A Thyroid Genetic Classifier Correctly Predicts Benign Nodules with Indeterminate Cytology: Two Independent,...

Area of Science:

• Endocrinology
• Oncology
• Molecular Diagnostics

Background:

• Most thyroid nodules with indeterminate cytology are benign, yet surgery is the common diagnostic approach globally.
• Diagnostic surgery for indeterminate thyroid nodules carries risks and costs.
• A validated genetic classifier could improve diagnostic accuracy and patient management.

Purpose of the Study:

• To clinically validate a 10-gene thyroid genetic classifier for predicting benign thyroid nodules.
• To assess the classifier's performance in diverse, independent patient cohorts.
• To determine the potential of the classifier in reducing unnecessary diagnostic surgeries.

Main Methods:

• Prospective, multicenter trials (TGCT-1 and TGCT-2) involving 4061 fine-needle aspirations.
• Testing of a 10-gene thyroid genetic classifier on 270 samples with matched surgical pathology.
• Independent validation across ethnically diverse populations and clinical settings.

Main Results:

• The classifier demonstrated high accuracy (89%) in predicting benign thyroid nodules across both cohorts.
• Sensitivity ranged from 90-91%, specificity from 87-89%, and negative predictive values from 94-96%.
• The classifier performed consistently in ethnically diverse populations, showing no significant differences in histopathological profiles.

Conclusions:

• Clinical validation confirms the 10-gene thyroid genetic classifier's equivalent and accurate performance in diverse cohorts.
• The classifier accurately predicts benign thyroid nodules, supporting surveillance as an alternative to diagnostic surgery.
• This genetic assay offers a promising tool for reference laboratories to manage indeterminate thyroid cytology cases.