The gut microbiome and ovarian cysts: a mendelian randomization study.

Ultimately, a total of 190 SNPs associated with 211 bacterial taxa were extracted under the genome-wide significance threshold ( P  < 1.0 × 10⁻ 5 ). The number of SNPs associated with each bacterial taxon ranged from 4 to 17, with no taxa containing only a single SNP in each outcome dataset. In this study, all F-statistics exceeded 10, indicating the absence of weak instrumental variables. Detailed information, including the effect allele, other allele, Beta, SE, P -value, and F-statistic for the instrumental variables, is provided in Supplementary Table 1. Using IVW as the primary MR analysis method, the analysis revealed suggestive causal relationships between certain gut microbiota and a reduced risk of ovarian cysts. Nine bacterial taxa were associated with a decreased risk of ovarian cysts, specifically class_Actinobacteria(OR = 0.916,95%CI = 0.858–0.978, P  = 0.009), family_Bifidobacteriaceae(OR = 0.917, 95% CI = 0.863–0.974, P  = 0.005), genus_Bifidobacterium (OR = 0.874, 95% CI = 0.819–0.932, P  = 0.000), genus_Ruminococcus (OR = 0.911, 95% CI = 0.846–0.981, P  = 0.014), species_Bacteroides_cellulosilyticus (OR = 0.944, 95% CI = 0.894–0.996, P  = 0.035), species_Bifidobacterium_adolescentis (OR = 0.931, 95% CI = 0.878–0.987, P  = 0.016), species_Escherichia_coli (OR = 0.920, 95% CI = 0.873–0.970, P  = 0.002), species_Lachnospiraceae_bacterium_8_1_57FAA (OR = 0.967, 95% CI = 0.938–0.997, P  = 0.030), and species_Phascolarctobacterium_succinatutens (OR = 0.947, 95% CI = 0.904–0.991, P  = 0.020). In contrast, 11 gut microbiota taxa were associated with an increased risk of ovarian cysts, including phylum_Firmicutes (OR = 1.159, 95%CI = 1.009–1.330, P  = 0.036), order_Clostridiales(OR = 1.145,95% CI = 1.024–1.280, P  = 0.018), order_Lactobacillales (OR = 1.091, 95% CI = 1.038–1.148, P  = 0.001), family_Clostridiales_noname (OR = 1.085, 95% CI = 1.008–1.167, P  = 0.029), genus_Oscillibacter (OR = 1.288, 95% CI = 1.156–1.435, P  = 0.000), genus_Pseudoflavonifractor (OR = 1.083, 95% CI = 1.022–1.147, P  = 0.007), species_Bacteroides_caccae (OR = 1.068, 95% CI = 1.000–1.141, P  = 0.048), species_Bacteroides_massiliensis (OR = 1.100, 95% CI = 1.022–1.185, P  = 0.011), species_Bacteroides_plebeius (OR = 1.049, 95% CI = 1.003–1.097, P  = 0.038), species_Eubacterium_eligens (OR = 1.119, 95% CI = 1.047–1.197, P  = 0.001), and species_Oscillibacter_unclassified (OR = 1.294, 95% CI = 1.156–1.447, P  = 0.000), as shown in Table  1 . Table 1 Mendelian randomization estimates for the association between gut microbiota and ovarian cysts nSNPs  the number of SNPs being used as IVs, OR  odds ratio, IVW  Inverse variance weighted Mendelian randomization estimates for the association between gut microbiota and ovarian cysts nSNPs  the number of SNPs being used as IVs, OR  odds ratio, IVW  Inverse variance weighted The results of Cochran’s Q test and MR-Egger test using the IVW method showed no significant heterogeneity between the gut microbiome and ovarian cysts, with P > 0.05. The intercept analysis from MR-Egger regression indicated no evidence of horizontal pleiotropy between the gut microbiome and ovarian cysts, with P > 0.05. Additionally, the MR-PRESSO analysis did not detect any significant outliers (Table  2 ). The “leave-one-out” analysis further demonstrated that most of the identified causal associations were not driven by any single instrumental variable, further validating the robustness of the data. family_Clostridiales_noname, species_Bifidobacterium_adolescentis, species_ Bacteroides_plebeius failed the test. (Supplementary Fig. 3). Table 2 Sensitivity analysis results Sensitivity analysis results

We employed a two-sample MR approach to investigate the genetic association between the human gut microbiome and ovarian cysts. In this study design, Single nucleotide polymorphisms (SNPs) related to the gut microbiome were regarded as exposure variables, while SNPs associated with ovarian cysts were considered as outcome variables. Our primary focus was to analyze whether the gut microbiome exerts a causal effect on ovarian cysts. To ensure the validity of the MR analysis, we adhered to three key assumptions: (i) a robust and statistically significant association exists between the exposure variable and the instrumental variable, (ii) the exposure variable is independent of any potential confounding factors related to the outcome and is unaffected by them, and (iii) the exposure variable serves as the sole mediator in the association between the instrumental variable and the outcome, with no other variables intervening in this relationship [ 17 ]. The relevant data on genetic variants of the human gut microbiome were derived from the summary statistics of the MiBioGen study, the largest multi-ethnic meta-analysis of the gut microbiome to date [ 18 ]. This study included 18,340 participants from diverse populations, including European, Hispanic/Latino from the United States, and East Asian. These participants were recruited across 24 cohorts in the MiBioGen consortium, covering multiple countries: the United States, Canada, Israel, South Korea, Germany, Denmark, the Netherlands, Belgium, Sweden, Finland, and the United Kingdom. The study encompassed 211 bacterial taxa, including 9 phyla, 16 classes, 20 orders, 35 families, and 131 genera. After excluding 15 traits of unknown species (at unknown family or genus levels), the analysis focused on 196 bacterial traits. The GWAS data on the human gut microbiome are accessible at https://mibiogen.gcc.rug.nl/menu/main/home . The relevant data on genetic variants for ovarian cysts were obtained from FinnGen Biobank’s 11th round of analysis, which included 145,032 individuals of European descent, comprising 25,564 cases and 119,468 controls. Diagnoses of ovarian cysts primarily relied on the International Classification of Diseases, 10th Revision (ICD-10), specifically including ovarian follicular cysts (N83.0), corpus luteum cysts (N83.1), and other or unspecified ovarian cysts (N83.2) [ 19 ]. The GWAS data for ovarian cysts can be accessed at https://r11.finngen.fi/ . In this study, to ensure the accuracy and reliability of the causal relationship between the gut microbiome and ovarian cysts, we applied the following methods to select eligible SNPs from the gut microbiome data as IVs. First, SNPs meeting a genome-wide significance threshold ( P  < 1.0 × 10⁻ 5 ) were selected as potential IVs. Second, to ensure that the IVs for the gut microbiome were independent, we used PLINK software to exclude SNPs exhibiting linkage disequilibrium effects (clump_kb = 500 and clump_r2 = 0.1). Third, to ensure that the SNPs for exposure and outcome were affected by the same allele, palindromic SNPs were excluded from the IV set. The IVs for the gut microbiome were extracted from each outcome dataset. When specific SNPs were absent in the GWAS results, proxy SNPs were not sought by default. The strength of the included IVs was assessed using R^2 and the F-statistic R^2 reflects the proportion of variance in the exposure explained by the IVs, with the calculation formula as follows: R2 = 2 × EAF × (1 − EAF) × b2/[2 × EAF × (1 − EAF) × b2 + 2 × EAF × (1 − EAF) × N × se2] (EAF: effect allele frequency, se: standard error of the effect size, b: effect size, N: sample size). The F-statistic is calculated as F = R2 × (N − 2)/(1 − R2) (N: sample size). An F-statistic exceeding 10 indicates the absence of weak instrument bias [ 20 ]. Five MR methods were used to examine the causal relationship between the gut microbiome and ovarian cysts: IVW, Simple mode, MR-Egger, weighted median, and weighted mode. IVW was chosen as the primary analysis method, as it aims to estimate the causal effect of each SNP under the assumption of no pleiotropy, providing more conservative and reliable estimates under certain conditions compared to other methods [ 21 ]. The IVW results are more reliable in the absence of heterogeneity and pleiotropy. However, when heterogeneity is present but pleiotropy is absent, the weighted median method is more reliable. In cases of pleiotropy, the MR-Egger method performs better [ 22 ]. We used Cochran’s Q test to assess heterogeneity among SNPs, with a p-value less than 0.05 indicating significant heterogeneity [ 23 ]. MR-Egger regression was applied to evaluate potential pleiotropic bias [ 24 ]. MR-PRESSO was used to reduce horizontal pleiotropy by detecting and removing outliers, providing higher precision than MR-Egger [ 25 ]. Additionally, to assess the robustness of the results, we conducted a"leave-one-out"analysis to evaluate their stability [ 26 ]. MR analyses were carried out using the sofmGutMR (version 0.23) in R (version 4.3.3).

This study used MR analysis to investigate the potential causal relationship between the gut microbiome and ovarian cysts. Based on genome-wide significance levels, our findings suggest that 17 bacterial taxa exhibit suggestive causal relationships with ovarian cysts. Through analysis of OR values, we identified 9 gut microbiotas associated with a higher risk of ovarian cysts and 8 associated with a lower risk, indicating a regulatory role of the gut microbiome in the development of ovarian cysts. Most ovarian diseases are benign, primarily comprising functional ovarian cysts and benign tumors. Functional ovarian cysts include follicular cysts, corpus luteum cysts, and theca lutein cysts, all of which are benign and generally self-limiting [ 27 , 28 ]. The ovarian cysts studied in this research primarily include follicular cysts, corpus luteum cysts, and other unspecified ovarian cysts. Follicular cysts are the most common type of functional ovarian cysts, characterized by a smooth, thin wall and a unilocular structure. Currently, there is no consensus on the exact mechanism underlying the formation of ovarian cysts. Sex hormones are key factors in the development of ovarian cysts, and increasing evidence suggests an interaction between the gut microbiota and sex hormones. Participants were grouped according to their serum testosterone and estradiol levels, and the results indicated that individuals in the high-level group exhibited greater gut microbiota diversity. In the female cohort, those in the high group showed an increased abundance of Bacteroidetes and a reduced presence of Firmicutes [ 29 ]. Excess androgens promote apoptosis in the inner layer of granulosa cells in preantral follicles, leading to the formation of follicular cysts [ 30 ]. Our study found that three Bacteroides species were associated with an increased risk of ovarian cysts. Studies on mice have shown elevated levels of estradiol, progesterone, and corticosterone following microbial colonization [ 31 ]. The development of ovarian cysts has been associated with high levels of estradiol, [ 32 ] while exogenous progesterone injections have been used to treat ovarian cysts in cattle [ 33 ]. The gut microbiota can regulate bile acid metabolism and inhibit its synthesis in the liver by suppressing farnesoid X receptor signaling in the intestine, which modulates the expression of fibroblast growth factor 15 (FGF15) in the ileum and cholesterol 7α-hydroxylase (CYP7A1) in the liver [ 34 , 35 ]. Testosterone is synthesized from bile acids, [ 36 ] which are influenced by microbiota; thus, the microbiota may indirectly affect testosterone levels. The gut microbiome is also involved in modulating the free concentrations of dihydrotestosterone (DHT) and testosterone, which can be significant [ 37 ]. Additionally, the gut microbiota produces β-glucuronidase, an enzyme that can prevent the conjugation of estrogen with glucuronic acid, thereby increasing estrogen levels in the body [ 38 ]. Although we have thoroughly explored the mechanistic role of sex hormones, conclusive evidence regarding the causal relationship between specific microbial taxa and cyst development is still lacking, and the connection remains speculative. Therefore, we advocate for further validation through functional or clinical studies to elucidate the nature of these associations and assess their potential clinical significance. This study has several strengths and limitations. To our knowledge, it is the first MR analysis evaluating the causal relationship between the gut microbiome and ovarian cysts. The use of MR methods to analyze the causal link between ovarian cysts and gut microbiota minimizes the impact of confounding factors and may offer more convincing evidence than observational studies. However, due to the unknown biological functions of many genetic variants, we cannot entirely eliminate the potential influence of horizontal pleiotropy. Thus, the results should be interpreted cautiously. Techniques such as MR-PRESSO and MR-Egger were employed to mitigate the effects of horizontal pleiotropy and heterogeneity, ensuring result accuracy. This study classified the gut microbiome at the species level, providing a more comprehensive and detailed outcome than previous research. However, this study also has limitations. First, there is a lack of standardized methods and criteria for measuring gut microbiota, with considerable variation in sample extraction and management. Differences in sequencing platforms and analytical processes may lead to inconsistent results and limit comparability. Second, to reduce heterogeneity and pleiotropy in the selection of instrumental variables, the analysis was restricted to individuals of European ancestry. While this approach improves internal validity, it limits the generalizability of the findings to other populations. Notably, substantial differences in gut microbiome composition have been observed across ethnic groups, influenced by genetic background, diet, lifestyle, and environmental exposures. Therefore, the causal relationships identified in this study may not be directly applicable to non-European populations. Third, GWAS data from the MiBioGen and FinnGen consortia may introduce selection bias. Additionally, the outcome definition in this study was based on ICD-10 codes (N83.0, N83.1, N83.2), which encompass ovarian follicular cysts, corpus luteum cysts, and other or unspecified ovarian cysts. The broad scope of this definition may introduce heterogeneity in the outcomes, potentially diluting or obscuring more specific causal associations between certain gut microbial genera and distinct ovarian cyst subtypes, particularly given that some cysts are transient or non-pathological. Due to limitations in the available GWAS summary data, stratification by cyst subtype was not feasible in this study. Future research, incorporating more refined phenotypic data and larger sample sizes for specific cyst types, may provide further validation and extend the findings of this study. Overall, we conducted a comprehensive evaluation of the potential link between gut microbiota and ovarian cysts. These findings may offer new biomarker candidates for future research and provide innovative perspectives for therapeutic strategies.

Ovarian cysts are a common benign gynecological condition, with a reported prevalence of 34.9% and an incidence rate of 15.3% in premenopausal women; for postmenopausal women, the prevalence is 17.0%, and the incidence rate is 8.2% [ 1 ]. Serum markers CA125 and HE4 aid in differentiating between benign and malignant ovarian cysts [ 2 ]. Studies indicate that benign ovarian cysts with a diameter of less than 5 cm may regress spontaneously; however, surgical intervention may be an option for cysts larger than 5 cm or those with complex structures [ 3 ]. Combined oral contraceptives appear to offer no benefit for treating ovarian cysts [ 4 ]. The human gut harbors the most complex microbiome, significantly impacting host homeostasis and immune suppression, thus playing a crucial role in maintaining health [ 5 ]. It is estimated that the gut microbiome encodes approximately 100 times more genes than the entire human genome, though later studies suggest this number is closer to 10 times more [ 6 ]. The majority of gut microbiota belong to the phyla Firmicutes and Bacteroidetes, with smaller proportions of Proteobacteria, Actinobacteria, Fusobacteria, and Verrucomicrobia [ 7 ]. Currently, the gut microbiota has become a focal point of research in cancer, immune diseases, and metabolic disorders [ 8 , 9 ]. The gut microbiome can influence ovarian diseases through various pathways. The interactions between the gut microbiome and hormones, such as estrogen, androgen, and insulin, contribute to polycystic ovary syndrome (PCOS) [ 10 ]. There is also an association between the gut microbiome and endometriosis via mechanisms involving estrogen, immune inflammation, and tumor characteristics [ 11 ]. Evidence suggests that the gut microbiota plays a role in regulating systemic inflammation and immune responses, thus impacting the ovarian tumor microenvironment. Dysbiosis, characterized by microbial imbalance, is linked to an inflammatory environment conducive to cancer development [ 12 ]. Restoration of the gut microbiome may also help overcome platinum resistance in patients with epithelial ovarian cancer [ 13 ]. The findings indicate that gut microbiota and its metabolic products not only serve as potential biomarkers for the diagnosis of endometrial cancer but can also be modulated by the active components of traditional Chinese medicine to exert anti-inflammatory and anti-tumor effects [ 14 ]. Although current studies on the association between gut microbiota and gynecological cysts are relatively limited, the findings of this study are expected to provide critical reference points for future research directions and lay a foundation for in-depth exploration in this field. MR is an effective method for inferring causal relationships between exposure and outcomes by using genetic variants closely related to the exposure as instrumental variables (IVs) [ 15 ]. MR can be regarded as a natural randomized controlled trial (RCT) that is less susceptible to confounding factors, providing high levels of evidence. Moreover, the risk estimated through MR reflects a lifetime risk, which is longer than the follow-up period in RCTs [ 16 ]. MR has been widely cited across various disease fields. This study aims to analyze the potential causal relationship between different levels of gut microbiota and ovarian cysts using the MR approach, offering new insights for therapeutic strategies.

Supplementary Material 1. Supplementary Material 1. Supplementary Material 2. Supplementary Material 2.