Validating a Molecular Thyroid Test for Avoiding Diagnostic Surgery

Evidence suggests that thyroid nodules are common in the general population, with incidence increasing with age. Ultrasound-guided fine needle aspiration (FNA) is a commonly used biopsy tool for diagnosing thyroid nodules. However, thyroid FNA cytology yields an indeterminate result for about 20% of nodules, preventing a definitive diagnosis and hampering patient management.

 

In recent years, molecular testing has been developed to fill this diagnostic gap for indeterminate thyroid nodules. Despite advancements, molecular tests have lacked high specificity and positive predictive value, adequate clinical validation, and refined cancer risk by specific molecular alteration. A newly developed thyroid test—ThyroSeq v3 Genomic Classifier (GC)—was recently assessed in a prospective, blinded, multicenter clinical validation, and the results were published in JAMA Oncology.

Among 257 thyroid samples with indeterminate cytology, informative ThyroSeq test results, and final surgical pathology diagnosis, ThyroSeq GC demonstrated high sensitivity at 94% and high specificity at 82%, for Bethesda cytology categories III and IV.

Negative predictive value (NPV) was high at 97%, with a cancer/ non-invasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTP) prevalence of 28%. Also, 61% of nodules yielded a negative ThyroSeq result, with 3% residual cancer risk in those nodules, a residual risk similar to that of benign cytology. False-negative nodules were all low-risk tumors. The observed positive predictive value (PPV) was 66%. ThyroSeq correctly classified 100% of Hurthle cell carcinomas and NIFTP as positive.

In addition to the high NPV and reasonably high PPV observed, ThyroSeq GC also provided specific genetic alteration information that informed cancer probability. In the test-positive nodules, the probability of malignancy or NIFTP varied from 59% to 100%, depending on specific genetic alteration. The specific alteration found and subsequent risk may inform the extent of thyroid surgery.

Overall, the study demonstrated that utilizing ThyroSeq testing for patients with indeterminate thyroid nodules may allow up to 82% of patients with histologically benign nodules to avoid diagnostic surgery. For ThyroSeq test-positive cases, the provided detailed genetic information may assist physicians, in conjunction with imaging and relevant clinical information, to offer more individualized treatment to these patients.