The impact of sex hormones on the brain has been widely recognized. Polycystic ovary syndrome (PCOS) is frequently accompanied by abnormal hormone. Furthermore, psychiatric disorders are among the most common comorbidities associated with PCOS. However, whether PCOS induces alterations in brain structure remains unclear. Mendelian randomization (MR) was conducted to determine this causal association and pinpoint specific brain regions prone to PCOS. Genome wide association study (GWAS) data from the FinnGen database, for PCOS phenotypes from 39,004 PCOS patients and 731,830 controls were included. Outcomes were determined using GWAS data from the ENIGMA Consortium for brain structural traits, specifically cortical thickness and cortical surficial area, from 51,665 participants and for the volume of subcortical structures from 30,717 participants. The inverse-variance weighted (IVW) method was adopted as the primary estimate method. Additionally, several sensitivity and pleiotropy analyses were conducted to verify MR results. Genetically predicted PCOS was nominally associated with increased cortical thickness in the paracentral lobule (β = 0.0020 mm, pIVW = 0.0097) and decreased cortical thickness in the precentral gyrus (β = -0.0038 mm, pIVW = 0.0101). In addition, it was nominally associated with increased cortical surficial area in the precentral gyrus (β = 14.4026 mm2, pIVW = 0.0088), the pars opercularis (β = 5.6965 mm2, pIVW = 0.0491), and the posterior cingulate cortex (β = 3.529 mm2, pIVW = 0.0396) and reduced cortical surficial area in the inferior parietal lobule (β = -6.1279 mm2, pIVW = 0.0273) and the caudal middle frontal gyrus (β = -20.7354 mm2, pIVW = 0.0265). The results remained consistent throughout the sensitivity and pleiotropy analyses. This hypothesis-generating study provides suggestive evidence for possible associations between genetic predisposition to PCOS and alterations in cortical structure in specific brain regions. These findings highlight the need for confirmatory studies using independent datasets and alternative methodologies to validate the observed associations and elucidate underlying mechanisms. The results should be interpreted cautiously given the exploratory nature of the analyses and the limitations inherent to MR studies.

The causal relationship between polycystic ovary syndrome and structural changes in brain: a Mendelian randomization study

期刊论文
发表日期：

2026年4月29日

查看全文

Han Yang,
Wei Xie,
Camilo Alberto Pinzon Galvis,
Si Huang,
Xu He,
Si-yi Yu,
Qing You,
Fan-rong Liang,
Zhi-yong Xiao,
Ning Li

摘要

The impact of sex hormones on the brain has been widely recognized. Polycystic ovary syndrome (PCOS) is frequently accompanied by abnormal hormone. Furthermore, psychiatric disorders are among the most common comorbidities associated with PCOS. However, whether PCOS induces alterations in brain structure remains unclear. Mendelian randomization (MR) was conducted to determine this causal association and pinpoint specific brain regions prone to PCOS. Genome wide association study (GWAS) data from the FinnGen database, for PCOS phenotypes from 39,004 PCOS patients and 731,830 controls were included. Outcomes were determined using GWAS data from the ENIGMA Consortium for brain structural traits, specifically cortical thickness and cortical surficial area, from 51,665 participants and for the volume of subcortical structures from 30,717 participants. The inverse-variance weighted (IVW) method was adopted as the primary estimate method. Additionally, several sensitivity and pleiotropy analyses were conducted to verify MR results. Genetically predicted PCOS was nominally associated with increased cortical thickness in the paracentral lobule (β = 0.0020 mm, p_IVW = 0.0097) and decreased cortical thickness in the precentral gyrus (β = -0.0038 mm, p_IVW = 0.0101). In addition, it was nominally associated with increased cortical surficial area in the precentral gyrus (β = 14.4026 mm², p_IVW = 0.0088), the pars opercularis (β = 5.6965 mm², p_IVW = 0.0491), and the posterior cingulate cortex (β = 3.529 mm², p_IVW = 0.0396) and reduced cortical surficial area in the inferior parietal lobule (β = -6.1279 mm², p_IVW = 0.0273) and the caudal middle frontal gyrus (β = -20.7354 mm², p_IVW = 0.0265). The results remained consistent throughout the sensitivity and pleiotropy analyses. This hypothesis-generating study provides suggestive evidence for possible associations between genetic predisposition to PCOS and alterations in cortical structure in specific brain regions. These findings highlight the need for confirmatory studies using independent datasets and alternative methodologies to validate the observed associations and elucidate underlying mechanisms. The results should be interpreted cautiously given the exploratory nature of the analyses and the limitations inherent to MR studies.