[IC‐P‐188]: ON EVALUATION OF TAU ACCUMULATIONS IN LONGITUDINAL STUDIES OF ALZHEIMER's DISEASE (AD): IMPLICATIONS FROM A PET STUDY WITH [18F]RO6958948
Year: 2017
Type: article
Abstract: For an evaluation of anti-tau treatments of AD and other types of dementia, it is critical to identify regions of statistically-justifiable increases and/or decreases of tau accumulations in individual scans. Thus, we proposed and evaluated test-retest scan-based approaches for this purpose. Second, we examined a hypothesis that the pons (Po) which showed lower distribution volume than the cerebellum gray matter (Cb) in a cross-sectional study involving 11 clinically diagnosed AD subjects could show less changes of binding over time, thus is more appropriate for the denominator region in longitudinal studies than Cb. Five clinically diagnosed AD subjects had test and retest scans within a month apart with [18F]RO6958948. Separately, 4 AD subjects had baseline and follow-up scans from 8 to 18 months apart. Cb- and Po-reference SUV values were obtained in 80 brain regions using 60–90 min frames. First, 95% confidence intervals (CIs) of scatterplots of test-retest data, and the mean and standard deviation (SD) of test-retest variability (TRV) across all regions were obtained. Then, regions outside the 95% Cis or y = (1±(mean+2.5·SD of TRV))·x (‘By TRV’) in scatterplots of follow-up vs. baseline scans of individual AD subjects were identified to have increased (above) or decreased (below) binging. Data of 50–70 min frames were used for Subject A2 who had a 70 min follow-up scan. Cb-Po SUV ratios increased in the follow-up scans in 3 subjects. SUVR-Po had 42% larger areas between the 95% CIs than SUVR-Cb, and had greater mean TRV values (6.4±4.8%) than SUVR-Cb (7.8±5.1%) (p<0.001; t=-4.5; paired t-test). The two approaches identified similar numbers of increased regions (Table). SUVR-Po alone identified negligible increases in subjects with little changes in MMSE (A1 and A2), and substantial numbers of increased regions in subjects with sizable decreases of MMSE (A3 and A4). Proposed 95% CI and TRV-based approaches could be useful for identifying regions of increased tau accumulations in longitudinal studies. This study indicated that SUVR-Po could be advantageous over SUVR-Cb for evaluation of longitudinal studies using tau-tracers.
Source: Alzheimer s & Dementia
Language: English
Subfield: Psychiatry and Mental health
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W2765805155