iMetaOmics | Inflammatory proteins for diagnosing the papillary thyroid carcinoma and nodular goiter

RESEARCH ARTICLE
Open Access
Combined detection of inflammatory proteins is beneficial for diagnosing the papillary thyroid carcinoma and nodular goiter
Yongqin Pan, Mingxi Xu, Tsz Hong Chong, Siping Xie, Kunsong Huang, Guanghao Wang, Yuhua Ma, Jinyi Li, Wah Yang
First published: 02 July 2024 doi.org/10.100...
Yongqin Pan and Mingxi Xu contributed equally to this study.
Abstract
Fine-needle aspiration cytology and imaging examinations are commonly used diagnostic tools for papillary thyroid carcinoma (PTC). However, these methods have limitations. Inflammatory proteins have the potential to serve as diagnostic and prognostic markers, as well as treatment targets. The expression profile and diagnosis effect of inflammatory proteins in PTC are not well understood. Here, 18 healthy volunteers (as healthy control), 12 patients with nodular goiter, and 34 patients with PTC were collected to analyze serum inflammatory proteins by proximity extension assay. Receiver operating characteristic curve analysis was used to evaluate the diagnostic potential of differential expression of proteins via the area under the curve (AUC) analysis. A total of 36 differentially expressed inflammatory proteins were found among PTC, nodular goiter, and healthy control. The combination diagnosis derived from the logistic regression analysis exhibited promising diagnostic capabilities in distinguishing nodular goiter from healthy control (AUC = 0.88), distinguishing PTC from healthy control (AUC = 0.89), and distinguishing PTC from nodular goiter (AUC = 0.87). Whereas the combination diagnosis derived from the least absolute shrinkage and selection operator (LASSO) exhibited promising diagnostic capabilities in distinguishing nodular goiter from healthy control (AUC = 0.92), distinguishing PTC from healthy control (AUC = 0.93), and distinguishing PTC from nodular goiter (AUC = 0.93). Overall, this study offers potential biomarkers for distinguishing between PTC and nodular goiter in clinical practice. The combination derived from the LASSO algorithm outperforms logistic regression.
Highlights
A total 36 differentially expressed inflammatory proteins were found among PTC, nodular goiter, and healthy control.
TGF-α + CXCL11 exhibited promising diagnostic capabilities in distinguishing nodular goiter from healthy control.
TGF-α + CXCL11 exhibited promising diagnostic capabilities in distinguishing PTC from healthy control.
GDNF + CXCL5 + ARTN demonstrated optimal diagnostic effectiveness in distinguishing PTC from nodular goiter.
The combination derived from the LASSO algorithm outperforms logistic regression, albeit with a higher number of proteins.