Large-scale proteomic characterization reveals clinically relevant subtypes of early-stage papillary thyroid cancer for prognostic prediction

Abstract

Papillary thyroid cancer (PTC) has a rising incidence. The majority are diagnosed at an early stage, albeit with a minority exhibiting aggressive tendencies. Given its generally indolent nature, controversy persists regarding the management of early-stage PTCs, particularly surgical intervention and long-term surveillance strategies. Here, we conduct a large-scale proteomic characterization of early-stage PTC based on tumor samples from 797 patients. Proteomic profiling unveils clinically relevant modules and identifies six subtypes, capturing the spectrum of heterogeneity from indolent to progressive disease states. Notably, subtypes associated with higher progression risk display enhanced signatures of ribogenesis and epithelial-mesenchymal transition. Furthermore, integrating machine learning with protein co-expression network analysis, we identify predictive markers of PTC progression risk such as GPX3 and TMSB4X. Integration of clinical and proteomic data provides a valuable resource for a deeper understanding of early-stage PTC biology and may inform personalized management strategies for patients with indolent or progressive disease. Papillary thyroid cancer (PTC) is often slow growing, but a subset of tumors can be aggressive. Here, the authors profile tumors from 797 PTC patients to define six molecular subtypes, identify protein markers linked to low- and high-risk disease, and develop a proteomic classification framework that may improve risk stratification and treatment decisions.