Utility of Afirma Gene Expression Classifier for Evaluation of Indeterminate Thyroid Nodules and Correlation With Ultrasound Risk Assessment: Single Institutional Experience

Document Type


Publication Date


Publication Title

Endocrine practice


Objective: We assessed our experience with Afirma gene expression classifier (GEC) combined with sonographic risk assessment, using both the ATA and TI-RADS in evaluating indeterminate thyroid nodules. Methods: We identified 98 patients with 101 nodules who had a second fine needle aspiration biopsy (FNA) between 01/01/2014 and 09/30/2017 and sent to Veracyte for cytopathology and subsequent Afirma GEC testing. A second FNA biopsy was performed if the initial cytopathology was either Bethesda III or IV (n= 94) or non-diagnostic (n= 7). We correlated cytopathology, histopathology, and Afirma GEC results with sonographic risk assessment using both the ATA system and TI-RADS. Results: The mean age of the cohort was 57.4 +/- 12.3 years; 84% women and 60% white. Repeat FNA was benign in 51 of 101 nodules, and of the remaining 50 nodules, 18 (36%) were GEC-benign and 32 (64%) GEC-suspicious. Eighteen of the 32 GEC-suspicious nodules underwent surgery with the following results: 7 benign (39%), 1 follicular thyroid carcinoma (6%), 6 follicular variant of papillary thyroid cancer (33%), and 4 non-invasive follicular tumor with papillary-like nuclear features (22%). The malignancy rate among the surgical cohort was 39% (without NIFTP) and 61% (with NIFTP) and about 50% and 20% of this group scored in the high suspicion category by ATA and TR5 by TI-RADS respectively. Conclusion: Afirma GEC was useful in avoiding surgery in one-third of indeterminate nodules and performed similarly to ATA and TI-RADS. However, use of echogenicity in scoring may underestimate the risk of malignancy in patients with indeterminate nodules.

PubMed ID



ePub ahead of print