Distinct clinical and metabolic deficits in PCA and AD are not related to amyloid distribution

Michael Rosenbloom, Adi Alkalay, Neha Agarwal, Suzanne L. Baker, James P. O’Neil, Mustafa Janabi, Irene V. Yen, Matthew E. Growdon, Jae‐Won Jang, Cindee Madison, Elizabeth C. Mormino, Howard J. Rosen, Maria Luisa Gorno‐Tempini, Michael W. Weiner, Bruce L. Miller, William J. Jagust, G.d. RabinoviciNeurology 2011 · DOI

TL;DR

This study investigates the relationship between amyloid distribution and metabolic deficits in Posterior Cortical Atrophy (PCA) and Alzheimer's Disease (AD). It reveals that while both conditions show similar amyloid deposition patterns, PCA exhibits a more pronounced metabolic deficit in the occipital cortex, suggesting that the clinical presentation of these neurodegenerative diseases is not solely driven by amyloid burden.

What Problem Does This Paper Address?

Posterior Cortical Atrophy (PCA) is a neurodegenerative syndrome characterized by progressive visual processing and spatial praxis deficits, often with relative sparing of memory. Autopsy studies frequently reveal underlying Alzheimer's Disease (AD) pathology in PCA patients. However, the precise relationship between the distribution of amyloid pathology (a hallmark of AD) and the specific clinical and anatomical deficits seen in PCA, compared to typical AD, remains unclear. Some studies suggest differences in amyloid plaque distribution, while others find similarities. This paper aims to clarify this by using in vivo imaging techniques to compare amyloid burden and glucose metabolism in PCA and AD patients.

Methods

The researchers employed two key Positron Emission Tomography (PET) imaging techniques:

  • Pittsburgh Compound B (PiB-PET): This tracer binds to fibrillar amyloid-beta plaques in the brain, allowing researchers to visualize and quantify amyloid deposition.
  • Fluorodeoxyglucose (FDG-PET): This tracer measures glucose metabolism, a proxy for neuronal activity. Reduced FDG uptake indicates hypometabolism, suggesting impaired neuronal function.

They compared these measures in three groups:

  1. Patients with PCA (n=12)
  2. Patients with clinically diagnosed AD (n=14)
  3. Cognitively normal older adults (NC, n=30)

Voxel-wise comparisons were performed on the PET data to identify regional differences in PiB binding and FDG uptake. To account for potential differences in brain size and shape, the PET data was spatially normalized to a standard brain atlas (Montreal Neurological Institute space) and further corrected for atrophy using a partial volume correction method on a subset of participants with MRI data. Statistical analyses included ANCOVA models to compare groups, with education and sex as covariates. Regions of interest (ROIs) were also defined to quantify uptake in specific brain areas.

Datasets

The study utilized PET imaging data from three groups of participants:

  • Posterior Cortical Atrophy (PCA) cohort: 12 patients.
  • Alzheimer's Disease (AD) cohort: 14 patients.
  • Cognitively Normal (NC) cohort: 30 individuals.

These participants were recruited from cohorts followed at the University of California San Francisco Memory and Aging Center. The study did not introduce a new benchmark dataset.

Key Insights and Ideas

  • Amyloid is Diffuse in Both PCA and AD: Contrary to some earlier post-mortem findings, in vivo PiB-PET imaging revealed that fibrillar amyloid deposition in PCA is diffuse and widespread across the cortex, similar to what is observed in typical AD. This suggests that the presence and distribution of amyloid plaques alone may not fully explain the distinct clinical presentations of PCA and AD.
  • Metabolic Deficits Differ Significantly: While both PCA and AD patients showed hypometabolism in temporoparietal regions compared to controls, PCA patients exhibited a more posterior pattern of hypometabolism, extending into the occipital cortex. This occipital hypometabolism was more pronounced in PCA than in AD.
  • Clinical Phenotype and Metabolic Pattern are Linked: The observed posterior metabolic deficit in PCA aligns with their characteristic visual processing impairments, suggesting a strong link between regional brain function and clinical symptoms.
  • Network Vulnerability Hypothesis: The findings support the idea that distinct clinical syndromes in AD may arise from the differential vulnerability of specific neural networks, rather than solely from the distribution of amyloid pathology. This suggests that factors beyond amyloid accumulation, such as intrinsic network connectivity or metabolic demand, play a crucial role in determining the pattern of neurodegeneration.

Key Results

  • PiB Distribution: No significant regional differences in PiB binding were found between PCA and AD groups, even after correcting for atrophy. Both groups showed diffuse PiB uptake throughout the frontal, temporoparietal, and occipital cortex compared to normal controls.
  • FDG Hypometabolism:
    • Compared to NC, both AD and PCA showed hypometabolism in the temporoparietal cortex and precuneus/posterior cingulate.
  • PCA patients showed significantly greater hypometabolism in the inferior occipitotemporal cortex compared to both NC and AD (p < 0.05).
  • There were no regions where AD patients showed relative hypometabolism compared to PCA.
  • The occipital percentage of FDG uptake was lower in PCA than in AD and NC (p < 0.05).

Limitations and Future Work

  • Sample Size: The authors acknowledge that the sample sizes were relatively small, which may limit the power to detect subtle differences between AD and PCA.
  • Early-Onset AD Cohort: The AD control group was largely composed of early age-at-onset patients, which might have minimized clinical and anatomical differences between AD and PCA.
  • Soluble Amyloid: The study focused on fibrillar amyloid (detected by PiB). Soluble amyloid oligomers, which are thought to be more neurotoxic, are not bound by PiB, and their distribution was not directly assessed.
  • NFT Distribution: While fibrillar amyloid distribution was similar, the authors note that neurofibrillary tangle (NFT) distribution, which is more closely related to neuronal death, might differ and could be a key driver of the distinct phenotypes.
  • Future Directions: The authors suggest that future studies should investigate premorbid "reserve" in specific cognitive domains, assess the integrity of associated neural networks, and explore the mechanisms of selective network degeneration in focal variants of AD. Further research is needed to determine if the observed metabolic differences are related to the distribution of neurofibrillary tangles or other pathological processes.

Summary updated

Abstract

Fibrillar amyloid deposition in PCA is diffuse and similar to AD, while glucose hypometabolism extends more posteriorly into occipital cortex. Further studies are needed to determine the mechanisms of selective network degeneration in focal variants of AD.
BibTeX
@inproceedings{rosenbloom2011distinct,
  author = {Michael Rosenbloom and Adi Alkalay and Neha Agarwal and Suzanne L. Baker and James P. O’Neil and Mustafa Janabi and Irene V. Yen and Matthew E. Growdon and Jae‐Won Jang and Cindee Madison and Elizabeth C. Mormino and Howard J. Rosen and Maria Luisa Gorno‐Tempini and Michael W. Weiner and Bruce L. Miller and William J. Jagust and G.D. Rabinovici},
  title = {{Distinct clinical and metabolic deficits in PCA and AD are not related to amyloid distribution}},
  booktitle = {Neurology},
  year = {2011},
  doi = {10.1212/wnl.0b013e31821cccad},
}
Figure