Summary 533 & practice changing rule-out test for cytologically indeterminate thyroid nodules prior to surgery. Its application should be limited to non-oncocytic nodules for optimal therapeutic yield. Next, in Chapter 5, the additional diagnostic value of quantitative standardized uptake value (SUV) based analysis and radiomic analysis of [18F]FDG-PET/CT images were evaluated. Including only the 123 patients whose [18F]FDG-PET/CT scan was acquired with strict adherence to the EANM guidelines, receiver operating characteristic (ROC) curve analysis was performed for SUVs and SUVratios, including determination of SUV cut-offs at which a malignant/borderline neoplasm was reliably ruled out (≥95% sensitivity). The ROC curve analysis showed that quantitative [18F]FDG-PET/ CT assessment also accurately ruled out malignancy in indeterminate thyroid nodules. At SUVmax and SUVmax-ratio cut-offs of 2.1 g/mL and 1.2, sensitivity of [ 18F]FDG-PET/CT was 95.8% (95% CI, 78.9%-99.9%) in non-oncocytic nodules. In oncocytic thyroid nodules, distinctive, higher SUV cutoffs should be applied to optimize rule-out ability in these lesions. In oncocytic cell nodules, SUVmax and SUVmax-ratio cut-offs of 5.2 g/mL and 3.4 resulted in 100% sensitivity (95% CI, 66.4%-100%). If these cut-offs can be externally validated in future studies, quantitative [18F]FDG-PET/CT assessment may provide a useful rule-out test in oncocytic indeterminate thyroid nodules. Radiomic analysis, which was performed on the 84 (68%) [18F]FDG positive nodules using 107 extracted radiomic features from [18F]FDG-PET and low-dose CT images, elastic net regression classifiers trained in a 20-times repeated random split with incorporated dimensionality reduction, did not contribute to the additional differentiation of the [18F]FDG positive nodules. The mean area under the ROC curve of 0.445 (95% CI, 0.290-0.600) for the PET model. Chapter 6 evaluated cost-utility of an [18F]FDG-PET/CT-driven workup as compared to diagnostic surgery for indeterminate thyroid nodules. The observed 1-year data from the EfFECTS trial were used to calculate the 1-year societal costs and quality adjusted life years (QALYs). The 1-year difference in total societal costs was -€4,500 (-€9,200 to €150, p=0.06). The difference in QALYs was 0.019 (-0.045 to 0.083) at one year (p=0.57). Subsequently, these data were extrapolated using a 12-health state Markov Model to estimate lifelong societal costs and QALYs. This comprehensive model showed that the lifelong cost difference was -€9,900 (-€23,100 to €3,200, p=0.14), the difference in QALYs was 0.402 (-0.581 to 1.385) over the lifelong period (p=0.42). Cost-effectiveness acceptability curves were used to estimate the probability of cost-effectiveness. At a willingness to pay threshold of €50,000 per QALY, an [18F]FDG-PET/CT-driven work-up was very likely the costeffective strategy with 84% certainty. Univariate sensitivity analysis additionally demonstrated that the disutility after hemithyroidectomy for a benign nodule, the probability of a missed malignancy after initial surveillance for an [18F]FDG negative nodule (representing the false-negative rate or NPV of [18F]FDG-PET/CT), the disutility of active surveillance of an [18F]FDG negative nodule, and the price of the [18F]FDG-PET/CT had the largest influence on cost-effectiveness to the detriment of [18F]FDGPET/CT-driven management. None of these parameters, however, actually overturned the estimated cost-effectiveness of [18F]FDG-PET/CT for the included parameter ranges.
RkJQdWJsaXNoZXIy MTk4NDMw