16 chapter 1 Diagnostic RCTs are also called ‘test-treatment trials’: patients are randomized to the experimental or standard diagnostic strategy, but health outcomes are only measured after the patients have also undergone the subsequent treatment, which is typically predefined in the trial protocol. As such, entire test-treatment strategies are actually evaluated rather than merely diagnostic strategies, and the effects of the diagnostic strategy will not just depend on the test itself but also on the effectiveness of subsequent management [30, 35]. Various designs are possible for diagnostic RCTs. Trials may assess a single test (i.e., comparing the experimental diagnostic intervention to standard patient management, ‘no test’) or compare multiple (experimental) diagnostics to each other. Next, the timing of the randomization may differ, that is, either prior to the diagnostic intervention or prior to the treatment based on the result of the diagnostic. These designs all have pros and cons with regard to methodological complexity, feasibility, efficacy, and validity (i.e., the degree to which the effect of the test itself is evaluated as compared to evaluation of the consequent treatment), among others depending on the desired outcome measures and the type, capacity, and costs of the diagnostic intervention(s) [34]. The most suitable trial design may therefore vary for each research aim. The EfFECTS trial This thesis is structured around the Efficacy of [18F]FDG-PET in Evaluation of Cytological indeterminate Thyroid nodules prior to Surgery (EfFECTS) trial. The EfFECTS trial was a triple blinded, randomised controlled multicentre trial that investigated the implementation of positron emission tomography/ computed tomography using 2-[18F]fluoro-2-deoxy-D-glucose ([18F]FDG-PET/CT) as a rule-out test in the diagnostic workup of cytologically indeterminate (Bethesda III and IV) thyroid nodules, aiming to reduce the number of futile diagnostic surgeries for benign nodules. [18F]FDG-PET/CT visualizes metabolic activity in tissues and can be used in the diagnosis, staging and therapeutic response monitoring of many malignancies. It utilizes the basic principle that the metabolism of (malignant) neoplasms and inflammation is upregulated as compared to that of normal tissues, with up to 200 times higher glycolytic rates and preferential lactic acid fermentation, even in abundance of oxygen (the Warburg effect) [36]. The EfFECTS trial was founded on previous work by our group: a 2006 prospective study that demonstrated that [18F]FDG-PET/CT accurately ruled out malignancy with 100% sensitivity in 44 patients with a thyroid nodule with inconclusive cytology, a 2011 meta-analysis of six earlier nonrandomized studies that demonstrated 95% sensitivity for [18F]FDG-PET/CT in indeterminate thyroid nodules, increasing to 100% for nodules above 15 mm in diameter, and a 2014 cost-effectiveness analysis based on a Markov decision model with a 5-year horizon that showed that [18F]FDG-PET/CTdriven management may cost-effectively reduce the fraction of futile surgeries from ~ 75% to ~ 40%, with an expected reduction in direct healthcare costs while preserving HRQoL [29, 37, 38]. Following these and other studies that confirmed the safety (i.e., high negative predictive value) of [18F]FDG-
RkJQdWJsaXNoZXIy MTk4NDMw