Background Many factors, such as lifestyle, medication, and environmental exposures, are reported to cause thyroid hormone system disruption (THSD) in humans, however studies linking THSD to health effects are sparse. Adverse Outcome Pathways (AOPs) provide mechanistic links from molecular events to adverse outcomes, with effect biomarkers serving as a tool to empirically anchor key events and health effects and to assess biological relevance. Aim This review aims to identify and evaluate effect biomarkers for thyroid hormone system-related AOPs for further validation in experimental and epidemiological studies. Methods Using AOP-wiki, we extracted and analysed thyroid-related AOPs, focusing on the eleven AOPs with mammalian evidence. We did systematic literature search to identify potential effect biomarkers for future epidemiological studies. Results In an AOP network analysis of the eleven thyroid-related AOPs, we identified four AOP clusters, including hippocampal alterations, impaired learning and memory, thyroid follicular cell adenomas/carcinomas, and kidney toxicity. For the clusters on hippocampal alterations and impaired learning and memory, brain-derived neurotrophic factor emerged as a promising effect biomarker. For the cluster on thyroid follicular cell adenomas/carcinomas, no promising effect biomarkers with high specificity were identified, but interleukin-34, oxidative stress, and expression of several genes were found to be related to the adverse outcome. For kidney toxicity, a panel of effect biomarkers were identified, such as clusterin, cystatin-C, kidney injury molecule-1, N-acetyl-beta-d-glucosaminidase, neutrophil gelatinase-associated lipocalin, and osteopontin. Conclusion This review operationalizes the AOP framework to support the use of mechanistically anchored effect biomarkers in human studies on THSD. By aligning key biological events with measurable endpoints, human matrices, and feasibility considerations, it provides a scientifically grounded path from mechanistic understanding to population research application. This enables more targeted biomonitoring, strengthens interpretation of epidemiological findings, and informs research and regulatory priorities for future validation efforts.
Wielsøe et al. (Wed,) studied this question.