Molecular markers for the classification of cytologically indeterminate thyroid nodules

Abstract

Background The diagnosis of indeterminate lesions of the thyroid is a challenge in cytopathology practice. Indeed, up to 30% of cases lack the morphological features needed to provide definitive classification. Molecular tests have been developed to assist in the diagnosis of these indeterminate cases. The first studies dealing with the preoperative molecular evaluation of FNA samples focused on the analysis of BRAF V600E or on the combined evaluation of two or three genetic alterations. The sensitivity of molecular testing was then improved through the introduction of gene panels, which became available for clinical use in the late 2000s. Two different categories of molecular tests have been developed, the 'rule-out' methods, which aim to reduce the avoidable treatment of benign nodules, and the 'rule-in' tests that have the purpose to optimize surgical management. The genetic evaluation of indeterminate thyroid nodules is predicted to improve patient care, particularly if molecular tests are used appropriately and with the awareness of their advantages and weaknesses. The main disadvantage of these tests is the cost, which makes them rarely used in Europe. To overcome this limitation, customized panels have been set up, which are able to detect the most frequent genetic alterations of thyroid cancer. Conclusions In the present review, the most recent available versions of commercial molecular tests and of custom, noncommercial panels are described. Their characteristics and accuracy in the differential diagnosis of indeterminate nodules, namely Bethesda classes III (Atypical follicular lesion of undetermined significance, AUS/FLUS) and IV (Suspicious for follicular neoplasm, FN/SFN) are fully analyzed and discussed.

Association of Bethesda category and molecular mutation in patients undergoing thyroidectomy

“…An increasing number of studies have concluded that molecular testing of thyroid nodules has a significant diagnostic as well as prognostic value and can help improve management of patients. [19][20][21][22][23][24][25] Other studies, however, question the clinical utility and costbenefit analysis of molecular testing. [20,24] The Afirma Gene Expression Classifier (Afirma®, Veracyte, Inc., South San Francisco, CA) measures mRNA transcript levels of 142 genes and has shown to have a high sensitivity of 91% and a moderate specificity of 68%.…”

“…[20,24] The Afirma Gene Expression Classifier (Afirma®, Veracyte, Inc., South San Francisco, CA) measures mRNA transcript levels of 142 genes and has shown to have a high sensitivity of 91% and a moderate specificity of 68%. [18,22] ThyGeNEXT (ThyGeNEXT®, Interpace Diagnostics, PDI Inc, Parsippany, NJ) and ThyroSeq v3 (ThyroSeq® joint partnership between UPMC and CBLPath, Pittsburgh, PA) are two comprehensive tests, designed to improve the overall accuracy of molecular testing, with sensitivity of 89% and 94% and specificity of 85% and 82%, respectively. [20,24] The data regarding the clinical utility of molecular testing in patients with a Bethesda V and VI thyroid nodules remains unclear.…”

Molecular Markers Guiding Thyroid Cancer Management

“…The choice of MT is based on performance factors such as specificity, sensitivity, PPV, Negative predictive value (NPV), and prevalence. It should either provide the information if a test can "rule out" cancer (likelihood that a nodule is benign) and reduce diagnostic surgery for benign nodules or "rule in" cancer (likelihood that the nodule is malignant) to optimize extent of surgical treatment [82,117]. Therefore, high sensitivity and NPV would be ideal to "rule out" malignancy, while high specificity and high PPV would help to "rule in" malignancies [148].…”

The Combination of Sonographic Features and the Seven-Gene Panel May be Useful in the Management of Thyroid Nodules With Indeterminate Cytology

“…The availability of molecular markers enhanced was significantly increased by the advent of next generation sequencing technology, and the gene expression classifier has also added a valuable contribution in the diagnosis of thyroid nodules ( 8 , 9 ). However, these tools are not routinely used in Europe because of the high cost ( 10 ). Recently, the availability of high-frequency ultrasonography and the development of risk scores that can quantify the risk of malignancy have significantly implemented the decision-making in the thyroid nodules ( 11 ).…”