Cobalt Mediates Stage-Specific Toxicity of Metal Mixtures in Cardiovascular-Kidney-Metabolic Syndrome

Wei Zhang, Guoqin Jiang, Ziyan Liu, Lian Duan, Jiayi Liang, Ziyan Wang, Huiwen Kang, Danyang Huang, Ai GaoToxicological 2025 · DOI

TL;DR

This study reveals that exposure to a mixture of metals is associated with the progression of Cardiovascular-Kidney-Metabolic (CKM) syndrome. Using advanced machine learning and statistical models, the research identifies cobalt as a key driver of CKM, with distinct toxicological pathways influencing early versus late stages of the syndrome.

What Problem Does This Paper Address?

Cardiovascular-Kidney-Metabolic (CKM) syndrome is a growing global health concern characterized by the interconnected dysfunction of the cardiovascular system, kidneys, and metabolic processes. While environmental metal exposure is known to contribute to various health issues, the specific impact of mixtures of metals on the progression of CKM syndrome, and the underlying mechanisms, remain poorly understood. This paper aims to fill this knowledge gap by investigating the joint effects of multiple metal exposures on CKM staging and pinpointing the critical toxic drivers.

Methods

The researchers employed a multi-pronged analytical approach to dissect the complex relationship between metal mixtures and CKM syndrome:

  • Weighted Quantile Sum (WQS) Regression: This statistical method is used to assess the combined effect of multiple correlated exposures (in this case, various metals) on an outcome (CKM syndrome). It helps identify which metals contribute most significantly to the overall mixture effect.
  • Generalized Linear Models (GLMs) and Ridge Regression: These models were used to examine the associations between individual metal exposures and CKM syndrome, allowing for the identification of specific metals that might be independently linked to the condition. Ridge regression is particularly useful for handling multicollinearity among predictors.
  • Shapley Additive Explanations (SHAP) Analysis: A powerful machine learning technique, SHAP was employed to understand the contribution of each metal to the model's predictions. It provides a measure of feature importance, helping to validate which metals are the most influential drivers of CKM.
  • Polynomial Regression: This method was used to explore potential non-linear dose-response relationships between individual metal exposures and CKM syndrome, looking for thresholds or more complex patterns of toxicity.
  • Adverse Outcome Pathway (AOP) Framework: This conceptual framework was utilized to map out the mechanistic pathways through which cobalt, identified as a key toxic driver, might exert its effects on CKM syndrome at different stages. This involved identifying molecular initiating events, key events, and ultimately, adverse outcomes.

Datasets

The study utilized data from the National Health and Nutrition Examination Survey (NHANES), specifically the 2011-2016 cycles. This dataset provided information on a large population of US adults, including their exposure levels to various metals (measured in urine) and their CKM syndrome status, along with a comprehensive set of covariates. The final analytical sample consisted of 1,816 participants.

Key Insights and Ideas

  • Cobalt as a Dominant Driver: The consistent identification of cobalt as a key toxic driver across multiple analytical models (WQS, GLM, Ridge, SHAP) is a significant finding. This suggests that cobalt plays a disproportionately large role in the metal mixture's impact on CKM syndrome compared to other metals.
  • Stage-Specific Mechanisms: The application of the AOP framework revealed that cobalt's toxicity is not uniform across CKM progression. Different biological pathways are implicated in the early stages (metabolic and immune dysregulation) versus the late stages (disruptions in calcium homeostasis, lipid metabolism, and cell apoptosis). This highlights the dynamic nature of metal-induced toxicity.
  • Linear Dose-Response for Cobalt: The finding that cobalt exhibits a linear dose-response relationship with CKM syndrome is crucial. This implies that there may be no "safe" threshold for cobalt exposure in the context of CKM, underscoring the importance of minimizing exposure.
  • Mixture Effects Matter: The WQS regression analysis demonstrated a significant association between the mixture of metals and CKM syndrome, emphasizing that the combined effect of multiple metals is important and cannot be overlooked.

Key Results

  • The WQS regression model showed a significant positive association between the mixed metal exposure index and CKM syndrome (β = 0.502, p = 0.013).
  • Cobalt was consistently identified as the most significant contributor to CKM syndrome across WQS, GLM, and Ridge regression models.
  • SHAP analysis validated cobalt's dominant role in predicting CKM syndrome.
  • Polynomial regression indicated a significant linear dose-response relationship between cobalt exposure and CKM syndrome. Non-linear relationships were observed for cadmium, strontium, and thallium.
  • The AOP framework identified distinct mechanisms for early CKM (e.g., metabolic homeostasis, immune response) and late CKM (e.g., calcium homeostasis, lipid metabolism, cell apoptosis-survival balance) driven by cobalt.

Limitations and Future Work

  • Cross-Sectional Design: The study is based on cross-sectional data, which limits the ability to establish definitive causality. While the AOP framework provides mechanistic insights, longitudinal studies are needed to confirm the causal links between metal exposure and CKM progression.
  • Urinary Metal Measurement: Urinary metal concentrations reflect recent exposure and can be influenced by various factors, including hydration status. While creatinine adjustment was performed, it's a proxy for dilution.
  • Unmeasured Confounders: Despite adjusting for numerous covariates, there's always a possibility of unmeasured confounders influencing the observed associations.
  • Future Work: The authors suggest that future research should focus on longitudinal studies to confirm causality and further investigate the specific molecular mechanisms identified by the AOP framework. Additionally, exploring the role of cobalt in new energy materials and its ubiquitous presence in the environment warrants further attention regarding CKM syndrome.

Summary updated

Abstract

Cardiovascular-Kidney-Metabolic (CKM) syndrome imposes a rising global health burden, yet the link between environmental metal mixtures and CKM progression remains unclear. To assess the joint effects of metal mixtures on CKM syndrome staging and identify critical toxic drivers through advanced mixture analysis. NHANES data (2011-2016) from 1,816 participants were analyzed via Weighted Quantile Sum (WQS) regression, generalized linear models (GLMs), ridge regression, Shapley Additive exPlanations (SHAP) analysis, and polynomial regression. An Adverse Outcome Pathway (AOP) framework was utilized to characterize the mechanisms of metal-mediated CKM. The WQS model revealed an association between mixed metal exposure and CKM (β = 0.502, p = 0.013). Subsequently, GLMs and ridge regression further identified the associative characteristics of individual metals, with all three models pointing to cobalt as the key driver. The SHAP model validated cobalt's dominant contribution from the perspective of marginal feature importance. Additionally, a polynomial equation analysis showed that cobalt exhibited a linear dose-response relationship with CKM syndrome. Based on these findings, the AOP framework furtherly identified that early CKM stages are linked with cobalt-related metabolic and immune dysregulation. In contrast, late stages involve disruptions in calcium homeostasis, lipid metabolism, and cell apoptosis-survival balance. Our findings highlight the impact of metal exposure on the progression of CKM syndrome, the AOP framework has deciphered stage-specific mechanisms of cobalt, revealing distinct toxicological pathways in early versus late CKM.
BibTeX
@inproceedings{zhang2025cobalt,
  author = {Wei Zhang and Guoqin Jiang and Ziyan Liu and Lian Duan and Jiayi Liang and Ziyan Wang and Huiwen Kang and Danyang Huang and Ai Gao},
  title = {{Cobalt Mediates Stage-Specific Toxicity of Metal Mixtures in Cardiovascular-Kidney-Metabolic Syndrome}},
  booktitle = {Toxicological},
  year = {2025},
  doi = {10.1093/toxsci/kfaf172},
}
Figure