Premium
IC‐P‐024: EFFECTIVE CONNECTIVITY WITHIN THE LEFT AND RIGHT EXECUTIVE CONTROL NETWORKS IN MCI AND AD
Author(s) -
Cook Cole John,
Hwang Gyujoon,
Nair Veena A.,
Pasquesi Mary-Elizabeth,
Alexander Andrew L.,
Antuono Piero G.,
Asthana Sanjay,
Birn Rasmus,
Carlsson Cynthia,
Chen Guangyu,
Edwards Dorothy Farrar,
Franczak Malgorzata,
Goveas Joseph S.,
Johnson Sterling C.,
Kecskemeti Steven,
Kulkarni Arman P.,
Mohanty Rosaleena,
Moreno Brittany,
Nencka Andrew S.,
Okonkwo Ozioma C.,
Rivera-Bonet Charlene N.,
Spees Morgen,
Taylor Ingrid Kristine,
Tellapragada Neelima,
McDonald Williams Leroy,
Li Shi-Jiang,
Bendlin Barbara B.,
Prabhakaran Vivek
Publication year - 2019
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2019.06.4186
Subject(s) - default mode network , human connectome project , connectome , neuroscience , connectomics , resting state fmri , psychology , voxel , medicine , cognition , artificial intelligence , computer science , functional connectivity
Background: APOE ε4 has been linked to accelerated amyloid deposition and tau-tangle formation. We hypothesized that cognitively normal AD biomarker-negative individuals would exhibit synaptic dysfunction and subsequent resting-state functional connectivity (rs-fc) network disruptions related to the presence of APOE ε4. Methods: Using cross-sectional analyses, we examined rs-fc from 120 cognitively-normal adults (CDR 0) with negative CSFAb42, CSF ptau/Ab42, and PET (PiB-SUVR) according to established criteria (Hansson et al 2018). The cohort had a mean age of 76.4 years 67.59 years, 15% were APOE e4 carriers and 65% were female. All participants had cerebrospinal fluid (CSF) and positron emission tomography (PET) evaluations. Rs-fc were aggregated into canonical cortical networks based on defined criteria (Power et al 2011). A 298x298 connectivity matrix was generated and was then masked to examine only intra-hemispheric (i.e. lateralized) and inter-hemispheric (i.e. callosal) connections between APOE ε4 carriers/non-carriers, with particular focus on the default-mode, memory, and salience (i.e. DMS) networks. Results: Qualitative rs-fc differences observed in the group difference matrix (Fig. 1A) were quantified by averaging, with differences observed between (p1⁄4 0.03), not within (p1⁄4 0.08), the DMS cluster (Fig. 1B). Observed rs-fc differences were primarily due to a significant strengthening of lateralized versus callosal connections both within (p 1⁄4 0.02) and between (p 1⁄4 0.02) the DMS networks (Fig. 1C, Fig. 1D). We next modeled this dysfunction as a function of imaging and CSF biomarkers. First, participants were categorized using the amyloid, tau, neurodegeneration (A/T/N) CSF and imaging criteria (Jack et al, 2016), with subthreshold levels noted across all biomarkers (Fig. 2A). Next, a linear regression model of the rs-fc dysfunction revealed significant weighting for CSF ptau (p 1⁄4 0.05) and t-tau (p 1⁄4 0.05) but not for CSF Ab42 or PETPiB (Fig. 2B). Conclusions: Early changes in key rs-fc networks in cognitively-normal, AD biomarker-negative APOE ε4 carriers are driven by changes in local lateralized connections relative to long-range callosal connections. Linear modeling results are suggestive of an early role of tau in early network dysfunction in elderly asymptomatic APOE ε4 carriers without biomarker evidence of AD.