Dynamics of Gray Matter Loss in Alzheimer's Disease, Mapped with a Population Based Brain Atlas

Paul M. Thompson1, Kiralee M. Hayashi1, Greig de Zubicaray2, Andrew L. Janke2, Stephen E. Rose2, James Semple3, Michael S. Hong1, David H. Herman1, David Gravano1, Stephanie Dittmer1, David M. Doddrell2, Arthur W. Toga1
1Laboratory of Neuro Imaging, Brain Mapping Division, Department of Neurology, UCLA School of Medicine
2Centre for Magnetic Resonance, University of Queensland, Brisbane 4072, Australia
3GlaxoSmithKline Pharmaceuticals plc, Cambridge, UK

Proceedings of the 19th Colloque Medecine et Recherche,
"The Living Brain and Alzheimer's Disease", [eds. Hyman, B., Demonet, J.F., and Christen, Y.],
Fondation IPSEN, Paris, France, March 17, 2003.
[Full Program, HTML]

Due to recent innovations in brain mapping, the dynamic spread of Alzheimer’s disease can now be charted in the living brain (see, e.g., [1,2,3]). Effects of drug treatment, risk genes, and brain deficits can be tracked as they emerge, in patients and those at risk. These maps can be warehoused in statistical brain atlases, which encode dynamic [4,5] and genetic [6] information on brain change in entire populations, across the human lifespan. We show how these tools help to explore and map the disease process, revealing factors that affect it. Novel mathematics are described for visualizing therapeutic and gene effects.

Dynamic Spread of AD. As an illustrative example, we report the mapping of a dynamically spreading wave of gray matter loss in the brains of Alzheimer’s patients, scanned repeatedly with MRI. The loss pattern is visualized, in 3D video data, as it spreads from temporal cortices into frontal and cingulate brain regions as the disease progresses. Deficit patterns are resolved with a novel 4D cortical pattern matching strategy, which resolves the dynamic path of the disease as it spreads in the human cortex over a period of several years.

Methods. A population-based brain atlas containing 6840 anatomical surface models was created from 3D MRI scans of 43 AD patients and 34 controls, scanned longitudinally, with 3 month repeat scans for 2-4 years. Gyral pattern and shape variations were encoded using high-dimensional elastic deformation mappings [7] driving each subject’s cortical anatomy into a group average configuration. Dynamic maps of atrophic rates, with millions of degrees of freedom, were generated by computing continuum-mechanical warping fields to match anatomy across time ([7], cf. [2]). Annualized 4D maps of tissue loss rates within each subject were elastically realigned for averaging across diagnostic groups, and covaried for age, sex, hemisphere, and MMSE score. Maps of brain change, and factors that influence it, were visualized using color-coded maps, in 3D video format. We assessed the significance of these effects using covariant partial differential equations, random effects models, and anisotropic random field theory.

Results. A dynamically spreading wave of cortical gray matter loss (up to 4-5%/year locally) was detected in the AD patients, beginning in temporal cortices. Its rate was tightly linked with cognitive decline, as measured by MMSE loss rate. Primary sensorimotor cortices were comparatively spared as the loss pattern shifted into frontal cortices, and intensified in the cingulate and paralimbic belts; transient divisions occurred in the patterns of loss. A second technique was developed, based on skeletonizing 3D anatomical surface models, to map rates of hippocampal atrophy and ventricular expansion. This statistical mapping technique was also found to be much more sensitive to medial temporal lobe change than conventional volume measures [8].

Conclusion. These video maps chart the dynamic progress of Alzheimer’s disease. They reveal a changing pattern of deficits. We are now using them urgently to detect where deficit patterns are modified by drug treatment and known risk genotypes.

References: [1] Reiman ER et al., PNAS 98(6):3334-9(2001). [2]. Fox NC et al., Lancet 358(9277):201-5(2001). [3] Thompson PM et al. J. Neuroscience, 23(3):(2003). [4] Thompson PM et al., Nature 404:190-193(2000). [5] Thompson PM et al., Nature Neuroscience 4(12):1253-8(2001). [6] Thompson PM et al., PNAS 98(20):11650-11655(2001). [7] Thompson PM et al., HBM 9(2):81-92(2000); [8]. Thompson PM et al., HBM(2003). Available at: http://www.loni.ucla.edu/~thompson/thompson_pubs.html