肿瘤防治研究  2015, Vol. 42 Issue (8): 824-828
本刊由国家卫生和计划生育委员会主管,湖北省卫生厅、中国抗癌协会、湖北省肿瘤医院主办。
0

文章信息

马利敏,阮林海,刘洪超,冯艳铭,杨海平. 2015.
MA Limin, RUAN Linhai, LIU Hongchao, FENG Yanming, YANG Haiping. 2015.
MTRR基因A66G多态性与儿童急性淋巴细胞白血病关系的Meta分析
MTRR Gene A66G Polymorphism and Pediatric Acute Lymphoblastic Leukemia: A Meta-analysis
肿瘤防治研究, 2015, 42(08): 824-828
Cancer Research on Prevention and Treatment, 2015, 42(08): 824-828
http://www.zlfzyj.com/CN/10.3971/j.issn.1000-8578.2015.08.016

文章历史

收稿日期:2014-11-24
修回日期:2015-04-09
MTRR基因A66G多态性与儿童急性淋巴细胞白血病关系的Meta分析
马利敏1, 阮林海1, 刘洪超2, 冯艳铭2, 杨海平1    
1. 471003 洛阳,河南科技大学第一附属医院血液科;
2. 471003 洛阳, 河南科技大学医学院
摘要目的 评估甲硫氨酸合成酶还原酶(MTRR)基因A66G多态性与儿童急性淋巴细胞白血病(ALL)发生风险的关系。方法 全面检索PubMed、Elsevier、Embase、中文期刊全文数据库(CNKI)和万方数据库,收集探索MTRR基因A66G多态性与儿童ALL发生关系的病例对照研究,纳入符合入选标准的文献并评估其质量。优势比(ORs)及95%可信区间(CIs)评估关联强度。应用RevMan 5.2软件对纳入研究进行异质性检验和效应值合并,漏斗图评估发表性偏倚,敏感性分析采用逐一排除的方法以评估结果的稳定性。结果 共纳入7篇文献,包括儿童ALL患者2 326例,对照3 090例。异质性检验结果表明纳入研究间无显著异质性,采用固定效应模型合并数据。Meta分析结果示,在整体人群纯合子模型和显性模型发现MTRR A66G多态性与儿童ALL风险有关联(GG vs.AA:OR=0.81,95% CI:0.69~0.95,P=0.009;AG+GG vs.AA:OR=0.87,95% CI:0.77~0.98,P=0.03);根据种族进行亚组分析时在白种人群中发现显著性关联(AG vs.AA:OR=0.84,95% CI:0.72~0.99,P=0.04;GG vs.AA:OR=0.79,95% CI:0.66~0.95,P=0.01;AG+GG vs.AA:OR=0.82,95% CI:0.71~0.96,P=0.01)。漏斗图未检测出显著性发表性偏倚,敏感性分析表明结果稳定可靠。结论 目前Meta分析表明MTRR基因A66G多态性与儿童ALL发生风险存在关联,尤其在白种人群。
关键词甲硫氨酸合成酶还原酶     基因多态性     急性淋巴细胞白血病     风险     Meta分析    
MTRR Gene A66G Polymorphism and Pediatric Acute Lymphoblastic Leukemia: A Meta-analysis
MA Limin1, RUAN Linhai1, LIU Hongchao2, FENG Yanming2, YANG Haiping1    
1. Department of Hematology, The First Affiliated Hospital, He’nan University of Science and Technology, Luoyang 471003, China;
2. Medical College, He’nan University of Science and Technology, Luoyang 471003, China
AbstractObjective To evaluate the relationship between methionine synthase reductase (MTRR) A66Ggenetic polymorphism and the risk of pediatric acute lymphoblastic leukemia (ALL).Methods Relevantliteratures were extensively searched in PubMed,Elsevier,Embase,China National Knowledge Infrastructureand Wanfang Databases for collecting the case-control studies investigating the relationship between MTRRA66G genetic polymorphism and pediatric ALL.Odds ratios (ORs) with 95% confidence intervals (CIs) wereapplied to assess the strength of association.The RevMan 5.2 software was applied for heterogeneity testand combined ORs and their 95% CIs calculation.Publication bias was assessed through funnel plot and sensitivity analysis was performed by sequential remove individual studies to assess the stability of the results.Results Seven studies bearing 2,326 cases and 3,090 controls met the inclusion criteria and were included inthe Meta-analysis.There was no significant heterogeneity among the included studies and fixed-effects modelwas applied to combine the data.The results suggested that there was significant association between MTRRA66G polymorphism and pediatric ALL risk in overall comparisons under homozygote and dominant geneticmodels (GG vs.AA:OR=0.81,95% CI:0.69-0.95,P=0.009;AG+GG vs.AA:OR=0.87,95% CI:0.77-0.98,P=0.03).In the subgroup analysis by ethnicity,significant association was found in Caucasians (AG vs.AA:OR=0.84,95% CI:0.72-0.99,P=0.04;GG vs.AA:OR=0.79,95% CI:0.66-0.95,P=0.01;AG+GG vs.AA:OR=0.82,95% CI:0.71-0.96,P=0.01).No significant publication bias was detected by funnel plot and sensitivity analysis suggested the robustness of the results.Conclusion The present Meta-analysis suggests that MTRR A66G polymorphism is associated with pediatric ALL risk,especially in Caucasian populations.
Key words: Methionine synthase reductase(MTRR)     Genetic polymorphism     Acute lymphoblastic leukemia(ALL)     Risk     Meta-analysis    
0 引言

急性淋巴细胞白血病(acute lymphoblastic leukemia,ALL)是儿童时期最常见的恶性肿瘤,约占儿童肿瘤的30%[1]。尽管儿童ALL的临床表现、病理特点和免疫表型等已经清楚,然而其病因学尚未阐明,通常认为其发生是环境暴露因素与内在遗传易感因素相互作用的结果[2, 3, 4]。临床流行病学研究表明孕妇产前补充叶酸能够降低儿童ALL发生风险[5]。叶酸代谢网络是由一系列相互关联的生物学反应组成,不仅为DNA甲基化提供主要的甲基供体,也为DNA合成供给甲基基团,因此叶酸代谢连接了遗传学与表观遗传学[6]。甲硫氨酸合成酶还原酶(methionine synthase reductase,MTRR)为叶酸代谢网络的关键酶之一,其催化甲基钴胺再生,作为甲硫氨酸合成酶(methioninesynthase,MTR)的辅助因子而维持MTR的活性形式,因此MTRR在叶酸和钴胺素依赖的同型半胱氨酸重新甲基化为甲硫氨酸过程中发挥着重要作用[7]MTRR基因在人群中的分布呈多态性,其中A66G单核苷酸多态性最为常见,该多态性引起甲硫氨酸被异亮氨酸替代,影响了MTRR与甲硫氨酸合成酶的相互作用[8, 9],导致MTR酶活性降低,改变机体内同型半胱氨酸水平[10, 11],进而导致心血管疾病、畸形和肿瘤等疾病的发生[12, 13, 14]。目前已有关于MTRR基因A66G多态性与儿童ALL发生风险关系的研究,但各研究结果间存在一定争议。Meta分析对于存在差异的不同研究,能够汇总多项研究结果,增大样本量,提高效应值估计精度,定量分析合并效应量,成为循证医学中有力的研究工具。本研究采用Meta分析评估MTRR基因A66G多态性与儿童ALL发生风险的关系,为儿童ALL的早期诊断和筛选易感人群提供理论依据。

1 资料与方法 1.1 文献检索

计算机检索PubMed、Elsevier、Embase、中文期刊全文数据库(CNKI)和万方数据库,并辅以附加检索和手工检索,收集探索MTRR基因A66G多态性与儿童ALL发生关系的文献资料。检索使用的中文主题词或关键词为:"蛋氨酸合成酶还原酶"或"甲硫氨酸合成酶还原酶","多态性" ,"白血病","儿童" ;英文主题词或关键词为: "methionine synthase reductase" or "MTRR" ,polymorphism "or" variant "or" mutant "," childhoodor "children" or "pediatric" ,"leukemia "or" acute leukemia"。最后检索时间为2013年11月,检索语种不限。各纳入文献有用的参考文献亦作为本研究入选文献。两位系统评价员独立进行文献检索与筛选,如有争议则协商决定。

1.2 文献纳入标准

纳入本次Meta分析的文献均符合以下标准:(1)原始资料为已公开发表的文献;(2)评估MTRR基因A66G多态性与儿童ALL发生关系的病例对照研究;(3)病例组按照儿童ALL通用诊断标准诊断明确;(4)病例组和对照组各基因型例数表达明确,当纳入研究未提供充足数据时,我们首先通过邮件向文章作者索要原始数据,如未收到相关数据则排除研究;(5)对照组各基因型分布符合Hardy-Weinberg平衡。病例报道、评论、综述和系统评价类文献不作为本研究纳入文献。

1.3 数据提取

两位系统评价员按照STREGA标准独立评价纳入文献的质量,评价内容包括:(1)样本量是否充分;(2)诊断标准交待是否清楚;(3)分组匹配情况如何;(4)对照组是否与病例组具有可比性,对照组基因型分布是否符合Hardy-Weinberg遗传平衡定律;(5)基因检测方法是否合理;(6)数据是否充分。以上6项,每满足一项计1分,总分≥3分者视为质量可靠。并按照标准数据收集表提取文献基本资料,包括第一作者、发表年份、国家、种族、对照来源、对照匹配标准、样本量、病例组与对照组各基因型分布情况等。如有争议则共同复查原文后协商或由第三位评价员决定。

1.4 统计学方法

本研究采用优势比(odds ratios,ORs)及95%可信区间(confidence intervals,CIs)为效应指标来评价MTRR基因A66G多态性与儿童ALL风险关系的关联强度,合并OR的统计学意义采用Z检验,P<0.05为差异有统计学意义。采用χ2检验评估纳入研究间的异质性,检验水准为α=00.05:如纳入研究间无明显异质性(P>0.05)则采用固定效应模型(Mantel-Haenszel);如异质性明显(P<0.05),则采用随机效应模型(DerSimonianLaird)合并数据。发表性偏倚采用可视化的漏斗图评估,敏感性分析采用逐一排除的方法进行,重新估计合并效应量,并与排除前的合并效应量进行比较。纳入研究对照组基因型分布是否符合Hardy-Weinberg平衡采用χ2检验。对纳入研究的数据应用RevMan 5.2软件(The Cochrane Collaboration,Oxford,UK)进行分析。

2 结果 2.1 纳入文献的基本情况

根据文献检索策略共检出19篇可能相关的文献,初筛时排除10篇;仔细阅读文献全文后排除2篇,其中一篇非病例对照研究[15],另一篇未提供充足的原始数据[16]。因此,本研究最终共纳入7篇文献,包括儿童ALL患者2 326例,对照3 090例[17, 18, 19, 20, 21, 22, 23]。文献筛选的流程,见图 1。纳入文献质量评价结果显示:大多数纳入研究样本量充分、儿童ALL的诊断标准明确、分组匹配良好、组间可比性良好、基因检测方法合理、数据完整明确。因此,整体来看纳入研究的质量良好。在纳入的研究中,5项研究来自白种人群[19, 20, 21, 22, 23],一项来自亚洲人群[17],另一项为混合种人群[18]。纳入研究的对照组来自健康人群和血液捐献人群,主要以年龄、性别和种族与儿童ALL病例组匹配。纳入研究的对照组各基因型分布均符合Hardy-Weinberg平衡。各纳入研究的基本特征和原始数据见表 1

图 1 文献筛选流程图 Figure 1 Flow diagram of the process of studies selection
表 1 纳入研究的基本特征 Table 1 Main characteristics of the studies included in Meta-analysis
2.2 Meta分析结果

异质性检验结果和Meta分析结果见表 2。在整体人群四种遗传模型中均未发现纳入研究间存在显著异质性(AG vs.AA,χ2=8.84,Ph=0.18,I2=32%;GG vs.AA,χ2=4.56,Ph=0.60,I2=0%;GG vs.AA+AG,χ2=1.44,Ph=0.96,I2=0%;AG+GG vs.AA,χ2=8.73,Ph=0.19,I2=31%),因此采用固定效应模型合并效应值OR及其95% CI,见图 2表 2。合并结果表明,在整体人群纯合子模型和显性模型发现MTRR基因A66G多态性与儿童ALL风险有显著性关联(GG vs.AA:OR=0.81,95% CI:0.69~0.95,P=0.009;AG+GG vs.AA:OR=0.87,95% CI:0.77~0.98,P=0.03),见图 2表 2。根据种族进行亚组分析时在白种人群杂合子模型、纯合子模型和显性模型发现显著性关联(AG vs.AA:OR=0.84,95% CI:0.72~0.99,P=0.04;GG vs.AA:OR=0.79,95% CI:0.66~0.95,P=0.01;AG+GG vs.AA:OR=0.82,95% CI:0.71~0.96,P=0.01),在亚洲人群和混合人群四种遗传模型均未发现显著性关联,见图 2表 2

图 2 MTRR A66G多态性与儿童ALL风险的Meta分析森林图 (GG vs. AA) Figure 2 Forest plot of the association between MTRR A66G polymorphism and pediatric ALL risk (GG vs. AA)
表 2 MTRR A66G多态性与儿童ALL风险不同遗传模型的Meta分析结果 Table 2 Meta-analysis results of MTRR A66G polymorphism and pediatric ALL under different genetic models
2.3 发表性偏倚与敏感性分析

发表性偏倚采用漏斗图评估,所有漏斗图中各点分布对称性均很好,说明不存在显著的发表性偏倚,见图 3。在敏感性分析过程,逐一排除纳入的研究,重新计算合并效应量,并与排除前的合并效应量进行比较。结果表明单独的研究没有显著影响合并效应量,说明目前所得结果是稳定可靠的。

图 3 MTRR A66G多态性与儿童ALL风险的漏斗图 (GG vs.AA+AG) Figure 3 Funnel plot of the association between MTRR A66G polymorphism and pediatric ALL risk (GG vs. AA+AG)
3 讨论

人类MTRR基因定位于5p15.2~p15.3,包括15个外显子和14个内含子,编码相对分子质量为78kDa的甲硫氨酸合成酶还原酶[24]。MTRR属于双黄素蛋白家族成员,从C端到N端依次为NADPH结合结构域、FAD结合结构域、连接结构域、铰链区和FMN结合结构域。在叶酸代谢过程中,MTRR通过其FMN结合结构域直接与MTR结合形成活化复合物,NADPH来源的电子通过MTRR的FAD和FMN辅助因子转移到失活的钴胺素(Ⅱ),同时转移S-腺苷甲硫氨酸的甲基,将钴胺素(Ⅱ)转变为甲基钴胺素(Ⅲ),作为MTR的辅助因子进而活化MTR。MTR催化同型半胱氨酸重新甲基化为甲硫氨酸,同时甲基四氢叶酸脱甲基为四氢叶酸,在维持细胞内合适的甲硫氨酸浓度和阻止同型半胱氨酸的积累起重要作用[25]。这些酶功能活性的改变可能影响DNA甲基化、DNA合成和修复等过程。目前已发现MTRR基因有200多种单核苷酸多态性,其中以A66G多态性最为常见。MTRR基因第66位点腺嘌呤被鸟嘌呤置换,导致第22位密码子甲硫氨酸被异亮氨酸替代,即I22M。该突变位于MTRR的FMN结合结构域,MTRR对MTR的亲和力降低,影响MTRR与MTR的相互作用,MTR酶活性降低,改变了同型半胱氨酸水平,进而导致心血管疾病、畸形和肿瘤等疾病的发生[8, 9, 10, 11]。近年来许多研究者对MTRR基因A66G多态性与人类多种肿瘤的发生关系进行了广泛探索,结果发现A66G多态性可能与结直肠癌和肝细胞癌风险升高有关[14, 26],而与非霍奇金淋巴瘤和乳腺癌发生风险无关[27, 28]

目前已有研究者对MTRR基因A66G多态性与儿童ALL发生风险的关系进行了研究,但各研究结果间存在一定差异。de Jonge等[21]研究表明MTRR基因A66G多态性与儿童ALL的发生风险无关联,而Gast等[23]研究发现MTRR基因66G等位基因频率在健康对照组的分布高于儿童ALL病例组(P=0.01);AG杂合子和GG纯合子携带者白血病发生风险降低(AG vs.AA,OR=0.7,95% CI:0.5~1.0;GG vs.AA,OR=0.6,95% CI:0.4~0.9),Amigou等[20]则报道了MTRR基因A66G多态性与整体急性白血病风险无关联,但与儿童ALL发生风险存在临界统计学意义的反向关联。本研究全面检索相关文献,严格筛选符合纳入标准的研究共7篇,结果显示,在整体人群纯合子模型和显性模型发现MTRR A66G多态性与儿童ALL风险有显著性关联(GG vs.AA:OR=0.81,95% CI:0.69~0.95,P=0.009;AG+GG vs.AA:OR=0.87,95% CI:0.77~0.98,P=0.03);根据种族进行亚组分析时在白种人群杂合子模型、纯合子模型和显性模型发现显著性关联(AG vs.AA:OR=0.84,95% CI:0.72~0.99,P=0.04;GG vs.AA:OR=0.79,95% CI:0.66~0.95,P=0.01;AG+GG vs.AA:OR=0.82,95% CI:0.71~0.96,P=0.01)。因此,MTRR基因A66G多态性与儿童ALL的发生风险存在关联,66G等位基因可能是ALL的保护因素,G等位基因携带者ALL发生风险降低。漏斗图未检测出显著的发表性偏倚,敏感性分析说明目前的结果是稳定可靠的。需要指出的是,本研究排除了未提供充足数据的研究,可能会引起选择偏倚;纳入研究的原始结果并未都根据相同的混杂因素进行校正,因此,本研究采用的效应指标均为未调整值,可能存在某些混杂因素,会在一定程度上影响Meta分析的结果;另外叶酸摄入水平对DNA甲基化与合成都有着重要影响,可能通过基因-饮食相互作用而影响肿瘤发生风险。然而纳入的研究中仅2篇文献探索了叶酸摄入水平、基因多态性与白血病风险之间的相互关系,所提供的信息有限,本文未对基因-环境因素的交互作用进行分析。

总之,本研究表明MTRR基因A66G多态性与儿童ALL的发生风险有关联,尤其在白种人群,G等位基因可能是ALL的保护因素。将来需要开展大样本量、多中心、设计良好的临床研究,同时考虑基因-基因、基因-环境之间的交互作用,更新系统评价,不仅丰富儿童ALL的病因学与遗传背景,也为筛选和保护易感人群提供依据。

参考文献
[1] ???Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013[J]. CA Cancer J Clin, 2013, 63(1): 11-30.
[2] Pui CH, Evans WE. Treatment of acute lymphoblastic leukemia[J]. N Engl J Med, 2006, 354(2): 166-178.
[3] Wiemels J. Perspectives on the causes of childhood leukemia[J]. Chem Biol Interact, 2012, 196(3): 59-67.
[4] ? ? Inaba H, Greaves M, Mullighan CG. Acute lymphoblastic leukaemia[J]. Lancet, 2013, 381(9881): 1943-55.
[5] Metayer C, Milne E, Dockerty JD, et al. Maternal supplementation with folic Acid and other vitamins and risk of leukemia in offspring: a childhood leukemia international consortium study[J]. Epidemiology, 2014, 25(6): 811-22.
[6] Stern LL, Mason JB, Selhub J, et al. Genomic DNA hypomethylation, a characteristic of most cancers, is present in peripheral leukocytes of individuals who are homozygous for the C677T polymorphism in the methylenetetrahydrofolate reductase gene[J]. Cancer Epidemiol Biomarkers Prev, 2000, 9(8): 849-53.
[7] Yamada K, Gravel RA, Toraya T, et al. Human methionine synthase reductase is a molecular chaperone for human methionine synthase[J]. Proc Natl Acad Sci USA, 2006, 103(25): 9476-81.
[8] ???Wilson A, Platt R, Wu Q, et al. A common variant in methionine synthase reductase combined with low cobalamin (vitamin B12) increases risk for spina bifida[J]. Mol Genet Metab, 1999, 67(4): 317-23.
[9] Olteanu H, Munson T, Banerjee R. Differences in the efficiency of reductive activation of methionine synthase and exogenous electron acceptors between the common polymorphic variants of human methionine synthase reductase[J]. Biochemistry, 2002, 41(45): 13378-85.
[10] Gaughan DJ, Kluijtmans LA, Barbaux S, et al. The methionine synthase reductase (MTRR) A66G polymorphism is a novel genetic determinant of plasma homocysteine concentrations[J]. Atherosclerosis, 2001, 157(2): 451-6.
[11] Geisel J, Zimbelmann I, Schorr H, et al. Genetic defects as important factors for moderate hyperhomocysteinemia[J]. Clin Chem Lab Med, 2001, 39(8): 698-704.
[12] Jiang S, Zhao R, Pan M, et al. Associations of MTHFR and MTRR polymorphisms with serum lipid levels in Chinese hypertensive patients[J]. Clin Appl Thromb Hemost, 2014, 20(4): 400-10.
[13] Yu D, Yang L, Shen S, et al. Association between methionine synthase reductase A66G polymorphism and the risk of congenital heart defects: evidence from eight case-control studies[J]. Pediatr Cardiol, 2014, 35(7): 1091-8.
[14] ??Zhou D, Mei Q, Luo H, et al. The polymorphisms in methylenetetr ahydrofolate reductase, methionine synthase, methionine synthase reductase, and the risk of colorectal cancer[J]. Int J Biol Sci, 2012, 8(6): 819-30.
[15] Lupo PJ, Nousome D, Kamdar KY, et al. A case-parent triad assessment of folate metabolic genes and the risk of childhood acute lymphoblastic leukemia[J]. Cancer Causes Control, 2012, 23(11): 1797-803.
[16] Gra OA, Glotov AS, Kozhekbaeva Zhm, et al. Genetic polymorphism in GST, NAT2, and MTRR and susceptibility to childhood acute leukemia[J]. Mol Biol (Mosk), 2008, 42(2): 214-25.
[17] ??Yang L, Liu L, Wang J, et al. Polymorphisms in folate-related genes: impact on risk of adult acute lymphoblastic leukemia rather than pediatric in Han Chinese[J]. Leuk Lymphoma, 2011, 52(9): 1770-6.
[18] Metayer C, Scélo G, Chokkalingam AP, et al. Genetic variants in the folate pathway and risk of childhood acute lymphoblastic leukemia[J]. Cancer Causes Control, 2011, 22(9): 1243-58.
[19] Lautner-Csorba O, Gézsi A, Erdélyi DJ, et al. Roles of genetic polymorphisms in the folate pathway in childhood acute lymphoblastic leukemia evaluated by Bayesian relevance and effect size analysis[J]. PLoS One, 2013, 8(8): e69843.
[20] Amigou A, Rudant J, Orsi L, et al. Folic acid supplementation, MTHFR and MTRR polymorphisms, and the risk of childhood leukemia: the ESCALE study (SFCE)[J]. Cancer Causes Control, 2012, 23(8): 1265-77.
[21] de Jonge R, Tissing WJ, Hooijberg JH, et al. Polymorphisms in folate-related genes and risk of pediatric acute lymphoblastic leukemia[J]. Blood, 2009, 113(10): 2284-9.
[22] Petra BG, Janez J, Vita D. Gene-gene interactions in the folate metabolic pathway influence the risk for acute lymphoblastic leukemia in children[J]. Leuk Lymphoma, 2007, 48(4): 786-92.
[23] Gast A, Bermejo JL, Flohr T, et al. Folate metabolic gene polymorphisms and childhood acute lymphoblastic leukemia: a case-control study[J]. Leukemia, 2007, 21(2): 320-5.
[24] Leclerc D, Wilson A, Dumas R, et al. Cloning and mapping of a cDNA for methionine synthase reductase, a flavoprotein defective in patients with homocystinuria[J]. Proc Natl Acad Sci USA, 1998, 95(6): 3059-64.
[25] Olteanu H, Banerjee R. Human methionine synthase reductase, a soluble P-450 reductase-like dual flavoprotein, is sufficient for NADPH-dependent methionine synthase activation[J]. J Biol Chem, 2001, 276(38): 35558-63.
[26] Kwak SY, Kim UK, Cho HJ, et al. Methylenetetrahydrofolate reductase (MTHFR) and methionine synthase reductase (MTRR) gene polymorphisms as risk factors for hepatocellular carcinoma in a Korean population[J]. Anticancer Res, 2008, 28(5A): 2807-11.
[27] Weiner AS, Beresina OV, Voronina EN, et al. Polymorphisms in folate-metabolizing genes and risk of non-Hodgkin’s lymphoma[J]. Leuk Res, 2011, 35(4): 508-15.
[28] Hu S, Liu HC, Xi SM. Methionine synthase reductase A66G polymorphism is not associated with breast cancer susceptibility - a meta-analysis[J]. Asian Pac J Cancer Prev, 2014, 15(7): 3267-71.