Basic Medicine College, Navy Military Medical University, Shanghai 200433, PR China.
Center for Clinical Laboratory, The 302nd Hospital of Chinese PLA, Beijing 100039, PR China.
Gene. 2018 Nov 30;677:211-227. doi: 10.1016/j.gene.2018.07.070. Epub 2018 Jul 29.
To identify the most suitable genetic model for detecting the risk of breast cancer (BC)/ovarian cancer (OC) in specific populations.
Databases were searched for related studies published up to October 2017. First, VDR genetic polymorphisms were compared in patients with and without cancer. Second, a network meta-analysis was used to reveal the relation between VDR genetic polymorphisms with disease outcomes. Subgroup analyses and a meta-regression were performed according to cancer types, ethnicity and genotypic method. The study is registered in PROSPERO with an ID: CRD42017075505.
Forty-five studies were eligible, which included 65,754 patients and 55 clinical analyses. Of genetic models, results suggested that the recessive model with the CDX2 polymorphism predicted the risk of BC in all cases. The recessive polymorphism model with the rs2228570 (FokI) polymorphism seemed to the best predictor of BC in Caucasian patients, whereas the homozygote model with the CDX2 polymorphism appeared to best predict BC in African-American patients. The homozygote model with the rs2228570 (FokI) polymorphism model appeared to detect the risk of OC in all cases, whereas the heterozygote model with the rs1544410 (BsmI) polymorphism seemed to detect the risk of OC in Caucasian patients.
By detecting the risk of BC, the recessive model with the rs2228570 (FokI) polymorphism is likely the best genetic model in Caucasian patients, and the homozygote model with the CDX2 polymorphism appears to be best genetic model in African-American patients. Moreover, for detecting clinical risk of OC, heterozygote models with the rs1544410 (BsmI) polymorphism is likely the best genetic model for detecting the risk of OC in Caucasian patients.
为了确定最适合检测特定人群乳腺癌（BC）/卵巢癌（OC）风险的遗传模型。
检索了截至 2017 年 10 月发表的相关研究数据库。首先，比较了癌症患者和非癌症患者的 VDR 基因多态性。其次，采用网络荟萃分析揭示 VDR 基因多态性与疾病结局的关系。根据癌症类型、种族和基因分型方法进行亚组分析和荟萃回归。该研究在 PROSPERO 注册，注册号为：CRD42017075505。
45 项研究符合条件，共纳入 65754 例患者和 55 项临床分析。在遗传模型中，结果表明，CDX2 多态性的隐性模型预测所有病例的 BC 风险。FokI 多态性的隐性多态模型似乎是预测白人患者 BC 的最佳预测因子，而 CDX2 多态性的纯合子模型似乎是预测非洲裔美国患者 BC 的最佳预测因子。FokI 多态性的 rs2228570 （FokI）的纯合子模型似乎可以检测所有病例的 OC 风险，而 BsmI 多态性的 rs1544410 （BsmI）的杂合子模型似乎可以检测白人患者的 OC 风险。
通过检测 BC 的风险，FokI 多态性的 rs2228570 （FokI）的隐性模型可能是白人患者中最佳的遗传模型，而 CDX2 多态性的纯合子模型似乎是非洲裔美国患者中最佳的遗传模型。此外，对于检测 OC 的临床风险，rs1544410 （BsmI）多态性的杂合子模型可能是检测白人患者 OC 风险的最佳遗传模型。
