Comparative analysis of diagnostic performance, feasibility and cost of different test-methods for thyroid nodules with indeterminate cytology.

Salvatore Sciacchitano,Carlo Bellotti,Armando Bartolazzi,Maria Trovato,Alessandra Ulivieri,Leila B Salehi,Fiorenza Magi,Gian Paolo De Francesco,Carlo Drago,Luca Lavra

doi:10.18632/oncotarget.17220

Abstract

Since it is impossible to recognize malignancy at fine needle aspiration (FNA) cytology in indeterminate thyroid nodules, surgery is recommended for all of them. However, cancer rate at final histology is <30%. Many different test-methods have been proposed to increase diagnostic accuracy in such lesions, including Galectin-3-ICC (GAL-3-ICC), BRAF mutation analysis (BRAF), Gene Expression Classifier (GEC) alone and GEC+BRAF, mutation/fusion (M/F) panel, alone, M/F panel+miRNA GEC, and M/F panel by next generation sequencing (NGS), FDG-PET/CT, MIBI-Scan and TSHR mRNA blood assay.We performed systematic reviews and meta-analyses to compare their features, feasibility, diagnostic performance and cost. GEC, GEC+BRAF, M/F panel+miRNA GEC and M/F panel by NGS were the best in ruling-out malignancy (sensitivity = 90%, 89%, 89% and 90% respectively). BRAF and M/F panel alone and by NGS were the best in ruling-in malignancy (specificity = 100%, 93% and 93%). The M/F by NGS showed the highest accuracy (92%) and BRAF the highest diagnostic odds ratio (DOR) (247). GAL-3-ICC performed well as rule-out (sensitivity = 83%) and rule-in test (specificity = 85%), with good accuracy (84%) and high DOR (27) and is one of the cheapest (113 USD) and easiest one to be performed in different clinical settings.In conclusion, the more accurate molecular-based test-methods are still expensive and restricted to few, highly specialized and centralized laboratories. GAL-3-ICC, although limited by some false negatives, represents the most suitable screening test-method to be applied on a large-scale basis in the diagnostic algorithm of indeterminate thyroid lesions.

Highlights

Follicular thyroid nodules with indeterminate pattern at fine needle aspiration (FNA) cytology are called in different ways, according to the different classification systems adopted
Our comparative analysis is based on data obtained from different populations in different settings and in different Countries and, it should not be considered as a formal comparison among cost-effectiveness analyses
In the same algorithm the use of M/F panel has been restricted to confirm malignancy in nodules suspicious for malignancy, commonly referred to surgery, for a more appropriate planning of the extent of www.impactjournals.com/oncotarget surgery

Summary

Introduction

Follicular thyroid nodules with indeterminate pattern at fine needle aspiration (FNA) cytology are called in different ways, according to the different classification systems adopted. They are classified as thy3a in the presence of atypical features and as thy3f when a follicular neoplasm is suspected, according to the British Thyroid Association (BTA) [1]. Of the classification system used, thyroid nodules classified in these categories represent the gray zone of conventional FNA-cytology [4, 5] They are diagnosed in 15% - 30% of the total FNA cases and are currently referred for surgery more for diagnosis rather than for a real therapeutic necessity. Cancer prevalence in such indeterminate nodules varies according to the larger studies, performed in different Countries (Table 1)

Methods

Results

Conclusion