Standard Centiloid Research Articles

Tau‐PET to predict conversion from mild cognitive impairment to dementia; a head‐to‐head comparison against amyloid‐PET and MRI

AbstractBackgroundIn contrast to amyloid‐PET and MRI, tau‐PET imaging has not been integrated into clinical practice even though it shows a close relationship with clinical symptoms. With the present study, we aim to assess the performance of tau‐PET to predict conversion to dementia (all‐cause and AD) in individuals with mild cognitive impairment (MCI), and compare it with amyloid‐PET and MRI.MethodWe selected participants from an ongoing multicenter study (n = 408, Table 1). The following ROIs were used to define quantified thresholds for positivity using an independent cognitively unimpaired reference group: a temporal meta‐region for tau‐PET and an AD‐susceptible temporoparietal region for MRI‐based cortical thickness. For amyloid‐PET, we used the standard Centiloid mask and a published threshold of 27 Centiloids to determine positivity. We additionally performed visual reads for all imaging modalities. We used Cox proportional hazard, logistic regression, and receiver operating curve (ROC) analyses to assess the performance of amyloid‐PET (A), tau‐PET (T), and MRI (N) to predict progression from MCI to dementia over a mean follow‐up period of 2 years.ResultHazard ratios indicated that tau‐PET and amyloid‐PET positivity (both when assessed quantitatively and using visual reads) conveyed a greater risk of progression to dementia than MRI, and that a combination of A+ and T+ (or N+) conveyed a greater risk than A‐T‐N‐ or isolated A+ or T+ (Figure 1). ROC analyses revealed that only tau‐PET performed better than demographics alone when predicting all‐cause dementia while both amyloid‐ and tau‐PET performed better than demographics and MRI when predicting AD dementia. LASSO logistic regression revealed that the optimal model to predict conversion from MCI to all‐cause dementia included a quantitative MRI marker and tau‐PET visual read. For optimal prediction of AD dementia specifically, the quantitative measure of amyloid‐PET was added in addition to the quantitative MRI marker and the tau‐PET visual read (Table 2).ConclusionAmong individuals with MCI, tau‐PET performs best when predicting dementia, and can be optimized when combined with MRI. For AD dementia, tau‐PET and amyloid‐PET are the best‐performing modalities, but an optimal prediction is again achieved when combined with MRI.

Tau‐PET to predict conversion from mild cognitive impairment to dementia; a head‐to‐head comparison against amyloid‐PET and MRI

AbstractBackgroundIn contrast to amyloid‐PET and MRI, tau‐PET imaging has not been integrated into clinical practice even though it shows a close relationship with clinical symptoms. With the present study, we aim to assess the performance of tau‐PET to predict conversion to dementia (all‐cause and AD) in individuals with mild cognitive impairment (MCI), and compare it with amyloid‐PET and MRI.MethodWe selected participants from an ongoing multicenter study (n = 408, Table 1). The following ROIs were used to define quantified thresholds for positivity using an independent cognitively unimpaired reference group: a temporal meta‐region for tau‐PET and an AD‐susceptible temporoparietal region for MRI‐based cortical thickness. For amyloid‐PET, we used the standard Centiloid mask and a published threshold of 27 Centiloids to determine positivity. We additionally performed visual reads for all imaging modalities. We used Cox proportional hazard, logistic regression, and receiver operating curve (ROC) analyses to assess the performance of amyloid‐PET (A), tau‐PET (T), and MRI (N) to predict progression from MCI to dementia over a mean follow‐up period of 2 years.ResultHazard ratios indicated that tau‐PET and amyloid‐PET positivity (both when assessed quantitatively and using visual reads) conveyed a greater risk of progression to dementia than MRI, and that a combination of A+ and T+ (or N+) conveyed a greater risk than A‐T‐N‐ or isolated A+ or T+ (Figure 1). ROC analyses revealed that only tau‐PET performed better than demographics alone when predicting all‐cause dementia while both amyloid‐ and tau‐PET performed better than demographics and MRI when predicting AD dementia. LASSO logistic regression revealed that the optimal model to predict conversion from MCI to all‐cause dementia included a quantitative MRI marker and tau‐PET visual read. For optimal prediction of AD dementia specifically, the quantitative measure of amyloid‐PET was added in addition to the quantitative MRI marker and the tau‐PET visual read (Table 2).ConclusionAmong individuals with MCI, tau‐PET performs best when predicting dementia, and can be optimized when combined with MRI. For AD dementia, tau‐PET and amyloid‐PET are the best‐performing modalities, but an optimal prediction is again achieved when combined with MRI.

A universal neocortical mask for Centiloid quantification.

The Centiloid (CL) project was developed to harmonize the quantification of amyloid beta (Aβ) positron emission tomography (PET) scans to a unified scale. The CL neocortical mask was defined using 11C Pittsburgh compound B (PiB), overlooking potential differences in regional distribution among Aβ tracers. We created a universal mask using an independent dataset of five Aβ tracers, and investigated its impact on inter-tracer agreement, tracer variability, and group separation. Using data from the Alzheimer's Dementia Onset and Progression in International Cohorts (ADOPIC) study (Australian Imaging Biomarkers and Lifestyle + Alzheimer's Disease Neuroimaging Initiative + Open Access Series of Imaging Studies), age-matched pairs of mild Alzheimer's disease (AD) and healthy controls (HC) were selected: 18F-florbetapir (N=147 pairs), 18F-florbetaben (N=22), 18F-flutemetamol (N=10), 18F-NAV (N=42), 11C-PiB (N=63). The images were spatially and standardized uptake value ratio normalized. For each tracer, the mean AD-HC difference image was thresholded to maximize the overlap with the standard neocortical mask. The universal mask was defined as the intersection of all five masks. It was evaluated on the Global Alzheimer's Association Interactive Network (GAAIN) head-to-head datasets in terms of inter-tracer agreement and variance in the young controls (YC) and on the ADOPIC dataset comparing separation between HC/AD and HC/mild cognitive impairment (MCI). In the GAAIN dataset, the universal mask led to a small reduction in the variance of the YC, and a small increase in the inter-tracer agreement. In the ADOPIC dataset, it led to a better separation between HC/AD and HC/MCI at baseline. The universal CL mask led to an increase in inter-tracer agreement and group separation. Those increases were, however, very small, and do not provide sufficient benefits to support departing from the existing standard CL mask, which is suitable for the quantification of all Aβ tracers. This study built an amyloid universal mask using a matched cohort for the five most commonly used amyloid positron emission tomography tracers.There was a high overlap between each tracer-specific mask.Differences in quantification and group separation between the standard and universal mask were small.The existing standard Centiloid mask is suitable for the quantification of all amyloid beta tracers.

Validation of quantitative assessment of florbetaben PET scans as an adjunct to the visual assessment across 15 software methods.

Amyloid positron emission tomography (PET) with [18F]florbetaben (FBB) is an established tool for detecting Aβ deposition in the brain in vivo based on visual assessment of PET scans. Quantitative measures are commonly used in the research context and allow continuous measurement of amyloid burden. The aim of this study was to demonstrate the robustness of FBB PET quantification. This is a retrospective analysis of FBB PET images from 589 subjects. PET scans were quantified with 15 analytical methods using nine software packages (MIMneuro, Hermes BRASS, Neurocloud, Neurology Toolkit, statistical parametric mapping (SPM8), PMOD Neuro, CapAIBL, non-negative matrix factorization (NMF), AmyloidIQ) that used several metrics to estimate Aβ load (SUVR, centiloid, amyloid load, and amyloid index). Six analytical methods reported centiloid (MIMneuro, standard centiloid, Neurology Toolkit, SPM8 (PET only), CapAIBL, NMF). All results were quality controlled. The mean sensitivity, specificity, and accuracy were 96.1 ± 1.6%, 96.9 ± 1.0%, and 96.4 ± 1.1%, respectively, for all quantitative methods tested when compared to histopathology, where available. The mean percentage of agreement between binary quantitative assessment across all 15 methods and visual majority assessment was 92.4 ± 1.5%. Assessments of reliability, correlation analyses, and comparisons across software packages showed excellent performance and consistent results between analytical methods. This study demonstrated that quantitative methods using both CE marked software and other widely available processing tools provided comparable results to visual assessments of FBB PET scans. Software quantification methods, such as centiloid analysis, can complement visual assessment of FBB PET images and could be used in the future for identification of early amyloid deposition, monitoring disease progression and treatment effectiveness.

Head‐to‐Head comparison between Philips Gemini TF64 and Siemens Biograph Vision 600 for brain amyloid Centiloid quantitation

AbstractBackgroundThe Centiloid (CL) scale calibrates the beta‐amyloid (Aβ) deposition from different PET tracers to a standardised 0‐100 CL unit scale. As imaging sites update their PET cameras, most are switching to digital detector systems with superior resolution and sensitivity that may affect quantitation. This has significant implications for dementia clinical trials. In this study, we examine the impact on CL quantification between Philips Gemini TF64 and Siemens Biograph Vision 600.MethodSeven subjects (76.4±2.2 yo) were imaged with 18F‐NAV4694 on both Gemini TF64 and Biograph Vision consecutively with an average scan interval of 25.1±11.2 weeks. The injected doses were 200MBq and 100MBq, respectively. On the Gemini TF64, the PET images were reconstructed by LOR‐RAMLA algorithm with smoothing parameter setup as ‘SHARP’. On Biograph Vision, the PET images were reconstructed by OSEM‐3D (8 iterations and 5 subsets, TOF enabled) with 3mm post Gaussian smoothing. A T1 MRI image was acquired for each subject. As per the standard Centiloid method the whole cerebellum was used as the reference in SUVR images, and all images were processed using CapAIBL to calculate the CL using both MR‐based and MR‐Less spatial normalisation.ResultFigure 1 shows the CL images of a subject scanned on Gemini TF64 and Biograph Vision within sixteen weeks. The Biograph Vision images have higher contrast and higher spatial resolution despite using half of the dose. Figure 2 shows the linear regression plot of the scanner comparison. Biograph Vision CL are progressively higher than those obtained from the Gemini TF64 as the CL value rises (Table 1). There were no significant differences between the MR‐based and MR‐less results.ConclusionBiograph Vision yields higher SUVR and therefore CL values compared to Gemini TF64 in a head‐to‐head comparison. These results show that the selection of PET camera has a significant impact on CL quantification, which needs to be considered when merging cohorts from different studies or changing cameras during longitudinal studies or trials. These initial results indicate that the CL difference could be corrected by a linear transform.

Head‐to‐Head comparison between Philips Gemini TF64 and Siemens Biograph Vision 600 for brain amyloid Centiloid quantitation

AbstractBackgroundThe Centiloid (CL) scale calibrates the beta‐amyloid (Aβ) deposition from different PET tracers to a standardised 0‐100 CL unit scale. As imaging sites update their PET cameras, most are switching to digital detector systems with superior resolution and sensitivity that may affect quantitation. This has significant implications for dementia clinical trials. In this study, we examine the impact on CL quantification between Philips Gemini TF64 and Siemens Biograph Vision 600.MethodSeven subjects (76.4±2.2 yo) were imaged with 18F‐NAV4694 on both Gemini TF64 and Biograph Vision consecutively with an average scan interval of 25.1±11.2 weeks. The injected doses were 200MBq and 100MBq, respectively. On the Gemini TF64, the PET images were reconstructed by LOR‐RAMLA algorithm with smoothing parameter setup as ‘SHARP’. On Biograph Vision, the PET images were reconstructed by OSEM‐3D (8 iterations and 5 subsets, TOF enabled) with 3mm post Gaussian smoothing. A T1 MRI image was acquired for each subject. As per the standard Centiloid method the whole cerebellum was used as the reference in SUVR images, and all images were processed using CapAIBL to calculate the CL using both MR‐based and MR‐Less spatial normalisation.ResultFigure 1 shows the CL images of a subject scanned on Gemini TF64 and Biograph Vision within sixteen weeks. The Biograph Vision images have higher contrast and higher spatial resolution despite using half of the dose. Figure 2 shows the linear regression plot of the scanner comparison. Biograph Vision CL are progressively higher than those obtained from the Gemini TF64 as the CL value rises (Table 1). There were no significant differences between the MR‐based and MR‐less results.ConclusionBiograph Vision yields higher SUVR and therefore CL values compared to Gemini TF64 in a head‐to‐head comparison. These results show that the selection of PET camera has a significant impact on CL quantification, which needs to be considered when merging cohorts from different studies or changing cameras during longitudinal studies or trials. These initial results indicate that the CL difference could be corrected by a linear transform.

Development and clinical validation of CT-based regional modified Centiloid method for amyloid PET

BackgroundThe standard Centiloid (CL) method was proposed to harmonize and quantify global 18F-labeled amyloid beta (Aβ) PET ligands using MRI as an anatomical reference. However, there is need for harmonizing and quantifying regional Aβ uptakes between ligands using CT as an anatomical reference. In the present study, we developed and validated a CT-based regional direct comparison of 18F-florbetaben (FBB) and 18F-flutemetamol (FMM) Centiloid (rdcCL).MethodsFor development of MRI-based or CT-based rdcCLs, the cohort consisted of 63 subjects (20 young controls (YC) and 18 old controls (OC), and 25 participants with Alzheimer’s disease dementia (ADD)). We performed a direct comparison of the FMM-FBB rdcCL method using MRI and CT images to define a common target region and the six regional VOIs of frontal, temporal, parietal, posterior cingulate, occipital, and striatal regions. Global and regional rdcCL scales were compared between MRI-based and CT-based methods. For clinical validation, the cohort consisted of 2245 subjects (627 CN, 933 MCI, and 685 ADD).ResultsBoth MRI-based and CT-based rdcCL scales showed that FMM and FBB were highly correlated with each other, globally and regionally (R2 = 0.96~0.99). Both FMM and FBB showed that CT-based rdcCL scales were highly correlated with MRI-based rdcCL scales (R2 = 0.97~0.99). Regarding the absolute difference of rdcCLs between FMM and FBB, the CT-based method was not different from the MRI-based method, globally or regionally (p value = 0.07~0.95). In our clinical validation study, the global negative group showed that the regional positive subgroup had worse neuropsychological performance than the regional negative subgroup (p < 0.05). The global positive group also showed that the striatal positive subgroup had worse neuropsychological performance than the striatal negative subgroup (p < 0.05).ConclusionsOur findings suggest that it is feasible to convert regional FMM or FBB rdcSUVR values into rdcCL scales without additional MRI scans. This allows a more easily accessible method for researchers that can be applicable to a variety of different conditions.

Baseline amyloid and tau PET imaging characteristics in the ongoing phase 2 study of tilavonemab in early Alzheimer’s disease

AbstractBackgroundAmyloid and tau are recognized hallmarks of Alzheimer’s disease (AD). Tilavonemab is a monoclonal antibody that binds to the N terminus of human tau and is being developed as a treatment for early AD. The goal of this work is to present the baseline amyloid and tau PET imaging characteristics of patients enrolled in the ongoing phase 2 study of tilavonemab.Method453 patients who met the clinical criteria for early AD and had a positive amyloid PET scan (by visual read) at screening were randomized into the study. Amyloid PET was quantified on the standard Centiloid (CL) scale utilizing a processing pipeline developed in‐house and validated using recommended procedures [Klunk, et al., 2015]. Tau PET imaging using 18F‐MK6240 was also performed in a subset of patients to assess presence of tau pathology, the target of tilavonemab, and to monitor disease progression and treatment effects. Here we report baseline tau PET data from 59 subjects randomized into the trial (out of 70 scans collected, not including screen fails or out of baseline window scans). Tau PET was quantified in Braak regions I‐VI using standardized uptake value ratio (SUVR), extent (percentage of voxels within a region above threshold) and load (extent * mean SUVR of voxels within a region above threshold) metrics. Ventral cerebellar gray matter was used as a reference region for SUVR calculation and to determine the subject‐specific threshold needed for the extent and load calculation.ResultOver 99% of subjects (449/453) had amyloid measurements greater than 20CL, a threshold used for positivity [Burnham, et al., 2020], consistent with visual read; mean±s.d. was 99.0±31.1 CL. For the 59 patients with baseline tau PET data 95% (56/59) were considered to have elevated tau pathology in the brain (SUVR ≥1.27 in the entorhinal cortex [Betthauser, et al., 2020]). The tau metrics for Braak regions I‐VI are summarized in Table 1.ConclusionThis analysis confirmed the presence of amyloid pathology in the patients randomized into the trial and suggested a high proportion of patients have tau pathology in the brain. Future analyses will examine how tilavonemab effects tau pathology.

A novel regional centiloid atlas to investigate the relationship between the quantification of [18F]flutemetamol amyloid PET and CSF Aβ42/40 in patients with SCD or MCI

AbstractBackgroundVisual assessment of [18F]flutemetamol amyloid PET images, performed regionally (frontal, lateral temporal, posterior cingulate/precuneus, temporo‐parietal, striatum), is recommended, with a positive (abnormal) classification if any one of these regions is clearly positive. Changes in the striatum, known to contain extensive fibrillar amyloid deposits in AD, could identify likely converters in Subjective Cognitive Decline (SDC) or Mild Cognitive Impairment (MCI) subjects. CSF biomarkers are proven to detect abnormal amyloid metabolism in early stages of AD, with Aβ42/40 more reliable than Aβ42 alone. We applied a novel regional Centiloid Atlas, based on standard Centiloid (CL) cortical VOIs with additional striatal , aiming to better stratify amyloid abnormalities. Correlation between CL and Aβ42/40 was also investigated.MethodsThe regional Centiloid atlas, developed by overlaying standard CL cortical VOI on AAL atlas and incorporating additional striatal regions, was applied to 394 [18F]flutemetamol subjects from the Swedish BioFINDER I study following the methods of Klunk et.al. (Alzheimer’s &amp; Dementia, 2015). In total, 387 Subjects with CSF Aβ42/40 information from the ADx analysis platform available, were included for analysis.ResultsAverage Aβ42/40 and CL values are presented (Table 1). Statistically significant differences in Aβ42/40 between control and MCI (p=0.0004), and SCD and MCI (p=0.0031) were observed (Fig. 1). Statistically significant differences in CL were also found in the cortical (p=0.007 or less) and striatal (p=0.028 or less) regions across all groups. Correlation between CL and Aβ42/40 for controls was moderate in both cortical (R2=0.35, Fig. 1) and striatal (R2=0.33, Fig. 2) regions. Similar correlations were found for SCD (Cortical: R2=0.52, Striatal: R2=0.48) and MCI (Cortical: R2=0.52, Striatal: R2=0.54).ConclusionsStandard Centiloid PET quantification provides global values across the cortical region. Identifying combined striatal differences in [18F]flutemetamol retention may provide additional information, aiding . Detecting these abnormalities and understanding their relationship to CSF biomarker changes may be useful in identifying those subjects for progression to AD. Further work to identify differences between cortical and striatal uptake in within subjects is ongoing.

Classification of AD/MCI/HC based on amyloid‐PET using Random Forest Ensemble

AbstractBackgroundPositron emission topography (PET) and magnetic resonance imaging (MRI) are two common in‐vivo techniques for supporting clinical diagnosis of dementia. The correlated neuropathological changes including amyloid‐beta (Aβ) deposition, and cortical atrophy may have the potential to predict conversion from cognitively normal elderly adults (HC), to mild cognitive impairment (MCI) and eventually Alzheimer’s disease (AD). In order to overcome the complex interactions among these biomarker datasets, random forest (RF) as one of the machine learning algorithms is usually used in data classification. In current study, we aim at evaluating the accuracy of RF model in distinguishing HC, MCI from AD and the importance of various neuropathological features in selection.MethodThree cohorts were included in our study. We recruited 94 AD, 82 MCI and 85 HC from GAAIN (The Global Alzheimer’s Association Interactive Network) database, AIBL (Australian imaging, biomarkers and lifestyle) database, and our memory clinic database. Based on Centiloid pipeline, we processed the co‐registered MRI and PET images of each subject. Additionally, to compare quantitative amyloid load across different tracers, we converted SUVR (Standardized Uptake Value Ratio) values into standard Centiloid units. RF algorithm was performed on Python via scikit‐learn package. Finally, 122 regional volumes, 68 regional cortical thicknesses, 16 small regional amyloid distribution and 1 global amyloid load were input as features.ResultIn Table 1, the AUC value was highest (AUC=0.82) in the classification between HC and AD, intermediate (AUC=0.78) between HC and MCI and lowest (AUC=0.65) between AD and MCI. In each binary classification, sensitivity‐71%, specificity‐85%, accuracy‐78%; sensitivity‐88%, specificity‐76%, accuracy‐81% and sensitivity‐86%, specificity‐44%, accuracy‐66% were achieved in differentiating MCI from HC, AD from HC and AD from MCI respectively. For importance ranking of features within the model, 6 features of regional cortical thickness and 4 features of regional volume were found in the classification between HC and MCI, while all 10 essential features belong to regional Aβ ‐ROI in classification between HC and AD as well as MCI and AD (Table 2).ConclusionRandom forest model using regional volume, regional cortical thickness and amyloid load (in Centiloid unit) had moderate to high accuracy in differentiating AD from HC and MCI.

Centiloid scale analysis for 18F-THK5351 PET imaging in Alzheimer's disease.

A standardized method for quantification is required for analyzing PET data, but such standards have not been established for tau PET imaging. The Centiloid scale has recently been proposed as a standard method for quantifying amyloid deposition on PET imaging. Therefore, the present study aimed to apply the Centiloid scale to 18F-THK5351 PET imaging in Alzheimer's disease (AD). We acquired 18F-THK5351 PET, 11C-PiB PET, and MR images from 47 cognitively normal (CN) individuals and 28 patients with AD with mild to moderate dementia. PET images were spatially normalized to Montreal Neurological Institute space. The PET signals were then normalized using the signal in the reference volume of interest (VOI). Target VOI for specific 18F-THK5351 retention in AD was extracted by voxel-wise comparison of PET images between the 47 CN individuals and 16 AD patients with moderate dementia. Scale anchor points were defined by the CN individuals as 0-anchor points and by that of the average of the typical AD patients as 100-anchor points. Specific retention of 18F-THK5351 was predominant in the angular gyrus, inferior temporal cortex, and parieto-occipital regions in patients with AD. Standardized uptake value ratio (SUVR) of 1.227 and 1.797 were defined as 0- and 100-anchor points, respectively. 18F-THK5351 PET data could be expressed using the Centiloid scale, with the SUVR of the 18F-THK5351 PET images converted to Centiloid using our VOI, the standard Centiloid reference VOI, and the following equation: Centiloid=169.0×SUVR-204.6. Centiloid methods can be applied to tau PET imaging using 18F-THK5351.

Non-negative matrix factorisation improves Centiloid robustness in longitudinal studies

Background: Centiloid was introduced to harmonise β-Amyloid (Aβ) PET quantification across different tracers, scanners and analysis techniques. Unfortunately, Centiloid still suffers from some quantification disparities in longitudinal analysis when normalising data from different tracers or scanners. In this work, we aim to reduce this variability using a different analysis technique applied to the existing calibration data. Method: All PET images from the Centiloid calibration dataset, along with 3762 PET images from the AIBL study were analysed using the recommended SPM pipeline. The PET images were SUVR normalised using the whole cerebellum. All SUVR normalised PiB images from the calibration dataset were decomposed using non-negative matrix factorisation (NMF). The NMF coefficients related to the first component were strongly correlated with global SUVR and were subsequently used as a surrogate for Aβ retention. For each tracer of the calibration dataset, the components of the NMF were computed in a way such that the coefficients of the first component would match those of the corresponding PiB. Given the strong correlations between the SUVR and the NMF coefficients on the calibration dataset, all PET images from AIBL were subsequently decomposed using the computed NMF, and their coefficients transformed into Centiloids. Results: Using the AIBL data, the correlation between the standard Centiloid and the novel NMF-based Centiloid was high in each tracer. The NMF-based Centiloids showed a reduction of outliers, and improved longitudinal consistency. Furthermore, it removed the effects of switching tracers from the longitudinal variance of the Centiloid measure, when assessed using a linear mixed effects model. Conclusion: We here propose a novel image driven method to perform the Centiloid quantification. The methods is highly correlated with standard Centiloids while improving the longitudinal reliability when switching tracers. Implementation of this method across multiple studies may lend to more robust and comparable data for future research.

CENTILOID SCALE IN PRACTICE: EFFECT OF DIFFERENT SUVR REFERENCE REGIONS AND COMPARISON OF CENTILOID CUT-OFFS

Centiloid (CL) conversion aligns amyloid PET measures across tracers1. A CL pipeline converts PET measures to their PiB equivalent and linearly transforms them onto a scale anchored by calibrated means of young controls (YC) and Alzheimer's disease (AD) subjects. We evaluated the impact of the choice in reference region and compared positivity cut-offs for FDA-approved amyloid tracers. Calibration datasets were processed via Bioclinica (BC) SUVR method with whole-cerebellum (WC) and subcortical white matter (SWM) reference regions. FreeSurfer MRI segmentation was used to calculate SUVR from co-registered, motion-corrected, averaged, smoothed PET. Correlation coefficient (R2) from linear regression between published SUVR and BC-SUVR was used to evaluate reliability. If R2 was >0.7, BC-method was considered well-correlated and linear regression between BC-SUVR and published CL was used. Standard of truth SUVR (SOT-SUVR)2,3 cut-offs were converted to CLs and compared to BC-SUVR cut-offs defined as 3 SD above YC CL mean. Standard and BC-SUVRs strongly correlated for WC (Florbetapir R2=0.96, Florbetaben R2=0.98, Flutemetamol R2=0.98). SWM yielded weaker correlations that still met the reliability criterion (Florbetapir R2=0.72, Florbetaben R2=0.88, Flutemetamol R2=0.91). Mean WC-CL in YC was 0.4±9.8 for Florbetapir, -2.0±5.4 for Florbetaben and -1.6±7.4 for Flutemetamol versus published CL values of 1.8±10.8, -0.3±6.8 and -1.0±7.2 respectively, yielding variance ratios of 0.91, 0.79 and 1.03. Mean WC-CL in AD was 53.7±51.7 for Florbetapir, 68.1±54.2 for Florbetaben, and 51.2±49.4 for Flutemetamol versus published CL values of 56.2±55.2, 64.8±56 and 51.6±50.3 respectively, yielding variance ratios of 0.94, 0.97 and 0.98. SOT-SUVR cut-off of 1.112 for Florbetapir converted to 25 CL, Florbetaben 1.453 converted to 62 CL and Flutemetamol 1.202 converted to 37 CL. 3 SD above YC CL mean led to BC-SUVR cut-offs of 30 for Florbetapir, 14 for Florbetaben, and 21 for Flutemetamol. Bioclinica's pipeline produced results concordant with the standard Centiloid method. The whole cerebellum was a more suitable reference region for Centiloid conversion. Centiloids align tracers on a linear scale and therefore tracer related differences are still noticeable. Positivity cut-offs differ among tracers even after Centiloid conversion.

Comparison of 18F-florbetaben quantification results using the standard Centiloid, MR-based, and MR-less CapAIBL® approaches: Validation against histopathology

Introduction18F-florbetaben is currently approved for the visual rule out of β-amyloid (Aβ) pathology. It is also used for recruitment and as an outcome measure in therapeutic trials, requiring accurate and reproducible quantification of Aβ burden in the brain. MethodsData from eighty-eight subjects (52 male subjects, aged 79.8 ± 10.6 years) who underwent antemortem 18F-florbetaben positron emission tomography scan and magnetic resonance imaging less than a year before neuropathological assessment at autopsy were evaluated. Image analysis was performed using the standard Centiloid (CL) statistical parametric mapping approach and CapAIBL®. Imaging results were compared against autopsy data. ResultsAgainst combined Bielschowsky silver staining and immunohistochemistry histopathological scores, statistical parametric mapping had 96% sensitivity, 96% specificity, and 95% accuracy, whereas magnetic resonance–less CapAIBL standardized uptake value ratioWhole Cerebellum had 94% sensitivity, 96% specificity, and 95% accuracy. Based on the combined histopathological scores, a CL threshold band of 19 ± 7 CL was determined. DiscussionQuantification of 18F-florbetaben positron emission tomography scans using magnetic resonance–based and magnetic resonance–less CapAIBL® approaches showed high agreement, establishing a pathology-based threshold in CL.

Centiloid scaling for quantification of brain amyloid with [18F]flutemetamol using multiple processing methods

IntroductionA standardised method for quantifying β-amyloid PET tracers would allow comparison across different tracers and different sites. The development of the Centiloid scale has aimed to achieve this, applying a common scale to better aid the diagnosis and prognosis of Alzheimer’s disease (AD) and to monitor anti-amyloid therapeutic interventions. Here, we apply the Centiloid method to [18F]flutemetamol and [11C]PiB (PiB, Pittsburgh compound B) PET images and derive the scaling factor to express their binding in Centiloids.MethodsPaired PiB and [18F]flutemetamol scans for 74 subjects, including 24 young healthy controls (37 ± 5 years), were analysed using the standard Centiloid method. The same subjects were also analysed using PMOD- and FSL-based pipelines as well as SPM8. Test-retest analysis of 10 AD subjects was also performed with each pipeline.ResultsThe standard uptake value ratios (SUVR), determined using the standard SPM8 Centiloid process, showed a strong correlation between [18F]flutemetamol (Flute) and PiB binding (SUVR-Flute = 0.77 × SUVR-PiB + 0.22, R2 = 0.96). Application of the standard Centiloid process allowed the calculation of a direct conversion equation for SUVR-Flute to Centiloid units (CL) (CL = (121.42*SUVR-Flute) − 121.16). Analysis of the data via the two alternate Centiloid pipelines allowed us to derive standardised, SPM8-equivalent equations for both PMOD (CL = (115.24*SUVR-Flute) − 107.86) and FSL (CL = (120.32*SUVR-Flute) − 112.75) respectively. Test-retest analysis of 10 AD subjects showed an approximate 2% difference for each pipeline.Conclusions[18F]flutemetamol data can now be expressed in Centiloid units, enhancing its utility in clinical and research applications for β-amyloid imaging. The standard Centiloid method also demonstrates that [18F]flutemetamol has favourable performance compared with PiB and other β-amyloid tracers. Test-retest difference averaged 2%, with no difference between image processing pipelines. Centiloid scaling is robust and can be implemented on a number of platforms.

Standardization of amyloid quantitation with florbetapir standardized uptake value ratios to the Centiloid scale

IntroductionKlunk et al. recently proposed a means of standardizing quantitation of amyloid burden from positron emission tomography scans to a common Centiloid scale, and we have applied that method to florbetapir. MethodsFlorbetapir and Pittsburgh compound B scans were acquired for 46 mixed clinical presentation subjects within 18 ± 20 days. Florbetapir and Pittsburgh compound B cortical standardized uptake value ratio (SUVr) values were well correlated for both standard Centiloid (R2 = 0.894) and Avid (R2 = 0.901) volume of interests (VOIs). The methods of Klunk et al. were applied to establish a conversion first from florbetapir SUVr values obtained using standard Centiloid VOIs to Centiloids and then from Avid VOIs (Joshi et al.) to Centiloids. ResultsThe equation for conversion of florbetapir SUVr from Avid VOIs to the Centiloid scale was as follows: Florbetapir Centiloids = 183 × SUVrAvid − 177. The threshold that discriminated neuropathologically verified none or sparse versus moderate to frequent plaques in autopsy-confirmed data is 24.1 Centiloids. DiscussionThese findings may allow improved tracer-independent amyloid quantitation.

Considerations for Performing Level-2 Centiloid Transformations for Amyloid PET SUVR values

The Centiloid Project describes a post-hoc data transformation to standardize amyloid PET measurements to enable direct data comparisons across sites and studies using differing acquisition/analysis methods. It uses linear regression that transforms values using different measurement scales to match those from a standard Centiloid unit scale. Our group’s measurement method differs from the Centiloid’s standard method in both acquisition and analysis methods. In this work we examine multiple variations for performing these transformations and compare several approaches. We hypothesized that using Deming regression, which accounts for error on both axes, would produce a more optimal transformation than the recommended standard linear regression. We also examined the effects of performing separate regressions for differences in acquisition and analysis methods, rather than a direct single-regression approach. Our results found that all transformation approaches had very similar performance and were within the recommended tolerance thresholds.

Cryptoexplosion structures caused from without or from within the earth? ("astroblemes" or "geoblemes?")

<h3>ABSTRACT</h3> <h3>Purpose</h3> The Centiloid scale provides a systematic means of harmonising amyloid-β PET measures across different acquisition and processing methodologies. This work explores the Centiloid transformation of [¹⁸F]florbetapir PET data acquired on a combined PET/MR scanner and processed with methods that differ from the standard Centiloid pipeline. <h3>Methods</h3> The Standard PiB and Florbetapir Calibration datasets were processed using a standardised uptake value ratio (SUVR) pipeline with MRI parcellations from the Geodesic Information Flow (GIF) algorithm in native PET space. We generated SUVRs using whole cerebellum (GIF_WC_SUVR) and eroded white matter (GIF_WM_SUVR) reference regions, with and without partial volume correction (PVC). Linear regression was used to calibrate these processing pipelines to the standard Centiloid approach. We then applied the resulting transformation to 432 florbetapir scans from the Insight 46 study of mostly cognitively normal individuals aged ∼70 years, and defined Centiloid cutpoints for amyloid-β positivity using Gaussian-mixture modelling. <h3>Results</h3> GIF-based SUVR processing pipelines were suitable for conversion according to Centiloid criteria. For GIF_WC_SUVR, cutpoints translated to 14.2 Centiloids, or 11.8 with PVC. There was a differential relationship between florbetapir uptake in WM and WC regions in Florbetapir Calibration and Insight 46 datasets, causing implausibly low Centiloid values for GIF_WM_SUVR. Linear adjustment to account for this difference resulted in Centiloid cutpoints of 18.1 for GIF_WM_SUVR (17.0 with PVC). <h3>Conclusion</h3> Our results show florbetapir SUVRs acquired on PET/MR scanners can be reliably converted to Centiloids. Acquisition or biological factors can have large effects on Centiloid values from different datasets, we propose a correction to account for these effects.