Association between XRCC3 rs861539(Thr241Met) polymorphism and thyroid cancer risk (a Meta-analysis)

Objective To explore the association between single nucleotide polymorphism (SNP) in the XRCC3 rs861539(Thr241Met) locus and thyroid cancer risk. Methods Studies investigating the association between SNP in the XRCC3 gene and thyroid cancer susceptibility were retrieved from the PubMed, Embase, Web of Science, CNKI (Chinese National Knowledge Infrastructure), WanFang, and CBM (China Biology Medicine) databases. Eligible studies were screened according to inclusion/exclusion criteria and principles of quality evaluation. Meta-analysis was performed using Stata 14.0 software. Odds ratios with their corresponding 95% condence intervals were pooled to assess the association between SNP in the XRCC3 gene rs861539 locus and thyroid cancer susceptibility. Results 10 articles(11 studies) were eligible for this meta-analysis. Meta-analysis results were shown as follows: No signicant association was found between XRCC3 rs861539 polymorphism and thyroid cancer risk in Dominant and Overdominant models (cid:0) Dominant model: CT+TT vs CC, OR=1.231, 95% CI(0.998, 1.474); Overdominant model: CT vs TT+CC, OR=1.05, 95% CI(0.94, 1.18) (cid:0) . Signicant associations were found in Recessive and Allelic models (cid:0) Recessive model: TT vs CC+CT, OR=1.632, 95% CI(1.349, 1.974); Allelic model: T vs C: OR=1.263, 95% CI(1.091, 1.462) (cid:0) . Conclusion The results of this study suggest that the XRCC3 rs861539(Thr241Met) polymorphism may be associated with an increased thyroid cancer risk in overall population, and a tendency for signicantly increased thyroid cancer risk in TT(Met/Met) genotype population.


Introduction
Thyroid cancer(TC) is one of the most common endocrine malignancies in human, which accounts for 3.1% of all new cancer cases and 0.4% of cancer deaths worldwide annually [1] . The incidence of thyroid cancer has been increasing rapidly in recent years, and the largest increase in incidence of all cancers was seen for thyroid cancer in China [2] . The reason might due to the rapid development of imaging detection technologies and increasing awareness of people's health [3] , especially considering that thyroid cancer mortality remained stable at a rate of approximately 0.5 cases per 100000 persons [4] . Thyroid cancers are divided into four main sub-types(papillary, follicular, medullary, and undifferentiated cancers) [5] .To date, the mechanism of thyroid carcinogenesis remains incompletely understood, and the only well-established risk factor for thyroid cancer might be exposure to ionizing radiation [6] . DNA damaging which caused by ionizing radiation may lead to mutations, genomic instability. The XRCC3(X-ray repair cross-complementing group 3) located on chromosome 14q32.3 is structurally and functionally related to the RAD51 gene [7] , which encodes a member of the RecA/Rad51-related protein family involved in homologous recombination to preserve chromosome stability and repair DNA damage caused by endogenous and exogenous factors [8] .Over the past decade, several studies have reported the association regarding XRCC3 rs861539 polymorphism and thyroid cancer risk [9][10][11][12][13][14] , however, the results remained inconclusive due to some potential limitations, such as small sample size, different ethnicity, and phenotypic heterogeneity. Therefore, this meta-analysis was conducted to make this discrepancy clear and to create a comprehensive picture of the association between XRCC3 rs861539 polymorphism and thyroid cancer susceptibility. or rs861539 or X-ray repair cross-complementing group 3 or Thr241Met or C241T or T241M and thyroid carcinoma or thyroid cancer. Furthermore, to retrieve as many articles meeting our criteria as possible, we also investigated the reference literature listed in the articles we had found.

Data extraction
Literature selection was completed by two investigators independently. For con icting evaluations, a third reviewer assessed the articles until an agreement was reached. The following information was recorded for each study: rst author, year of publication, country, ethnicity, number of cases, number of controls, genotype distribution, genotyping methods(all of the data are shown in Table 1).

Assessment of the risk of bias in the included literature
The Newcastle-Ottawa quality assessment Scales (NOS) was used to assess all case-control studies included in our present meta-analysis. The NOS contains 8 items,and the possible scores is ranged from 0 up to 9, a study is considered high quality if it gets more than 5 scores. Finally, the quality of all the studies included in this meta-analysis was acceptable.

Statistical analysis
All statistical analysis were carried out using the STATA version 14.0 software(StataCorp, College Station, TX, USA).χ 2 analysis with a signi cance level of P<0.05 was used to evaluate whether rs861539(Thr241Met) polymorphism distribution of the XRCC3 gene in controls ts HWE (Hardy Weinberg equilibrium). Heterogeneity among the studies in each genetic model was assessed by using Cochran's Q and I 2 statistics.
A P value of <0.05 or I 2 value of >50% was interpreted as having signifcant heterogeneity, Random effect model was used to summarize all the studies. Otherwise, the xed effects model was chosen.We used odds ratios (ORs) and their corresponding 95% con dence intervals (CIs) to evaluate relationships between rs861539(Thr241Met) polymorphism and any predisposition to thyroid cancer. Publication bias was assessed by performing Peters' test. The signi cance of the intercept was determined by the t-test suggested by Peters, where P<0.05 was considered representative of statistically signi cant publication bias. For all analyses, statistically signi cance was assumed at P<0.05, unless otherwise stated.

Retrieval of studies and their characteristics
The initial search strategy resulted in the identi cation of 96 records (PubMed, N = 19; Web of Science, N = 23; Embase, N = 20; CNKI, N = 23; CBM, N = 2; Wanfang, N = 9), of which 85 were excluded after reading the title, abstract, full text. One relevant article was excluded by reason of the low quality evaluation [15] . Finally, 10 articles (11 studies) met the inclusion criteria (the detailed selection process has been illustrated in Fig.   1), consisting of 1 mixed, 5 Asian and 5 Caucasian populations, and these included eligible studies published between 2005 and 2019. Extracted data of eligible studies are summarized in Table 1. The genotype frequency among cases and controls for rs861539 polymorphism are also shown in Table 1. The distributions of genotype in the control group were in HWE for most studies (P > 0.05).  Table 3). The results suggested that no individual study signi cantly affected the pooled ORs in all genetic models.

Publication bias
Publication bias was assessed using Harbord's tests for each genetic model, and there was no evidence of publication bias for all of the genetic models (as is shown in Table 2).

Discussion
The X-ray repair cross-complementing group 3 gene (XRCC3) belongs to a family of genes responsible for repairing DNA double-strand breaks [23] , and it encodes a 346 amino acid polypeptide that participates in DNA double-strand break repair [24] . Among several SNPs identi ed in the XRCC3 gene, the most extensively studied is Thr241Met on exon 7(rs861539), which can in uence the ability to repair DNA [25] . The rs861539(C>T) polymorphism is a nonsynonymous substitution (C→T) resulting in an amino acid change from threonine to methionine at position 241 (Thr241Met) [26,27] . In the past decade, Numerous studies have investigated the potential biological signi cance of the XRCC-3 rs861539(Thr241Met) [28][29][30][31] , and it has been implicated as a causative risk factor for the development of different types of malignancies. The malignancy types included breast cancer [7,32,33] , lung cancer [34] , osteosarcoma [35] , melanoma [36] , thyroid cancer [14,16] . However, the correlation between XRCC3 rs861539 polymorphism and thyroid cancer risk still remained controversial [9,13,16,17] . To derive a more precise estimation, we performed this meta-analysis to investigate the association between rs861539 polymorphism on XRCC3 gene and the risk of thyroid cancer.
In this study based on results of 11 studies containing 1904 thyroid cancer cases and 2960 controls, we observed an association of XRCC3 rs861539 polymorphism with thyroid cancer risk in the overall analysis under recessive and allelic genetic models, and this result was inconsistent with the previous meta-analysis by Wang [37] . indicate that the effect of XRCC3 rs861539 polymorphism on the risk of thyroid cancer is ethnically different and the rs861539 polymorphism has its obvious effect on the development of thyroid cancer in Asians. However, these results above by subgroup analysis were inconsistent with the previous meta-analysis by Lu [11] .
To sum up, our meta-analysis indicates that rs861539 polymorphism may be associated with an increased thyroid cancer risk in overall population, and a tendency for signi cantly increased thyroid cancer risk in TT(Met/Met) genotype population., and this association was obviously exists in Asians.
Individuals with polymorphisms in XRCC3 rs861539 locus have a potential risk of thyroid cancer, but the carcinogenesis of rs861539 polymorphism to thyroid is still not completely understood. It is largely accepted that ionizing radiation would stimulate DNA damage, which would then induct genomic instability, and previous study indicated that variant genotypes of XRCC3 codon241 are associated with increased DNA damage with increased sensitivity to DNA toxic agents [38] .
To our best knowledge, this is the most comprehensive meta-analysis with the largest sample size tried to explore the association between single nucleotide polymorphism (SNP) in the XRCC3 rs861539(Thr241Met) locus and thyroid cancer risk. Our study had the advantage of including higher numbers of cases and controls and assessment of sensitivity analysis. However, there were still some limitations in this meta-analysis. First, these results are based on unadjusted estimates that lack original data from the included studies, and a more precise analysis should be conducted. Second, the controls were not uniformly de ned, and not all the controls from included studies in consistent with Hardy Weinberg equilibrium. Third, Additional functional as well as association studies investigating gene-gene and gene-environment interactions are required to elucidate this issue. Lastly, lack of available data prevented us from performing additional subgroup analyses by age, gender, ionizing radiation and other risk factors. For these limitations, the results should be treated with caution. Certainly, our discovery warrant con rmation by further investigations with larger sample size.