Abstract
PURPOSE: Papillary thyroid cancer (PTC) is the most common thyroid malignancy and is often difficult to distinguish from benign nodules using current diagnostic tools, resulting in unnecessary surgical interventions. This study aimed to identify robust, reliable, and non-invasive circulating biomarkers capable of accurately discriminating benign from malignant PTC nodules by elucidating the relationship between thyroid tissue and circulating microRNA/cytokine profiles. METHODS: Paired arterial plasma supplying the thyroid gland and venous plasma draining the gland were collected from patients with benign thyroid nodules and malignant PTC. In addition, preoperative and postoperative peripheral plasma samples were obtained. MicroRNA profiles were compared between arterial inflow and venous outflow to assess thyroid-derived contributions to circulating biomarkers. Differentially expressed microRNAs were validated in preoperative and postoperative peripheral plasma samples and formalin-fixed, paraffin-embedded (FFPE) thyroid tissue sections. Subsequently, cytokine profiles associated with target microRNAs were investigated using cytokine arrays, and selected target proteins were further validated by ELISA. RESULTS: Arterial–venous comparisons revealed distinct microRNA and cytokine signatures associated with malignant PTC compared with benign nodules, reflecting tumor-specific release into the circulation. These findings were consistently validated in FFPE tissues and peripheral plasma samples. Importantly, the combined panel of hsa-miR-21-3p, sTNF-RI, IL-12p40, and CCL25 demonstrated high diagnostic performance in distinguishing malignant PTC from benign thyroid lesions. DISCUSSION: This integrated tissue-to-circulation approach provides direct evidence that the thyroid microenvironment significantly contributes to the circulating biomarker landscape. The identified biomarker panel represents a promising non-invasive diagnostic tool with the potential to improve PTC diagnosis, reduce unnecessary surgical procedures, and support clinical decision-making. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s12020-026-04612-9.