189 Diagnostic utility of molecular and imaging biomarkers 2 GEC in indeterminate thyroid nodules with Hürthle cell cytology Meta-analysis was performed of the eight studies that separately reported on GEC test performance in 209 indeterminate thyroid nodules with Hürthle cell cytology [164, 169-171, 173, 180, 182]. Two studies did not report the distribution of GEC-positive and GEC-negative tests prior to selection for surgery [168, 169]. The others mention negative test results in 0% to 36%, and positive results in 63% to 100% [164, 170, 171, 173, 180, 182]. In total, 42 SHCN nodules were GEC-negative (21%). Histopathology was available for 160 SHCN nodules (76.6%), including only 12 of the GEC-negative SHCN nodules and 148 of the 161 reported GEC-positive (28% versus 91.9%, p<0.0001, Pearson chi-squared). Eleven of these 12 GEC-negative SHCN nodules were benign, one was a 3.5 cm FTC-OV [164]. Only 35 of the 148 GEC-positive SHCN nodules were histopathologically malignant. Prevalence of thyroid carcinoma was 17% (36/209). Estimated pooled sensitivity, specificity, positive and negative LR are 98.4% (95% CI: 38.6%-100%), 8.3% (95% CI: 3.9%-16.7%), 1.07 (95% CI: 0.99-1.16) and 0.19 (95% CI: 0.00-14.77), respectively. The AUC is 0.34 (95% CI: 0.30-0.38) (Table 41, Figure 51 and Figure 52). Together these findings demonstrate that the Afirma® GEC has poor diagnostic efficacy in SHCN nodules. Although pooled sensitivity in SHCN is similar to the result for all indeterminate thyroid nodules, the benign call rate in SHCN nodules is unfavourable. The majority of the SHCN nodules were GEC-positive, and less than one in four of these were histopathologically malignant. This increases the number needed to test and negatively affects possible cost-effectiveness of a GEC-driven management. These conclusions support previous presumptions that oncocytic thyroid nodules are a completely separate entity with a unique molecular profile [339-342]. In its current form, the Afirma® GEC may not be sufficiently trained to distinguish between benign and malignant Hürthle cell neoplasms. Diagnostic accuracy could improve if oncocytic indeterminate thyroid nodules are excluded from GEC testing. Otherwise, similar to the additional testing for medullary carcinoma, adaptations should be made to the Afirma® GEC to improve its clinical utility for Hürthle cell lesions. Figure 50. SROC curve of the Afirma® GEC – second-best-case scenario Summary receiver operating characteristic plot showing sensitivity versus 1-specificity of the Afirma® GEC in a hypothetical secondbest-case scenario in which we presumed 6% (missed) malignancies in indeterminate thyroid nodules without available histopathological correlation. In all other cases without histopathological correlation, we assumed that the nodule was benign. AUC = 0.84 (95% CI: 0.81-0.87). AUC, area under the curve; HSROC, hierarchical summary receiver operating characteristic.
RkJQdWJsaXNoZXIy MTk4NDMw