ALZHEIMERS RESEARCH PROJECT: The mean apparent propagator (MAP) MRI indices as neuroimaging biomarkers of cerebral aging, mild cognitive impairment, Alzheimers disease, and associated risk factors

National Institute On Aging

Investigators

Abstract

One of the current critical goals of biomedical research, in general, and the NIA’s mission, in particular, is to establish sensitive and specific biomarkers as indicators of Alzheimer’s disease (AD), especially during the prodromal phase of the disease, for accurate and early diagnosis, as well as for the design of targeted and rational interventions. These biomarkers are expected to represent quantifiable metrics of biological processes related to AD that are directly linked to clinical outcomes and can, therefore, be used as surrogates for the disease process. AD has historically been regarded as a gray matter (GM) disease, with cortical atrophy and ventricular enlargement on the macroscopic scale, and development of amyloid-β plaques and neurofibrillary tau tangles on the microscopic scale. However, this conventional paradigm has recently been challenged for several reasons. First, mounting evidence revealed a weak correlation between the amyloid load and cognitive impairment. Second, more than third of cognitively normal individuals exhibit abnormal levels of tau and amyloid-β proteins’ aggregations. Third, and perhaps the most puzzling and striking reason, is the continuous failure of various anti-amyloid interventions to reverse, or at least to stop, the progression of the cognitive decline in subjects with AD. Therefore, researchers are now questioning this classical, long-standing, amyloid-β/tau paradigm, while wondering whether the amyloid-β and tau aggregations are simply byproducts of other primary drivers of neurodegeneration, rather than the main causative suspects. Based on these cumulative observations, in the recent years, it has been suggested that alterations in GM and white matter (WM), including demyelination, axonal loss and synaptic degeneration, may precede amyloid-β and tau depositions, while demonstrating much stronger relationships to dementias’ severity and cognitive deficits in AD. Interestingly, in addition to tangles and plaques description, these microstructural changes, especially WM degeneration in subjects’ brain with AD, have been described by Dr. Alois Alzheimer’s himself over a century ago in several instances of his seminal papers. Curiously, these aspects of cerebral microstructural changes have gained little, to no attention, until recently. Recent advances in quantitative magnetic resonance imaging (MRI) methodology have enabled probing tissue microstructure with exquisite specificity and sensitivity in human and system models. Using this technology to examine changes in cerebral tissue microstructure in aging and neuropathology, including mild cognitive impairments (MCI) and dementias, has the potential to provide a window into the underlying age-related diseases’ biology and mechanisms, and to nominate new MR biomarkers for longitudinal assessment and targets for intervention. Indeed, the last three decades have seen a myriad of advanced methods to characterize cerebral microstructure with unprecedented sensitivity, particularly using diffusion tensor imaging (DTI). This extensive work has revealed complex and nonlinear age-related trajectories of cerebral tissues during brain maturation and degeneration. DTI describes the distribution of diffusion displacements using a simplistic Gaussian model. However, water molecular diffusion follows a non-Gaussian distribution in the complex biological tissues due to the restriction of cell membranes, organelles, and liquid compartments. Therefore, despite the promising sensitivity of DTI, it has inherent limitations that hamper utility and specific interpretation. To improve specificity, multicomponent diffusion approaches have been introduced. However, while compelling, biophysical tissue models, such as Neurite Orientation Distribution and Density Imaging (NODDI), are constructed based on strict assumptions about the underlying tissue compartments, and incorporate fixed parameters to avoid fitting degeneracy. Therefore, despite their potential greater specificity, derived parameters are dramatically dependent on these underlying assumptions, which could impede their sensitivity to early changes due to pathology. For the sake of compromise between the highly sensitive, but simplistic, DTI modeling, and the theoretically more specific, but challenging, multicomponent diffusion modeling such as NODDI, the mean apparent propagator (MAP) MRI technique has recently been introduced. The MAP model avoids any prior assumptions about the behavior of the water molecules diffusion in the tissues, and offers a mathematical framework to assess the dispersion distribution of water molecules through a probabilistic approach that allows a comprehensive measurement of water molecules’ displacements in complex tissues. Aside from the conventional DTI indices, MAP additionally provides a wide range of complementary indices including the return-to-origin probability (RTOP), a measure of the likelihood of water molecules undergoing zero net displacements due to restricting barriers, as well as the return-to-axis probability (RTAP) and the return-to-plane probability (RTPP), measures of the presence of restrictive barriers in the radial and axial orientations, respectively. In addition, the non-Gaussianity index (NG) metric reflects the extent of deviation from a homogeneous/Gaussian diffusivity, and the propagator anisotropy (PA) provides an indicator of the anisotropy and dispersion of the fibers. Consequently, MAP comprehensively and accurately captures the microstructural complexity of cerebral tissues with various underlying structural and architectural characteristics. However, unlike DTI, MAP requires specialized MRI protocols with data acquired at different degrees of water attenuations, which is referred to as multishell diffusion imaging in the MRI jargon. The Alzheimer’s Disease Neuroimaging Initiative (ADNI), an ongoing large-scale longitudinal study designed to improve methods for clinical trials and identify various biomarkers of AD and aging, offers the possibility to undertake our study. Indeed, the ADNI’s third phase (ADNI3) incorporates a multishell MRI diffusion protocol with over 400 datasets already acquired from over 150 participants that include cognitively normal, MCI and AD subjects. For these participants, a variety of vascular, genetic, metabolic, and functional measures were also collected.