333 [18F]FDG-PET/CT to prevent futile surgery: a blinded, randomised controlled multicentre trial 4 setting, with relatively limited costs for surgery and hospital admission, cost-effectiveness of these commercial molecular tests seems unattainable [29]. Locally developed and less comprehensive European molecular panels are available, but their diagnostic accuracy appears too limited for routine application in daily practice [492-494]. Due to large global variations in local health care expenses, case-mix, and availability of techniques, cost-utility and convenience of any diagnostic workup will greatly vary among different health care systems [25, 29, 495, 496]. Following previous model-based assumptions and the significant difference in costs that is demonstrated in the current study, life-long real-world cost-effectiveness of [18F]FDG-PET/CT is currently being modelled using the results of the EfFECTS trial [29]. Implementability of [18F]FDG-PET/CT was assessed in patients with an [18F]FDG-negative nodule in the [18F]FDG-PET/CT-driven group, who could deduce their allocation and [18F]FDG-PET/CT result from the surveillance advice. In spite of the suspense of participating in a clinical trial, the observed high therapy compliance reflects the patients’ and physicians’ diagnostic confidence and adoption of [18F]FDG-PET/CT as a trustworthy diagnostic tool. During study participation as well as afterwards, compressive symptoms were the principal reason for surgery in patients with a negative [18F]FDGPET/CT result and surveillance advice. This demonstrated that shared decision-making remains crucial to select patients for [18F]FDG-PET/CT who would not prefer surgery for discomfort from compressive symptoms, fear of malignancy or other reasons, optimize the (long-term) therapeutic yield of [18F]FDG-PET/CT, and limit unbeneficial use of resources. A potential limitation of our study is its per protocol one-year follow-up period for [18F]FDG-negative nodules. Concerns about missed cancer diagnoses were ameliorated by the extended median ultrasound follow-up of 29 months. Whether any very slow-growing malignant or borderline thyroid tumours in these nodules will lower diagnostic accuracy of [18F]FDG-PET/CT and limit the [18F]FDG-PET/CT-driven reduction in futile diagnostic surgeries remains to be established. Similarly, additional long-term false-negative results were recently reported for a well-known molecular classifier [497]. It seems clinically unlikely that a delayed diagnosis of any missed slow-growing malignancies or borderline tumours following [18F]FDG-PET/CT-driven management will alter the patients’ prognosis, as such tumours are likely indolent in nature. False-negative results in previous [18F]FDG-PET/CT studies in indeterminate nodules mostly concerned low-risk (T1) cancers [39, 40, 485]. In differentiated thyroid carcinoma, [18F]FDG-uptake is inversely related to prognosis, and [18F] FDG-negative carcinomas showed fewer aggressive characteristics on histopathology [498, 499]. In the current study, the two reported false-negative nodules were difficult to establish and only classified as malignant after extensive assessments including molecular analysis by multiple expert thyroid pathologists. Limitations of the routine use of [18F]FDG-PET/CT include the limited worldwide availability of PET/ CT scanners and adherence to standardised international scanning protocols, the use of low levels
RkJQdWJsaXNoZXIy MTk4NDMw