Regional specificity of aberrant thalamocortical connectivity in autism

Abstract

Preliminary evidence suggests aberrant (mostly reduced) thalamocortical (TC) connectivity in autism spectrum disorder (ASD), but despite the crucial role of thalamus in sensorimotor functions and its extensive connectivity with cerebral cortex, relevant evidence remains limited. We performed a comprehensive investigation of region-specific TC connectivity in ASD. Resting-state functional MRI and diffusion tensor imaging (DTI) data were acquired for 60 children and adolescents with ASD (ages 7–17 years) and 45 age, sex, and IQ-matched typically developing (TD) participants. We examined intrinsic functional connectivity (iFC) and anatomical connectivity (probabilistic tractography) with thalamus, using 68 unilateral cerebral cortical regions of interest (ROIs). For frontal and parietal lobes, iFC was atypically reduced in the ASD group for supramodal association cortices, but was increased for cingulate gyri and motor cortex. Temporal iFC was characterized by overconnectivity for auditory cortices, but underconnectivity for amygdalae. Occipital iFC was broadly reduced in the ASD group. DTI indices (such as increased radial diffusion) for regions with group differences in iFC further indicated compromised anatomical connectivity, especially for frontal ROIs, in the ASD group. Our findings highlight the regional specificity of aberrant TC connectivity in ASD. Their overall pattern can be largely accounted for by functional overconnectivity with limbic and sensorimotor regions, but underconnectivity with supramodal association cortices. This could be related to comparatively early maturation of limbic and sensorimotor regions in the context of early overgrowth in ASD, at the expense of TC connectivity with later maturing cortical regions.

Underconnected, But Not Broken? Dynamic Functional Connectivity MRI Shows Underconnectivity in Autism Is Linked to Increased Intra-Individual Variability Across Time

“…More specifically, convergent evidence from both datasets showed that reduced static iFC in ASD (''underconnectivity'') was related to increased temporal variability. Selecting an ROI pair (PCC-mPFC), for which underconnectivity findings have converged in the literature (Abbott et al, 2015;Assaf et al, 2010;Doyle-Thomas et al, 2015;Monk et al, 2009;von dem Hagen et al, 2012;Washington et al, 2013), mediation analysis showed that reduced static iFC was significantly impacted by increased temporal variability. The research questions motivating the current study were thus mostly affirmed by our findings: (1) iFC variability across time was atypically increased in ASD for several ROI pairings, that is, for PCC-mPFC, L LP-mPFC, and L Thal-R Thal pairs in Study 1 and for PCC-mPFC pair in Study 2; and (2) group differences in static iFC were significantly impacted by differences in the variability of iFC across time.…”

“…Coordinates from previous studies were used to identify the seeds for the following regions: posterior cingulate cortex (PCC), medial prefrontal cortex (mPFC), left (L)/right (R) hippocampal formation (Hipp), L/R lateral parietal cortex (LP) (Van Dijk et al, 2010), pregenual anterior cingulate cortex (Prg ACC) (Di Martino et al, 2009), dorsal anterior cingulate cortex (Abbott et al, 2015), and L/R anterior insula (Ins) (Ebisch et al, 2011). All ROI masks, except for the Prg ACC, were defined as 10 mm radius spheres centered on the seed coordinates.…”

“…However, in its application to ASD, fcMRI has generated highly inconsistent findings, ranging from exclusive underconnectivity (Gotts et al, 2012;Just et al, 2004;Kana et al, 2007;Kleinhans et al, 2008) to mixed effects (Abbott et al, 2015;Di Martino et al, 2014;Doyle-Thomas et al, 2015;Fishman et al, 2014Fishman et al, , 2015Lynch et al, 2013;Monk et al, 2009;Washington et al, 2013) and even predominant overconnectivity (Cerliani et al, 2015;Di Martino et al, 2011;Shih et al, 2011;Supekar et al, 2013). Some of these inconsistencies may be due to methodological differences Nair et al, 2014) and age-related changes (Uddin et al, 2013b), as well as cohort effects due to heterogeneity within the disorder.…”

“…We focused on several networks with previous reports of anomalous connectivity in ASD. These included regions of the default mode network (DMN) (Doyle-Thomas et al, 2015;Kennedy and Courchesne, 2008;Monk et al, 2009;Washington et al, 2013), the salience network (Abbott et al, 2015;Ebisch et al, 2011;Uddin et al, 2013a), as well as amygdala (Abrams et al, 2013;Grelotti et al, 2005;Kleinhans et al, 2011;Murphy et al, 2012) and thalamus (Cerliani et al, 2015;Hardan et al, 2008;Nair et al, 2013Nair et al, , 2015. These regions of interest (ROIs) were selected to test the general questions described below, without any assumption of exclusive impairment in ASD (which would be unwarranted, given the breadth of regional findings implicating almost every network and brain region in ASD).…”

Functional SPECT neuroimaging using machine learning algorithms distinguishes autism spectrum disorder from healthy subjects

“…These include structural MRI [11][12][13], intrinsic functional connectivity MRI (fcMRI) [14][15][16], diffusion tensor imaging (DTI) [17,18] and electroencephalography (EEG) [19,20]. Reported ASD abnormalities have been identified in the cerebellum [21] and cerebellar vermis [22], anterior cingulate gyrus [23], amygdala [24][25][26], hippocampus [24,27,28], and areas of the frontal [29][30][31], temporal [29,32] parietal lobes [30], caudate and putamen [29,33]. Additional imaging abnormalities in autism include impaired brain growth [24], cortical thickness [12,34], alterations in white matter architecture [18,35], and aberrant connectivity within the somatosensory, visual and default mode network [36].…”