CASC15 gene polymorphisms reduce neuroblastoma risk in Chinese children

In this case-control study, we analyzed the association between three single nucleotide polymorphisms (SNPs) in the CASC15 gene (rs6939340 A>G, rs4712653 T>C, and rs9295536 C>A) and neuroblastoma susceptibility in the Guangdong and Henan populations of China. We genotyped and analyzed 118 cases and 281 control subjects from Henan province and combined them with previously published data from the Guangdong population. In the Henan population, only the rs6939340 G>A variant homozygote AA was associated with decreased neuroblastoma risk [AA vs. GG: adjusted odds ratio (OR) = 0.47, 95% confidence interval (CI) = 0.23-0.98; P=0.045]. All three polymorphisms, individually and in combination, were associated with decreased neuroblastoma susceptibility in the Guangdong population. Moreover, subjects carrying 1-3 of these protective genotypes had lower neuroblastoma susceptibility than non-carriers (adjusted OR=0.65, 95% CI=0.51-0.84, P=0.0007). These results show that all three genetic variants of CASC15 identified in a genome-wide association study (GWAS) decrease neuroblastoma risk in two distinct Chinese populations.


INTRODUCTION
Neuroblastoma is an embryonal cancer that arises from primordial sympathetic neural precursors in infants [1]. The incidence rate of neuroblastoma is about 1 in 7000 in the USA [2] and 7.7 per million in China [3]. Neuroblastoma is the most common solid tumor in childhood and despite marked improvements in its treatment, it still accounts for 15% of the cancer-related mortality in children [4]. Neuroblastoma is characterized by diverse clinical behaviors based on which it is classified into low risk, intermediate risk and high risk groups [5]. Typically, high-risk patients have widely disseminated disease at diagnosis and poor survival rates, whereas most of the low-risk patients spontaneously regress without chemotherapy [6,7]. Nearly 50% of the neuroblastoma patients are diagnosed as high-risk and less than 40% of them survive despite intensive therapies [4,8]. Moreover, survivors face lifelong serious co-existing health issues that affect their social life [9].
Only 1% neuroblastoma cases are familial, and in most cases are associated with ALK gene mutations [10,11]. However, the exact etiology of sporadic neuroblastoma is obscure [1]. Extrinsic risk factors that influence neuroblastoma like exposure to radiation and other environmental factors have not yet been identified. Few children develop neuroblastoma because their parents are exposed to radiation sources, wood dust www.impactjournals.com/oncotarget/ Oncotarget, 2017, Vol. 8, (No. 53), pp: 91343-91349 Research Paper www.impactjournals.com/oncotarget and hydrocarbons [12,13]. Therefore, genetic factors combined with environmental factors can influence neuroblastoma outcomes.
The availability of the human genome sequence and high density single nucleotide polymorphism (SNP) arrays have resulted in genome-wide association studies (GWASs), which shed more light into possible genetic mechanisms of human diseases including cancers [14]. GWAS on neuroblastoma identified several susceptibility loci. SNPs in HSD17B12, DUSP12, IL31RA and DDX4 genes were enriched in the low-risk group of patients [15]. SNPs in BARD1 [16], cancer susceptibility candidate 15 (CASC15) [17], and LMO1 [18] genes were enriched in high-risk neuroblastoma patients. Also, some common variants were associated with neuroblastoma risk, but were not correlated with the disease phenotype [19]. Since its possible to miss relevant risk variants because of stringent threshold parameters in GWAS studies, alternative approaches were used to identify potential variants associated with neuroblastoma including NEFL [20] and TP53 [21]. Most GWAS-identified SNPs in neuroblastoma have been verified in other population samples [22][23][24]. Yet, only four studies have assessed the association of polymorphisms in CASC15 gene with neuroblastoma risk [22][23][24][25].
In our previous study, we reported that CASC15 gene polymorphisms were associated with neuroblastoma risk in Southern Chinese population [25]. In this study, we evaluate the association between CASC15 gene polymorphisms and neuroblastoma risk in two combined cohorts of neuroblastoma patients from Northern China.

Population characteristics
The detailed demographic characteristics of Guangdong population were reported previously [25][26][27]. For Henan population, a summary of demographic characteristics of the 118 cases and 281 controls are shown in Table 1  .0008) were all associated with reduced neuroblastoma risk. When the protective genotypes were combined, the subjects with 1-3 protective genotypes were less likely to develop neuroblastoma than those not carrying the protective genotypes (adjusted OR=0.62, 95% CI=0.46-0.84, P=0.002).

False positive report probability (FPRP) analysis
The prior probabilities ranging from 0.25 to 0.0001 were used to weigh the significant findings (Table 3). Except for rs6939340 AA vs. GG/AG and 2 protective carriers, all the significant findings remained true when we used a FPRP threshold of 0.2 as suggested by Wacholder et al. [29]. www.impactjournals.com/oncotarget

DISCUSSION
In the current case-control study, we investigated the potential association between CASC15 gene polymorphisms and neuroblastoma risk in two distinct populations from China. Given the relative paucity of somatic mutations in high-risk neuroblastoma, the genetic basis of sporadic neuroblastoma still needs to be elucidated. The CASC15 gene, also referred as FLJ22536 or LINC00340 gene spans ~530 kilobases on chromosome 6p22. It was originally identified in silico as a highly active long non-coding RNA (lncRNA) [30]. In 2008, the first GWAS was conducted in the European population to identify common DNA variations that were predisposed to sporadic neuroblastoma [17]. This GWAS consisted of the discovery stage that included 1032 cases and 2043 controls from Europe and the replication stage that included 720 cases and 2128 controls from Northern Europe. Three SNPs were identified in the CASC15 gene that associated with clinically aggressive neuroblastoma. Since then, these SNPs have been investigated in African-American and Italian populations. Latorre et al. studied 390 cases and 2500 controls in an African-American cohort and did not find any association between these 3 CASC15 gene SNPs with neuroblastoma susceptibility [22]. Capasso et al. showed that two CASC15 gene SNPs (rs6939340 A> G and rs4712653 T>C) were risk factors for neuroblastoma by evaluating 370 cases and 809 controls in an Italian cohort [23]. Mike et al. showed that decreased expression of the truncated CASC15-S isoform was associated with more advanced neuroblastoma [31]. We previously conducted a case-control study with 201 cases and 531 controls in southern China and showed that all three SNPs were protective [25]. Moreover, subjects carrying one or more protective genotypes showed lower neuroblastoma susceptibility than non-carriers. However, the findings of our previous study were limited by selection bias and small sample size. Therefore, in this study we investigated the role of the three GWAS-identified SNPs in neuroblastoma risk in the two distinct populations.
The study subjects from Henan showed significant association with neuroblastoma risk only in the rs6939340 G>A variant and not in the rs4712653 C>T and rs9295536 A>C. On the other hand, all three SNPs were associated with decreased neuroblastoma risk in the Guangdong population. Combining the two populations also resulted in similar results. The differences in the relationship between CASC15 gene polymorphisms and neuroblastoma risk among the two different populations might be ascribed to different environmental exposures as well as different genetic backgrounds. We also found that subjects with 1-3 protective genotypes showed decreased neuroblastoma  risk. Further, FPRP analysis excluded the possibility of false positive results in our study, thereby confirming our data. The discrepancy of the role of CASC15 gene polymorphisms in neuroblastoma risk might be attributed to (1) relative small sample sizes in the validation studies and (2) the variation in allele frequencies, linkage disequilibrium patterns, and environmental exposures among different ethnic populations that influence cancer risk [32]. Although this study is the first to verify the genetic role of CASC15 gene polymorphisms in two independent populations in China, it has several limitations. First, although this is the largest case-control study conducted to-date, the sample size was still not large enough. The low incidence rate of neuroblastoma in China that makes it challenging to recruit eligible subjects. Second, only three GWAS-identified SNPs in the CASC15 gene were analyzed and other potential functional SNPs were not tested. Third, the present study focused on the subjects recruited from single hospital rather than from the entire community, which may result in selection bias. Fourth, we only analyzed the genetic factors in neuroblastoma risk, whereas other environmental factors (living environment, dietary intake, and paternal exposures) were not available to conduct gene-environmental interaction analysis.
In summary, our results showed the protective role of three CASC15 gene polymorphisms in neuroblastoma susceptibility in two independent populations in China. Larger and multicenter-based case-control studies are warranted to further confirm the role of CASC15 gene polymorphisms in decreasing neuroblastoma risk and determine the molecular mechanism of action.

Study subjects
This study involved two independent case-control populations. The first was derived from our previous study, which was conducted in Guangdong province [25,26]. The second was conducted in Henan province, which included 118 cases and 281 controls. The recruitment procedure and criteria were as described previously [27,33]. In brief, we identified 118 histologically confirmed neuroblastoma cases among Henan residents from The First Affiliated Hospital of Zhengzhou University between August 2011 and April 2017. We also recruited 281 control subjects from the same hospital during the same period. The case and control subjects were age-and gender matched. The response rate was approximately 80% for cases and 90% for controls enrolled from Henan province. All subjects or their guardians were informed about the project and gave written consent. This study was approved by the Institutional Review Board of The First Affiliated Hospital of Zhengzhou University.

SNP selection and genotyping
Three SNPs, rs6939340 G > A, rs4712653 C > T, rs9295536 A > C in the CASC15 gene were chosen from the previous GWAS study [17]. The genomic DNA was isolated from venous blood samples using TIANamp Blood DNA Kit (TianGen Biotech Co. Ltd., Beijing, China) according to manufacturer's instructions. Genotyping was performed by Taqman real-time PCR as published previously [26,33]. To ensure the accuracy of genotyping results, 10% of randomly selected samples were genotyped by DNA sequencing method. The concordance rate for the quality control samples was 100%.

Statistical analysis
The goodness-of-fit χ 2 test was used to assess if the selected SNPs deviated from Hardy-Weinberg equilibrium among controls. The two-sided chi-squared test was used to compare demographic variables and genotype frequencies of the cases and controls. ORs and their corresponding 95% CIs were computed by unconditional logistic regression analyses with or without adjusting for age and gender. The FPRP analysis was performed as described previously [29]. All reported P values were two sided and a P value < 0.05 was considered statistically significant. The SAS statistical package (version 9.1; SAS Institute, Cary, NC) was used to perform all statistical analysis.