Integrated proteomic and metabolomic analysis of plasma reveals regulatory pathways and key elements in thyroid cancer.

Zijian Sun,Jingkui Tian,D. Feng,Wei Zhu,Jia-feng Wang,Liehao Jiang

DOI: https://doi.org/10.1039/d3mo00142c

2023-08-29

Molecular Omics

Abstract:Thyroid cancer (TC) is the most common endocrine malignancy with increasing incidence in recent years. Fine-needle aspiration biopsy (FNAB), as a gold standard for the initial evaluation of thyroid nodules, fails to cover all the cytopathologic conditions resulting in overdiagnosis. There is an urgent need for a better classification of thyroid cancer from benign thyroid nodules (BTNs). Here, data independent acquisition (DIA)-based proteomics and untargeted metabolomics in plasma samples of 10 patients with TC and 15 patients with BTNs were performed. Key proteins and metabolites were identified specific to TC, and an independent cohort was used to validate the potential biomarkers using enzyme-linked immunosorbent assay (ELISA). In total, 1429 proteins and 1172 metabolites were identified. Principal component analysis showed a strong overlap at the proteomic level and a significant discrimination at the metabolomic level between the two groups, indicating a more drastic disturbance in the metabolome of thyroid cancer. Integrated analysis of proteomics and metabolomics shows glycerophospholipid metabolism and arachidonic acid metabolism as key regulatory pathways. Furthermore, a multi-omics biomarker panel was developed consisting of LCAT, GPX3 and leukotriene B4. Based on the AUC value for the discovery set, the classification performance was 0.960. The AUC value of the external validation set was 0.930. Altogether, our results will contribute to the clinical application of potential biomarkers in the diagnosis of thyroid cancer.

Medicine,Biology