Methodological challenges in mendelian randomization studies of immune cell phenotypes and autoimmune liver diseases

Dear Editors,

We read with great interest the article by Zou et al. (1), which investigated the potential causal relationships between diverse immune cell phenotypes and autoimmune liver diseases (AILDs) using a bidirectional two-sample Mendelian randomization (MR) framework. The authors should be commended for addressing an important gap in AILD pathogenesis by integrating a large-scale immunophenotype genome-wide association study (GWAS) with disease-specific genetic datasets. Their work contributes to the growing field of immune-genetic causal inference and provides valuable hypotheses regarding natural killer (NK) cells, B-cell subsets, dendritic cells, and monocyte populations. Nevertheless, several methodological considerations may influence the reliability and interpretability of the study’s conclusions.

First, although the study draws upon large publicly available GWAS datasets, the exposure and outcome sources differ substantially in terms of cohort characteristics, genotyping platforms, and sample structures. The immune phenotype GWAS was conducted in a Sardinian founder population, whereas AILD GWAS datasets were derived from broader European cohorts with distinct demographic and genetic profiles. This population mismatch raises concerns about instrument transportability and the validity of causal estimates. Prior MR literature has demonstrated that allele frequency differences and linkage disequilibrium (LD) architecture divergence between exposure and outcome populations can bias causal inference (2). Sensitivity analyses stratified by ancestry-matched subsets, or the use of ancestry-harmonized LD panels, would strengthen the robustness of the findings.

Second, although the authors implemented multiple pleiotropy and heterogeneity tests, the immune system’s inherently interconnected nature increases the likelihood of horizontal pleiotropy—particularly given the functional overlap among innate and adaptive immune subsets. Some significant instruments were excluded due to pleiotropy or insufficient single-nucleotide polymorphism (SNP) counts, which underscores the analytical challenges in dealing with immune-related exposures. Importantly, the study did not conduct colocalization analysis to confirm that the same causal variants drive both immune phenotypes and AILDs. Colocalization is essential in immunogenetic MR studies because overlapping signals may simply reflect shared genomic regions rather than direct causal pathways (3). Without these additional layers of validation, the biological interpretation of immune cell causality warrants caution.

Third, the statistical power of some analyses may be limited by the low number of available instruments for certain immune phenotypes. Many phenotypes were represented by only one or two SNPs, limiting the ability to perform MR-Egger regression, weighted median analyses, or MR-PRESSO outlier correction. Sparse-instrument MR is known to produce unstable estimates and inflated type I error rates (4). The authors attempted to mitigate weak-instrument bias by using F-statistics >10, but weak-instrument bias can still persist when instrument counts are low. Reporting conditional F-statistics or applying robust MR methods designed for sparse instruments (e.g., MR-RAPS) would help confirm result stability.

Finally, although the study highlights potential immune subsets (e.g., HLA-DR-expressing plasmacytoid dendritic cells, CD27⁺ memory B cells, and specific NK-cell populations) as causal contributors to AILD susceptibility, the analysis does not incorporate broader immunological or clinical covariates. Key determinants of autoimmune disease activity—such as cytokine profiles, autoantibody titers, epigenetic modifications, gut microbiome features, or environmental exposures—were not considered (5). Additionally, immune cell phenotypes measured in peripheral blood may not fully represent liver-resident immune microenvironments, which play a decisive role in AILD pathogenesis. Integration of liver-specific expression quantitative trait loci (eQTL), single-cell immune atlases, or functional genomic datasets would help clarify whether the observed associations reflect systemic immunity or liver-specific mechanisms.

In summary, while the study by Zou et al. provides valuable genetic evidence supporting the involvement of specific immune phenotypes in AILD susceptibility, several issues—including population mismatch, incomplete pleiotropy control, limited SNP instrument strength, and absence of tissue-specific validation—should be addressed to enhance confidence in the causal claims. Incorporating ancestry-matched analyses, colocalization, robust MR methods, and liver-focused functional data would further strengthen the conclusions and facilitate translation of these findings into clinically meaningful insights for AILD pathogenesis and therapy development.

Acknowledgments

None.

Footnote

Provenance and Peer Review: This article was a standard submission to the journal. The article did not undergo external peer review.

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tgh.amegroups.com/article/view/10.21037/tgh-20251-157/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Zou H, Liang X, Luo M, et al. Causal effects of immune cell phenotypes on the risk of autoimmune liver diseases: a bidirectional two-sample Mendelian randomization study. Transl Gastroenterol Hepatol 2025;10:68. [Crossref] [PubMed]
2. Swerdlow DI, Kuchenbaecker KB, Shah S, et al. Selecting instruments for Mendelian randomization in the wake of genome-wide association studies. Int J Epidemiol 2016;45:1600-16. [Crossref] [PubMed]
3. Giambartolomei C, Vukcevic D, Schadt EE, et al. Bayesian test for colocalisation between pairs of genetic association studies using summary statistics. PLoS Genet 2014;10:e1004383. [Crossref] [PubMed]
4. Zhao Q, Wang J, Hemani G, et al. Statistical inference in two-sample summary-data Mendelian randomization using robust adjusted profile score. Ann Statist 2020;48:1742-69.
5. Trivedi PJ, Hirschfield GM, Adams DH, et al. Immunopathogenesis of Primary Biliary Cholangitis, Primary Sclerosing Cholangitis and Autoimmune Hepatitis: Themes and Concepts. Gastroenterology 2024;166:995-1019. [Crossref] [PubMed]