Quantitative amyloid PET imaging in Alzheimer’s disease: Walking the line between accuracy and simplicity

Research output: PhD ThesisPhd-Thesis - Research and graduation internal


Positron emission tomography (PET) allows for in vivo visualisation and quantification of amyloid-β (Aβ) plaques, which are known to play a major role in the development of Alzheimer’s disease (AD). However, it remains unclear whether these plaques have a causal relationship with AD dementia. Quantitatively tracking of Aβ plaques using PET may 1) enhance our understanding of AD, 2) improve our ability to predict who will develop AD dementia and 3) improve participant selection in secondary prevention trials. Typically, quantification requires a dynamic scan and measurement of the available PET tracer in plasma. As these are time consuming and complex procedures, widespread implementation in clinical or research practise is not feasible and warrants simplified approaches. The overall aim of this thesis was to evaluate to what extent approaches for quantification of Aβ pathology using PET can be simplified, within the context of different applications. In short, the main findings of this thesis were that, in case of the [18F]flutemetamol, [18F]florbetaben and [11C]PiB amyloid tracers, the semi-quantitative standardized uptake value ratio (SUVR), calculated from a static PET scan, showed a bias that was proportional to underlying Aβ pathology compared with the quantitative distribution volume ratio (DVR). The bias was smallest in participants (or regions) with very low levels of Aβ pathology and the effect of cerebral blood flow changes on bias in SUVR appeared to be minimal in cognitively unimpaired individuals. The proportional bias in SUVR suggests that for highly accurate quantification of Aβ pathology, such as when measuring small changes in longitudinal studies, dynamic or validated dual-time window scanning protocols may be required. More specifically, the present work showed that dual-time window protocols can substitute dynamic protocols with only limited loss of accuracy, while increasing patient comfort and throughput. Clearly, highly accurate and robust quantification can only be achieved if these scanning protocols are used in combination with a validated pharmacokinetic model and reference tissue, if applicable. Another advantage of dynamic and dual-time window scanning protocols is that a measure of relative tracer delivery (R1) can be extracted. In terms of its methodological characteristics, it was demonstrated that R1 is a stable parameter and, it has the potential to serve as an additional biomarker for measuring disease progression in AD. Finally, considering that Aβ plaque accumulation starts focally, research aiming to measure small changes in Aβ pathology over time should focus on regional instead of global measures of Aβ pathology. This strategy has the potential to improve participant selection in clinical trials and predict cognitive decline. In addition, regional quantification of amyloid burden and the use of dynamic scanning protocols can reduce the required sample sizes in primary and secondary AD prevention trials.
Original languageEnglish
QualificationDoctor of Philosophy
Awarding Institution
  • Vrije Universiteit Amsterdam
  • Lammertsma, Adriaan, Supervisor
  • Yaqub, Mohammed, Co-supervisor
  • Lopes Alves, Isadora, Co-supervisor
Award date3 Nov 2021
Place of PublicationEnschede
Print ISBNs9789464214826
Publication statusPublished - 3 Nov 2021


  • Alzheimer's disease
  • amyloid
  • positron emission tomography
  • quantification, imaging

Cite this