Exploratory Two-Sample Mendelian Randomization Study of Genetically Predicted IgA Nephropathy Susceptibility and Risk of Epithelial Ovarian Cancer Across Histological Subtypes
Xiaolin Liu, Sen Li, Yingjie Wang, Xiaomi Li, Peng Jiang, Yishuo Wang, Yuzhu Chen, Yixuan Wang, Fei Qi, Tongmei ZhangBackground
Given the widespread prevalence of epithelial ovarian cancer as a global malignant tumor, there is an urgent need to deeply explore its risk determinants. This study used cluster analysis to summarize the association patterns of different histological subtypes of epithelial ovarian cancer, and employed the two-sample Mendelian randomization method to evaluate the potential causal association between genetic prediction of IgA nephropathy susceptibility and the risk of epithelial ovarian cancer. This method, under the condition of satisfying the correlation, independence, and exclusion of restrictive assumptions of instrumental variables, helps to reduce the influence of traditional confounding and reverse causation.
Methods
Genome-wide association data from 15,587 European IgA nephropathy patients served as the primary exposure, with GWAS data from epithelial ovarian cancer subtypes. Three Mendelian randomization(MR) methods including inverse variance weighting MR-Egger regression and weighted median were adopted for causal inference. Sensitivity analyses included Cochran’s Q test, MR-Egger intercept, MR-PRESSO, and Leave-one-out analysis. Six subtype-specific UVMR analyses were conducted, and the results were reported separately. Because the subtype outcome datasets were overlapping or nested, no formal meta-analysis or pooled estimate was performed.
Results
In the subtype-specific UVMR analyses, the effect estimates were generally below the null value, suggesting a directionally concordant pattern towards a lower risk of epithelial ovarian cancer. The association was statistically significant for high-grade serous ovarian cancer (OR =0.81, 95% CI: 0.75- 0.87, P <0.05). The estimates for clear cell ovarian cancer and the remaining evaluated histological subtypes were not statistically significant, although some showed similar inverse directions. Because the subtype-specific outcome datasets were overlapping or nested and shared the same IgA nephropathy exposure GWAS, no formal pooled estimate was calculated. Sensitivity analyses [including MR-Egger intercept, MR-PRESSO, leave-one-out analysis, and Steiger directionality testing] yielded results consistent with the primary subtype-specific analyses.
Conclusion
Genetically predicted liability to IgA nephropathy may be associated with a lower risk of epithelial ovarian cancer, particularly high-grade serous ovarian cancer, although this finding requires validation in independent and sex-matched datasets.