Early axonal degeneration linked to clinical decline in Alzheimer’s disease progression revealed with diffusion MRI
Zhaoyuan Gong, John P. Laporte, Alexander Y. Guo, Murat Bilgel, Jonghyun Bae, Noam Y. Fox, Angelique de Rouen, Nathan Zhang, Aaliya Taranath, Rafael de Cabo, Josephine M. Egan, Luigi Ferrucci, Mustapha Bouhrara, Alzheimer’s Disease Neuroimaging Initiative
Zhaoyuan Gong, John P. Laporte, Alexander Y. Guo, Murat Bilgel, Jonghyun Bae, Noam Y. Fox, Angelique de Rouen, Nathan Zhang, Aaliya Taranath, Rafael de Cabo, Josephine M. Egan, Luigi Ferrucci, Mustapha Bouhrara, Alzheimer’s Disease Neuroimaging Initiative
View: Text | PDF
Clinical Research and Public Health Clinical Research Neuroscience Public Health

Early axonal degeneration linked to clinical decline in Alzheimer’s disease progression revealed with diffusion MRI

Abstract

BACKGROUND Axonal degeneration is believed to be an early hallmark of Alzheimer’s disease (AD). This study investigated the temporal trajectory of axonal loss and its association with cognitive and functional decline using a dMRI-derived axonal density index (ADI).METHODS Longitudinal dMRI, CSF, and PET data from the ADNI study were analyzed, including 117 subjects that were cognitively normal (CN) and 88 that were cognitively impaired (CI), consisting of 74 individuals with mild cognitive impairment (MCI) and 14 with AD. Linear mixed-effects models examined group differences and associations between baseline and longitudinal changes in ADI, CSF, or PET biomarkers and clinical outcomes. Results derived from larger CSF (n = 527) and PET (tau-PET: n = 870; amyloid-PET: n = 1,581) datasets are also presented.RESULTS Compared with the CN group, the CI group exhibited significantly lower baseline ADI values and steeper longitudinal decline (P < 10–6). Lower baseline ADI predicted faster cognitive and functional decline in the CI group (MMSE: P = 0.03; CDR-SB: P < 10–4), and longitudinal decreases in ADI were associated with worsening clinical outcomes (MMSE: P = 0.001; CDR-SB: P < 10–12). Compared with CSF and PET biomarkers, ADI demonstrated superior sensitivity in tracking disease progression and matched these biomarkers in predicting future cognitive and functional decline. Furthermore, decreases in ADI were significantly associated with declines in clinical outcomes; this association was observed only with amyloid-PET, but not CSF, biomarkers.CONCLUSION Axonal degeneration is an early and clinically meaningful feature of AD. ADI is a promising noninvasive biomarker for early detection, prognosis, and disease monitoring.TRIAL REGISTRATION ClinicalTrials.gov NCT00106899.FUNDING This work was supported by the National Institute on Aging Intramural Research Program.

Authors

Zhaoyuan Gong, John P. Laporte, Alexander Y. Guo, Murat Bilgel, Jonghyun Bae, Noam Y. Fox, Angelique de Rouen, Nathan Zhang, Aaliya Taranath, Rafael de Cabo, Josephine M. Egan, Luigi Ferrucci, Mustapha Bouhrara, Alzheimer’s Disease Neuroimaging Initiative

×

Figure 3

Baseline MRI measurements of axonal density/integrity, as measured using the ADI, predict prospective changes in cognition and function, as measured using the MMSE and CDR-SB scores.

Options: View larger image (or click on image) Download as PowerPoint
Baseline MRI measurements of axonal density/integrity, as measured using...
Analyses were conducted using the following linear mixed-effects models: MMSE/CDR-SBij ~ β0 + βage × agei + βsex × sexi + βtime × timeij + βdiagnosis × diagnosisi + βADI × ADIi + βtime×ADI × timeij × ADIi + βtime×diagnosis × timeij × diagnosisi + βADI×diagnosis × ADIi × diagnosisi + βtime×diagnosis×ADI × timeij × diagnosisi × ADIi + bi + εij. Low, middle, and high values are the 25th, 50th, and 75th quantiles of the ADI z score. (A and B) The longitudinal distributions of MMSE and CDR-SB anchored at the first dMRI scan of each subject, with CN and CI subjects color-coded. Both the CN and CI groups have similar longitudinal distributions, with many of the participants having over 4 years of cognitive measurement follow-up from baseline. (C) Predicted longitudinal changes in MMSE for CN and CI groups based on ADINODDI, ADIC-NODDI, or ADISMI values. (D) Predicted longitudinal changes in CDR-SB for CN and CI groups based on ADINODDI, ADIC-NODDI, or ADISMI values. While the CN group exhibits relatively stable MMSE and CDR-SB, the CI group shows significant decline in MMSE and an increase in CDR-SB scores. Among ADI biomarkers, ADIC-NODDI uniquely predicts differential trajectories within CI subjects, with higher baseline ADIC-NODDI values predicting slower decline in cognition and function (middle row, green prediction lines). These findings highlight the importance of maintaining axonal integrity and the sensitivity of ADIC-NODDI as an imaging biomarker for axonal integrity determination and prediction of cognitive and functional decline. Full results are shown in Supplemental Table 2.