Grey matter in adults with ASD : The effect of age, symptom severity and lateralization on vertex-wise analysis of cortical volume, cortical thickness, and surface area

Grey Matter in Adults with ASD

The effect of age, symptom severity and lateralization on vertex-wise analysis of cortical volume, cortical thickness, and surface area

Jelena Šušnja

Student number: 5975999

Masters dissertation Clinical Neuropsychology Faculty: Behavioral and Social Sciences University of Amsterdam

Supervisor: dhr. Dr P.C.M.P. Koolschijn External examiner: mw. A.G Lever Date: 7 July

2

Abstract

The current body of research has shown that infants with Autism Spectrum Disorder (ASD) show age-related differences with regard to cortical volume compared to typically developing infants. However it remains unclear if these differences continue into adulthood and old age and if they can be traced back to changes in cortical surface area or cortical thickness. Therefore the objective of this study was to examine the difference in grey matter volume of subcortical areas and differences in cortical thickness, surface area and cortical volume of the whole brain and regions-of-interest between ASD participants aged 30-75 years compared to controls. Second, the influence of age and association with symptom severity with regard to cortical volume, surface area and grey matter volume of the STS, FFA, insula, amygdale, hippocampus and putamen based were investigated based on prior research. Finally, lateralization of the temporal lobe and the mentioned regions of interest in ASD and controls were assessed. Here we employed a cross-sectional vertex-wise analysis of 21 controls and 49 participants with ASD. There were no differences found in cortical thickness, surface area or cortical volume between ASD and controls. Regression analysis demonstrated an age-by-group interaction for the grey matter volume of the left insula, showing age-related decrease in volume in ASD participants compared to an increase in controls. No association of symptom severity with cortical thickness, surface area and cortical volume was found. Differences in laterality were found for cortical thickness in the STS between controls and ASD participants, showing a thicker cortex for the right superior temporal sulcus in ASD compared to a thicker left superior temporal sulcus in controls. These effects did not survive Bonferroni correction for multiple comparisons. Furthermore, exploratory findings of a whole brain analysis demonstrated in ASD larger surface area in the left hemisphere for females compared to males for locations in the precentral gyrus, the post central gyrus, superior temporal gyrus and superior parietal sulcus, the lateral occipital lobe, and the precuneus. A sex-by-group interaction was found for cortical thickness in the right inferiorparietal cortex, showing a decrease in cortical thickness for females with ASD with increased symptom severity compared to an increase in cortical thickness for males with ASD. Moreover, in controls men demonstrated larger hippocampal volume than females when controlled for total intracranial volume, IQ and age. Altogether it seems that differences between ASD and controls in grey matter volume, thickness and surface area subside in adulthood and old age. Sex differences between and within groups should be further investigated

Keywords: Autism, adults, brain, MRI, cortical thickness, surface area, cortical volume, age-related, lateralization

3

Introduction

Autism spectrum disorder (ASD) is a pervasive neurodevelopmental disorder characterized by deficits in communication, impairments in social interaction and the use of repetitive stereotypical patterns in behavior, interests and activities (APA, 2001). The diagnosis of ASD remains to be one purely based on behavior with biological markers remaining unknown. Although the direction of the data from structural research shows great inconsistency, neuroanatomical abnormalities in ASD seem consistent (Anagnostou &Taylor, 2011). One of the most consistent findings is increased brain volume during early childhood (Courchesne et al., 2007; Courchesne, Carper, Akshoomoff, 2003; Sparks et al., 2002). The years

following early childhood show mixed results with regard to localization and direction (increase, decrease) of grey matter and cortical thickness compared to controls (Hadjikhani, Jospeph, Snyder,

&Tager-Flusberg, 2006; Hardan et al., 2006, 2009; Hyde, Samson, Evans, Motron, 2010).This has led to the suggestion of a second period, from middle childhood onwards, characteristic for ASD with abnormal negative cortical growth (Courchesne et al., 2001; Wallace, Dankner, Kenworthy, Giedd, & Martin, 2010). However it remains unclear if these differences in grey matter normalize or persist in middle- and old age. Previous research on structural brain differences in ASD have mainly focused on differences in cortical volume. Cortical volume consists of two components, namely the cortical surface area and cortical thickness. It has been suggested that surface area and cortical thickness are both heritable but independent from each other and genetically uncorrelated with different genetic determinants

(Huttenlocher, 1990; Panizzon et al., 2009). For instance, the surface area is assumed to demonstrate the amount of minicolumns (stacked cells) in the cortical layer, while the cortical thickness is linked to dendritic branching (Huttenlocher, 1990; Panizzon et al., 2009). When merely measuring cortical volume these distinct genetic and cellular etiologies are combined and will plausibly show less clear cut

associations with ASD. This is different when the two components are measured separately, giving more understanding of the etiology of the grey matter abnormalities.

Only a few studies researched the relationship between surface area and cortical thickness in ASD. A significant decrease in cortical volume of the bilateral orbitofrontal cortex in ASD participants ranging from 18 to 42 years compared to controls has been reported (Ecker, Ginestet, Feng, Johnston, Lombardo, 2013). ASD participants demonstrated significantly increased cortical thickness in the frontal lobe, mixed results of decreased and increased cortical thickness in the temporal lobe and reduced surface area in the orbitofrontal region and the posterior cingulum. Furthermore, the significant difference found in grey matter volume between ASD and controls was mostly explained by surface area (56%) and only a small amount by cortical thickness (8%) and cortical thickness and surface area taken together (3%). However surface area is a two-dimensional measure and differences in surface area will show a bigger impact on grey matter volume than the one-dimensional cortical thickness (Ecker et al., 2013). In this study age-related differences were not taken into account, while previous studies showed that between-group

4

difference in grey matter volume, cortical thickness, and surface area depend on age-group interaction (Courchesne et al., 2001; Wallace et al., 2010; Mak-Fan, Taylor, Roberts, & Lerch, 2011). In another study, Ecker et al. (2014) reported an age-related decrease in individuals with ASD, for surface area in the

prefrontal cortex and cortical thickness in fronto-temporal regions compared to controls (sample ranging 7-25 years). This effect was mostly linear for the surface area, while cortical thickness was peculiarly best predicted by a quadratic age term meaning; reduced cortical thickness during childhood yet increased thickness in adulthood. Moreover, in a study spanning into old age (10-60 years), the temporal lobe demonstrated an age-by-group interaction for cortical thickness and grey matter volume, showing less decrease in ASD with age compared to controls (Raznahan et al., 2010). No differences were reported for surface area between ASD and controls. These findings suggest that the previously found linear reduction of surface area normalizes in adulthood and old age while the grey matter and quadratic reduction of cortical thickness abate and reduce less compared to controls.This is a curious suggestion knowing that the two-dimensional surface area has a bigger effect on grey matter volume than the one-dimensional cortical thickness.

Moreover, the previous findings suggest that ASD has region-specific associations with grey matter. Brain regions that have been associated with communication, socio-emotional processing and repetitive

behavior are assumed to play an important role in ASD, for deficits in these domains are characteristic for the disorder (APA, 2001). The fusiform face area (FFA) and superior temporal sulcus (STS) play an important role in face perception and in the interpretation of facial expression (Yowel & Kanwisher, 2004). The insula is assumed to play a role in the mapping of internal bodily states leading to the experience of emotions and with that also empathizing with the pain of others (Singer, Critchley, & Preuschoff, 2009). These brain regions have often demonstrated reduced activation or abnormal connectivity in ASD compared to controls (Pierce, Müller, Ambrose, Allen, Courchesne, 2001; Sherf, 2009; Singer et al., 2009; Boddaert et al., 2004) It is likely that these functional deficits will also show structural differences in ASD. Deficits in communicative skills and an increase in repetitive behavior scores on the Autism Diagnostic Interview-Revised (ADI-R) showed associations with increased volume of respectively the amygdala and putamen (Cauda et al., 2011; Hollander et al., 2005; McAlonan et al., 2005; Stanfield et al., 2008; Shaw et al., 2004). The hippocampus has also often been found enlarged in ASD, which is an area that is associated with emotional memory and affective behavior by inhibitory connections with the amygdala and other structures involved in emotional processing (Groen, Teluij, Buitelaar, & Tendolkar, 2010; Shumann et al., 2004). Furthermore, thicker cortex of the left temporal lobe also portrayed increase of repetitive behavior of the ADI-R symptoms of the repetitive domain (Ecker et al., 2013). However, research has shown that extensive behavioral training can significantly reduce symptom severity in ASD, though not to the level of typically developing individuals (Sheinkopf, & Siegel, 1998). It may be the case that acquired socially accepted behavior in time reduces the severity of symptoms in ASD. No research has yet sought to find region-specific associations of the morphometry of grey with the severity of autistic symptoms in adults and elderly.

5

Furthermore, it has been shown that reduced leftward language lateralization shows is common among individuals with ASD (Bodaert et al., 2003; Gervais et al., 2004; Knaus et al., 2010; Redcay &Courchesne, 2008). Research showed that toddlers that were later diagnosed with ASD demonstrated a

right-lateralized temporal cortex response to language compared to left- right-lateralized response in typically developing children. This effect became stronger with age and reached its peak at 3-4 years of age (Eyler, Pierce & Courchesne, 2011). Furthermore, the left hemisphere showed to distinguish more accurately between ASD and controls participants (79%) than the right hemisphere (65%), with cortical thickness providing the best classification accuracy (Ecker et al., 2010). It is plausible that atypical cerebral anatomical asymmetry in ASD underlies these abnormalities.

In this study we employed a cross-sectional design to investigate the difference in grey matter volume of subcortical areas and differences in cortical thickness, surface area and grey matter volume in ASD participants aged 30-75 years compared to matched controls. We hypothesized that average surface area and grey matter volume of the FFA, STS and the insula together with the volume of the hippocampus, amygdala and the putamen will show no difference between ASD and controls, cortical thickness will show to be thinner in controls compared to ASD (Ecker et al., 2013). Second, the influence of age and symptom severity in ASD with regard to cortical thickness, surface area and grey matter volume in cortical and subcortical areas was assessed. We expected that grey matter volume and surface area will decrease in ASD and controls with age and that the two will not differ (Ecker et al., 2013, 2014; Raznahan, et al., 2010). Cortical thickness will show less reduction with age in ASD compared to controls and there will be a age-by-group interaction for FFA, STS and insula (Ecker et al., 2014). Also, we hypothesize that symptom severity in general will lessen with age (Sheinkopf, & Siegel, 1998) in ASD and that decreased severity of symptoms will be associated with a decrease in volume of the subcortical areas, for these areas are increased in young individuals with ASD (Cauda et al., 2011; Hollander et al., 2005; McAlonan et al., 2005; Stanfield et al., 2008; Shaw et al., 2004) and with decrease of CT in the regions of interest (Ecker et al., 2013). Finally, differences in lateralization of the temporal lobe and regions-of-interest between controls and ASD were investigated. Based on the previous findings of Ecker et al. (2010) and concerning lateralization differences between ASD and controls, we predict a rightward lateralization in the temporal lobe and the regions-of-interest with regard to cortical thickness for the cortical areas and cortical volume for the subcortical areas.

Materials and method

The participants were part of an ongoing behavioral study and were asked to participate in a MRI study. For this study 50 participants with ASD and 25 controls ranging from 30-75 years of age were recruited by advertisement. The diagnosis of ASD participants was made by a multidisciplinary team prior to the study. In addition, participants filled in the Autism-Spectrum Quotient (AQ) questionnaire to assess

6

autistic traits for adults and elderly with normal intelligence (Woodbury-Smith, Robinson, Wheelwright, & Baron-Cohen, 2005). The AQ contains 4 subdomains namely; social skills, communication, switching, imagination and attention to detail (Hoekstra, Bartels, Cath, &Boomsma, 2008). An estimate of the intelligence quotient (IQ) was obtained beforehand as part of the behavioral study and was carried out using the subtests ‘Vocabulary’ measuring word comprehension and ‘Matrix Reasoning’ for perceptual reasoning of the Adult Intelligent Scale IV-NL (WAIS-IV-NL). All participants had an IQ≥80 and showed no signs of neurocognitive decline, neurological diseases, psychosis, schizophrenia, substance abuse, and no MRI contraindications. Controls were additionally excluded if they had ASD running in the family. Participants received travel reimbursement and a monetary reward. All participants gave informed written consent in accordance with the ethics approval of The Psychology Department of The University of Amsterdam.

Magnetic Resonance imaging data acquisition

Scans were acquired with a standard whole-head coil on a 3-Tesla Philips Achieva MRI scanner(Best, The Netherlands), situated at the Spinoza Centre in Amsterdam. High resolution T1-3D weighted anatomical scans were obtained: TR= 8.2 ms; TE = 3.8 ms; flip angle = 8º; 220 slices; FOV 240*188; Voxelsize 1 mm3_{; matrix 240.}

Cortical reconstruction using FreeSurfer

Standardized procedures and methods are used for cortical reconstruction and volumetric segmentation and were performed by FreeSurfer image analysis suite. These procedures have been documented on the FreeSurfer website to ensure a correct representation of the procedure and are. The procedure has directly been cited from the following website

(http://surfer.nmr.mgh.harvard.edu/fswiki/FreeSurferMethodsCitation).

‘The technical details of these procedures are described in prior publications (Dale et al., 1999; Dale and Sereno, 1993; Fischl and Dale, 2000; Fischl et al., 2001; Fischl et al., 2002; Fischl et al., 2004a; Fischl et al., 1999a; Fischl et al., 1999b; Fischl et al., 2004b; Han et al., 2006; Jovicich et al., 2006; Segonne et al., 2004, Reuter et al. 2010, Reuter et al. 2012). Briefly, this processing includes motion correction and averaging (Reuter et al. 2010) of multiple volumetric T1 weighted images (when more than one is available), removal of non-brain tissue using a hybrid watershed/surface deformation procedure (Segonne et al., 2004), automated Talairach transformation, segmentation of the subcortical white matter and deep grey matter volumetric structures (including hippocampus, amygdala, caudate, putamen, ventricles) (Fischl et al., 2002; Fischl et al., 2004a) intensity normalization (Sled et al., 1998), tessellation of the grey matter white matter boundary, automated topology correction (Fischl et al., 2001; Segonne et al., 2007), and surface deformation following intensity gradients to optimally place the grey/white and

grey/cerebrospinal fluid borders at the location where the greatest shift in intensity defines the transition to the other tissue class (Dale et al., 1999; Dale and Sereno, 1993; Fischl and Dale, 2000). Once the

7

cortical models are complete, a number of deformable procedures can be performed for in further data processing and analysis including surface inflation (Fischl et al., 1999a), registration to a spherical atlas which utilized individual cortical folding patterns to match cortical geometry across subjects (Fischl et al., 1999b), parcellation of the cerebral cortex into units based on gyral and sulcal structure (Desikan et al., 2006; Fischl et al., 2004b), and creation of a variety of surface based data including maps of curvature and sulcal depth. This method uses both intensity and continuity information from the entire three

dimensional MR volume in segmentation and deformation procedures to produce representations of cortical thickness, calculated as the closest distance from the grey/white boundary to the grey/CSF boundary at each vertex on the tessellated surface (Fischl and Dale, 2000). The maps are created using spatial intensity gradients across tissue classes and are therefore not simply reliant on absolute signal intensity. The maps produced are not restricted to the voxel resolution of the original data thus are capable of detecting sub millimeter differences between groups. Procedures for the measurement of cortical thickness have been validated against histological analysis (Rosas et al., 2002) and manual

measurements (Kuperberg et al., 2003; Salat et al., 2004). Freesurfer morphometric procedures have been demonstrated to show good test-retest reliability across scanner manufacturers and across field strengths (Han et al., 2006; Reuter et al., 2012).

An approximate mapping of regions- of-interest assigned by Desikan-Killiany atlas (Desikan et al., 2006) was applied for the creation of a temporal lobe used to investigate lateralization effects between ASD and controls. These ROI’s consist of superior parietal, inferior parietal, supramarginal, postcentral and the precuneus. The surface area, cortical thickness and grey matter volume were automatically extracted. Statistical procedure

First an analysis of grey matter volume, cortical thickness and surface area of the whole brain was applied. A vertex-wise analysis was carried out using a general linear model in QDEC to analyse grey matter volume, cortical thickness and surface area of the whole brain. Then, age, sex, IQ and total scores on AQ (measuring symptom severity) were applied as regressors to measure associations with cortical thickness, grey matter volume and surface area between groups and sex. Differences were reported being significant below a FDR corrected p-value of 0.05.

Furthermore, all regions-of-interest including the amygdale, hippocampus, putamen, FFA, STS and insula were compared for the left and right hemisphere between groups due to possible difference in lateralization between groups (Eyler et al., 2011). All measures were corrected for total intracranial volume as an estimate for head size, due to an average 10% larger head size in males (Giedd&Rapoport, 2010). A multivariate analysis of variance (MANOVA) was applied to examine the difference in grey matter volume, cortical thickness and surface area of cortical and subcortical areas between ASD and controls while controlling for sex, total intracranial volume and IQ (α=.05). Furthermore, a forced entry regression analysis was performed on the region-of-interest with total intracranial volume, sex, IQ, group as predictors for the first model. Age and group (controls or ASD) or AQ scores and group were utilized

8

as predictors in the second model. For the third model a group-age interaction term and group-AQ scores interaction term was entered. Regions-of-interest are specialized in certain functions (Groen et al., 2010; Shumann et al., 2004; Singer et al., 2009; Yowel & Kanwisher, 2004) that presumably associate better with certain subdomains of the AQ than with the total AQ, therefore we utilized the subdomains as predictors and in the third model also entered the subdomains as an interaction term with group. The subdomains are: attention for detail, communication, imagination, social skills and divided attention. Every parameter (grey matter volume, cortical thickness and surface area) was analyzed sseparately. A Bonferroni correction for multiple comparison was applied due to the amount of analysis α = .05/6 ROI’s = 0.0083 Lateralization in grey matter volume and cortical thickness was assessed by applying a formula calculating the difference between specific regions in the left and in the right hemisphere and thereby calculating Laterality index (LI )that ranges from -1 to 1, i.e., LI=(left-right)/(left+right). A negative LI implies relatively more right hemispheric asymmetry, whereas a positive LI implies more left hemispheric asymmetry (left larger than right). The data for the temporal lobe as for the regions-of-interest was analyzed with a MANOVA while controlling for, sex, total intracranial volume, IQ and additionally also handedness for it may influence hemispheric lateralization (Pujol, Deus, Losilla & Capdevilla, 1999).

Results

Subject characteristics

Participants’ demographic characteristics are shown in Table 1. There were no differences (2-tailed) found between ASD and controls participants in age (t(68)=1.27; p=.93), IQ (t(68)= .206; p= .84), sex

(t(68)=.604; p=.55) or total intracranial volume (t(68)=.017; p=.68). However age showed to be highly skewed to the left in the control sample compared to the ASD sample (γ1= -0,728; γ1=-0,008).

Furthermore, ASD participants scored significantly higher on symptom severity (AQ) than those in control group (t(68)=.601; p≤.001).

9 Table 1. Subject characteristics

Whole Brain Vertex-wise Analysis of grey matter volume, cortical thickness and

surface area

No significant difference was found for cortical thickness, surface area and grey matter volume between ASD and controls (applying FDR correction for multiple comparisons). Further exploration did reveal a larger surface area of the left hemisphere in females compared to males in regions of the precentral gyrus, the post central gyrus, superior temporal gyrus and superior parietal sulcus, regions of the lateral occipital lobe and the precuneus (see Table 2A, Fig. 1). Also, a sex-by-group interaction was found for the mean thickness of the right inferioparietal cortex (covariate: AQ-score; nuisance factor: age and IQ), that showed cortical thinning in females with ASD with increase in symptom severity compared to thickening in males with ASD (Table2B, Fig. 2).

Description Group

Autism Spectrum Disorder

(N=49) Controls (N=21) Total (N=70) Age Mean 52.4 56.6 53.64 Standard deviation 12.5 13.4 12.79 Range 30-73.9 30.6-73.7 30.03-73.98 Gender Male 34 13 47 Female 15 8 23 IQ Mean 115.8 114.8 115.60 Standard deviation 15.96 13.33 15.28 Range 89-150 96-139 89-150 AQ Total Mean 16 13.3 30.03 Standard deviation 6.86 5.94 11.79 Range 19-47 5-23 5-47 Handedness Left 7 0 7 Right 41 18 44 Ambidexterity 1 3 19

10

Table 2. Significant clusters after FDR correcting of multiple comparison (<0.05)

A. Left hemispheric sex differences in surface area

Brain area max _{Size(mm2) Talx} Taly TalZ

precentral 4.4155 154.44 -29.2 -22.8 60.5 superiortemporal 4.868 151.16 -56.2 -4.2 -3.0 precentral 4.1859 113.45 -27.3 -8.1 44.3 lateraloccipital 4.0580 174.99 -29.0 -93.4 -8.1 lateraloccipital 4.0061 163.43 -24.5 -89.2 -12.9 superiorparietal 3.8264 26.16 -32.1 -49.5 33.3 precuneus 3.7000 51.79 -21.7 -65.2 15.6 postcentral 3.4732 42.81 -59.8 -10.5 20.7

B. Right hemispheric group-by-sex interaction in cortical thickness

Brain area max Size(mm2) Talx Taly TalZ

inferiorparietal -5.9436 111.54 48.7 -60.3 24.7

Fig. 1 Inflated surface maps (dark grey = sulci; light grey gyri) of the left hemisphere showing larger surface area in women than men in the A lateral, B medial, C occipital views. FDR-corrected, p<0.05; only clusters exceeding a threshold of 3.29514 are shown

Fig. 2 Findings of group-by-sex interaction for the cortical thickness of the right hemisphere (covariate: AQ-score; nuisance factor: Age and IQ) in the lateral view showing an association of decrease of cortical thickness with increased symptom severity in females with ASD compared to an increase in cortical thickness in males with ASD. FDR-corrected p<.05, only clusters exceeding a threshold of 4.1856 are shown

11

Region-Specific Between-Groups Differences in Surface Area, Cortical Thickness

and Cortical Volume

Differences between ASD and controls in average cortical thickness, surface area and cortical volume

In this section a region-specific approach was used to investigate differences between ASD participants and controls while controlling for total intracranial volume, sex and IQ. MANOVA did not reveal any differences between ASD and controls in average grey matter volume for hippocampus, amygdale or putamen nor did the surface area, cortical thickness or cortical volume of the STS, FFA or insula differ while controlling for sex, IQ and total intracranial volume. Other exploratory findings revealed a sex-by-group interaction in hippocampal volume, showing for controls a significantly larger hippocampus in males than females (F (1, 9)=7,302; p=0.024) when controlled for IQ, total intracranial volume and age. Putamen was larger for men but when controlled for age, or when looked at controls and ASD separately this effect diminished. There were no significant between-sex differences in individuals with ASD. Regression with Group and Age on cortical volume, cortical thickness and surface area of regions-of –interest

All regions-of interest showed main effects of age except for the right insula. After correction for multiple comparisons the surface area of the left FFA and of the STS, the surface area and cortical thickness of the left insula did not survive. Only the left STS showed a main effect of group, but after correction for multiple comparisons it diminished. There was no age-by-group interaction for any region-of-interest. Only the left insula volume demonstrated an age-by-group interaction effect (β= -.859; t= -2.110; p=.040), showing a decrease in volume for ASD participants compared to an increase in controls (Figure 1). Here however only ASD participants showed significant effects of age (β=-.406; t=-3.886; p≤.001).

Regression with Group and Symptom Severity on cortical volume, cortical thickness and surface area of regions-of interest

Symptom severity did not decrease with age in ASD and controls. No group-by-symptom severity interaction effects on cortical thickness, surface area or cortical volume were found. Also, no main effect of group or symptom severity (AQ-scores) on cortical thickness, surface area or cortical volume was detected. Due to possible lateralization effects (Eyler et al., 2011) the hemispheres were further separately explored but no effects were found. Furthermore, the subdomains of the AQ were also separately explored, for certain regions- of-interest plausibly associate better with sub-domains of the AQ. Here no main- or interaction effects were found either.

12

Lateralization of the temporal lobe and the regions-of-interest

There was no difference in lateralization for the temporal lobe between ASD and controls in cortical thickness, surface area or cortical volume. As predicted the STS did show a significant difference in lateralization for cortical thickness in the STS (F= 4.862; p=.032) when controlled for handedness, total intracranial volume sex and IQ. ASD participants demonstrated thicker cortex of the right STS (LI = -.0086) compared to a thicker left STS in controls (LI = .0035). After correcting for Bonferroni multiple comparison this effect diminished.

4,00 5,00 6,00 7,00 8,00 9,00 25 35 45 55 65 75 85 Volume left i nsula in mm 3

Age-by-Group Interaction in Cortical Volume of Left Insula

Linear (Controls) Linear (ASD)

Age

Discussion

The main goal of this study was to investigate differences in grey matter between ASD and controls. This was achieved by looking at the difference in grey matter volume, cortical thickness and surface area of the whole brain and of the region-of-interest. Furthermore, the influence of age and association of symptom severity with the regions-of-interest namely; FFA, STS, insula, amygdale, hippocampus and putamen were assessed. Also, lateralization differences between ASD and controls of the temporal lobe and regions-of interest were investigated. After correcting for multiple comparisons no between-group differences in cortical volume, cortical thickness or surface area were found, this was contrary to our expectations. Age did not have an effect on the right insula, the surface area of the left FFA and of the STS and on the surface area and cortical thickness of the left insula. The left insula showed a between-group interaction effect of the grey matter volume, showing increased volume with age for controls compared to a decrease in ASD. However, this effect did not survive the Bonferroni correction for multiple comparissons. Also, no interaction effects of symptom severity on the grey matter volume, cortical thickness, and surface area were detected. Furthermore, no lateralization differences of the temporal lobe and regions-of interest

Fig. 3 Age by group interaction in cortical volume of the left insula showing a non-significant increase of

cortical volume in controls with age compared to a significant decrease in ASD participants. The interaction did not survive the Bonferroni correction (α= .0083)

13

were found that survived correction for multiple comparisons. Our findings suggest thus, that differences in cortical thickness, grey matter volume and surface between ASD and controls subside and that

symptom severity in ASD commensurate with age.

The absence of differences between ASD and controls in average grey matter volume, cortical thickness or surface area of the whole brain or regions-of-interest, does not completely confirm our expectations. Ecker et al. (2014) found a quadratic age term as the best predictor of cortical thickness in ASD and consequently indicated an increase in cortical thickness for adults and elderly with ASD compared to controls. It is plausible that the found quadratic increase in childhood and adolescence reaches its peak around middle age and becomes less apparent or normalizes in elderly.

Moreover, age demonstrated a significant between-group interaction effect of the grey matter volume of the left insula, showing increased volume with age for controls compared to a decrease in ASD (see Fig.3). However, when the effect of age on ASD and controls was looked at separately, there was no significant age effect found in controls, only in ASD. These findings contradict the findings of Raznahan (2010), who found an age-related decrease for the temporal lobe in controls compared to ASD. This contradiction could be explained by the age-span of the groups used in Raznahan’s study (10-60 years) which was considerably bigger than the age-span used in this study (30-75 years). Since more changes occur in puberty and adolescence compared to adulthood and old age, a larger age-span may show more significant effects.

ASD participants scored significantly higher on symptom severity than controls and the severity of symptoms did not show a decrease with age as would be expected from compensatory training (Sheinkopf, & Siegel, 1998). Today, people diagnosed with ASD receive special education and training, yet this was not the case for elderly in our sample. It is plausible that the younger participants in our sample profited more from social behavioural training than the older participants, showing lack of decrease in symptom severity. Furthermore, the symptom severity did not show any association with grey matter volume, cortical thickness or surface area. The possibility of test unreliability can be excluded, for the Dutch translation of the AQ demonstrated to be a reliable instrument to assess autistic traits showing high internal consistency and retest reliability (Hoekstra, Bartels, Cath, &Boomsma, 2008). However, others have argued for the possibility in discrepancies in self- and parent-perception of autistic traits in ASD, which show lower scores on autistic traits in self-report in the AQ compared to higher parental reports (Johnston, Filliter, & Murphy, 2009). This distorted self-insight may well be the case for adults and elderly resulting in weaker and non-significant associations with cortical volume, thickness and surface area. Also here, acquired social behavior by training (Sheinkopf, & Siegel, 1998) could have influenced the association with grey matter making it more difficult to find clear cut associations with symptom severity.

14

Furthermore, lateralization differences between ASD and controls were found, showing that individuals with ASD have a thicker cortex of the right STS compared to a thinner left STS in controls. However after Bonferroni correction for multiple the effect diminished. These results are in line with the research of Ecker et al. (2010), whose findings suggested that there are lateralization differences between ASD and controls and that cortical thickness provides the best classification accuracy to distinguish between the two. Other research showed that with increased age individuals with ASD demonstrate a rightward lateralization in neural activity for speech compared to a leftward lateralization in controls (Flagg et al., 2005). The STS plays an important role in interpreting social and speech input (Redcay, 2008). Reduced cortical thickness in the left STS may undergo or be the effect of a different processing mechanism of social and communicational input than typical developing individuals. Consequently this may lead to many of the abnormalities in social interaction and communication observed in ASD.

Explorative findings revealed an unexpected sex-by-group interaction for the cortical thickness of the inferior parietal region, showing a decrease in cortical thickness with increased symptom severity in females compared to men with ASD (see Fig.2). Little research has been conducted in women with ASD due to a four times lower prevalence of the disorder in women than men (Kogan et al., 2009). Further research in sex-differences within the disorder is recommended as these may differ inherently from those in healthy individuals. Other explorative findings indicated an increased volume in controls for men compared to women (controlled for age, intracranial volume and IQ), however no sex-by-group interactions were found. It is possible that spatial knowledge or experience, in which men may possibly outperform women, play a role here (Maguire, Woollet, & Spiers, 2006). Yet, the hippocampal volume did not differ between men and women with ASD and plausibly suggest abnormal sex differences in ASD compared to controls. Furthermore, the left hemisphere demonstrated larger surface area in females for both ASD and controls compared to males for two regions of the precentral gyrus, the post central gyrus, superior temporal gyrus and superior parietal sulcus, two regions of the lateral occipital lobe and the precuneus (see Fig. 1). These sex differences may be explained by an increased gyrification in females compared to males due to brain size related compensation (Luders et al., 2004). An increased complexity of the gyrification and fysuration increases the cortical surface area and possibly underlies gender specific behavior. Altogether, sex differences in ASD seem to differ for specific brain regions from those in controls. Other research confirms this and describe a ‘female shielding effect’, meaning that more genetic mutations are required to give rise in autism in women than in men (Jacquemont et al., 2014). ASD in women may be inherently different from that in men.

Methodological considerations of this study should be taken into account for all previous findings. An important shortcoming of this paper is the highly positive skew in the control sample, showing that age is not equally distributed around the mean. As a consequence the outcome of the applied multiple analysis of variance and regression may be misleading, for it assumes equal distributions. Furthermore,

15

representative for highly functioning individuals. An estimate of 70% of individuals with autism have IQs in the mentally retarded range (Fombonne, 2003), showing that this study is excluding a large population of lower functioning individuals with ASD, consequently it is possible that valuable information may be lost. Due to time restraints and lack of participants for the control group this investigation had a lower statistical power than the recommended power of 80% (Cohen, 1988). This means there is an increased chance of making a type II error, namely failing to reject a false null hypothesis. Furthermore, a cross-sectional design was applied to study age related differences. While a cross-cross-sectional design has

advantages of being time efficient permitting the study of participants over a wide age range, individual difference may confound the data. Therefore, a longitudinal study is recommended to study

neurodevelopmental trajectories within individuals with ASD.

Conclusion

Altogether, in a sample of adult and elderly we researched developmental differences between ASD and controls in cortical thickness, cortical volume and surface area and associations with symptom severity. This research shows that with age differences in grey matter volume, cortical thickness and surface area between controls and ASD diminish. No interaction effects of age or symptom severity on the grey matter volume, cortical thickness, and surface area were detected after Bonferroni correction for multiple comparisons. Also, no lateralization effects were present after correction for multiple comparisons. Explorative findings demonstrated that sex differences in ASD seem to differ from those in controls and more research is recommended on women with ASD, for this group is underrepresented in the disorder. Furthermore, to control for individual differences longitudinal research for the study neurodevelopmental trajectories in ASD should be considered.

References

APA, 1994 American Psychiatric Association, 1994. Diagnostic and statistical manual of mental disorders, fourth ed. (DSM-IV). American Psychiatric Association, Washington, DC.

Anagnostou, E., & Taylor, M.J. (2011). Review of Neuroimaging in Autism Spectrum Disorders: What Have We Learned and Where We Go from Here. Molecular Autism, 2:4.

Barnea-Goraly, N., Kwon, H., Menon, V., Eliez, S., Lotspeich, L., Reiss, A.L. (2004). White Matter Structure in Autism: Preliminary Evidence from Diffusion Tensor Imaging. Biological Psychiary, 55, 323-326

Belmonte, M.K., Allen, G., Beckel-Mitchener, A., Boulanger, L.M., Carper, R.A., & Webb, S.J. (2004).Autism and Abnormal Development of Brain Connectivity. Journal of Neuroscience, 24, 9228-9231

16

Boddaert, N., Belin, P., Chabane, N., Poline, J.B., Barthéléme, C., Mouren-Simeoni, M.C., et al. (2003). Perception of Complex Sounds in Autism: Abnormal Patterns of Cortical Activation in Autism. American Journal of Psychiatry, 160, 2057-2060

Boddaert, N., Chabane, N., Gervais, H., Good, C.D., Bourgeois, M., Plumet, M.H., et al. (2004). Superior Temporal Sulcus Anatomical Abnormalities in Childhood Autism: A Voxel-Based Morphometry MRI Study. NeuroImage, 23, 364-369

Cauda, F., Geda, E., Sacco, K., D’Agata, F., Duca. S., Geminiani, G. (2011). Grey Matter Abnormalities in Autism Spectrum Disorder: an Activation Likelihood Estimation Meta-Analysis. Journal of

Neurology, Neurosurgery & Psychiatry, 82, 1304-1313

Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences, 2nd _{ed., Hilsdale, NJ;Erlbaum} Courchesne, E., Carper, R., Akshoomoff, N. (2003) Evidence of Brain Overgrowth of the First Years in

Life. The Journal of the American Medical Association, 290, 337-344

Courchesne, E., Karns, C.M., Davis, H.R., Ziccardi, R., Carper, R.A., Tigue, Z.D., et al. (2001). Unusual Brain Growth patterns in early life in patients with Autistic Disorder: An MRI Study. Neurology, 57, 245-254

Courchesne, E., & Pierce, K. (2005) Why The Frontal Cortex in Autism Might Be Talking to Itself: Local Over-Connectivity but Long-Distance disconnection. Current Opinion in Neurobiology, 15, 225-230 Courchesne, E., Pierce, K., Shumann, C.M., Redcay, E., Buckwalter, J.A., Kennedy, D.P., et al. (2007)

Mapping Early Brain Development in Autism. Neuron, 56, 399-413

Critchley, H.D., Daly, E.M., Bullmore, E.T., Williams, C.R., Van Amelsvoort, T., Robertson, D.M., et al. (2000). The Functional Neuroanatomy of Social Behavior: Changes in Cerebral Blood Flow When People With Autism Spectrum Disorder Process Facial Expressions. Brain, 123, 2203-2212 Desikan, R.S., Segonne, F., Fischl, B., Quinn, B.T., Dickerson, B.C., et al. (2006).An Automated Labeling

System for Subdividing the Human CerebralCorex on MRI Scans into Gyral Based Regions of Interest.Neuroimage, 31, 968-980

Ecker, C., Ginestet, C., Feng, Y., Johnston, P., Lombardo, M.V. (2013). Brain Anatomy in Adults with Autism: The Relationship between Surface Area, Cortical Thickness, and Autistic Symptoms. JAMA Psychiatry, 70, 59-70

Ecker, C., Marquand, A., Mourão-Miranda, J., Johnston, P., Daly, E.M., Brammer, M.J., et al. (2010).Describing the Brain in Autism in Five Dimensions-Magnetic Resonance Imaging-Assisted Diagnosis of Autism Spectrum Disorder Using a Multiparameter Classification Approach.The Journal of Neuroscience, 30(32), 10612-10623.

17

Ecker, C., Shahidiani, A., Feng, Y., Daly, E., Murphy, C., D’Almeida, V., et al. (2014). The Effect of Age, Diagnosis, and Their Interaction on Vertex-Based Measures of Cortical Thickness and Surface Area in Autism Spectrum Disorder.Journal of Neiural Transmission, Epub ahead of print Eyler,L.T., Pierce, K., & Courchesne, E., (2012). A failure of left temporal cortex to specialize for

language is an early emerging and fundamental property of autism, Brain, 3, 949-960

Flagg, E.J., Cardy, J.E.O., Robers, W., Roberts, T.P.L.(2005). Language Lateralization in Children With Autism: Insight from The Late Field of Magetoencephalogram. Neuroscience Letters, 386, 82-87 Floris, D.L., Chra, L.R., Holt, R.J., Suckling, J., Bullmore, E.T., Baron-Cohen, S., et al.(2012).

Psychological Correlates of Handedness and Corpus Callosum Asymmetry in Autism: The Left Hemisphere Dysfunction Theory Revisited. Journal of Autism and Developmental Disorders, 43, 1758-1772

Fombonne E. (2003). Epidemiological surveys of autism and other pervasive developmental disorders: An update. Journal of Autism and Developmental Disorders, 33, 365–382.

Gervais, H., Belin, P., Boddaert, N., Leboyer, M., Coez, A., Sfaello, I., et al. (2004).Abnormal Cortical Voice Processing in Autism.Nature Neuroscience, 7, 801-802

Giedd, J.N., &Rapoport, J.L.(2010). Structural MRI of Pediatric Brain Development: what have we learned and where are we going? Neuron, 67,728-734

Groen, W., Teluij, M., Buitelaar, J.,& Tendolkar (2010). Amygdala and Hippocampus enlargement in adolescence. Journal of the American Academy of Child & Adolescent Psychiatry, Volume 49, 533-538 Hadjikhani, N., Joseph, R.M., Snyder, J., Tager-Flusbeg, H.(2005). Anatomical Differences in The Mirror

Neuron System and Social Cognition Network in Autism. Cerebral Cortex, 16, 1276-1282 Hoekstra, R.A., Bartels, M., Cath, D.C., &Boomsma, D.I. (2008). Factor Structure, Reliability and

Criterion Validity of The Autism-Spectrum Quotient (AQ): A study in Dutch Population and Patient Group. Journal of Autism and Developmental Disorders, 38, 1555- 1566

Hollander, E., Anagnoustou, E., Chaplin, W., Esposito, K., Haznedar, M. M., Licalzi, E., et al. (2005). Striatal Volume on Magnetic Resonance Imaging and Repetitive Behavior in Autism.Biological Psychiatry, 55, 226-232

Hyde, K.L., Samson, F., Evans, A.C., Motron, L. (2010). Neuroanatomical Differences in Brain Areas Implicated in Perceptual and Other Core Features of Autism Revealed by Cortical Thickness Analysis and Voxel Based Morphometry. Human Brain Mapping, 31, 556-566

18

Jacquemont, S., Coe, B.P., Hersch, M., Duyzend, M.H., Krumm, N., Bergmann, S., et al. (2014)

.

Kogan, M.D., Blumberg, S.J., Schieve, L.A., Boyle, C.A., Perrin, M., Ghandour, R.M., et al.(2009). Prevalence of Parent-Reported Diagnosis ofAutism Spectrum disorder Among Children in the US. Official Journal of The American Academy of Pediatrics, 124, 1395-1403

Knaus, T.A., Silver, A.M., Kennedy, M., Lindgren, K.A., Dominick, K.C., Siegel, J., et al.(2010). Language Laterality in Autism Disorder and Typical Controls: A Functional, Volumetric, and Diffusion Tensor MRI Study. Brain & Language 112, 113-120

Luders, E., Narrl, K.L., Thompson, P.M., Rex, D.E., Jancke, L., Steinmetz, H., et al. (2004). Gender Differences in cortical complexity. Nature Neuroscience, 7, 799-798

Mak-Fan, K.M., Taylor, M.J., Roberts, W., Lerch, J.P. et al. (2012). Measures of Cortical Grey Matter Structure and Development in Children with Autism Spectrum Disorder.Journal of Autism of Autism and Developmental Disorders, 42, 419-427

Maquiere, E.A., Woollet K., & Spiers, H.J.(2006). London Taxi Drivers and Bus Drivers : A structural MRI and Neuropsychological Analysis. Hippocampus, 16, 1091-101

McAlonan, G.M., Cheung, V., Cheung, C., Suckling, J., Lam, G.Y., Tai, K.S., et al. (2005). Mapping The Brain in Autism: A Voxel Based MRI Study of Volumetric Differences and Intercorrelations in Autism, Brain, 128, 268-276

Nakagawa,S.(2004) A farewell to Bonferroni: The Problems of Low Statistical Powers and Publication Bias. Behavioral Ecology, 15, 1044-1045

Pierce, K., Muller, R.A., Ambrose, J., Allen, G (2001)Face processing occurs outside the fusiform `face area' in autism: evidence from functional MRI. Brain, 124, 2059-2073

Pujol, J., Deus, J., Losilla, J.M., & Capdevilla, A. (1999). Cerebral lateralization of language in normal left-handed people studied by functional MRI. Neurology, 52

Raznahan, A., Toro, R., Daly, E., Robertson, D., Murphy, C., Quinton, D., Bolton, P.F., et al. (2009). Cortical Anatomy in Autism Spectrum Disorder: An In Vivo MRI Study on The Effect of Age. Cerebral Cortex, 20, 1332-1340.

Redcay, E. (2008). The Superior Temporal Sulcus Performs a Common Function for Social and Speech Perception: Implications for the Emergence of Autism. Neurosciece and Viobehavioral Reviews, 32, 123-142

19

Redcay, E., &Courchesne, E. (2008).Deviant Functional Magnetic Resonance Imaging Patterns of Brain Activity to Speech in 2-3-Year-Old Children with Autism Spectrum Disorder. Biological Psychiatry, 64, 589-598

Salat, D.H., Buckner, R.L., Snyder, A.Z., Greve, D.N., Desikan., R.S.R, Busa., E., et al. (2004). Cerebral Cortex, 14, 721-730

Shannon, A.J., Filliter, J.H., Murphy, R.R. (2009). Discrepancies Between Self- and Parent-Perceptions of Autistic Traits and Empathy in High Functioning Children and adolescents on the Autism Spectrum. Journal of Autism Developmental disorders, 39, 1706–1714

Shaw, P., Lawrence, E.J., Radbourne, C., Bramham, J., Polkey, C.E., David, A.S. (2004). The Impact of Early and Late Damage to The Human Amygdala on ‘Theory of Mind’ Reasoning. Brain, 127, 1535-1548

Sheinkopf, S.J., & Sieg, B. (1998). Home-Based Behavioral Treatment of Young Children with Autism. Journal of Autism and Developmental Disorders, 28, 15-23

Singer, T., Critchley, H.D., & Preuschoff, K. (2009). A common role of insula in feelings, empathy and uncertainty. Trends in cognitive science, 13, 334-340

Sherf, S.(2009). A typical Development of Face-Related Activation in Autism. Conference on Neurocognitive development. Doi: 10.3389/conf.neuro.10.005

Shumann, C.M., Hamstra, J., Goodline-Jones, B.L., Lotspeich, L.J., Kwon, H., Buonocore, M.H., et al. (2004). The Amygdala is Enlarged in Children But Not in Adolescents with Autism; The Hippocampus is Enlarged in All Ages. The Journal of Neuroscience, 24, 6392-6401

Sparks, B.F., Friedman, S.D., Shaw, D.W., Aylward, E.W., Echelard, D., Asrtru, A.A., et al. (2002).Brain Structural Abnormalities in Young Children with Autism Spectrum Disorder.Neurology, 59, 184-192

Stanfield, A.C., McIntosh, A.M., Spencer, M.D., Philip, R., Gaur, S., Lawrie, S.M. (2008). Towards a Neuroanatomy of Autism: A Systematic Review and Meta-Analysis of Structural Magnetic Resonance Imaging Studies. European Psychiatry, 23, 289-299

Wallace, G.L., Dankner, N., Kenworthy, L., Giedd, J.N, & Martin, A. (2010).Age-Related Temporal and Parietal Cortical Thinning in Autism Spectrum Disorders.Brain, 133, 3745-3754.

Woodbury-Smith, M.R., Robinson, J., Wheelwright, S., & Baron-Cohen, S. (2005). Screening Adults for the Asperger Syndrome using the AQ: A Preliminary Study of its Diagnostic Validity in Clinical Practice. Journal of autism and Developmental Disorders, 35, 331-335

20

Yovel, G., Kanwisher, N. (2004). Face Perception; Domain Specific, Not Process Specific. Neuron, 44 889-898