Abstract
Context
Molecular testing to refine the diagnosis of cytologically indeterminate thyroid nodules has become increasingly popular, but data on long-term durability of test results and the rate of delayed operation are limited.
Objective
Determine the delayed rate of surgical resection in indeterminate nodules with benign/negative molecular testing and the risk of false-negative molecular test results.
Design
Prospective follow-up of the Gene Expression Classifier vs Targeted Next-Generation Sequencing in the Management of Indeterminate Thyroid Nodules randomized controlled trial comparing the diagnostic test performance of Afirma Gene Expression Classifier and ThyroSeq v2.
Setting
University of California, Los Angeles.
Participants
Patients who underwent thyroid biopsy with indeterminate (Bethesda III/IV) cytology (April 2016 to July 2017).
Intervention
Ultrasound surveillance.
Main Outcome Measure
False-negative rate of molecular testing.
Results
Of 95 indeterminate nodules with negative/benign molecular test results, 12 nodules underwent immediate resection (11 benign nodules, 1 noninvasive follicular thyroid neoplasm nodule with papillary-like nuclear features). Nonoperative management was pursued for 83 (87.4%) nodules. The median surveillance was 26.7 months. Ten nodules were resected during surveillance and malignancy was identified in 4 nodules (overall false-negative rate of 5.8%). In the 4 malignant nodules that underwent delayed operation, surgery was prompted by sonographic changes during surveillance.
Conclusions
The majority of indeterminate nodules with negative molecular testing have a stable clinical course over 3 years of follow-up, but our finding of a 6% false-negative rate highlights the importance of continuing sonographic surveillance. Long-term studies are needed to determine the optimal length of follow-up.