Impact of PET reconstruction on Aβ-amyloid quantitation in cross-sectional and longitudinal analyses

Abstract

AbstractAmyloid beta (Aβ) accumulation in Alzheimer’s disease (AD) is typically measured using standardized uptake value ratio (SUVR) and the Centiloid scale (CL). The low spatial resolution of PET images is known to degrade quantitative metrics due to the partial volume effect (PVE). This paper examines the impact of spatial resolution, as determined by the reconstruction configuration, on the Aβ-PET quantitation in both cross-sectional and longitudinal data.MethodsCross-sectional Study-89 subjects with [18F]-florbetapir scans (44 Aβ-, 45 Aβ+) were reconstructed using 69 reconstruction configurations. For each reconstruction, Aβ SUVR was calculated and the spatial resolution was calculated as full-width-at-half-maximum (FWHM) using the barrel phantom method (Lodgeet al, 2018). The change of SUVR and the effect size of the difference in SUVR between Aβ- and Aβ+ groups with FWHM were examined.Longitudinal study-79 subjects (46 Aβ-, 33 Aβ+) with three [18F]-flutemetamol scans were analysed. All scans were reconstructed using low-, medium- and high-resolution reconstruction configurations and Aβ CLs were calculated. Since linear Aβ accumulation was assumed over a 10-year interval, for each reconstruction configuration, Aβ accumulation rate differences (ARD) between the second and first periods were calculated for all the subjects and compared. Zero ARD was used as a consistency metric. The number of Aβ-accumulators was also used to compare reconstruction configurations.ResultsCross-sectional-SUVRs in both Aβ- and Aβ+ groups were impacted by the FWHM of the reconstruction method; Aβ- SUVRs increased for FWHM ≥ 4.5 mm, while Aβ+ SUVRs decreased across the FWHM range. High-resolution reconstructions provided the best statistical separation between groups.Longitudinal study-In the Aβ-group, the median ARD of low-resolution reconstructed data was greater than zero whereas the ARDs of higher-resolution reconstructions were not significantly different to zero, indicating less consistent rates in the low-than the higher-resolution data. Higher-resolution reconstructions identified 10 additional Aβ-accumulators in the Aβ-group, resulting in a 22% increased group size compared to the low-resolution reconstructions. Higher-resolution reconstructions reduced the average CL values of the negative group by 12 points.ConclusionsHigh-resolution PET reconstructions, inherently less impacted by PVE, may improve Aβ-PET quantitation in both cross-sectional and longitudinal data. In the cross-sectional analysis, separation of Aβ groups’ SUVRs increased with spatial resolution. Longitudinal analysis showed better Aβ accumulation consistency in higher-resolution compared to low-resolution reconstructions. The identification of more Aβ-accumulators from the higher-resolution reconstruction may be helpful in early-stage AD therapies.