Validation of a novel PET imaging marker for dopaminergic degeneration in Parkinson’s disease patients

Abstract

Background: There is a need for an objective disease progression marker for Parkinson’s disease (PD). DAT measurement with PET-radioligand [18F]FE-PE2I has so far shown to be promising with its favorable kinetics, low production of metabolites, and ability to image the whole nigrostriatal pathway.

Aim: The aim of this thesis was to continue the validation of [18F]FE-PE2I measures of DAT in subjects with Parkinson’s disease, addressing the following knowledge gaps and research questions: reliability and repeatability in PD subjects (paper 1); longitudinal measurements within PD (paper 4), correlations of DAT and clinical motor symptoms with improved methods (paper 3), and further exploration of different shortened scan times and simplified quantification methods, to assess feasibility of implementation in a clinical context (paper 2). Methods In total 41 subjects with non-advanced idiopathic PD and 37 age and sex matched healthy controls (HC) had 93-minute [18F]FE-PE2I PET examinations, collected with a high resolution research tomograph. Twenty-five PD subjects had follow-up PET after ~2 years. Nigrostriatal regions of interest were automatically delineated, and [18F]FE-PE2I measurements were quantified with wavelet-aided parametric imaging using Logan ref and cerebellum as reference region.

Results/conclusions/discussion: [18F]FE-PE2I test-retest data showed good-to-excellent reliability and low variability of striatal [18F]FE-PE2I PET measures in PD subjects, with moderate results for the substantia nigra (paper 1). The test-retest differences allowed for effect size and sample size calculations for future studies. The cross-sectional study (paper 3) replicated the finding of PD-related nigrostriatal, rostro-caudal pattern of DAT decline. DAT in hypothesized striatal areas correlated significantly with symptom duration, Hoehn and Yahr stage, and motor symptom score (MDS-UDPRS-III) especially when subtracting tremor. One outlier subject influenced the latter analysis results. Longitudinal [18F]FE-PE2I PET measures in PD (paper 4) proved to have large effect sizes (>0.88), allowing for 2-3 times lower sample size needed in a treatment trial when comparing to estimates made with [123I]FP-CIT SPECT data. Unfortunately, the sample with standardized longitudinal assessments of ‘OFF’ MDS- UPDRS-motor score was too low for robust analysis on correlations between DAT change versus motor symptom change in PD. A future study could address this remaining question. Both paper 3 and 4 showed a negative exponential relation between striatal DAT and symptom duration, fitting with literature. The DAT in the substantia nigra showed a flat relation to symptom duration and HY stage, possibly due to floor effect, later progression, or compensatory mechanisms in still non-advanced PD. A future study with patients with more advanced PD could shine more light on this, or a study combining [18F]FE-PE2I PET with other imaging markers. Paper 2 combined baseline, test-retest, and longitudinal data and showed that simplified quantification of shorter scan times (18 or 30 minutes) at late pseudo- equilibrium give DAT estimates that suffice to discriminate subjects with PD from HC, but for a clinical trial on disease progression or treatment effect, the early time-window is recommended, and 30 minutes is preferred over 18 minutes.

In conclusion, these additional validation studies show that [18F]FE-PE2I PET is reliable and precise for following DAT changes in PD subjects and non-inferior to other DAT-radioligands. It can be used in a simplified manner, enabling wider use. Further studies on DAT in substantia nigra in later stage PD, in combination with other biomarkers, and larger studies on longitudinal correlations with MDS-UPDRS motor scores would be interesting next studies.