Background The knowledge gap regarding whether the correlation between atrial fibrillation (AF) and dementia in observational studies is causation or driven by other shared risk factors remains substantially unfilled. Methods and Results We performed a comprehensive 2‐sample Mendelian randomization study to evaluate the causal effect of AF on overall dementia and its subtypes, including vascular dementia, Alzheimer dementia, Lewy body dementia, and frontotemporal dementia. The primary results in inverse variance–weighted analyses were further validated by various Mendelian randomization sensitivity analyses. Additionally, we conducted multivariable Mendelian randomization to examine 10 candidate mediators of the causal association of AF and dementia. Genetic predisposition to AF was modestly associated with an increased risk of overall dementia (odds ratio, 1.140 [95% CI, 1.023–1.271]; P =0.018) and strongly associated with vascular dementia (odds ratio, 1.350 [95% CI, 1.076–1.695]; P =0.010). Genetically predicted AF indicated neutral effects on Alzheimer dementia, Lewy body dementia, and frontotemporal dementia. In multivariable Mendelian randomization analysis, the total effect of AF on overall dementia was remarkably attenuated by adjusting for genetic effect for ischemic stroke (odds ratio, 1.068 [95% CI, 0.953–1.197]; P =0.259) and low cardiac output (odds ratio, 1.046 [95% CI, 0.926–1.181]; P =0.475), indicating that the causal association of genetically predicted AF with dementia was potentially mediated by ischemic stroke and low cardiac output. The causal effect of genetically predicted AF on dementia was independent of cerebral small‐vessel disease and brain volume phenotypes. Conclusions Our findings provided novel evidence supporting the causal effect of genetically predicted AF on dementia mediated by ischemic stroke and low cardiac output. Future clinical trials are warranted to evaluate the potential role of appropriate AF management in dementia prevention.

Background. We planned to uncover the cancer stemness-related genes (SRGs) in prostate cancer (PCa) and its underlying mechanism in PCa metastasis. Methods. We acquired the RNA-seq data of 406 patients with PCa from the TCGA database. Based on the mRNA stemness index (mRNAsi) calculated by one-class logistic regression (OCLR) algorithm, SRGs in PCa were extracted by WGCNA. Univariate and multivariate regression analyses were applied to uncover OS-associated SRGs. Gene Set Variation Analysis (GSVA), Gene Set Enrichment Analysis (GSEA), and Pearson’s correlation analysis were performed to discover the possible mechanism of PCa metastasis. The significantly correlated transcription factors of OS-associated SRGs were also identified by Pearson’s correlation analysis. ChIP-seq was applied to validate the binding relationship of TFs and OS-associated SRGs and spatial transcriptome and single-cell sequencing were performed to uncover the location of key biomarkers expression. Lastly, we explored the specific inhibitors for SRGs using CMap algorithm. Results. We identified 538 differentially expressed genes (DEGs) between non-metastatic and metastatic PCa. Furthermore, OS-associated SRGs were identified. The Pearson correlation analysis revealed that FOXM1 was significantly correlated with NEIL3 (correlation efficient =0.89, p < 0.001 ) and identified hallmark_E2F_targets as the potential pathway mechanism of NEIL3 promoting PCa metastasis (correlation efficient =0.58, p < 0.001 ). Single-cell sequencing results indicated that FOXM1 regulating NEIL3 may get involved in the antiandrogen resistance of PCa. Rottlerin was discovered to be a potential target drug for PCa. Conclusion. We constructed a regulatory network based on SRGs associated with PCa metastasis and explored possible mechanism.