A commentary on ‘Complex causal association between genetically predicted 731 immunocyte phenotype and osteonecrosis: a bidirectional two-sample Mendelian randomization analysis’ (Int J Surg. 2024 Mar 18)

Guang Tan, Qi Zhang, Xu Luo, Yanfei YangInternational Journal of Surgery 2024 · DOI

TL;DR

This commentary discusses a Mendelian randomization (MR) study investigating the causal link between 731 immune cell phenotypes and osteonecrosis. The authors of the commentary commend the original study's findings but propose two crucial methodological improvements: more rigorous instrumental variable screening for SNPs and explicit removal of confounding SNPs to enhance the reliability of the bidirectional MR analysis.

What Problem Does This Paper Address?

The original study, which this commentary addresses, aimed to explore the complex, potentially bidirectional causal relationship between a large number of immune cell phenotypes and osteonecrosis using Mendelian randomization (MR). Osteonecrosis is a condition where bone tissue dies due to a lack of blood supply, and understanding its potential links to immune system variations is crucial for developing better diagnostic and therapeutic strategies. MR is a powerful statistical technique that uses genetic variants as instrumental variables to infer causality from observational data, helping to overcome limitations of traditional observational studies that can be confounded by various factors. The challenge lies in performing such analyses with a high number of exposures (731 immune cell phenotypes) and a complex outcome, requiring robust statistical methods and careful consideration of potential biases.

Methods

This paper is a commentary, not an original research study. It discusses the methodology of a previously published MR study. The original study employed a bidirectional two-sample Mendelian randomization analysis. This involves using genetic variants (Single Nucleotide Polymorphisms or SNPs) associated with immune cell phenotypes as instrumental variables to assess their causal effect on osteonecrosis, and vice-versa. The commentary highlights that the original study used Inverse Variable Weighting (IVW), MR-Egger, Weighted Mode, Simple Mode, and Weighted Median methods, which are standard approaches in MR analysis to account for different potential biases.

The commentary's key methodological suggestions for improvement are:

  • Instrumental Variable Screening: The authors suggest that for each of the 731 MR analyses, the threshold for selecting statistically significant SNPs should be adjusted based on the specific P-value for that individual screening. This is proposed to reduce the inclusion of weakly correlated SNPs and improve reliability.
  • Confounding SNP Removal: The commentary emphasizes the need to explicitly search for and remove confounding SNPs from the exposure (immune cell phenotypes) before conducting the MR analysis. While the original study mentioned verifying confounding SNPs, the commentary stresses the importance of their removal to prevent interference with the results.

Datasets

Not discussed in the paper. The commentary refers to a previously published study by Li et al. [1] which likely utilized publicly available genetic association summary statistics for immune cell phenotypes and osteonecrosis.

Key Insights and Ideas

  • Confirmation of Associations: The commentary strongly agrees with the original study's findings that several immune cell phenotypes are significantly associated with osteonecrosis, both positively and negatively. This reinforces the biological plausibility of immune system involvement in the disease.
  • Methodological Rigor is Paramount: The core insight is that even with powerful techniques like MR, meticulous attention to methodological details, particularly in SNP selection and confounding factor management, is critical for ensuring the validity and reliability of the findings.
  • The Challenge of High-Dimensionality: The commentary implicitly highlights the complexity of analyzing a large number of exposures (731 immune cell phenotypes) and the increased potential for statistical noise and bias, underscoring the need for stringent analytical approaches.

Key Results

The commentary does not present new quantitative results. Instead, it summarizes the findings of the original study:

  • Positive Correlations: 8 immune cell phenotypes were positively correlated with osteonecrosis (P < 0.05, OR > 1) in the MR analysis.
  • Negative Correlations: 18 immune cell phenotypes were negatively correlated with osteonecrosis (P < 0.05, OR < 1) in the MR analysis.
  • Inverse MR Results: In the inverse MR analysis (osteonecrosis as exposure), 10 immune cell phenotypes were positively correlated with osteonecrosis (P < 0.05, OR > 1), and 14 were negatively correlated (P < 0.05, OR < 1).
  • Absence of Bias: The original study reported no significant heterogeneity, horizontal pleiotropy, or reverse causality.

The commentary agrees with these findings but suggests improvements to enhance their robustness.

Limitations and Future Work

Limitations Acknowledged by the Commentary:

  • SNP Screening Threshold: The commentary suggests the original study's SNP screening threshold might not be optimal for the large number of analyses performed.
  • Confounding SNP Management: The commentary points out that the original study did not explicitly state the removal of confounding SNPs, which could impact the results.

Future Work Suggested by the Commentary:

  • Refined SNP Screening: Implementing a more dynamic and specific P-value threshold for SNP screening in each of the 731 MR analyses.
  • Explicit Confounding SNP Removal: Researchers should clearly document and perform the removal of confounding SNPs before conducting subsequent MR analyses.
  • Further Exploration: The authors of the commentary encourage the original researchers to address these points to further improve the reference value and reliability of their research.

Summary updated

Abstract

Department of Hepatobiliary Surgery, The First Affiliated Hospital of Dalian Medical University, Dalian 116000, China Guang Tan, Qi Zhang and Xu Luo: These authors have contributed equally to this work. Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article. Published online ■ ■ *Corresponding author. Address: First Affiliated Hospital of Dalian Medical University Department of General Surgery, Dalian, Liaoning CHINA. E-mail address: [email protected] (Y. Yang). This is an open access article distributed under the Creative Commons Attribution-NoDerivatives License 4.0, which allows for redistribution, commercial and non-commercial, as long as it is passed along unchanged and in whole, with credit to the author. http://creativecommons.org/licenses/by-nd/4.0/
BibTeX
@article{tan2024commentary,
  author = {Guang Tan and Qi Zhang and Xu Luo and Yanfei Yang},
  title = {{A commentary on ‘Complex causal association between genetically predicted 731 immunocyte phenotype and osteonecrosis: a bidirectional two-sample Mendelian randomization analysis’ (Int J Surg. 2024 Mar 18)}},
  journal = {International Journal of Surgery},
  year = {2024},
  doi = {10.1097/js9.0000000000001656},
}
Figure