	317 (40.33)	716 (44.09)	1.00		1.00
CT	50 (6.36)	94 (5.79)	1.20 (0.83-1.74)	0.328	1.20 (0.83-1.74)	0.328
TG	368 (46.82)	722 (44.46)	1.15 (0.96-1.38)	0.130	1.15 (0.96-1.38)	0.134
TT	51 (6.49)	92 (5.67)	1.25 (0.87-1.81)	0.230	1.26 (0.87-1.81)	0.226

^a The haplotypes order were rs1800975 and rs3176752.

^b Obtained in logistic regression models with adjustment for age and gender.

Discussion

In order to identify the association between XPA SNPs and neuroblastoma risk, we conducted a two-center case-control study by recruiting 393 neuroblastoma patients and 812 healthy control subjects from Chinese population. Our results indicated that variations in XPA gene had no significant association with neuroblastoma risk. To the best of our knowledge, this is the first study investigating the associations in Chinese children.

XPA gene is mapped to chromosome 9p22.3. It encodes a relatively small 273-residue protein without enzymatic activity. Alternatively, XPA functions as a scaffold by interacting with many other NER proteins such as replication protein A, transcription factor IIH and excision repair cross complementing group 1-xeroderma pigmentosum group F protein complex ⁵⁰. Through interacting with these proteins, XPA protein facilitates the formation of the assembly and structural organization of human NER incision complexes ⁵¹. Current evidence indicates that mutations in XPA gene impairs normal function of NER ⁵¹. Thus, it is biologically reasonable that functional XPA gene variants may influence the DNA repair capacity, and thereby modify cancer risk. A vast number of studies has been performed to investigate the association of XPA gene SNPs with cancer risk. For example, the rs1800975 T>C polymerphism of XPA gene has been reported to be associated with increased risk of colorectal cancer in a study conducted in Poland ³⁶. Another study conducted in Norway indicated that XPA gene rs1800975 T>C was associated with a significant reduction in the risk of lung cancer ⁵². The discrepancy results suggested that the same polymorphism might have different roles in cancer susceptibility depending on cancer sites and ethnicities, not to mention the different polymorphisms in candidate gene.

Herein, for the first time we explored whether XPA gene SNPs could affect neuroblastoma susceptibility in Chinese children. Unexpectedly, neither of the two polymorphisms of XPA gene was associated with neuroblastoma risk, in the overall analysis and stratified analysis. The null association might be partially attributed to the limited statistical power caused by relatively small sample size. Moreover, the low-penetrance of single polymorphism might also account for such a null association. Neuroblastoma is a multifactorial and complex disease resulting from interplay between multiple genetic and environmental factors ^{1, 5, 53}. Previous study has demonstrated that polymorphisms in individual genes might not have enough impact on the risk of cancer ⁵⁴. Thus, it would be expected that the combined SNPs might have greater effects. Indeed, combined analysis revealed that 1-2 combined risk genotypes was markedly associated with an increased risk of neuroblastoma, compared with 0 risk genotypes. Consistent with our results, Tse et al. ⁵⁵, found that compared with one variant allele alone, the combined four NER SNPs could significantly increase risk of esophageal adenocarcinoma. Similar variant-dosage effect was also observed in our former case-control study in other NER genes ⁵⁶.

Although this study has its own merits, limitation accompanies. First, as a hospital-based case-control study, selection bias is inevitable. Second, the sample size of this case-control study is relatively small, especially for the stratification analysis. As a result, the statistical power was compromised. Thus, these findings call for replication studies in larger sample size. Third, the number of the analyzed SNPs is limited. Fourth, neuroblastoma is a heterogeneous disease with complex etiology; thus, genetic analysis alone is far more enough to elucidate its etiology. Yet due to the nature of retrospective investigations, we failed to obtain other risk exposing factors, such as paternal exposures, living environment, and dietary intake. Further studies including environmental factors analysis and gene-environment interaction analysis are warranted. Fifth, the conclusion obtained from this study could not be directly extrapolated to other ethnicities, as all the included subjects were restricted to unrelated Chinese Han ethnicity.

In summary, we first investigated the association of XPA gene polymorphisms with neuroblastoma risk. Our results indicated that no single XPA gene polymorphism could influence neuroblastoma risk. Ongoing epidemiological studies with larger samples from different ethnicities are needed to further elucidate the role of XPA gene polymorphisms in neuroblastoma tumorigenesis.

Abbreviations

GWAS

genome-wide association study

SNP

single nucleotide polymorphism

NER

nucleotide excision repair

XPA

xeroderma pigmentosum complementary group A, HWE, Hardy-Weinberg equilibrium

OR

odds ratio

CI

confidence interval.