Subacute thyroiditis (SAT) is an inflammatory thyroid disease characterized by neck pain, tenderness, general symptoms, and thyroid dysfunction. Despite gaining new insights into the epidemiology, pathogenesis, and treatment of SAT in recent years, the exact pathogenesis and determinants of its clinical progression remain unclear. Here, we profiled thyroid in situ protein alterations in fine needle aspiration biopsy samples from SAT patients using proteomic analysis and uncovered 57 differentially abundant proteins. Gene Ontology and KEGG enrichment analyses identified that these proteins were enriched in processes involving infection, inflammatory response, and cell adhesion and junction, which likely contribute to the pathogenesis. Moreover, the top three high-abundance proteins (NNMT, FTL, and TYMP) were further validated in the plasma from a larger SAT cohort using an enzyme-linked immunosorbent assay. After adjusting for sex, Spearman correlation analysis showed that NNMT, FTL, and TYMP levels were positively correlated with FT3, FT4, T3, T4, Tg, and ESR and negatively correlated with TSH. Furthermore, binary logistic regression analyses revealed that NNMT, FTL, and TYMP were independent factors of SAT. We also conducted a receiver operating characteristic (ROC) curve analysis to assess the diagnostic accuracy of NNMT, FTL, and TYMP for SAT. The results revealed that each factor demonstrated an area under the curve (AUC) score above 0.8. Thus, these high-abundance proteins can potentially serve as biomarkers for SAT diagnosis and outcome prediction. Our findings provide valuable insights into SAT biomarkers, and shed light on the potential pathogenesis and therapeutic targets of SAT.