To the Editor,
With great interest, I read the article by Junhao Chen et al.,1 recently published in Sexual Medicine. The authors analyzed genetic instruments for 731 immune cell types identified through genome-wide association studies and obtained erectile dysfunction (ED) data from the FinnGen database. They conducted univariable and multivariable bidirectional Mendelian randomization (MR) studies to explore the potential causal relationships between these immune cells and ED. They concluded that MR analysis can confirm a potential bidirectional causal relationship between immune cells and ED while providing new insights into the underlying pathophysiological mechanisms and subsequent therapeutic strategies. However, I would like to discuss some concerns regarding their findings.
First, I noticed that the FinnGen database utilized in this study is based on an older version of the IEU, while the latest update, FinnGen R11, has already been released, which includes a larger population sample. This discrepancy may affect the accuracy and relevance of the findings derived from this database, as more current data could better elucidate fundamental trends and associations within the population.
Additionally, I suggest that the authors incorporate data from other sources, such as the GCST90044277 database (https://www.ebi.ac.uk/gwas/studies/GCST90044277), which originates from the UK. Including this database would facilitate a meta-analysis, enhancing the robustness of their findings by integrating diverse datasets.
Regarding the reliability of the results, conducting a meta-analysis that incorporates ED data from multiple databases would allow for a more comprehensive synthesis of evidence, leading to stronger conclusions and greater confidence in the analysis results.
The authors identified numerous immune cell types in their study. While we acknowledge that the authors used a stricter P-value threshold (<0.01) for accuracy, I recommend further utilizing coloc colocalization analysis to examine the presence of multiple signals at the same genomic loci. This approach would indicate that these phenotypes may share common biological mechanisms, thereby supporting the existence of causal relationships. Such analysis would be valuable for elucidating the relationships between different immune cells and their shared genomic influences.2 By integrating colocalization analysis, stronger evidence could be provided than with Bayesian weighted MR alone, thereby enhancing the overall validity of the findings.
Finally, we believe that for large-scale MR, the authors should apply False Discovery Rate correction to address the potential for multiple comparisons, ensuring the robustness of their findings.
Contributor Information
Zhihao Huang, Department of Urology, Zhangjiagang Hospital Affiliated to Soochow University, Suzhou, Jiangsu, China.
Rong Wang, Department of Urology, Zhangjiagang Hospital Affiliated to Soochow University, Suzhou, Jiangsu, China.
Author contributions
Z. Huang and R. Wang contributed equally to this work.
Funding
There is no funding role in this study.
Conflicts of interest
None declared.
References
- 1. Chen J, Liu Y, Zhan P, et al. Bayesian-based analysis of the causality between 731 immune cells and erectile dysfunction: a two-sample, bidirectional, and multivariable Mendelian randomization study. Sex Med. 2024;12(4):qfae062. 10.1093/sexmed/qfae062 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Sanderson E. Multivariable Mendelian randomization and mediation. Cold Spring Harb Perspect Med. 2021;11(2):a038984. 10.1101/cshperspect.a038984 [DOI] [PMC free article] [PubMed] [Google Scholar]