Causality of immune cells and endometriosis: a bidirectional mendelian randomization study

Abstract

Abstract Background Endometriosis, a prevalent chronic condition, afflicts approximately 10% of women in their reproductive years. Emerging evidence implicates immune cells in the pathogenesis of endometriosis, particularly in angiogenesis, tissue proliferation, and lesion invasion. This investigation employs two-sample Mendelian Randomization (MR) to dissect the bidirectional causal relationships between immune cell profiles and endometriosis. Methods We leveraged publicly available genome-wide association study (GWAS) data to elucidate the causal interplay between immune cell traits and endometriosis. Utilizing GWAS summary statistics ranging from accession numbers GCST90001391 to GCST90002121 and endometriosis data from the FinnGen study GWAS (8,288 endometriosis cases and 68,969 controls), we adopted stringent criteria for instrumental variable selection. We applied MR-Egger, weighted median, inverse variance weighted (IVW), and weighted mode methods to derive causal estimates. To address potential heterogeneity and pleiotropy, Cochran’s Q test, MR-Egger intercept, and leave-one-out analyses were executed. Reverse-direction MR and bidirectional MR analyses evaluated potential reciprocal causation and the influence of endometriosis on immune cell composition. Results Our analysis identified five immune phenotypes inversely associated with endometriosis risk. These phenotypes comprise: a percentage of CD11c + HLA-DR + + monocytes, CD25 expression on CD39 + CD4 + T cells, elevated CD25 on CD45RA + CD4 + non-regulatory T cells, HLA-DR intensity on HLA-DR + CD8 bright (CD8br) T cells, and the proportion of naïve double-negative (CD4 − CD8− %DN) T cells. In contrast, eleven phenotypes were positively correlated with endometriosis risk, including: CD127 expression on T cells, the proportion of CD24 + CD27 + B cells within lymphocytes, CD25 expression on CD28 + CD4 + T cells, CD28 expression on CD39 + activated regulatory T cells (activated Tregs), the frequency of bright CD33 HLA-DR + CD14 − cells within the CD33br HLA-DR + compartment, CD45 expression on lymphocytes and natural killer (NK) cells, activation status of central memory CD8 bright (CM CD8br) T cells, CX3CR1 expression on monocytes, and the percentage of HLA-DR + NK cells within the NK cell subset. Sensitivity assessments that excluded significant heterogeneity and pleiotropy confirmed the stability of these associations, thereby reinforcing the validity of our findings. Conclusion This study provides novel evidence of the potential causal impact of specific immune cells on the risk of developing endometriosis. These findings enhance our understanding of endometriosis pathophysiology and may inform innovative approaches for its diagnosis and management. While our findings provide novel insights, limitations such as potential horizontal pleiotropy and reliance on European ancestry data should be considered. Future research should expand to diverse populations and incorporate individual-level data to refine these findings.

Keywords

• Endometriosis
• Immunity
• Mendelian randomization analysis
• Genome-wide association study
• Causal inference
• Genetics