Multi-center multi-omics integration predicts individualized prognosis in medullary thyroid carcinoma

ABSTRACT: Medullary thyroid carcinoma (MTC) is a rare and aggressive neuroendocrine tumor. Our study involves 482 retrospective MTC formalin-fixed, paraffin-embedded (FFPE) samples from 452 patients, collected from 10 Chinese clinical centers. Quantification of 10,092 proteins were achieved via diaPASEF, and 87.8% patients were found to harbor at least one mutation. International MTC grading system, concurrent papillary thyroid carcinoma (PTC), and lymph node metastasis were identified as significant risk factors. Notably, RET mutations M918T and S891A were associated with high recurrence risk in sporadic and hereditary MTC, respectively. Pathway analyses highlighted enhanced collagen biosynthesis linked to poor prognosis. Ubiquitinomics showed downregulated E3 ligases CUL4B and TRIM32 linked to structural recurrence. Unsupervised clustering identified three molecular subtypes with distinct clinical outcomes and characteristics. To address the need for precise risk stratification, we developed a machine learning model using clinical, genomic, and proteomic data to predict individualized recurrence risk. Our integrated model, comprising 20 features (2 clinical factors and 18 proteins), achieved 84.8% accuracy and an AUC of 0.87 in the independent test dataset. As a comprehensive multi-center, multi-omics study of MTC, our work provides critical insights into MTC heterogeneity and aggressiveness while offering a robust framework for personalized patient management and follow-up strategies.

INSTRUMENT(S):

ORGANISM(S): Homo Sapiens (human)

TISSUE(S): Thyroid Nodule

DISEASE(S): Medullary Thyroid Carcinoma

SUBMITTER: Yan Zhou  

LAB HEAD: Tiannan Guo

PROVIDER: PXD063677 | Pride | 2025-11-13

REPOSITORIES: Pride