The association between metabolic syndrome and lung cancer risk: a Mendelian randomization study Authors: Zhicheng Wei, Yunyun Hu, Fang Zuo, Xiushu Wen, Desheng Wu, Xiaodong Sun, Conghai Liu Published: 2024 ABSTRACT Metabolic syndrome (MetS) is closely linked to cancer development, with emerging evidence suggesting its association with pulmonary carcinoma. However, causal relationships remain unclear due to observational study limitations. Employing Mendelian randomization, we investigated the causal link between MetS and lung cancer (LC) susceptibility. The data utilized in this study were obtained from the publicly available genetic variation summary database. The causal relationship was assessed using the inverse variance weighting method (IVW), weighted median method, and MR-Egger regression. A sensitivity analysis was carried out to confirm the robustness of the findings. Furthermore, risk factor analyses were conducted to explore potential mediators. Utilizing various analyses, MetS demonstrated a significant positive association with LC (OR, 1.22; 95% CI, 1.09–1.37, p = 7.57 × 10− 4), lung squamous cell carcinoma (LUSC) (OR, 1.47; 95%, 1.23–1.75, p = 2.22 × 10− 5), and small cell lung cancer (SCLC) (OR, 1.76; 95% CI, 1.37–2.26, p = 8.20 × 10− 6) but not lung adenocarcinoma (LUAD) (OR, 1.08; 95% CI, 0.94–1.24, p = 0.28). Risk factor analyses indicated that smoking, alcohol, body mass index, education, and type 2 diabetes might mediate the association. This study genetically validates and reinforces the evidence of MetS increasing the incidence of LC, including both LUSC) and SCLC, especially among individuals with abdominal obesity. It provides valuable insights for the development of lung cancer prevention strategies and directions for future research.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-024-79260-y. Introduction Lung cancer (LC) has emerged as the leading malignancy posing a significant threat to human health. LC is a major global health concern with high mortality rates and a significant economic burden1,2. According to the World Health Organization’s findings, LC is the predominant cause of global cancer mortality, accounting for approximately 1.8 million annual fatalities3. LC is responsible for a significant number of disability-adjusted life years lost4. With the progression of population aging, numerous countries continue to experience a steady increase in the absolute number of LC cases and mortality, projecting an estimated 2.1 million male and 0.9 million female deaths from LC by 2035 on a global scale5. Given the alarming trend of LC prevalence and the substantial disease burden it entails, there is an urgent imperative to foster widespread awareness and prevention and control of LC. Metabolic syndrome (MetS), delineated as a clinical composite, is defined by an array of perturbed metabolic aberrations, encompassing central adiposity, high blood pressure, high blood glucose, and disordered lipometabolism6,7. The ubiquity of MetS is progressively surging globally, emerging as a cardinal etiological facet within the spectrum of cardiovascular pathologies, type 2 diabetes (T2D), and a spectrum of metabolic disorders6,8. In recent years, research has gradually unveiled that MetS might transcend its associations not exclusively with cardiovascular ailments and diabetes but also align with an augmented proclivity toward select malignancies9. Despite the fledgling status of inquiry within this domain, mounting empirical evidence indicates a potential nexus between MetS and an array of malignancies, including but not limited to LC10–12. However, existing academic research predominantly pivots upon observational methodologies, and their accuracy in revealing the definitive causal rapport between MetS and oncogenic diathesis needs to be improved. The Mendelian randomization (MR) research methodology serves as a potent tool in epidemiological studies, where its fundamental concept revolves around utilizing single nucleotide polymorphisms (SNPs) as instrumental variables to assess causal relationships between risk factors and specific diseases13. The application of the MR research methodology effectively circumvents issues encountered in observational studies, such as confounding factors, reverse causality, and challenges in achieving representativeness and feasibility seen in randomized controlled trials (RCTs)14. After thorough literature research, we have yet to find any reports utilizing MR to investigate the causal relationship between MetS and LC. Against this research backdrop, our aim is to employ the MR method to delve deeper into the potential causal chain between MetS and LC risk. This approach seeks to accurately assess the impact of MetS as an exposure factor on LC occurrence, thereby circumventing the interference of confounding factors and reverse causation commonly encountered in traditional observational studies. This study not only provides novel and robust evidence supporting the causal relationship between MetS and LC, effectively supplementing and reinforcing the existing evidence base but also enhances our understanding of the complex interplay between metabolic dysregulation and LC risk. It offers new perspectives and insights for the formulation of LC prevention strategies. Additionally, the findings of this study offer valuable insights for future research directions and clinical practices. Methods Design Figure 1 shows the study design. By employing Genome-wide association studies (GWAS) summary statistics, we adopted a two-sample MR analysis and assessed the causative association between MetS and susceptibility to LC and its subtypes. Furthermore, we conducted sensitivity analyses to ascertain the robustness of our results. The tenets of MR studies necessitate the fulfillment of three pivotal assumptions: (a) genetic instruments exhibit predictive capacity for focal exposure (P < 5 × 10− 8); (b) genetic instruments remain orthogonal to latent confounders; and (c) genetic instruments solely impact the outcome through mediating risk factors15. It is important to note that all data referenced from the GWASs cited in this work have undergone transparent publication and are publicly accessible, thereby obviating ethical scrutiny concerns. Fig. 1Study design overview. MetS Metabolic syndrome, WC Waist circumference, HTN Hypertension, FG Fasting glucose, TG Triglycerides, HDL-C High-density lipoprotein cholesterol, LC Lung cancer, LUAD Lung adenocarcinoma, LUSC Squamous cell lung carcinoma, SCLC Small cell lung carcinoma, BMI Body mass index, T2D Type-2 diabetes, SNP Single nucleotide polymorphism. Study design overview. MetS Metabolic syndrome, WC Waist circumference, HTN Hypertension, FG Fasting glucose, TG Triglycerides, HDL-C High-density lipoprotein cholesterol, LC Lung cancer, LUAD Lung adenocarcinoma, LUSC Squamous cell lung carcinoma, SCLC Small cell lung carcinoma, BMI Body mass index, T2D Type-2 diabetes, SNP Single nucleotide polymorphism. Cohort selection Our study has specifically chosen a publicly accessible, Europe-focused genetic variation aggregation database, with the selection informed by the following considerations: (a) European Ethnic Specificity: The database is dedicated to collating genetic information from individuals of European descent, thereby inherently limiting the racial generalizability of our study. However, this design ensures a deeper investigation of the causal relationship between metabolic syndrome and lung cancer risk within a higher homogeneity population, mitigating potential confounding influences arising from interethnic genetic heterogeneity; (b) Data Quality: Genetic data within the selected database have undergone rigorous quality control procedures, guaranteeing the precision and uniformity of SNP data. Such high-quality data provide reliable genetic markers for Mendelian randomization analysis, underpinning sound causal inference; (c) Genetic Variant Coverage: The database encompasses a wide range of genetic variations intimately tied to both metabolic syndrome as an entity and its constitutive elements. This rich repository furnishes our study with a wealth of genetic instrumental variables, facilitating a comprehensive unraveling of latent causal connections; (d) Purposeful Study Design: Despite its exclusive focus on the European population, our study aims to definitively elucidate the causal effect of metabolic syndrome on lung cancer risk within a specific genetic milieu. This effort serves as a theoretical underpinning and a reference point for analogous inquiries to be conducted in other ethnic groups in subsequent research. Source of exposure data and instrument variable selection This study adopts the internationally recognized definition of metabolic syndrome, which is jointly supported by authoritative organizations such as the International Diabetes Federation (IDF) and the American Heart Association (AHA). It integrates multiple metabolic abnormalities, including central obesity, hypertension, hyperglycemia, and lipid metabolism disorders16,17. Specifically, individuals who meet at least three of the following criteria are diagnosed with metabolic syndrome: (a) Central obesity, typically measured by waist circumference (WC), with a threshold of ≥ 90 cm for men and ≥ 80 cm for women (note that specific thresholds may vary depending on ethnicity and region); (b) Hypertension, defined as having a clinic blood pressure measurement of ≥ 130/85 mmHg or a known history of hypertension under treatment; (c) Hyperglycemia, defined as having a fasting glucose (FG) level of ≥ 100 mg/dL or a known history of type 2 diabetes (T2D) under treatment; (d) Lipid abnormalities, characterized by plasma triglyceride (TG) levels of ≥ 150 mg/dL or high-density lipoprotein cholesterol (HDL-C) levels of < 40 mg/dL for men and < 50 mg/dL for women. In this study, the selection of exposure variables encompassed MetS and its components: WC, FG, HTN, TG and HDL-C, all serving as focal exposure factors. For MetS, genetic instruments are based on the latest data from the Complex Trait Genetics Lab (CTG), which contains 461,920 valid subjects of European ancestry and identifies genetic variation in MetS through the structural equation modeling method18. For WC, we extracted the GWAS summary data from the Medical Research Council Integrative Epidemiology Unit (MRC-IEU) UK Biobank GWAS Pipeline, which included 462,166 subjects with European ancestry19. For HTN, the summary statistics were also available from the MRC-IEU UK Biobank pipeline, which included 462,933 subjects20. For FG, we obtained summary-level data from the most comprehensive GWAS in the meta-analyses of glucose and insulin-related traits Consortium (MAGIC), which included 200,622 subjects with European ancestry21. For TG and HDL-C, summary-level statistics were extracted from the UK Biobank with 441,016 and 115,082 participants22,23. Table 1 provides details of the data sources, and an F-statistic greater than 10 indicates that there is no weak instrumental variable problem24. Table 1Details of studies included in MR analyses.CategoryData sourcesPMIDAncestrySample sizeIdentified SNPs R 2 F-statisticWeb sourceMetSCTG35,983,957European461,9204013.53%42.08 https://ctg.cncr.nl/documents/p1651/CF_MetS_results_adapt.txt.gz WCMRC-IEU25,673,412European462,1663743.45%44.11 https://gwas.mrcieu.ac.uk/datasets/ukb-b-9405/ HTNMRC-IEU21,909,115European462,9332250.57%11.84 https://gwas.mrcieu.ac.uk/datasets/ukb-b-14057/ FGMAGIC34,059,833European200,622661.29%39.77 https://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST90002232/ TGUK Biobank32,203,549European441,0163138.90%137.47 https://gwas.mrcieu.ac.uk/datasets/ieu-b-111/ HDL-CUK Biobank35,213,538European115,082918.16%112.29 https://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST90092822/ LCGWAS Catalog28,604,730European85,716––– https://www.ebi.ac.uk/gwas/studies/GCST004748 LUADGWAS Catalog28,604,730European66,756––– https://www.ebi.ac.uk/gwas/studies/GCST004744 LUSCGWAS Catalog28,604,730European63,053––– https://www.ebi.ac.uk/gwas/studies/GCST004750 SCLCGWAS Catalog28,604,730European24,108––– https://www.ebi.ac.uk/gwas/studies/GCST004746 MetS Metabolic syndrome, WC Waist circumference, HTN Hypertension, FG Fasting glucose, TG Triglycerides, HDL-C High-density lipoprotein cholesterol, LC Lung cancer, LUAD Lung adenocarcinoma, LUSC Squamous cell lung carcinoma, SCLC Small cell lung carcinoma. Table 1Details of studies included in MR analyses.CategoryData sourcesPMIDAncestrySample sizeIdentified SNPs R 2 F-statisticWeb sourceMetSCTG35,983,957European461,9204013.53%42.08 https://ctg.cncr.nl/documents/p1651/CF_MetS_results_adapt.txt.gz WCMRC-IEU25,673,412European462,1663743.45%44.11 https://gwas.mrcieu.ac.uk/datasets/ukb-b-9405/ HTNMRC-IEU21,909,115European462,9332250.57%11.84 https://gwas.mrcieu.ac.uk/datasets/ukb-b-14057/ FGMAGIC34,059,833European200,622661.29%39.77 https://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST90002232/ TGUK Biobank32,203,549European441,0163138.90%137.47 https://gwas.mrcieu.ac.uk/datasets/ieu-b-111/ HDL-CUK Biobank35,213,538European115,082918.16%112.29 https://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST90092822/ LCGWAS Catalog28,604,730European85,716––– https://www.ebi.ac.uk/gwas/studies/GCST004748 LUADGWAS Catalog28,604,730European66,756––– https://www.ebi.ac.uk/gwas/studies/GCST004744 LUSCGWAS Catalog28,604,730European63,053––– https://www.ebi.ac.uk/gwas/studies/GCST004750 SCLCGWAS Catalog28,604,730European24,108––– https://www.ebi.ac.uk/gwas/studies/GCST004746 MetS Metabolic syndrome, WC Waist circumference, HTN Hypertension, FG Fasting glucose, TG Triglycerides, HDL-C High-density lipoprotein cholesterol, LC Lung cancer, LUAD Lung adenocarcinoma, LUSC Squamous cell lung carcinoma, SCLC Small cell lung carcinoma. Details of studies included in MR analyses. MetS Metabolic syndrome, WC Waist circumference, HTN Hypertension, FG Fasting glucose, TG Triglycerides, HDL-C High-density lipoprotein cholesterol, LC Lung cancer, LUAD Lung adenocarcinoma, LUSC Squamous cell lung carcinoma, SCLC Small cell lung carcinoma. In order to guarantee the reliability of the findings derived from our analyses, we adhered to rigorous criteria for the selection of instrumental variables25. (a) Only SNPs with genome-wide significance (P < 5 × 10− 8) were selected as instrumental variables. This criterion ensured a robust association between the selected SNPs and MetS or its components, thereby enhancing the validity of causal inference. (b) To circumvent the potential influence of linkage disequilibrium (LD) on the results, the SNPs were clustered using the PLINK algorithm (r2 threshold = 0.001 and window size = 10 mB), and the most representative SNPs were selected from each cluster. This step ensured the independence of the selected SNPs and reduced statistical bias due to LD. (c) SNPs that may have pleiotropic effects were further excluded. To ensure that the analysis results were not affected by pleiotropy, we conducted a comprehensive review of the literature to identify and exclude SNPs with known pleiotropic effects. Source of outcome data We sourced summary data on LC from the GWAS Catalog database (https://www.ebi.ac.uk/gwas/home). The data encompassed various subtypes, including LC with 29,266 cases and 56,450 controls, lung adenocarcinoma (LUAD) with 11,273 cases and 55,483 controls, squamous cell lung carcinoma (LUSC) with 7,426 cases and 55,627 controls, and small cell lung carcinoma (SCLC) with 2,664 cases of European descent and 21,444 controls26. Given that all GWAS data were obtained from individuals of European ancestry, our analysis was specifically limited to individuals of European descent. Table 1 provides details of the data sources. Statistical analysis Initially, we conducted univariable two-sample MR analysis, predominantly employing the inverse variance weighted (IVW) model. Additionally, we incorporated supplementary MR approaches, such as MR-Egger regression, weighted median, MR pleiotropy residual sum, and outlier (MR-PRESSO) test, to offer more reliable estimates across broader scenarios. The results were expressed as odds ratios (ORs) along with their corresponding 95% confidence intervals (CIs). We employed Cochran’s Q statistic to assess heterogeneity among individual genetic variations. The MR-Egger intercept test was used to evaluate potential horizontal pleiotropy. Moreover, we implemented the MR pleiotropy residual and outlier (MR-PRESSO) test to identify outlier SNPs and verify whether the causal effect direction changed after removing these outliers. Sensitivity analyses were carried out to evaluate whether the MR estimates were driven by individual SNPs or influenced by bias, and we calculated the residual SNP heritability. Asymmetry in the funnel plots could indicate a violation of the second assumption due to horizontal pleiotropy. To account for multiple testing in our primary analyses, a Bonferroni-corrected threshold of P<0.002(α = 0.05/24). All statistical tests were two-sided and executed using the TwoSampleMR, MR-PRESSO, and MR packages in R software (version 4.2.3). Risk factors To explore the underlying mechanisms genetically linking MetS with LC, we further calculated several potential risk factors for analysis, including smoking consumption (IEU GWAS ID: “ieu-a-961”), alcohol consumption (IEU GWAS ID: “ukb-b-5779”), educational attainment (IEU GWAS ID: “ieu-b-4836”), body mass index (BMI) (IEU GWAS ID: “ieu-b-40”), and type 2 diabetes (T2D) (IEU GWAS ID: “ebi-a-GCST007516”). These data are available in the IEU Open GWAS Project database27. Detailed information on each data source of risk factors is displayed in Supplementary Table 1. We used MR analysis to consider these risk factors as potential mediating variables, and for each risk factor we performed separate analyses to explore how MetS and WC mediate the association between MetS and lung cancer risk by assessing the effects of MetS and WC on these risk factors, and then exploring how they mediate the association between MetS and lung cancer risk. Specifically, we calculated the effects of MetS and WC on smoking, alcohol consumption, educational level, BMI and T2D using relevant data from the IEU Open GWAS Project database. For each risk factor, we performed MR analyses using IVW as the main method. MR-Egger regression and MR-PRESSO tests were used to assess heterogeneity and horizontal pleiotropy. Results MR estimates We conducted an analysis of the associations between MetS and its components (WC, FG, TG, HDL-C, HTN) and LC as well as its three subtypes (LUAD, LUSC, and SCLC). Forest plots depicting the associations between lifestyle characteristics and each LC subtype, generated using the IVW method, are shown in Fig. 2. Scatter plots displaying the associations between each genetic variant and the corresponding significantly (P < 0.05) correlated trait results are shown in Fig. 3. Complete results can be found in Supplementary Data Table 2. Fig. 2Forest plot of MetS and its components associated with the risk of LC (A), LUAD (B), LUSC (C), and SCLC (D). MetS Metabolic syndrome, WC Waist circumference, HTN Hypertension, FG Fasting glucose, TG Triglycerides, HDL-C High-density lipoprotein cholesterol, LC Lung cancer, LUAD Lung adenocarcinoma, LUSC Squamous cell lung carcinoma, SCLC Small cell lung carcinoma, SNP Single nucleotide polymorphism, OR Odds ratio, CI Confidence intervals. Forest plot of MetS and its components associated with the risk of LC (A), LUAD (B), LUSC (C), and SCLC (D). MetS Metabolic syndrome, WC Waist circumference, HTN Hypertension, FG Fasting glucose, TG Triglycerides, HDL-C High-density lipoprotein cholesterol, LC Lung cancer, LUAD Lung adenocarcinoma, LUSC Squamous cell lung carcinoma, SCLC Small cell lung carcinoma, SNP Single nucleotide polymorphism, OR Odds ratio, CI Confidence intervals. Fig. 3Scatter plots from genetically predicted MetS and WC on LC and its subtypes. A genetically predicted MetS on LC; B genetically predicted MetS on LUSC; C genetically predicted MetS on SCLC; D genetically predicted WC on LC; E genetically predicted WC on LUSC; F genetically predicted WC on SCLC. MetS Metabolic syndrome, WC Waist circumference, LC Lung cancer, LUAD Lung adenocarcinoma, LUSC Squamous cell lung carcinoma, SCLC Small cell lung carcinoma, SNP Single nucleotide polymorphism. Scatter plots from genetically predicted MetS and WC on LC and its subtypes. A genetically predicted MetS on LC; B genetically predicted MetS on LUSC; C genetically predicted MetS on SCLC; D genetically predicted WC on LC; E genetically predicted WC on LUSC; F genetically predicted WC on SCLC. MetS Metabolic syndrome, WC Waist circumference, LC Lung cancer, LUAD Lung adenocarcinoma, LUSC Squamous cell lung carcinoma, SCLC Small cell lung carcinoma, SNP Single nucleotide polymorphism. Table 2Heterogeneity and pleiotropy assessment. MetS Metabolic syndrome, WC Waist circumference, HTN Hypertension, FG Fasting glucose, TG Triglycerides, HDL-C High-density lipoprotein cholesterol, LC Lung cancer, LUAD Lung adenocarcinoma, LUSC Squamous cell lung carcinoma, SCLC Small cell lung carcinoma.ExposureOutcomeCochran Q testMR-EggerMR-PRESSOQQ P-valueInterceptPleiotropy test p-valueOutliersDistortion test p-valueOutlier-corrected p-valueMetSLC592.366.04E−130.0020.3730.821.21E−03WCLC547.868.03E−13-0.00070.8120.491.91E−03EHTNLC256.963.51E−030.0060.0710.920.81FGLC102.085.75E−04-0.0040.320––TGLC342.661.05E−040.0030.110––HDL-CLC101.390.0033442010.0030.470––MetSLUAD447.412.05E−030.0030.2910.880.19WCLUAD431.292.55E−04− 0.00030.9220.770.51HTNLUAD239.540.032558220.0040.3810.660.54FGLUAD75.471.17E−01− 0.0050.330––TGLUAD275.052.10E−010.0020.250––HDL-CLUAD57.790.69438740.0010.760––MetSLUSC517.742.81E−07− 0.0020.6820.951.93E−05WCLUSC479.331.69E−07− 0.0070.1320.99784.66E−04HTNLUSC284.777.63E−050.0040.430––FGLUSC78.073.34E−02− 0.0040.500––TGLUSC352.726.75E−050.0040.1620.600.21HDL-CLUSC63.530.5632387− 0.00040.930––MetSSCLC437.807.14E−030.0020.700––WCSCLC368.549.37E−02− 0.0020.710––HTNSCLC231.696.17E−020.0110.130––FGSCLC84.331.08E−02− 0.010.290––TGSCLC318.712.44E−030.00030.9310.910.67HDL-CSCLC75.650.17222460.0010.890–– Table 2Heterogeneity and pleiotropy assessment. MetS Metabolic syndrome, WC Waist circumference, HTN Hypertension, FG Fasting glucose, TG Triglycerides, HDL-C High-density lipoprotein cholesterol, LC Lung cancer, LUAD Lung adenocarcinoma, LUSC Squamous cell lung carcinoma, SCLC Small cell lung carcinoma.ExposureOutcomeCochran Q testMR-EggerMR-PRESSOQQ P-valueInterceptPleiotropy test p-valueOutliersDistortion test p-valueOutlier-corrected p-valueMetSLC592.366.04E−130.0020.3730.821.21E−03WCLC547.868.03E−13-0.00070.8120.491.91E−03EHTNLC256.963.51E−030.0060.0710.920.81FGLC102.085.75E−04-0.0040.320––TGLC342.661.05E−040.0030.110––HDL-CLC101.390.0033442010.0030.470––MetSLUAD447.412.05E−030.0030.2910.880.19WCLUAD431.292.55E−04− 0.00030.9220.770.51HTNLUAD239.540.032558220.0040.3810.660.54FGLUAD75.471.17E−01− 0.0050.330––TGLUAD275.052.10E−010.0020.250––HDL-CLUAD57.790.69438740.0010.760––MetSLUSC517.742.81E−07− 0.0020.6820.951.93E−05WCLUSC479.331.69E−07− 0.0070.1320.99784.66E−04HTNLUSC284.777.63E−050.0040.430––FGLUSC78.073.34E−02− 0.0040.500––TGLUSC352.726.75E−050.0040.1620.600.21HDL-CLUSC63.530.5632387− 0.00040.930––MetSSCLC437.807.14E−030.0020.700––WCSCLC368.549.37E−02− 0.0020.710––HTNSCLC231.696.17E−020.0110.130––FGSCLC84.331.08E−02− 0.010.290––TGSCLC318.712.44E−030.00030.9310.910.67HDL-CSCLC75.650.17222460.0010.890–– Heterogeneity and pleiotropy assessment. MetS Metabolic syndrome, WC Waist circumference, HTN Hypertension, FG Fasting glucose, TG Triglycerides, HDL-C High-density lipoprotein cholesterol, LC Lung cancer, LUAD Lung adenocarcinoma, LUSC Squamous cell lung carcinoma, SCLC Small cell lung carcinoma. We found a significant association between the genetic susceptibility of MetS and LC (OR, 1.22; 95% CI, 1.09–1.37, p = 7.57 × 10− 4), LUSC (OR, 1.47; 95% CI, 1.23–1.75, p = 2.22 × 10− 5), and SCLC (OR, 1.76; 95% CI, 1.37–2.26, p = 8.20 × 10− 6). Nevertheless, there was no association observed with LUAD (OR, 1.08; 95% CI, 0.94–1.24, p = 0.28). Genetically ascertained WC exhibited a substantial correlation with LC (OR, 1.25; 95% CI, 1.11–1.41, p = 3.27 × 10− 4), LUSC (OR, 1.38; 95% CI, 1.15–1.65, p = 6.46 × 10− 4), and SCLC (OR, 2.02, 95% CI, 1.57–2.59, p = 3.06 × 10− 8), while no significant association was observed with LUAD (OR, 1.06; 95% CI, 0.92–1.23, p = 0.40). FG, TG, HDL-C, and HTN showed no significant association with LC or its subtypes. Sensitivity analyses To assess the reliability of the above results, a battery of sensitivity analyses was conducted to identify any latent heterogeneity and horizontal pleiotropy, including Cochran’s Q test, MR-Egger intercept evaluation, and MR-PRESSO global assessment (as detailed in Table 2). Although heterogeneity was observed in most results of Cochran’s Q test (p < 0.001), it did not adversely affect the MR estimates in this study. This robustness can be ascribed to the application of the random-effects IVW method, which effectively harmonizes the aggregated heterogeneity. Furthermore, all MR-Egger intercept test p values exceeded 0.05, signifying an absence of horizontal pleiotropy. This further proves that even in the presence of heterogeneity, MR estimation is not affected by pleiotropic bias. MR-PRESSO analysis detected 1 to 3 outliers. However, after removing these outliers, the associations remained consistent, and there was no statistically significant difference in estimates before and after outlier exclusion (p value from distortion test > 0.05). Moreover, the significance of the results endured following the elimination of outliers (refer to Table 2). In addition, leave-one-out (LOO) analysis demonstrated that the estimates were not affected by individual SNPs, and the funnel plot exhibited a symmetrical configuration (Supplementary Material 2: Fig. 1), indicating that the estimates were unviolated (Supplementary Material 2: Figs. 2–7). Risk factor analysis To further dissect the connection between MetS and the increased risk of LC, we applied MR methods to assess how MetS and WC affect several prevalent LC risk factors. As shown in Table 3, the OR of MetS was 9.52 (95% confidence interval (CI), 4.94, 18.35; p = 1.67 × 10− 11) for a one standard deviation (SD) increase in smoking, 1.38 (95% CI, 1.32, 1.45; p = 4.86 × 10− 44) for a one SD increase in alcohol, 2.23 (95% CI, 2.10, 2.37; p = 9.06 × 10− 122) for one SD increase in BMI, 0.81 (95% CI, 0.77, 0.85; p = 4.15 × 10− 16) for one SD increase in education, and 1.64 (95% CI, 1.38, 1.96; p = 4.36 × 10− 8) for one SD increase in T2D. The OR of WC was 8.47 (95% CI, 3.11, 23.03; p = 2.86 × 10− 5) for one SD increase in smoking, 1.33 (95% CI, 1.26, 1.41; p = 2.58 × 10− 22) for one SD in alcohol, 2.77 (95% CI, 2.64, 2.91; p < 9.06 × 10− 122) for one SD in BMI, 0.80 (95% CI, 0.76, 0.84; p = 6.03 × 10− 18) for one SD in education, and 1.40 (95% CI, 1.14, 1.71; p = 1.08 × 10− 3) for one SD increase in T2D. In summary, based on risk factor analysis, our study suggests that smoking, alcohol consumption, BMI, education level, and T2D may contribute to susceptibility to MetS-related and WC-related LC. Table 3Risk factors analysis. IVW inverse variance weighted, OR odds ratio, CI confidence intervals, MetS metabolic syndrome, BMI body mass index, T2D type-2 diabetes, WC waist circumference. *P < 0.05.ExposureOutcomeIVWHeterogeneityMR-Egger methodOR (95% CI) P Q value P Intercept P MetSSmoking9.52(4.94, 18.35)1.67E−11*392.410.460.0220.13MetSAlcohol1.38(1.32, 1.45)4.86E−44*1606.42< 0.0010.00050.36MetSEducation0.81(0.77, 0.85)4.15E−16*762.15< 0.001− 0.0020.12MetSBMI2.23(2.10, 2.37)9.06E−122*8375.61< 0.001− 0.0020.06MetST2D1.64(1.38, 1.96)4.36E−08*366.68< 0.0010.00090.81WCSmoking8.47(3.11, 23.03)2.86E−05*178.000.23− 0.0070.77WCAlcohol1.33(1.26, 1.41)2.58E−22*2284.15< 0.001− 0.00090.50WCEducation0.80(0.76, 0.84)6.03E− 18*686.02< 0.001− 0.0020.16WCBMI2.77(2.64, 2.91)< 9.06E− 122*809.26< 0.001− 0.0020.06WCT2D1.40(1.14, 1.71)0.00107881*190.10< 0.001− 0.0030.51 Table 3Risk factors analysis. IVW inverse variance weighted, OR odds ratio, CI confidence intervals, MetS metabolic syndrome, BMI body mass index, T2D type-2 diabetes, WC waist circumference. *P < 0.05.ExposureOutcomeIVWHeterogeneityMR-Egger methodOR (95% CI) P Q value P Intercept P MetSSmoking9.52(4.94, 18.35)1.67E−11*392.410.460.0220.13MetSAlcohol1.38(1.32, 1.45)4.86E−44*1606.42< 0.0010.00050.36MetSEducation0.81(0.77, 0.85)4.15E−16*762.15< 0.001− 0.0020.12MetSBMI2.23(2.10, 2.37)9.06E−122*8375.61< 0.001− 0.0020.06MetST2D1.64(1.38, 1.96)4.36E−08*366.68< 0.0010.00090.81WCSmoking8.47(3.11, 23.03)2.86E−05*178.000.23− 0.0070.77WCAlcohol1.33(1.26, 1.41)2.58E−22*2284.15< 0.001− 0.00090.50WCEducation0.80(0.76, 0.84)6.03E− 18*686.02< 0.001− 0.0020.16WCBMI2.77(2.64, 2.91)< 9.06E− 122*809.26< 0.001− 0.0020.06WCT2D1.40(1.14, 1.71)0.00107881*190.10< 0.001− 0.0030.51 Risk factors analysis. IVW inverse variance weighted, OR odds ratio, CI confidence intervals, MetS metabolic syndrome, BMI body mass index, T2D type-2 diabetes, WC waist circumference. *P < 0.05. Discussion This study employs MR analysis to investigate the causal relationship between MetS and its components with LC and its subtypes from a genetic perspective. The results reveal a significant association between MetS, WC, and LC (including LUSC and SCLC). Risk factors such as smoking, alcohol consumption, BMI, education level, and T2D may mediate this association. Sensitivity and pleiotropy analyses further support the robustness of the findings. MetS is typically characterized by features such as increased adiposity, elevated blood pressure, high blood sugar levels, and abnormal lipid levels28. It is widely recognized as the main factor in the development of cardiovascular diseases29. However, the intricate influence of MetS on the trajectory of oncogenic diseases is an aspect that cannot be underestimated30. This encompasses pathways involving chronic low-grade inflammation31, insulin resistance32, cellular proliferation33, and immune dysregulation34. Thus, we postulate that MetS may augment the risk of LC through these mechanistic pathways. An investigation into the association between MetS and cancer susceptibility has revealed a moderate elevation in the risk of common neoplasms, including colorectal and bladder cancers10. Additionally, a population-based prospective cohort inquiry has unveiled a robust correlation between MetS and an overall increased susceptibility to LC, spanning adenocarcinoma and squamous cell carcinoma. Lower levels of high-density lipoprotein cholesterol (HDL-C), augmented WC, and elevated blood glucose levels positively correlate with LC risk35. This underscores the pivotal role of metabolic status and biomarkers in primary prevention and the screening of LC in high-risk cohorts. Despite these observations being derived from observational studies, they still lack direct evidence of causality between MetS and LC. Contrary to extensive and lengthy prospective clinical trials that require long-term observation, Mendelian randomization studies efficiently unearth the causal linkage between MetS/WC and LC, offering a cost-effective and expeditious investigative approach. It’s worth noting that this study did not find a significant association between MetS and LUAD. This could be attributed to the heterogeneity in the pathogenesis of different subtypes of lung cancer, and the interaction between components of MetS and LUAD may be influenced by various complex factors such as genetic factors, environmental factors, and lifestyle. These factors might obscure or weaken the potential association, resulting in the study findings showing no correlation. Additionally, our comprehensive analysis unveiled multiple potential explanations for the causal association between MetS (especially WC) and LC. Notably, it is estimated that a significant proportion of the most prevalent malignancies are causally linked to lifestyle factors that individuals can actively modify36. These factors encompass behavioral elements such as smoking, alcohol consumption, and various anthropometric parameters, including BMI36,37. Consequently, we conducted an investigation aimed at identifying lifestyle-related phenotypes. These variables included smoking, alcohol consumption, educational attainment, BMI, and T2D, which may serve as intermediary factors in the complex relationship between MetS and WC concerning the incidence of LC. It is essential to underscore that prior research has solidified the understanding that behavioral aspects such as smoking, educational attainment, and BMI are established etiological factors for LC38,39. In this context, we suggest that these elements may act mutually as mediators with MetS in lung cancer susceptibility. Furthermore, individuals afflicted by obesity and T2D often exhibit a protracted state of chronic inflammation and hormonal dysregulation31,40. These intricate biological modifications may substantially contribute to the advancement of the neoplastic process. It is paramount to acknowledge that further in-depth investigations are warranted to ascertain the precise causal relationship and the extent of mediation attributable to unhealthy lifestyles concerning LC41. In the absence of dedicated mediation analyses, the direct impact of MetS on LC remains indeterminate. Furthermore, this investigation furnishes fresh leads for preventive and interventional measures. Alleviating the perilous factors inherent in MetS, including smoking cessation, abstinence from alcohol, and diligent weight regulation, may play a pivotal role in diminishing susceptibility to lung neoplasms. A disciplined lifestyle and periodic medical surveillance continue to be central tenets in attenuating these susceptibilities. Timely detection and efficacious management of obesity and T2D can significantly contribute to abating the oncogenic proclivity linked to diabetes42. Furthermore, pharmacological modalities targeting MetS might have certain effects on LC prophylaxis; nevertheless, rigorous scrutiny is needed to corroborate this conjecture. The findings of this investigation bear substantial clinical and public health significance. Primarily, it accentuates the perilous nature of MetS not only limited to cardiovascular disease and diabetes but also LC. Given the significant global incidence and mortality rates associated with LC, this study illuminates the pivotal role MetS plays in elevating the risk of LC. This study provides novel insights into potential prophylactic strategies for reducing LC risk and lays the groundwork for future comprehensive investigations. Healthcare providers should consider intensified surveillance of patients with metabolic syndrome to discern potential LC risks early. However, it is essential to acknowledge that this study has certain limitations. Firstly, due to the study’s exclusive focus on participants of European ancestry, the findings may not be universally applicable to other ethnic groups. Secondly, the study did not comprehensively account for other potential risk factors, such as physical activity and dietary patterns, which might have distorted the results. Thirdly, the absence of mediation analyses precluded the determination of the precise proportion of these risk factors involved in the mediation process. Subsequent research endeavors should scrutinize the constituents of MetS in finer detail to explore their specific connections with LC. Additionally, another significant consideration is the potential presence of overlapping subjects between GWAS datasets. Despite our use of non-overlapping GWAS exposure and outcome data sourced from different origins and a thorough examination that confirmed no overlapping subjects, it is essential to acknowledge that the existence of overlapping subjects, if any, could potentially introduce bias into the estimation of the causal relationship between MetS and LC. Lastly, our study necessitates further probing into the underlying biological mechanisms, elucidating how MetS amplifies the risk of LC. Molecular and cellular biology experiments can facilitate the revelation of these intricate mechanisms. Conclusions Utilizing Mendelian Randomization techniques, we have confirmed and augmented the compelling and incontrovertible evidence for a causal relationship between MetS and LC. Our findings indicate a significant genetic susceptibility linking MetS and WC to LC, including both LUSC and SCLC. This motivates us to strengthen management and early intervention for patients with Metabolic Syndrome (MetS) in clinical practice, aiming to reduce the risk of lung cancer, especially among those with abdominal obesity. Furthermore, future research should delve deeper into the biological underpinnings of this association, elucidating the hidden mechanisms therein. Electronic supplementary material Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2