Association between childhood adiposity and gynecologic cancers: a mendelian randomization analysis

Prior studies have reported connections between cathepsins (CTS) and gynecological cancers; however, the exact causal links are yet to be fully understood. Leveraging publicly accessible genome-wide association study summary datasets, we performed a two-sample bidirectional Mendelian randomization (MR) and multivariate MR (MVMR) analysis, with the inverse variance weighted (IVW) method as the primary approach. MR analysis demonstrated inverse associations between CTSB and cervical cancer (IVW: odds ratio [OR] = 0.9995, 95% confidence interval [CI] = 0.9991–0.9999, P = .0418), CTSE and ovarian cancer (IVW: OR = 0.9197, 95% CI = 0.8505–0.9944, P = .0358), CTSZ and ovarian cancer (IVW: OR = 0.9449, 95% CI = 0.8938–0.9990, P = .0459), CTSE and high grade serous ovarian cancer (IVW: OR = 0.8939, 95% CI = 0.8248–0.9689, P = .0063), and CTSZ and high grade serous ovarian cancer (IVW: OR = 0.9269, 95% CI = 0.8667–0.9913, P = .0268). A positive correlation was identified between CTSH and clear cell ovarian cancer (IVW: OR = 1.1496, 95% CI = 1.0368–1.2745, P = .0081). Nevertheless, subsequent adjustment for the false discovery rate revealed that none of the P-values retained statistical significance (PFDR > 0.05). MVMR analysis results elucidated that CTSZ was inversely associated with cervical cancer (IVW: OR = 0.9988, 95% CI = 0.9981–0.9996, P = .0022). Moreover, a positive association was noted between CTSF and cervical cancer (IVW: OR = 1.0007, 95% CI = 1.0000–1.0014, P = .0364), and similarly, between CTSS and cervical cancer (IVW: OR = 1.0005, 95% CI = 1.0000–1.0011, P = .0490). CTSO exhibited a positive association with non-endometrioid endometrial cancer (IVW: OR = 1.4405, 95% CI = 1.1864–1.7490, P < .001), and CTSH was positively associated with clear cell ovarian cancer (IVW: OR = 1.1167, 95% CI = 1.0131–1.2310, P = .0263). The MVMR analysis findings reveal that CTSZ emerges as a protective element against cervical cancer, whereas CTSF and CTSS represent risk factors for this disease. CTSO stands out as a risk factor for non-endometrioid endometrial cancer, and CTSH acts as a risk factor for clear cell ovarian cancer. This study elucidates causative connections between CTS and gynecological cancers, providing innovative insights for diagnostic and therapeutic optimization.

Numerous observational studies have investigated the potential link between hypertensive disorders of pregnancy (HDPs) and the subsequent risks of gynecologic tumors, yet the findings have been inconsistent. In this study, we utilized Mendelian randomization (MR) approach to assess the influence of HDPs on the future risks of ovarian, cervical, endometrial, and breast cancer and uterine fibroids, controlling for confounding factors. The genome-wide association studies (GWAS) summary data relevant to HDPs was obtained from the FinnGen databases (10,736 cases and 136,325 controls). Gynecologic tumor outcomes were extracted from the IEU Open GWAS project and UK Biobank (47,690 cases and 1, 092,073 controls). The inverse variance weighted (IVW) approach was selected as the principal method for MR analysis, supplemented by MR-Egger, weighted median, weighted model, simple model methods, MR pleiotropy residual sum and outlier (MR-PRESSO) test, and leave-one-out method. Multivariate MR (MVMR) analysis was conducted after adjusting systolic blood pressure (SBP), body mass index (BMI) and type 2 diabetes mellitus (T2DM). Our univariate MR analysis (UVMR) results revealed no significant relationship between HDPs and the risks of ovarian cancer (odds ratio [OR] = 0.924, p = 0.360), cervical cancer (OR = 1.230, p = 0.738), endometrial cancer (OR = 1.006, p = 0.949), uterine fibroids (OR = 1.155, p = 0.158), and breast cancer (OR = 0.792, p = 0.241) by IVW test. Similar results were observed in gestational hypertension and preeclampsia/eclampsia. Additionally, our study detected neither heterogeneity nor pleiotropy. MVMR analysis also provided no evidence of a causal association between HDPs and common gynecologic tumors after adjusting SBP, BMI, and T2DM. We discovered no causal relationship between HDPs and ovarian, cervical, endometrial, breast cancer, and uterine fibroids in European populations. However, present analysis did not explore the effect of HDPs on the subtypes of gynecologic tumors across varied ethnic populations, which may require additional research.

High serum uric acid (UA) levels have been linked to cancer development through chronic inflammation and oxidative damage. Traditional epidemiological studies have shown inconsistent results regarding the relationship between uric acid and gynecological cancers. This study uses Mendelian randomization (MR) to explore the potential association between serum UA levels and various gynecological cancers. In this two-sample MR study, summary statistical data of the genome-wide association studies (GWASs) on serum UA levels were extracted from the UK Biobank (UKB), and those on gynecological cancers were obtained from the FinnGen consortium, the Epidemiology of Endometrial Cancer Consortium (E2C2), and the Ovarian Cancer Association Consortium (OCAC). Inverse variance weighted (IVW), weighted median, MR-Egger, weighted mode, MR-Pleiotropy RESidual Sum and Outlier (MR-PRESSO), and MR-Radial methods were utilized to investigate the bidirectional causal associations of serum UA levels with gynecological cancers. The evaluation indexes were odds ratios (ORs) and confidence intervals (CIs). Tests for horizontal pleiotropism and heterogeneity of instrumental variables (IVs) were performed, respectively using MR-Egger test and Cochran's Q statistics. In addition, leave-one-out and MR scatter plots were employed for sensitivity analyses. IVW estimates suggested that serum UA levels elevated 1 unit had a potential causal association with higher odds of both cervical cancer (CC) (OR=1.147, 95% CI: 1.020-1.290) and invasive mucinous ovarian cancer (IMOC) (OR=1.199, 95% CI: 1.033-1.393). Also, endometrial carcinoma (EC) had a potential causal association with it (OR=1.012, 95% CI: 1.000-1.024). Additionally, sensitivity analyses showed the potential causal associations between UA and CC/IMOC were relatively robust. An elevated serum UA level had potential associations with CC and IMOC, whereas patients with EC should pay attention to it in clinical practice, which may reduce the potential risk of gynecological cancers. However, further evidence is needed to clarify the true relationships between UA and gynecological cancers.

To explore the causal associations between genetically predicted telomere length and gynecologic and breast cancers. This Mendelian randomization study used data from genome-wide association studies on telomere length and breast (BC), cervical cancer, endometrial (EC), and ovarian (OC) cancers. The primary analysis was performed using the inverse variance weighted (IVW) method, with confirmation using the weighted median, weighted mode, and MR-Egger methods. Heterogeneity was detected using Cochran's Q-test, horizontal pleiotropy using MR-Egger regression, outliers using MR-PRESSO, and discordant single-nucleotide polymorphisms using the leave-one-out method. The genetic prediction results indicated causal associations between the risk of telomere length and EC [IVW; OR = 1.29, 95% confidence interval (95%CI): 1.05-1.59, P= .02], leukocyte telomere length and EC (IVW; OR = 1.23, 95%CI: 1.01-1.51, P = .04), telomere length and OC (IVW; OR = 1.27, 95%CI: 1.01-1.60, P = .04), telomere length and BC (IVW; OR = 1.12, 95%CI: 1.01-1.23, P = .03), and leukocyte telomere length and BC (IVW; OR = 1.12, 95%CI: 1.02-1.24, P = .02). Cochran's Q-test revealed heterogeneity for telomere length and BC (P < .001), leukocyte telomere length and BC (P < .001), and B-cell telomere length and BC (P = .012). The MR-Egger regression results suggest that the analyses of telomere length and BC (P = .014) and leukocyte telomere length and BC (P = .044) were influenced by horizontal pleiotropy. The MR-PRESSO analysis indicated the presence of outliers in the analyses of telomere length and BC and leukocyte telomere length and breast cancer. After removing the outliers, the statistical significance remained. This MR study suggests a causal association between telomere length and BC, EC, and OC, warranting additional study. Key message What is already known on this topic? Previous research has indicated an association between telomere length and the risk of various cancers, including breast and gynecologic cancers. However, the causal relationship remained unclear, necessitating further exploration to establish whether telomere length could be a modifiable risk factor for these cancers. What this study adds? This study provides robust evidence of a causal relationship between genetically predicted telomere length and an increased risk of breast cancer, endometrial cancer, and ovarian cancer, with specific odds ratios indicating a significant association. It highlights that both leukocyte and overall telomere length are important factors in cancer risk. How this study might affect research, practice, or policy? The findings could inform future research into telomere length as a biomarker for cancer risk, promote investigations into telomere-targeting interventions, and influence guidelines on screening and preventive strategies for at-risk populations based on genetic predispositions.

Childhood adiposity and novel subtypes of diabetes in adults: a Mendelian randomisation and genome-wide genetic correlation study

BackgroundAlzheimer’s disease (AD) manifests with a higher rate of occurrence in women. Previous epidemiological studies have suggested a potential association between AD and gynecological cancers, but the causal relationship between them remains unclear. This study aims to explore the causal link between 12 types of gynecological cancers and AD using a bidirectional Mendelian randomization (MR) approach.MethodsWe obtained genetic correlation tools for AD using data from the most extensive genome-wide association study. Genetic correlation data for 12 types of gynecological cancers were also sourced from the Finnish Biobank. These cancers include breast cancer (BC), cervical adenocarcinoma (CA), cervical squamous cell carcinoma (CSCC), cervical cancer (CC), endometrial cancer (EC), ovarian endometrioid carcinoma (OEC), ovarian cancer (OC), ovarian serous carcinoma (OSC), breast carcinoma in situ (BCIS), cervical carcinoma in situ (CCIS), endometrial carcinoma in situ (ECIS), and vulvar carcinoma in situ (VCIS). We used the inverse-variance weighted (IVW) model for causal analysis and conducted horizontal pleiotropy tests, heterogeneity tests, MR-PRESSO tests, and leave-one-out analyses to ensure the robustness of our results. We also applied replication analysis and meta-analysis to further validate our experimental results.ResultsThe study found that EC (P_IVW =0.037, OR [95% CI] = 1.032 [1.002, 1.064]) and CCIS (P_IVW = 0.046, OR [95% CI] = 1.032 [1.011, 1.064]) increase the risk of AD, whereas OC was negatively correlated with AD (P_IVW = 0.016, OR [95% CI] = 0.974[0.954, 0.995]). In reverse MR analysis, AD increased the risk of CC (P_IVW = 0.039, OR [95% CI] = 1.395 [1.017, 1.914]) and VCIS (P_IVW = 0.041, OR [95% CI] = 1.761 [1.027, 2.021]), but was negatively correlated with OEC (P_IVW = 0.034, OR [95% CI] = 0.634 [0.417, 0.966]). Sensitivity analysis results demonstrated robustness. These findings were further substantiated through replication and meta-analyses.ConclusionsOur MR study supports a causal relationship between AD and gynecological cancers. This encourages further research into the incidence of gynecological cancers in female Alzheimer’s patients and the active prevention of AD.

Objective: this study aimed to investigate if childhood body mass index (BMI) causally contributed to the risk of endometrial cancer (EC), which had not been well answered. Methods: genetic instruments were selected using single-nucleotide polymorphisms (SNPs) associated with childhood BMI in European population from a large-scale genome-wide association studies (GWAS, n = 39,620). A two-sample Mendelian randomization (MR) study was performed to evaluate the effect of higher childhood BMI on risk of EC. The data for endometrioid EC was obtained from a GWAS dataset comprising 54,884 individuals (8,758 cases and 46,126 controls). Inverse variance weighting (IVW), weighted median, weighted mode, and MR-Egger regression approaches were applied. Results: we selected 16 SNPs with genome-wide significance in childhood BMI for the analysis. The IVW analysis provided a causal link between childhood BMI and EC (beta = 0.408, standard error [SE] = 0.088, p < 0.001). Similarly, the weighted median method also provided robust evidence for the causal correlation (beta = 0.390, SE = 0.119, p < 0.001). Although the MR-Egger regression did not achieve the same significance (beta = 0.071, SE = 0.362, p = 0.848), it showed a minimal intercept value indicating small bias for directionality of pleiotropic effects (intercept = 0.024; p = 0.354). Through Cochran's Q test and visual inspection via funnel plot, the assessment of heterogeneity found no evidence of heterogeneity or asymmetry in our findings, further supporting the absence of directional pleiotropy. Conclusions: childhood BMI and risk of EC might be causally related, and early-life intervention on weight control might be considered for children to reduce the life-span risk of EC.

PurposeIn recent years, research interest in the potential link between female infertility (FI) and gynecological cancer (GC), including ovarian cancer (OC), endometrial cancer (EC), cervical cancer (CC), and breast cancer (BC), has grown, yet findings remain inconclusive. This study aims to explore the causal relationship between FI and GC using bidirectional two-sample Mendelian randomization (MR) analyses, thereby informing future strategies for FI and GC prevention.MethodsWe utilized SNPs identified from genome-wide association studies (GWAS) on FI and GC. The inverse variance weighted (IVW) method served as the primary approach to assess the causal association between FI and GC. Additionally, five other MR methods—Weighted median, Weighted mode, MR-Egger, Simple mode, and Robust-Adjusted Profile Score—were employed to enhance result robustness and credibility.ResultsIn the forward MR analysis, our IVW results indicated no significant association between FI and GC (FI-BC: OR = 0.95, 95% CI: 0.83–1.09, P = 0.47, P-FDR = 0.775; FI-OC: OR = 1.01, 95% CI: 0.84–1.24, P = 0.789, P-FDR = 0.896; FI-CC: OR = 0.80, 95% CI: 0.61–1.06, P = 0.118, P-FDR = 0.775; FI-EC: OR = 1.07, 95% CI: 0.88–1.30, P = 0.490, P-FDR = 0.775).In the reverse MR analysis, we found a marginal association between BC and FI. However, after adjusting for multiple testing using the FDR method, no significant causal relationship was found between BC and FI, suggesting a marginal association (OR = 1.054, 95% CI: 1.001–1.108, P = 0.043, P-FDR = 0.331). For other cancers, no significant causal relationships were observed between OC, CC and EC with FI(OC-FI: OR = 1.043, 95% CI: 0.999–1.087, P = 0.051, P-FDR = 0.331;CC-FI: OR = 0.992, 95% CI: 0.956–1.028, P = 0.654, P-FDR = 0.836; EC-FI: OR = 1.006, 95% CI: 0.956–1.055, P = 0.809, P-FDR = 0.885).ConclusionsOur study found no significant causal relationship between FI and GC. However, a potential marginal association between BC and FI was observed. These findings underscore the need for further research to confirm this association and emphasize the importance of reproductive protection for young breast cancer patients to preserve fertility.

While physical activity (PA) patterns are epidemiologically linked to gynecologic pathologies, establishing causality remains uncertain. This Mendelian randomization (MR) study evaluated causal relationships between PA phenotypes and gynecologic disorders. Using European-ancestry female genome-wide association study (GWAS) data, we analyzed eight gynecologic disorders and four PA instruments. Primary inverse-variance weighted (IVW) estimates were supplemented by sensitivity analyses (MR-Egger, weighted median, MR-PRESSO) to address pleiotropy and heterogeneity. Our MR revealed subtype-specific effects: walking for pleasure [OR = 0.63, 95% CI: 0.41-0.96], strenuous sports [OR = 0.34, 95% CI: 0.16-0.74], and other exercises [OR = 0.70, 95% CI: 0.49-0.99]) reduced ER+ breast cancer (BC) risk, while leisure screen time increased risk (OR = 1.08, 95% CI: 1.004-1.163). Only other exercises protected against ER- BC (OR = 0.60, 95% CI: 0.38-0.96). Exercise frequency inversely correlated with endometrial cancer (EC) (walking: OR = 0.20; strenuous sports: OR = 0.06; other exercises: OR = 0.41) and endometriosis (ES) (strenuous sports: OR = 0.17; other exercises: OR = 0.26), but was positively associated with sedentary behavior (EC: OR = 1.36). Ovarian cyst (OCS) risk decreased with walking (OR = 0.34) and other exercises (OR = 0.28), yet increased with screen time (OR = 1.23). Paradoxically, walking showed strong positive (OR = 31.48) and strenuous exercise inverse (OR = 0.004) associations with polycystic ovary syndrome (PCOS). No causal links emerged for cervical cancer (CC), ovarian cancer (OC), or uterine fibroids (UF). This first MR evidence demonstrates PA's differential causal effects on gynecologic disorders, particularly highlighting exercise modality-specific protection against ER+ BC, EC, and ES. Paradoxical PCOS associations warrant mechanistic investigation. Findings underscore PA as a modifiable preventive factor and advocate for subtype-tailored exercise guidelines in women's health.

Epidemic of Childhood Obesity: Implications for Kidney Disease

Previous studies investigated the associations between obesity and temporomandibular disorders (TMDs), but the evidence for the causal inferences was unclear. We aimed to investigate the causal link between life course adiposity and TMDs. Mendelian randomization (MR) studies were performed using genetic instruments for birth weight (BW) (N = 261 932), childhood body mass index (BMI) (N = 39 620), childhood body size (N = 454 718), adult BMI (N = 99 998), body fat percentage (N = 454 633) and TMDs (N = 211 023). We assessed the overall effect of each life course adiposity factor via inverse-variance weighted (IVW), weighted median, and MR-Egger methods and performed extensive sensitivity analyses. Additionally, multivariable MR was conducted to evaluate the direct and indirect effects of childhood BMI on TMDs while accounting for BW and adult BMI, and vice versa. Univariable MR analyses revealed a causal effect of low childhood adiposity on an increased risk of TMDs (childhood BMI: IVW OR: 0.65, 95% CI: 0.54-0.78, p < .001; childhood body size: IVW OR: 0.56, 95% CI: 0.43-0.73, p < .001). No causal association existed between genetically predicted BW, adult BMI, or body fat percentage and TMDs. In the multivariable MR analyses, the effects of childhood BMI on TMDs occurrence remained significant and direct, even after adjusting for BW and adult BMI (multivariable IVW OR: 0.78, 95% CI: 0.61-0.99, p = .048). No pleiotropy and heterogeneity were detected (p > .05). Low childhood BMI might causally increase the risk of TMDs through a direct pathway.

BackgroundChildhood obesity is a growing global concern and is associated with cardiometabolic comorbidities in adulthood. However, the association between childhood body size and age-related macular degeneration (AMD) in late life remains to be investigated. We aimed to explore the association between childhood obesity and incident AMD and the underlying anatomical and physiobiological mechanisms.MethodsWe investigated the association between childhood body size and incident AMD using multivariable Cox regression models and its relation to retinal layer thickness using linear regression. We performed four-way decomposition mediation analyses to explore the underlying mechanism. Lastly, we used univariable Mendelian randomisation (UVMR) and multivariable Mendelian randomisation (MVMR) to evaluate and differentiate the causal effect of childhood and adulthood body mass index (BMI).ResultsOver a median follow-up of 12.8 years, 5026 incident AMD cases occurred among 487 009 participants. Plumper childhood body size at age 10 conferred independent risk to incident AMD in later life (adjusted hazards ratio (aHR) = 1.13; 95% confidence interval (CI) = 1.03, 1.24, P = 0.007) and was associated with photoreceptor outer segment layer thinning. Adulthood BMI mediated the association between childhood plumper body size and incident AMD (pure indirect effect = 33%; 95% CI = 9.9, 56.9, P = 0.05). Mediation analysis of adulthood physiobiological and immuno-metabolic function showed that 17 peripheral biomarkers of 7 categories significantly mediated the aforementioned pathway, with HbA1c and cystatin C showing the two largest effects. Mendelian randomisation suggested a potential causal association between childhood BMI and AMD (UVMR inverse variance weighted (IVW) odd ratio (OR) = 1.50; 95% CI = 1.09, 2.08, P = 0.013), independent of adulthood BMI (MVMR adulthood BMI-adjusted IVW OR = 1.29; 95% CI = 1.03, 1.61, P = 0.024).ConclusionsChildhood obesity may be a causal risk factor for incident AMD in later life, partially mediated by persistent obesity and physiobiological memory. Prevention of retinal degenerative diseases should therefore begin in childhood, whereby children should be encouraged and supported to maintain a normal body size.

Despite adiposity's potential role in modulating immune responses, the causal effects of adiposity at different life stages on immune responses to COVID-19 vaccines remain poorly understood. Thus, we aimed to investigate the causal association between life-course adiposity and the immune response to COVID-19 vaccination, and to explore the metabolic mechanisms underlying this association. We used a life-course Mendelian randomization (MR) analytic framework, starting with univariable MR, to estimate the total effects of body mass index (BMI) and obesity in childhood and adulthood on the immune response to COVID-19 vaccination. We then used multivariable MR to distinguish the direct and indirect effects of childhood adiposity, adjusting for adulthood adiposity and lifestyle confounders. Furthermore, we conducted a metabolome-wide mediation MR analysis to explore the potential links between adiposity at different life stages and immune responses to COVID-19 vaccination, mediated by 249 metabolic traits. The univariable MR results revealed that genetically predicted childhood and adulthood BMIs were significantly associated with an increased risk of anti-spike IgG seronegativity after both the first and second doses of the COVID-19 vaccine. The multivariable MR analyzes further revealed that childhood BMI had a direct causal effect on anti-spike IgG seronegativity after the first vaccine dose (OR 1.13, 95% CI 1.00-1.27), whereas adulthood BMI did not (OR 0.99, 95% CI 0.87-1.13). In contrast, adulthood BMI had a direct causal effect on anti-spike IgG seronegativity after the second vaccine dose (OR 1.29, 95% CI 1.14-1.46). Mediation analyzes identified 71 metabolic traits that mediated the effects of childhood and adulthood BMI on the immune response to COVID-19 vaccination. Notably, branched-chain amino acids (valine, leucine, and isoleucine) mediated over 20% of the causal effect between adulthood BMI and the vaccine-induced immune response. Furthermore, the triglycerides to total lipids ratio in medium LDL was identified as a significant mediator of the causal relationship between childhood BMI and anti-spike IgG seronegativity after the second vaccine dose, but not for adulthood BMI. The findings highlight the potential for targeted interventions to improve vaccine responses, particularly in individuals with a high childhood BMI.

Growing evidence has suggested that gut microbiota is associated with gynecologic cancers. However, whether there is a causal relationship between these associations remains to be determined. A two-sample Mendelian randomization (MR) evaluation was carried out to investigate the mechanism associating gut microbiota and 3 prevalent gynecological cancers, ovarian cancer (OC), endometrial cancer, and cervical cancer as well as their subtypes in individuals of European ancestry. The Genome-wide association studies statistics, which are publically accessible, were used. Eligible instrumental single nucleotide polymorphisms that were significantly related to the gut microbiota were selected. Multiple MR analysis approaches were carried out, including inverse variance weighted, MR-Egger, Weighted Median methods, and a range of sensitivity analyses. Lastly, we undertook a reverse MR analysis to evaluate the potential of reverse causality. We sifted through 196 bacterial taxa and identified 33 suggestive causal relationships between genetic liability in the gut microbiota and gynecological cancers. We found that 11 of these genera could be pathogenic risk factors for gynecological cancers, while 19 could lessen the risk of cancer. In the other direction, gynecological cancers altered gut microbiota composition. Our MR analysis revealed that the gut microbiota was causally associated with OC, endometrial cancer, and cervical cancer. This may assist in providing new insights for further mechanistic and clinical studies of microbiota-mediated gynecological cancer.

Disparities in Oncofertility: Assisted reproductive technology utilization and fertility-sparing oncology care in women with a history of breast, ovarian, cervical or endometrial cancer (094)