Operationalizing the centiloid scale for [18F]florbetapir PET studies on PET/MRI

Abstract

Introduction: The Centiloid scale aims to harmonize amyloid beta (Aβ) positron emission tomography (PET) measures across different analysis methods. As Centiloids were created using PET/computerized tomography (CT) data and are influenced by scanner differences, we investigated the Centiloid transformation with data from Insight 46 acquired with PET/magnetic resonanceimaging (MRI).

Methods: We transformed standardized uptake value ratios (SUVRs) from 432 florbetapir PET/MRI scans processed using whole cerebellum (WC) and white matter (WM) references, with and without partial volume correction. Gaussian-mixture-modelling-derived cutpoints for Aβ PET positivity were converted.

Results: The Centiloid cutpoint was 14.2 for WC SUVRs. The relationship between WM and WC uptake differed between the calibration and testing datasets, producing implausibly low WM-based Centiloids. Linear adjustment produced a WM-based cutpoint of 18.1.

Discussion: Transformation of PET/MRI florbetapir data to Centiloids is valid. However, further understanding of the effects of acquisition or biological factors on the transformation using a WM reference is needed.

Highlights: Centiloid conversion of amyloid beta positron emission tomography (PET) data aims to standardize results.Centiloid values can be influenced by differences in acquisition.We converted florbetapir PET/magnetic resonance imaging data from a large birth cohort.Whole cerebellum referenced values could be reliably transformed to Centiloids.White matter referenced values may be less generalizable between datasets.

Keywords: Alzheimer's disease; amyloid beta; centiloid; florbetapir; positron emission tomography/magnetic resonance imaging.

FIGURE 1

The relationship between non‐standard SUVRs (y‐axis) with GIF_WC (A), GIF_WC_PVC (B), GIF_WM (C), and GIF_WM_PVC (D) processing and standard processing PiB SUVRs (all x‐axis) in the Standard PiB dataset. The dashed line represents x = y and the black line is the linear regression fit with gray area representing 95% confidence interval. AD‐100, Alzheimer's disease group; GIF, Geodesic Information Flows‐based pipeline; PiB, Pittsburgh compound B; PVC, partial volume corrected; STD, standard Centiloid pipeline; SUVR, standardized uptake value ratio; WC, whole cerebellum reference; WM, eroded white matter reference; YC‐0, young control group.

FIGURE 2

Paired FBP and PiB SUVR data from the Florbetapir Calibration dataset. Plots show the relationship between FBP SUVRs (y‐axis) processed using GIF_WC (A), GIF_WC_PVC (B), GIF_WM (C), and GIF_WM_PVC (D) pipelines and PiB SUVRs with STD_WC processing (all x‐axes). The dashed line represents x = y and the black line is the linear regression fit with gray area representing 95% confidence interval. Conversion equations and R ² are displayed on the plots. All non‐standard methods exceed the reliability threshold (R ² > 0.7) set by Klunk et al. and are therefore suitable for Centiloid conversion. ES, elder subjects group; FBP, florbetapir; GIF, Geodesic Information Flows‐based pipeline; PiB, Pittsburgh compound B; PVC, partial volume corrected; STD, standard Centiloid pipeline; SUVR, standardized uptake value ratio; WC, whole cerebellum reference; WM, eroded white matter reference; YCN, young cognitively normal group.

FIGURE 3

The Centiloid conversion of florbetapir PET/MRI data from Insight 46 (N = 432) processed using the whole cerebellum reference region. The direct linear transformation from SUVR method to Centiloid values and conversion equations are shown for FBP_STD_WC (A), FBP_GIF_WC (B), and FBP_GIF_WC_PVC (C). For comparison, the distribution of SUVR (D), and Centiloid (E) are shown for each processing method. Dashed lines represent cutpoints derived using Gaussian mixture modelling in Insight 46. CL, Centiloid units; FBP, florbetapir; GIF, Geodesic Information Flows‐based pipeline; MRI, magnetic resonance imaging; PET, positron emission tomography; PVC, partial volume corrected; SD, standard deviation; STD, standard Centiloid pipeline; SUVR, standardized uptake value ratio; WC, whole cerebellum reference.

FIGURE 4

The Centiloid conversion of florbetapir PET/MRI SUVRs from Insight 46 (N = 432) processed with an eroded white matter reference region, without (A‐B) and with PVC (C‐D), with both unadjusted (A, C) and adjusted Centiloid values (B, D), conversion equations and cutpoints are presented. SUVR values (E) can be compared to both adjusted and unadjusted Centiloid values (F). Dashed black lines represent cutpoint values derived using Gaussian mixture modelling in Insight 46. Dashed lines represent cutpoints derived using Gaussian mixture modelling in Insight 46. CL, Centiloid units; FBP, florbetapir; GIF, Geodesic Information Flows‐based pipeline; MRI, magnetic resonance imaging; PET, positron emission tomography; PVC, partial volume corrected; SD, standard deviation; STD, standard Centiloid pipeline; SUVR, standardized uptake value ratio; WC, whole cerebellum reference.

FIGURE 5

Diagram outlining the steps for calibration of non‐standard whole cerebellum referenced SUVRs to Centiloids (A). An additional adjustment is required for conversion of SUVRs using a white matter reference region (B). Dashed lines represent cutpoints derived using Gaussian mixture modelling in Insight 46. CL, Centiloid units; FBP, florbetapir; GIF, Geodesic Information Flows‐based pipeline; PVC, partial volume corrected; STD, standard Centiloid pipeline; SUVR, standardized uptake value ratio; WC, whole cerebellum reference.