Beyond Cytology: Integrating Molecular Testing into Diagnostic Algorithm of Indeterminate Thyroid Nodules

Abstract

Background: Thyroid nodules represent the most pervasive endocrine neoplasia, detectable in over 50% of the adult population via high-resolution ultrasonography. While the vast majority are benign, approximately 20% of fine-needle aspiration (FNA) biopsies yield indeterminate cytological results, carrying an unpredictable risk of malignancy (ROM) ranging from 10% to 40%. Historically, this diagnostic ambiguity necessitated repeat biopsies or diagnostic hemithyroidectomy, often exposing patients with benign disease to unnecessary surgical risks and lifelong supplementation. Main Body: Over the past decade, advancements in high-throughput sequencing have enabled the integration of molecular testing into the diagnostic algorithm of indeterminate thyroid nodules. Current molecular platforms primarily utilize DNA- and RNA-based next-generation sequencing (NGS) to detect oncogenic driver mutations (e.g., BRAF, RAS), gene fusions (RET/PTC, PAX8/PPARγ), copy number variations, and altered microRNA (miRNA) expression profiles. These assays function through two primary paradigms: "rule-out" tests characterized by high sensitivity and negative predictive value (NPV) to avoid unnecessary surgeries, and "rule-in" tests with high specificity and positive predictive value (PPV) to guide definitive oncological resections. Prominent commercial platforms, such as the Afirma Genomic Sequencing Classifier (GSC) and ThyroSeq v3, have revolutionized risk stratification by refining indeterminate categories (Bethesda III and IV) into distinct molecular phenotypes. Conclusion: Integrating molecular testing into clinical practice refines risk stratification, curtails diagnostic hemithyroidectomies, and personalizes surgical management. However, its implementation must be individualized, considering institutional cancer prevalence, cost-effectiveness, and ultrasound characteristics.