Abstract: While benign and malignant diagnoses rendered by fine-needle aspiration cytology (FNAC) are quite accurate, there remains a group of indeterminate lesions that are diagnostically challenging and difficult to preoperatively assign to a specific biologic behavior. These indeterminate lesions make up approximately 30% of thyroid FNAC and include atypia/follicular lesion of undetermined significance (AUS/FLUS), suggestive of follicular neoplasm (SFN), and suggestive of malignancy. With increasing knowledge about molecular mechanisms used in cancer, and lack of definitive cytologic characterization of indeterminate lesions, utilization of molecular techniques has become an attractive option to clinicians. The best available commercial tests today are miRInform (Asuragen) and Afirma (Veracyte), a 4-gene panel and gene expression classifier that are used to confirm or exclude the diagnosis of malignancy, respectively. Afirma test performs best in AUS/FLUS cases, but should be considered only if surgery is an option and only after repeat FNAC diagnosis of AUS/FLUS or SFN. Either or both molecular tests may be utilized in SFN cases, to confirm a benign or malignant clinical impression, and/or extent of surgery. Both of these molecular panels, however, are associated with significant false-positive and false-negative results. We believe that molecular testing can contribute to patient management, as an ancillary tool, in selected AUS/FLUS and SFN cases, but has no significant benefit to patients with lesions suggestive of malignancy. Molecular studies should not be ordered indiscriminately as a reflex test, and their results must not outweigh clinical judgment.