Integrating Tumor Genomics and Urinary Exosomal Proteomics to Establish a Multi-Layer Biomarker Framework for Early Risk Stratification and Post-Treatment Surveillance in Fresh Thyroid Cancer Patients

Overview

Thyroid cancer is the most common endocrine malignancy, and although differentiated thyroid cancer (DTC) generally confers favorable outcomes, 10-20% of patients still face substantial postoperative risks, including local recurrence, distant metastasis, and inadequate response to radioactive iodine therapy. Current risk stratification, largely based on tumor size, lymph node involvement, and histopathology, fails to adequately represent tumor heterogeneity and evolutionary changes, potentially resulting in both overtreatment and undertreatment. Next-generation sequencing (NGS) has revealed a stepwise accumulation of genomic alterations from early driver mutations (e.g., BRAF, RAS, RET/PTC, PAX8-PPARG) to late-stage progression events (e.g., TERT promoter, TP53, PI3K/AKT/mTOR), while metastatic lesions often harbor high-risk mutations absent in primary tumors, underscoring the limitations of single-time-point tissue sampling. Furthermore, serum thyroglobulin (Tg) surveillance is hindered in patients with anti-Tg antibodies. Extracellular vesicles (EVs), particularly those obtained from urine, provide a compelling liquid biopsy modality due to their non-invasiveness, repeatability, and reduced interference by abundant serum proteins. The investigators' previous findings demonstrate that urinary exosomal peptides-including U-Ex Tg, ANXA2, TIMP, and Angiopoietin-1-correlate with malignancy, capsular invasion, and nodal metastasis, and exhibit dynamic postoperative variation, suggesting their utility in detecting molecular residual disease. This prospective study will recruit 100 fresh thyroid cancer cases and integrate tumor genomic profiling, urinary exosomal proteomics via LC-MRM/MS, and clinical phenotype assessment-including nodal involvement, subsequent therapies, and long-term outcomes-to delineate causal links between genomic drivers, proteomic execution signals, and clinical progression. The overarching aim is to establish an early risk-stratification and molecular recurrence-alerting model capable of identifying high-risk trajectories earlier than conventional approaches, thereby enhancing surveillance precision and enabling timely intervention. This multi-layered biomarker framework holds strong potential to redefine postoperative monitoring standards and advance the clinical and policy implementation of precision medicine in thyroid cancer.

This prospective clinical study is designed to enroll patients, aged between 18 to 80 years, with newly diagnosed papillary thyroid carcinoma (PTC) or highly suspicious follicular-pattern neoplasms, including lesions with potential for poorly differentiated thyroid carcinoma. Eligible patients must have undergone fine-needle aspiration cytology (FNAC) as part of the diagnostic work-up. (This is a routine examination that the subject would undergo regardless of study participation.) Cytological reports and subsequent comprehensive surgical pathology assessments will be systematically documented and reviewed to ensure accurate classification of each case. Across the first two years of study implementation, we aim to recruit 100 consecutive patients presenting with untreated, or "fresh," thyroid cancer. For individuals undergoing surgery, urinary exosomal peptide profiling will be conducted at two standardized time points: (1) preoperatively, and (2) four weeks after thyroidectomy, allowing assessment of perioperative biomarker dynamics. Surgical treatment-anticipated to be total thyroidectomy in most cases-will be performed according to standard clinical judgment by experienced endocrine surgeons. Postoperative pathological evaluation will include meticulous characterization of tumor morphology, invasion patterns, and molecular features, based on finalized histopathological reports. In parallel, this study incorporates a robust tumor genomic investigation supported by our previously developed and validated next-generation sequencing (NGS) platforms. Under prior funding from the National Science and Technology Council (NSTC), the investigators established a thyroid cancer-specific DNA-based gene panel targeting 50 key oncogenic drivers and a complementary RNA-based fusion panel covering 19 clinically relevant gene rearrangements. These panels were curated through comprehensive interrogation of internationally recognized cancer genomic repositories, based on previous successfully completed somatic variant profiling in 51 patients with metastatic thyroid cancer, including 29 paired primary-metastatic tumor analyses, demonstrating strong capability for high-resolution comparative cancer genomics. Furthermore, the investigators have established technical proficiency in polyguanine marker analysis for phylogenetic inference of tumor evolution, and have previously performed whole-genome sequencing (WGS) and HLA genotyping in large-scale association studies. Notably, the investigators' WGS-based investigations identified genetic susceptibility to antithyroid drug-induced agranulocytosis through a rigorous, two-stage analytic strategy. Overall, this integrated study design enables simultaneous evaluation of genomic initiation and progression alongside urinary exosomal peptide expression, positioning this work to elucidate mechanistic relationships between tumor molecular biology and non-invasive biomarkers for disease surveillance and precision-guided management.