SUN-370 - Metabolomic Profiling of Papillary Thyroid Carcinoma Reveals Pathways Associated with Aggressive Disease

Introduction: Papillary thyroid carcinoma (PTC) is characterized by clinical, genomic and transcriptional heterogeneity. However, the metabolic programs associated with PTC’s molecular diversity remain not well characterized, in part due to high cost and complexity of direct assessment of metabolite levels. Here, we applied a transcriptomics-based metabolite imputation framework to determine PTC’s distinct metabolic profiles.

Methods: The “UnitedMet” Bayesian model was applied on bulk RNA-seq data from 497 PTC tumors in the TCGA cohort, trained on paired metabolomic and transcriptomic profiles of a renal cell carcinoma cohort, to infer relative abundances of approximately 1,700 metabolites. Clinical and molecular features were used to classify samples, including high vs. low-risk of recurrence disease, MACIS stage, BRAF mutation status, histologic subtype, and BRAF-like vs. RAS-like transcriptional phenotypes. Differential analysis was performed using FDR (false discovery rate) cutoff < 0.05 to identify metabolites significantly associated with different disease states, whereas pathway enrichment analysis was performed using MetaboAnalyst and applying multiple stringency criteria (FDR < 0.05, Holm < 0.05, and Impact > 0.1)

Results: Classical histotype and BRAF-mutant PTC were associated with upregulated pathways of glutathione metabolism and aromatic amino acid biosynthesis, including phenylalanine, tyrosine, and tryptophan, suggesting augmented oxidative stress responses and enhanced precursor availability for proliferative activity. BRAF-like PTC was characterized by enrichment in metabolic pathways linked to one-carbon cycling and γ-glutamyl amino acid metabolism, potentially driven by MAPK pathway activation.

Conclusions: This is, to our knowledge, the first study on transcriptomics-inferred metabolomic profiling of PTC and has identified clinically relevant metabolic vulnerabilities. Future studies are needed using paired metabolomic and transcriptomic thyroid cancer training datasets, as well as orthogonal validation with direct metabolite assessment (e.g. with LC-MS) to determine the prognostic and clinical utility of these metabolomic signatures.

*Unless otherwise noted, all abstracts presented at ENDO must not be released to the press or the public until the date and time of presentation. For oral presentations, the abstracts are embargoed until the session begins. The Endocrine Society reserves the right to lift the embargo on specific abstracts that are selected for promotion prior to or during ENDO.*