Department of Clinical Medicine, Translational Psychiatry Research Group, Faculty of Medicine, Federal University of Ceará, Fortaleza, CE, Brazil; Centre for Addiction & Mental Health (CAMH), Toronto, ON, Canada.
Department of Clinical Medicine, Translational Psychiatry Research Group, Faculty of Medicine, Federal University of Ceará, Fortaleza, CE, Brazil.
Prog Neuropsychopharmacol Biol Psychiatry. 2018 Aug 30;86:102-113. doi: 10.1016/j.pnpbp.2018.05.012. Epub 2018 May 17.
Imaging genetics studies involving participants with major depressive disorder (MDD) have expanded. Nevertheless, findings have been inconsistent. Thus, we conducted a systematic review and meta-analysis of imaging genetics studies that enrolled MDD participants across major databases through June 30th, 2017. Sixty-five studies met eligibility criteria (N = 4034 MDD participants and 3293 controls), and there was substantial between-study variability in the methodological quality of included studies. However, few replicated findings emerged from this literature with only 22 studies providing data for meta-analyses (882 participants with MDD and 616 controls). Total hippocampal volumes did not significantly vary in MDD participants or controls carrying either the BDNF Val66Met 'Met' (386 participants with MDD and 376 controls) or the 5-HTTLPR short 'S' (310 participants with MDD and 230 controls) risk alleles compared to non-carriers. Heterogeneity across studies was explored through meta-regression and subgroup analyses. Gender distribution, the use of medications, segmentation methods used to measure the hippocampus, and age emerged as potential sources of heterogeneity across studies that assessed the association of 5-HTTLPR short 'S' alleles and hippocampal volumes. Our data also suggest that the methodological quality of included studies, publication year, and the inclusion of brain volume as a covariate contributed to the heterogeneity of studies that assessed the association of the BDNF Val66Met 'Met' risk allele and hippocampal volumes. In exploratory voxel-wise meta-analyses, MDD participants carrying the 5-HTTLPR short 'S' allele had white matter microstructural abnormalities predominantly in the corpus callosum, while carriers of the BDNF Val66Met 'Met' allele had larger gray matter volumes and hyperactivation of the right middle frontal gyrus compared to non-carriers. In conclusion, few replicated findings emerged from imaging genetics studies that included participants with MDD. Nevertheless, we explored and identified specific sources of heterogeneity across studies, which could provide insights to enhance the reproducibility of this emerging field.
影像遗传学研究涉及到患有重度抑郁症（MDD）的参与者，此类研究不断增加。然而，研究结果却并不一致。因此，我们通过主要数据库，检索并分析了截止至 2017 年 6 月 30 日，涉及 MDD 参与者的影像遗传学研究。共有 65 项研究符合入选标准（N=4034 名 MDD 患者和 3293 名对照组），纳入研究的方法学质量存在较大差异。然而，该文献中仅出现了很少的复制结果，只有 22 项研究提供了元分析的数据（882 名 MDD 患者和 616 名对照组）。携带 BDNF Val66Met 'Met'（386 名 MDD 患者和 376 名对照组）或 5-HTTLPR 短 'S'（310 名 MDD 患者和 230 名对照组）风险等位基因的 MDD 患者和对照组的总海马体体积没有显著差异。通过元回归和亚组分析，探讨了研究间的异质性。研究间的异质性来源于性别分布、药物使用、用于测量海马体的分割方法以及年龄，这些因素与评估 5-HTTLPR 短 'S' 等位基因与海马体体积之间关联的研究有关。我们的数据还表明，纳入研究的方法学质量、发表年份以及将脑容量作为协变量纳入，是评估 BDNF Val66Met 'Met' 风险等位基因与海马体体积之间关联的研究异质性的原因。在探索性体素水平元分析中，携带 5-HTTLPR 短 'S' 等位基因的 MDD 患者的白质微观结构异常主要位于胼胝体，而携带 BDNF Val66Met 'Met' 等位基因的患者的灰质体积较大，右侧额中回过度激活。总之，纳入 MDD 患者的影像遗传学研究中很少出现复制结果。尽管如此，我们探索并确定了研究间异质性的具体来源，这可能为提高这一新兴领域的可重复性提供一些见解。
