Clinical variables have been investigated for a long time with regard to their ability to predict malignancy. High-risk history/examination features such as hoarseness, fixation of thyroid nodule to adjacent structures, nodal disease, strong family history or genetic conditions (e.g., medullary thyroid carcinoma, Cowden’s syndrome), and prior radiation exposure are well known but are relatively uncommon findings. Other major clinical factors that have been investigated are age, gender, and tumor size (Fiore et al. 2009):

There have been advances in associating sonographic features of thyroid nodules to malignancy risk. Hypoechogenicity, solid composition, microcalcifications, irregular or ill-defined margins, an absent sonolucent rim (or “halo”), and Doppler evidence of increased blood flow in the center of the nodule are associated with an increased risk of malignancy (Alexander 2008). True microcalcification appears to be the highest predictor of malignancy. Of course associated abnormal lymph nodes are highly suspicious for a primary thyroid malignancy.

Much effort has been made to identify molecular markers of thyroid malignancy. The most widely studied and most promising markers are BRAF, RAS, PAX8-PPAR-g, Galectin-3, and RET-PTC rearrangements. Recent research has shown in particular, BRAF mutational analysis to enhance diagnostic accuracy when used in conjunction with FNAC (Nikiforov and Nikiforova 2011). Several large centers in the USA now routinely test all thyroid FNAC samples for BRAF mutation as positivity provides both diagnostic and prognostic information. Further, in the USA (and perhaps shortly in Europe), there is increasing use of the commercial “genetic” testing Afirma (Veracyte Inc: http://​www.​veracyte.​com/​afirma/​) on indeterminate FNAC samples that appears to be very accurate. This panel of tests is presently very expensive but with eventual drop in costs its use may become widespread and points towards the expected future of molecular/gene testing in routine clinical practice.