2017Alzheimer s & DementiaRequires access

[IC‐01–01]: AV1451‐PET CORTICAL UPTAKE AND REGIONAL DISTRIBUTION PREDICTS LONGITUDINAL ATROPHY IN ALZHEIMER's DISEASE

Renaud La Joie, Adrienne Visani, Viktoriya Bourakova, Suzanne L. Baker, Nagehan Ayakta, Julie Pham, Ashley Mensing, Howard J. Rosen, Bruce L. Miller, William J. Jagust, Gil D. Rabinovici

Open publisher page 5 citations

Abstract

We aimed to assess whether β-amyloid (PIB) and tau (AV1451) PET predict longitudinal atrophy in patients with AD. A group of 10 patients fulfilling NIA-AA criteria for AD dementia likely due to AD were included (age = 63 ± 9, MMSE = 24 ± 4 at baseline). All patients underwent i) a baseline visit with structural MRI and PET imaging with both AV1451 and PIB, ii) a follow-up structural MRI (time between MRIs = 1.06 ± 0.15 years). Structural images were preprocessed using SPM12's longitudinal registration pipeline to obtain voxelwise maps of atrophy showing areas of local contractions and expansions (i.e. Jacobians, see figure 1A). Relationships between baseline PET data (Standardized Uptake Value Ratio (SUVR) images normalized to cerebellar gray matter) and subsequent atrophy were assessed in two complementary ways. First, a global cortical value was extracted for each patient and each modality (figure 1B) and correlations were computed at the group level. Second, we computed voxelwise correlations at the individual patient level to evaluate the similarity between baseline maps of PET uptake and the subsequent atrophy map (figure 1C). Using global cortical measures (figure 2A), a significant association was observed between steeper cortical atrophy (lower Jacobians) and higher baseline AV1451 cortical SUVR (rho = -0.76, p = 0.02) but not PIB-SUVR (rho = 0.13, p = 0.73), see Fig 2a. Similarly, voxelwise correlation analyses revealed that maps of atrophy resembled baseline AV1451-PET images (median rho = -0.57), i.e. voxels of higher baseline AV1451-uptake showed a steeper rate of atrophy. Correlation between PIB and atrophy maps were minimal (median rho = -0.14) and significantly different from the correlations with AV1451 (Wilcoxon signed-rank test: p=0.004, Figure 2B). These results support the hypothesis that neurodegeneration is more closely related to tau than to β-amyloid pathology, and further suggests that tau pathology precedes and drives neurodegeneration locally. Our results further suggest a potential prognostic role for AV1451 in predicting individual patient longitudinal trajectories. From a clinical perspective, our results suggest that AV1451-PET could have a major clinical utility to predict short-term outcomes in patients. Study design, image preprocessing, and analyses (All images & values are from a representative patient with close to median values). Results of the correlation analyses.

About this research paper

What this paper is about

We aimed to assess whether β-amyloid (PIB) and tau (AV1451) PET predict longitudinal atrophy in patients with AD. A group of 10 patients fulfilling NIA-AA criteria for AD dementia likely due to AD were included (age = 63 ± 9, MMSE = 24 ± 4 at baseline). All patients underwent i) a baseline visit with structural MRI and PET imaging with both AV1451 and PIB, ii) a follow-up structural MRI (time between MRIs = 1.06 ± 0.15 years). Structural images were preprocessed using SPM12's longitudinal registration pipeline to obtain voxelwise maps of atrophy showing areas of local contractions and expansions (i.e. Jacobians, see figure 1A). Relationships between baseline PET data (Standardized Uptake Value Ratio (SUVR) images normalized to cerebellar gray matter) and subsequent atrophy were assessed in two complementary ways. First, a global cortical value was extracted for each patient and each modality (figure 1B) and correlations were computed at the group level. Second, we computed voxelwise correlations at the individual patient level to evaluate the similarity between baseline maps of PET uptake and the subsequent atrophy map (figure 1C). Using global cortical measures (figure 2A), a significant association was observed between steeper cortical atrophy (lower Jacobians) and higher baseline AV1451 cortical SUVR (rho = -0.76, p = 0.02) but not PIB-SUVR (rho = 0.13, p = 0.73), see Fig 2a. Similarly, voxelwise correlation analyses revealed that maps of atrophy resembled baseline AV1451-PET images (median rho = -0.57), i.e. voxels of higher baseline AV1451-uptake showed a steeper rate of atrophy. Correlation between PIB and atrophy maps were minimal (median rho = -0.14) and significantly different from the correlations with AV1451 (Wilcoxon signed-rank test: p=0.004, Figure 2B). These results support the hypothesis that neurodegeneration is more closely related to tau than to β-amyloid pathology, and further suggests that tau pathology precedes and drives neurodegeneration locally. Our results further suggest a potential prognostic role for AV1451 in predicting individual patient longitudinal trajectories. From a clinical perspective, our results suggest that AV1451-PET could have a major clinical utility to predict short-term outcomes in patients. Study design, image preprocessing, and analyses (All images & values are from a representative patient with close to median values). Results of the correlation analyses.

Why it matters

OpenAlex reports 5 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

We aimed to assess whether β-amyloid (PIB) and tau (AV1451) PET predict longitudinal atrophy in patients with AD. A group of 10 patients fulfilling NIA-AA criteria for AD dementia likely due to AD were included (age = 63 ± 9, MMSE = 24 ± 4 at baseline). All patients underwent i) a baseline visit with structural MRI and PET imaging with both AV1451 and PIB, ii) a follow-up structural MRI (time between MRIs = 1.06 ± 0.15 years). Structural images were preprocessed using SPM12's longitudinal registration pipeline to obtain voxelwise maps of atrophy showing areas of local contractions and expansions (i.e. Jacobians, see figure 1A). Relationships between baseline PET data (Standardized Uptake Value Ratio (SUVR) images normalized to cerebellar gray matter) and subsequent atrophy were assessed in two complementary ways. First, a global cortical value was extracted for each patient and each modality (figure 1B) and correlations were computed at the group level. Second, we computed voxelwise correlations at the individual patient level to evaluate the similarity between baseline maps of PET uptake and the subsequent atrophy map (figure 1C). Using global cortical measures (figure 2A), a significant association was observed between steeper cortical atrophy (lower Jacobians) and higher baseline AV1451 cortical SUVR (rho = -0.76, p = 0.02) but not PIB-SUVR (rho = 0.13, p = 0.73), see Fig 2a. Similarly, voxelwise correlation analyses revealed that maps of atrophy resembled baseline AV1451-PET images (median rho = -0.57), i.e. voxels of higher baseline AV1451-uptake showed a steeper rate of atrophy. Correlation between PIB and atrophy maps were minimal (median rho = -0.14) and significantly different from the correlations with AV1451 (Wilcoxon signed-rank test: p=0.004, Figure 2B). These results support the hypothesis that neurodegeneration is more closely related to tau than to β-amyloid pathology, and further suggests that tau pathology precedes and drives neurodegeneration locally. Our results further suggest a potential prognostic role for AV1451 in predicting individual patient longitudinal trajectories. From a clinical perspective, our results suggest that AV1451-PET could have a major clinical utility to predict short-term outcomes in patients. Study design, image preprocessing, and analyses (All images & values are from a representative patient with close to median values). Results of the correlation analyses.

Key concepts: Atrophy, Pittsburgh compound B, Standardized uptake value, Voxel, Nuclear medicine, Medicine, Dementia, Psychology

Related papers

Back to paper searchBrowse research topicsOriginal source
[IC‐01–01]: AV1451‐PET CORTICAL UPTAKE AND REGIONAL DISTRIBUTION PREDICTS LONGITUDINAL ATROPHY IN ALZHEIMER's DISEASE — Research Paper | ScholarLens