35 Diagnostic utility of molecular and imaging biomarkers 2 hyaluronic acid. Overexpression is found in various human cancers, including thyroid [155, 156]. In indeterminate thyroid nodules, a positive test for either one of the two biomarkers resulted in 100% sensitivity and 60% specificity. It is presumed that similar results for these markers are achieved with the more economical immunohistochemistry techniques [156, 157]. 7-Gene mutation panel Ongoing research in the past years has demonstrated that assessment of individual oncogenic mutations generally has limited clinical utility in indeterminate thyroid cytology. Combining forces of individual genetic alterations into a gene mutation panel, however, likely improves diagnostic accuracy, especially as mutations are mutually exclusive in most cases. These gene mutation panels typically assess the seven genetic alterations – gene mutations as well as gene fusions – that occur most frequently in differentiated thyroid carcinoma, including BRAFV600E, BRAFK601E, NRAS codon 61, HRAS codon 61 and KRAS codon 12-13 point mutations and RET/PTC1, RET/PTC3 and PAX8/PPARγ gene rearrangements [69, 93]. The best-known panel is the commercially available miRInform® thyroid (Asuragen Inc., Austin, Texas, USA), currently rebranded as the ThyGenX® Thyroid Oncogene Panel (Interpace Diagnostics, Parsippany, NJ, USA). The miRInform® thyroid tests 17 specific genetic alterations in these seven genes [97]. It is marketed as a rule-in test for thyroid malignancy. The first large clinical utility study to investigate the miRInform® thyroid test was published in 2011. Nikiforov et al. prospectively included 1,056 FNAC samples, 92% of which had sufficient epithelial cells and nucleic acids to pursue molecular testing. Residual FNAC material was used for mutation analysis – no additional aspirates were required. Unfortunately, surgery was performed for only 461 of 900 (51%) indeterminate thyroid nodules, independent of the test outcome; these operated cases were included in their final analysis. It is not reported whether nonsurgically managed nodules were mutation-positive or -negative. Sensitivity and specificity were 63% and 99% in the 247 Bethesda III nodules, and 57% and 97% in the 214 Bethesda IV nodules, respectively. The authors suggested that the high PPV of the miRInform® thyroid in these indeterminate thyroid nodules (88% and 87%, respectively) warrants a direct total thyroidectomy instead of two-step surgery in patients with a positive test [69]. None of the subsequent studies matched the initially reported excellent specificity. The next industrysponsored prospective study by Beaudenon et al. reported 47% sensitivity and 88% specificity in 80 Bethesda III and IV nodules. Surprisingly, not a single BRAF mutation was detected [97]. Valderrabano et al. reported not a single mutation in 47 included Bethesda III nodules. Moreover, only 1 of 18 nodules with Hürthle cell cytology in this study tested positive, suggesting that Hürthle cell nodules may carry different mutations than the ones investigated by the miRInform® thyroid [114]. Ohori et al. demonstrated that genetic alterations less frequently occurred in the textbook colloid-poor Bethesda IV cytology compared to the less common colloid-rich variant. Differences in aetiology
RkJQdWJsaXNoZXIy MTk4NDMw