The Correlation Study on Ultrasonography, Fine Needle Aspiration Cytology, and Molecular Testing in Management of Thyroid Nodules

Summary: Introduction: As one of the initial tools in evaluation of thyroid nodules, ultrasonography (US) plays an important role in selecting patients with higher risk of malignancy for further investigation. The American College of Radiology Thyroid Imaging, Reporting and Data System (ACR TI-RADS) is most often used in the United States for risk stratification. Based on the features on ultrasound imaging, ACR TI-RAD is classified into TR1 through TR5, with the estimated risk of malignancy (ROM) from no more than 2% for TR1 and TR2 to at least 20% for TR5. However, significant discrepancy existed among the studies regarding the effectiveness of ACR TI-RAD system. The current study aimed in correlating ACR TI-RAD classification with corresponding FNA cytology, histology, and the results of molecular testing. Materials and Methods: Ultrasound (US)- guided thyroid FNA cases performed during 2022 to 2023 and the corresponding ultrasound reports that lead to the FNA were retrieved from the electric medical records of our institution. Cytologic diagnosis was based on Bethesda system in Reporting Thyroid Cytology (BSRTC) and the molecular testing (ThyroSeq) was performed at CBLPath (Rye Brook, NY). ACR TI-RAD score (TR1 to TR5) of the nodule and corresponding cytologic diagnosis, histological diagnosis, and the results of molecular testing were evaluated. Results: A total of 1343 cases that met the search criteria were identified. Correlation between ACR TI-RAD and cytologic diagnosis was shown in Table 1. The percentage of cytologically benign and abnormal (AUS and worse) cases in TR3, TR4, and TR5 categories was 74.2% and 15.8%, 70.7% and 16.5%, 66.6% and 20.6%, respectively. TR5 showed higher percentage of cases with cytologically malignant diagnosis. Correlation between ACR TI-RAD scores and the results of molecular testing was shown in Table 2. The percentage of Negative and Positive results of ThyroSeq testing in TR3, TR4, and TR5 was 64.5% and 35.5%, 62.5% and 37.5%, 44.4% and 55.6%, respectively. TR5 showed a higher percentage of molecular positive cases. Correlation between ACR TI-RAD scores and histological diagnosis was shown in Table 3. While the surgical rate was similar across ACR TI-RAD categories, the rate of malignant cases, including NIFTP in TR3, TR4, and TR5 was 20.3%, 30.7%, and 60.3%, respectively. Conclusions: While most of the thyroid nodules in each ACR-TI-RAD category were benign and ROM was lower than those reported in literature, ACR TI-RAD scores, to some extent, correlated with cytologic diagnosis. ACR TI-RAD scores also correlated with positive molecular testing and histological malignant diagnosis. Combination of ACR-TI-RAD scores and the results of molecular testing can improve risk stratification of thyroid nodules with indeterminate cytologic diagnosis.