Introduction

• Frontotemporal dementia (FTD) presents with heterogeneous, progressive

deficits in behavior, language, and cognition.1

• FTD is associated with structural changes in the brain,2 including grey matter

(GM) volume loss on MRI.

• However, GM volume loss lacks specificity to detect subtle microstructural

alterations observed in early-stage FTD, such as demyelination, inflammation,

and iron accumulation. The ratio of T1-weighted and T2-weighted images on

MRI (T1w/T2w) has been proposed as an in vivo proxy for overall

microstructural integrity.3

 

Hypothesis

We hypothesized that FTD patients would exhibit altered T1w/T2w in

subcortical and cortical GM compared to healthy controls.

 

Methods

• All participants consented to be a part of the Canadian Consortium on

Neurodegeneration in Aging (CCNA) COMPASS-ND Study.4 The participants

underwent neuroimaging, neurological, cognitive and neuropsychological

assessments.

• For this study, we only selected participants who were diagnosed with FTD

(N=38) or who were recruited as cognitively normal controls (N=118).

• Mean age ± SD at baseline: FTD (67y±12), Controls (65y±11).

• Sex: FTD (55% F), Controls (60% F).

• Specific FTD types included behavioral variant FTD (n=16), Agrammatic

Nonfluent Primary Progressive Aphasia (PPA) (n=10), Semantic PPA (n=3),

Logopenic variant PPA (n=3), Corticobasal syndrome (CBS) nonfluent PPA

(n=1), Unspecified PPA (n=1), and Progressive Supranuclear Palsy (PSP) (n=4).

• T1w and T2w MRI scans were acquired using 3T scanners according to the

Canadian Dementia Imaging Protocol.5

• T1w and T2w images were spatially coregistered, followed by intensity

standardization. Average T1w/T2w ratios were calculated within GM regions of

interest defined by the Harvard-Oxford Atlases.

• This was a cross-sectional analysis using baseline images.

• General linear models were used to conduct group comparisons. The models

were adjusted for age, sex, symptomatic status, and volumes of GM, white

matter (WM), and WM hyperintensities.

 

Results

• T1w/T2w was significantly higher in the FTD patient group compared to controls in the

following subcortical structures: the thalamus, putamen, and pallidum.

• Group differences in the caudate and hippocampus were nonsignificant.

• T1w/T2w was higher in cortical GM in the FTD patient group compared to controls,

particularly in the temporal, parietal, and occipital lobes.

• Additionally, T1w/T2w was higher in WM in the FTD patient group across all lobes (frontal,

temporal, parietal, and occipital).

 

Conclusions

• Microstructural integrity within GM may be altered in FTD patients compared to controls.

In the context of FTD, a feasible explanation is iron accumulation and neuroinflammatory

activity, processes that shorten T2 relaxation and elevate the T1w/T2w ratio.

• Future studies involving in vivo iron imaging (e.g., T2*-weighted MRI or quantitative

susceptibility mapping) are warranted to better understand GM alterations in FTD.

 

References

[1] Johnen et al., Front Neurol 2019, PMID: 31231305

[2] Cash et al., Neurobiol Aging 2018, PMID: 29172163

[3] Glasser et al., J Neurosci 2011, PMID: 21832190

[4] Chertkow et al., Can J Neurol Sci 2019, PMID: 31309917

[5] Duchesne et al., J Magn Reson Imaging, PMID: 30635988

 

Acknowledgements

• The CCNA is supported by an infrastructure and operating grant from the CIHR (Grant no. CNA-137794) and the following

partners: Alberta Prion Research Institute; Alzheimer’s Research UK; Alzheimer Society of Canada; Canadian Nurses

Foundation; Fonds de recherche du Québec – Santé; Michael Smith Foundation for Health Research; New Brunswick

Health Research Foundation; Nova Scotia Health Research Foundation; Ontario Brain Institute; Pfizer Inc.; Robin and

Barry Picov Family Foundation; Sanofi; Saskatchewan Health Research Foundation; Women’s Brain Health Initiative