Longitudinal iron in caudate related to control network connectivity and planning in older adults
Presented During:
Monday, June 24, 2024: 5:45 PM - 7:00 PM
COEX
Room:
Hall D 2
Poster No:
2410
Submission Type:
Abstract Submission
Authors:
jing zhou1, Colleen Hughes1, Alfie Wearn1, Julia Huck2, Giulia Baracchini1, Elisabeth Sylvain3, Jennifer Tremblay-Mercier3, Judes Poirier4, Sylvia Villeneuve4, Christine Tardif1, Claudine J. Gauthier2, Gary Turner5, R. Nathan Spreng1
Institutions:
1Montreal Neurological Institute, McGill University, Montreal, Québec, 2Physics Department, Concordia University, Montreal, Québec, 3Douglas Mental Health University Institute, Montreal, Québec, 4Department of Psychiatry, McGill University,, Montreal, Québec, 5York University, Toronto, Ontario
First Author:
Co-Author(s):
Introduction:
Cognitive functions attributed to the frontoparietal control network (FPCN) are central to goal-directed behaviors, and these abilities are vulnerable to decline with advancing age.[1] Age-related declines in executive function (EF) have been linked to dopaminergic striatal dysfunction[2]. Elevated striatal iron deposition has been associated with lower cognition [3], and differences in brain function, measured with fMRI[4]. Here, we examined the impact of longitudinal iron accumulation in the caudate, a core subcortical node of the FPCN, on resting state functional connectivity (RSFC) changes in the FPCN in a sample of older adults. We then examined caudate iron associations with brain function and impact on changes in EF.
Methods:
Participants and cognitive assessment: 132 older adults (mean age, 67.64.6y, 34 men) at familial risk for AD from the PREVENT-AD cohort underwent MRI scanning at baseline and follow-up (mean interval, 2.7y). 81 completed the Cambridge Brain Sciences test battery (Planning, Token Search); 127 completed the Trail Making Task and Stroop at each timepoint.
Data acquisition: MRI scanning was conducted on a Siemens Prisma 3T scanner with a 32-channel head coil array. QSM images were acquired from 3D GRE sequence with single-echo (TR=20ms; TE=7.29ms; Voxel size=0.8 x 0.8 x1.0mm3; scan time=5.31min). Resting-state multi-echo functional MRI were acquired from a multi-echo (ME) EPI sequence (TR=1000ms; TE1=12ms, TE2=30.11ms, TE3=48.22ms; Voxel size=3.0 x 3.0 x3.0mm3; scan time=10.24 min). T1w were acquired with 3D MP-RAGE sequence (TR=2300ms; TE=2.96ms; voxel size=1.0 x 1.0 x 1.2mm3; scan time=5.3min).
Image processing:
QSM: Phase images were combined to obtain offset-corrected images, using estimation from POEM[5]. QSM maps were reconstructed using TGV-based method[6].
FMRI: Functional images were submitted to ME-ICA[7]. Whole brain RSFC matrices were initialized to the 200-parcel Schaefer atlas[8] 7 networks. Parcels were then individualized to each participants' functional neuroanatomy using Group Prior Individualized Parcellation[9]. Product-moment correlation between parcels was computed for each timepoint, resulting in 2RSFC matrices for each participant (Fig.1), then restricted to the FPCN.
MP-RAGE: Caudate ROI was obtained through Freesurfer subcortical segmentation and registered the QSM space.
Analysis: We examined longitudinal change in iron and cognitive performance using a paired sample t-test predicting an increase in iron and decline in EF. The relationship between longitudinal change in iron and FPCN RSFC was conducted using a two condition (time 1, time 2) behavioral Partial Least Squares (bPLS) analysis. To test associations with EF, we used a linear mixed model. EF measures (planning, token search, Stroop inhibition time and TMT B-A RT) were assigned as dependent variables, with brain scores derived from bPLS and timepoint (baseline and follow-up) as fixed factors, and age, sex and education as covariates, allowing for random intercepts across subjects.
Data acquisition: MRI scanning was conducted on a Siemens Prisma 3T scanner with a 32-channel head coil array. QSM images were acquired from 3D GRE sequence with single-echo (TR=20ms; TE=7.29ms; Voxel size=0.8 x 0.8 x1.0mm3; scan time=5.31min). Resting-state multi-echo functional MRI were acquired from a multi-echo (ME) EPI sequence (TR=1000ms; TE1=12ms, TE2=30.11ms, TE3=48.22ms; Voxel size=3.0 x 3.0 x3.0mm3; scan time=10.24 min). T1w were acquired with 3D MP-RAGE sequence (TR=2300ms; TE=2.96ms; voxel size=1.0 x 1.0 x 1.2mm3; scan time=5.3min).
Image processing:
QSM: Phase images were combined to obtain offset-corrected images, using estimation from POEM[5]. QSM maps were reconstructed using TGV-based method[6].
FMRI: Functional images were submitted to ME-ICA[7]. Whole brain RSFC matrices were initialized to the 200-parcel Schaefer atlas[8] 7 networks. Parcels were then individualized to each participants' functional neuroanatomy using Group Prior Individualized Parcellation[9]. Product-moment correlation between parcels was computed for each timepoint, resulting in 2RSFC matrices for each participant (Fig.1), then restricted to the FPCN.
MP-RAGE: Caudate ROI was obtained through Freesurfer subcortical segmentation and registered the QSM space.
Analysis: We examined longitudinal change in iron and cognitive performance using a paired sample t-test predicting an increase in iron and decline in EF. The relationship between longitudinal change in iron and FPCN RSFC was conducted using a two condition (time 1, time 2) behavioral Partial Least Squares (bPLS) analysis. To test associations with EF, we used a linear mixed model. EF measures (planning, token search, Stroop inhibition time and TMT B-A RT) were assigned as dependent variables, with brain scores derived from bPLS and timepoint (baseline and follow-up) as fixed factors, and age, sex and education as covariates, allowing for random intercepts across subjects.
Results:
Iron level significantly increased over time (t(131)=2.4, p<.01, single-tail). Increasing iron content in caudate was significantly associated with longitudinal FPCN RSFC (p<.05). Both positive and negative RSFC were observed within the FPCN to covary with iron over time. This multivariate association between iron accumulation in caudate and FPCN RSFC was then related to EF, noting a specific decline in planning (t(80)=1.81, p<.05, single-tail). Caudate iron - FPCN longitudinal effects were significantly associated with declines in planning ability (r =.365, p<.005) but not other measures of EF (Fig.2).
Conclusions:
Iron deposition in the caudate in older adults is reliably associated with changes in brain connectivity within the FPCN. This relationship was associated with decline in planning ability. These findings suggest that iron accumulation may be a driving pathological factor in RSFC changes, and predicts EF decline with advancing age.
Higher Cognitive Functions:
Executive Function, Cognitive Control and Decision Making 2
Novel Imaging Acquisition Methods:
Multi-Modal Imaging 1
Keywords:
Other - Quantitative susceptibility mapping; Iron depostion; caudate nucleus; frontoparietal control network; executive function
1|2Indicates the priority used for review
Provide references using author date format
[2] Bäckman, L., Lindenberger, U., Li, S.-C., & Nyberg, L. (2010). Linking cognitive aging to alterations in dopamine neurotransmitter functioning: Recent data and future avenues. Neuroscience & Biobehavioral Reviews, 34(5), 670-677. doi:https://doi.org/10.1016/j.neubiorev.2009.12.008
[3] Spence, H., McNeil, C. J., & Waiter, G. D. (2020). The impact of brain iron accumulation on cognition: A systematic review. PLoS One, 15(10), e0240697. doi:10.1371/journal.pone.0240697
[4] Salami, A., Avelar-Pereira, B., Garzon, B., Sitnikov, R., & Kalpouzos, G. (2018). Functional coherence of striatal resting-state networks is modulated by striatal iron content. Neuroimage, 183, 495-503. doi:10.1016/j.neuroimage.2018.08.036
[5] Sun, H., Cleary, J. O., Glarin, R., Kolbe, S. C., Ordidge, R. J., Moffat, B. A., & Pike, G. B. (2020). Extracting more for less: multi-echo MP2RAGE for simultaneous T1-weighted imaging, T1 mapping, mapping, SWI, and QSM from a single acquisition. Magnetic Resonance in Medicine, 83(4), 1178-1191. doi:https://doi.org/10.1002/mrm.27975
[6] Langkammer, C., Bredies, K., Poser, B. A., Barth, M., Reishofer, G., Fan, A. P., . . . Ropele, S. (2015). Fast quantitative susceptibility mapping using 3D EPI and total generalized variation. Neuroimage, 111, 622-630. doi:https://doi.org/10.1016/j.neuroimage.2015.02.041
[7] Kundu, P., Brenowitz, N. D., Voon, V., Worbe, Y., Vértes, P. E., Inati, S. J., . . . Bullmore, E. T. (2013). Integrated strategy for improving functional connectivity mapping using multiecho fMRI. Proceedings of the National Academy of Sciences, 110(40), 16187-16192. doi:doi:10.1073/pnas.1301725110
[8] Schaefer, A., Kong, R., Gordon, E. M., Laumann, T. O., Zuo, X.-N., Holmes, A. J., . . . Yeo, B. T. T. (2017). Local-Global Parcellation of the Human Cerebral Cortex from Intrinsic Functional Connectivity MRI. Cerebral Cortex, 28(9), 3095-3114. doi:10.1093/cercor/bhx179
[9] Chong, M., Bhushan, C., Joshi, A.A., Choi, S.Y., Haldar, J.P., Shattuck, D.W., Spreng, R.N. & Leahy, R.M. (2017). Individual parcellation of resting fMRI with a group functional connectivity prior. NeuroImage, 156, 87–100.