Abstract Study question Is the risk of endometriosis associated with genetically predicted levels of different blood lipid and apolipoprotein traits? Summary answer Our Mendelian randomization (MR) analysis indicated that causal associations existed between some serum lipid and apolipoprotein levels and endometriosis occurrence in European ancestry. What is known already Endometriosis is a chronic gynecological disease with a high prevalence among reproductive-aged women. Recently, many observational, epidemiological studies have investigated the causal effect of different serum lipid and apolipoprotein parameters on the occurrence and progression of endometriosis. However, the results of these studies remained controversial. Moreover, no different subtypes of endometriosis were investigated in previous observational studies, thus detailed information about the associations was insufficient. Study design, size, duration We performed a two-sample MR analysis in the largest available genetic datasets. To correctly employ the MR method, the following assumptions are required to be met: (a) the genetic variants must be strongly associated with the exposures; (b) the single-nucleotide polymorphisms (SNPs) are not correlated with the confounders that are related to the exposure and the outcome; (c) the SNPs can influence the outcome solely through the exposure and no other biological pathways are concerned. Participants/materials, setting, methods Summary-level data for genetic variants of different serum lipid and apolipoprotein traits were derived from the UK Biobank and EBI database. Instrumental variables (IVs) associated with endometriosis and all the subtypes were acquired from the FinnGen biobank. The MR method is an accurate way to determine the underlying causality between the exposure and the outcome, which can minimize the effect of potential confounders and avoid reverse causality. Main results and the role of chance Our results demonstrated that five serum lipid-related traits were causally linked to lower odds of endometriosis, including very low-density lipoprotein (VLDL) cholesterol (OR: 1.231, 95% CI: 1.035-1.463, p = 0.019), cholesterol in chylomicron (CM) and extremely large VLDL (OR: 1.290, 95% CI: 1.013-1.644, p = 0.039), total cholesterol (TC) (OR: 1.250, 95% CI: 1.001-1.562, p = 0.049), apolipoprotein B (OR: 1.273, 95% CI: 1.022-1.587, p = 0.031), the ratio of apolipoprotein B to apolipoprotein A1 (OR: 1.163, 95% CI: 1.011-1.337, p = 0.035), while high-density lipoprotein cholesterol (HDL) (OR: 0.853, 95% CI: 0.754-0.963, p = 0.011) and apolipoprotein A1 (OR: 0.874, 95% CI: 0.770-0.993, p = 0.038) showed the opposite. Subgroup analyses about subtypes of endometriosis suggested the causal association of several serum lipid and apolipoprotein levels with peritoneal endometriosis. Neither pleiotropy nor heterogeneity was found in our study. No evidence was detected on the causal relations of low-density lipoprotein cholesterol (LDL) (OR: 1.014, 95% CI: 0.859-1.197, p = 0.868 and OR: 1.029, 95% CI: 0.906-1.169, p = 0.659, respectively) and triglyceride (TG) (OR: 1.138, 95% CI: 0.978-1.324, p = 0.095) with endometriosis. Limitations, reasons for caution Several limitations do exist in this study. We could not control the uniformity of sex between the exposure and the outcome, due to a lack of suitable summary-level data. Furthermore, it is possible that some of the null findings were due to limited statistical power from current genome-wide association study. Wider implications of the findings Our study found strong evidence for the possibility that aberrant lipid metabolism is causally involved in the pathogenesis of endometriosis, especially for endometriosis of pelvic peritoneum, and shed new insights on targeting serum lipid and apolipoprotein levels as a potential novel strategy for the prevention and treatment of endometriosis. Trial registration number not applicable
Read full abstract