Alzheimer's disease neuroimaging initiative: a one-year follow up study using tensor-based morphometry correlating degenerative rates, biomarkers and cognition

Abstract

Tensor-based morphometry can recover three-dimensional longitudinal brain changes over time by nonlinearly registering baseline to follow-up MRI scans of the same subject. Here, we compared the anatomical distribution of longitudinal brain structural changes, over 12 months, using a subset of the ADNI dataset consisting of 20 patients with Alzheimer's disease (AD), 40 healthy elderly controls, and 40 individuals with mild cognitive impairment (MCI). Each individual longitudinal change map (Jacobian map) was created using an unbiased registration technique, and spatially normalized to a geometrically-centered average image based on healthy controls. Voxelwise statistical analyses revealed regional differences in atrophy rates, and these differences were correlated with clinical measures and biomarkers. Consistent with prior studies, we detected widespread cerebral atrophy in AD, and a more restricted atrophic pattern in MCI. In MCI, temporal lobe atrophy rates were correlated with changes in mini-mental state exam (MMSE) scores, clinical dementia rating (CDR), and logical/verbal learning memory scores. In AD, temporal atrophy rates were correlated with several biomarker indices, including a higher CSF level of p-tau protein, and a greater CSF tau/beta amyloid 1-42 (ABeta42) ratio. Temporal lobe atrophy was significantly faster in MCI subjects who converted to AD than in non-converters. Serial MRI scans can therefore be analyzed with nonlinear image registration to relate ongoing neurodegeneration to a variety of pathological biomarkers, cognitive changes, and conversion from MCI to AD, tracking disease progression in 3-dimensional detail.

Fig. 1

Unbiased registration was performed on 100 pairs of serial MR images, acquired 12 months apart, from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset. The selected sample consisted of 20 patients with Alzheimer’s disease (AD), 40 individuals with mild cognitive impairment (MCI), and 40 healthy elderly controls (CTL). The mean of the resulting Jacobian maps in each group is superimposed on a brain volume.

Fig. 2

Voxel-wise Z-statistics (top row) comparing mean Jacobian values for AD (N =20) versus Controls (N=40) on the top, and MCI (N=40) versus Controls (N=40) on the bottom. Corresponding color-coded P maps also show the local significance of these differences (bottom row). There is widespread progressive atrophy in AD, at a faster mean rate than in normals — this difference in rates reaches the voxelwise significance level of 0.05 in most regions of the brain, and remains significant after corrected for multiple comparisons in ROIs including the temporal lobes, parietal lobes, occipital lobes, and frontal lobes. By contrast, for MCI versus Controls, only the parietal and temporal lobes reach ROI significance. Please refer to the Results section for more detailed discussions.

Fig. 3

Percent brain tissue loss from baseline to follow-up as determined by the average Jacobian value within each lobe (with CSF excluded). The diagnosis for the AD, MCI, and CTL groups was determined at baseline. Here, MCI at 12Mo denotes subjects diagnosed with MCI at baseline who did not convert to AD at 12-month follow-up, whereas MCI to AD signifies those who had converted to AD at 12-month follow-up. One MCI subject’s diagnosis converted back to control at follow-up, and thus was excluded from the MCI subgroups. The first bar in each group (colored white), shows mild but significant progressive atrophy in controls. The 2nd bar (turquoise) denotes MCI subjects at baseline, and is followed by bars denoting converters and non-converters. Mean rates are typically higher in the converters, comparable to subjects diagnosed as AD at both time-points, or AD at the last time point (last two bars). An * indicates p<0.01 for the comparison. Here, p values for MCI to AD group are not given, as there were only 7 subjects (significance levels were as follows: p<0.01 for all lobes in AD and AD at 12Mo; p=0.012, 0.054, and 0.02 in MCI for the temporal, frontal and occipital lobes; p=0.02 and 0.055 in controls for the occipital and temporal lobes; p=0.068 and 0.045 in MCI at 12Mo for occipital and temporal lobes; regions not reported here do not reach significance at the 0.05 level).

Fig. 4

(a) CDF plots for voxel-wise correlation of progressive temporal lobe tissue loss in MCI, AD, and pooled groups (ALL, N=100) with (a) various biomarker indices including ABeta42 (AB142), tau protein (TAU), phosphorylated-tau 181 (PTAU), tau/ABeta42 ratio (TAUAB), and PTAU/AB42 ratio (PTAUAB), and (b) various clinical measures corresponding to those in Tables 1 and 2. Here, biomarkers correlate better in AD and the pooled group (but not in MCI), while clinical measures manifest better correlations in the MCI group. For more sensitive biomarkers or clinical measures, the departure of the early part of the corresponding CDF curve (i.e., the upswing) will be larger. The lack of significant correlations between biomarkers and ongoing temporal lobe atrophy in the MCI group is most likely due to the heterogeneous nature of MCI. By contrast, the better correlations with clinical measures in the MCI group support the use of serial neuropsychiatric testing in monitoring disease progression. Please refer to text for more detailed discussions.