Abstract
Background: A suspicious thyroid nodule is defined as a thyroid lesion with potential risk
of malignancy based on clinical history, physical examination, ultrasound, and cytology, but without
definitive confirmation of cancer. This uncertainty often leads to unnecessary thyroid surgeries for
diagnostic purposes.
Objective: This study aimed to evaluate the diagnostic performance of BUU-Customized
Whole Exome Sequencing (BUU WES), using next generation sequencing (NGS) from preoperative
ultrasound-guided core needle aspiration biopsy specimens, compared with postsurgical
histopathology as the gold standard. Eligible patients included those with suspicious thyroid
nodules, classified as TI-RADS 4–5 or Bethesda I, III, IV, or V.
Results: Among 82 patients with 110 suspicious thyroid nodules, histopathology confirmed
malignancy in 21 cases (25.6%). The most frequent mutation identified was BRAF, followed by NRAS
and KRAS. Overall, BUU WES demonstrated sensitivity 57.1% (34.0–78.3%), specificity 86.9% (75.8–
94.2%), PPV 60.0% (36.1–80.9%), NPV 85.5% (74.2–93.1%), and accuracy 79.3% (68.9–87.4%).
Subgroup analysis of nodules >1 cm revealed improved diagnostic accuracy, with sensitivity 83.3%
(51.6–97.9%), specificity 89.8% (79.2–96.2%), PPV 62.5% (35.4–84.8%), NPV 96.4% (87.5–99.6%), and
accuracy 88.7% (79.0–95.0%).
Conclusion: Although BUU WES showed certain limitations in sensitivity and NPV compared
with established commercial panels, its diagnostic reliability significantly improved in nodules >1
cm. Moreover, the mutation profiles obtained from BUU WES provide essential data that may
facilitate the development of a customized genetic testing panel tailored to the context of thyroid
cancer in Thailand.
