ABSTRACTBackgroundAmyloid positron emission tomography (PET) is instrumental in achieving an accurate diagnosis and may help to limit health‐seeking behavior. Currently, amyloid‐PET is not routinely used in clinical practice due to lack of evidence on cost‐utility. We assessed the cost‐utility of early versus no amyloid‐PET in the diagnostic work‐up of memory clinic patients after 6 months.MethodsWe assessed cost‐utility of patients enrolled in AMYPAD‐DPMS (EudraCT Number: 2017‐002527‐21) from six European memory clinics and randomized in ARM1; early amyloid‐PET, ARM2; no amyloid‐PET or ARM3; (amyloid‐PET at request of the managing physician). ARM3 was not part of the cost‐utility analysis. The EuroQol classification system (EQ‐5D‐5L), visual analogue scale (VAS), and ICEpop Capability measure for older people (ICECAP‐O) were collected at baseline and 6 months. Costs were calculated from cost diaries at baseline, 3 and 6 months. The incremental cost‐effectiveness ratio (ICER) was calculated using EQ‐5D‐5L and a societal perspective.ResultsFrom April 2018, to October 2020, 844 participants were screened and 840 were randomized (290 ARM1; 270 ARM2 and 280 ARM3). N = 514 (250 ARM1; 264 ARM2) were included in the economic evaluation. Amyloid‐PET resulted in higher costs at 6 months (ARM1 vs. ARM2 ∆€1384, bootstrapped 95% CI [7, 2761]). No significant difference in EQ‐5D‐5L, VAS or ICECAP‐O was found. The incremental cost‐effectiveness ratio (ICER) was €461,333 per QALY.ConclusionAlthough patients receive an early etiological diagnosis, the cost‐utility after 6 months is not favorable for amyloid‐PET. The cost‐utility will need to be reassessed when considering amyloid‐PET to select patients for anti‐amyloid biologics.