The Diagnostic Value of Various Immunohistochemical Biomarkers in the Detection of Papillary and Follicular Thyroid Cancers: A Systematic Review and Meta-Analysis Publisher



Heidarpour M¹ ; Heshmaty Y¹ ; Rahimi A² ; Feizi A³ ; Rakhshan R¹ ; Ghasemi M⁴
Source: Journal of Health Sciences and Surveillance System Published:2024

Abstract

Background: Thyroid neoplasia is the most common endocrine malignancy worldwide. Fine-needle aspiration biopsy of thyroid nodules has a low sensitivity in distinguishing between benign and malignant lesions. Evaluation of the rate of expression and diagnostic value of immunohistochemical biomarkers in differentiating between benign and malignant thyroid lesions and different types of malignant lesions is the main purpose of this study. Methods: Sixty articles were reviewed in this systematic review and meta-analysis study. The rate of detection of various immunohistochemistry (IHC) biomarkers in several thyroid lesions was examined by meta-analysis. Specificity, sensitivity, positive and negative likelihood ratios, and confidence intervals (95% CI) were calculated for each marker. The accuracy of each test was evaluated by calculating the diagnostic odds ratio (DOR). ROC (receiver operating characteristic) analysis was performed for three markers. Results: Sensitivity and specificity of CK-19, Gal-3, and carcinoembryonic antigen (CEA) for detection of thyroid malignancies were 81% and 73%, 82% and 81%, and 77% and 83 %, respectively. The combination of these three markers showed the sensitivity of 85%, specificity of 97%, and diagnostic odds ratio of 95.1. Additionally, uPAR, Sialyl Lewis X, MIB-1, and Hector Battifora mesothelial-1. (HBME-1) can effectively differentiate the follicular variant of papillary thyroid carcinoma (FVPTC) from follicular thyroid carcinoma (FTC) as they are significantly more common in FVPTCs (P<0.05). Conclusion: We showed that CK-19, Gal-3, and CEA had an important and statistically significant role in differentiating between benign and malignant thyroid lesions. In addition, according to our results, urokinase-type plasminogen activator receptor (uPAR), Sialyl Lewis X, MIB-1, and HBME-1 can effectively differentiate FVPTC from FTC with acceptable sensitivity and specificity. © 2024 Shiraz University of Medical Sciences. All rights reserved.