Age-dependent alterations in the coordinated development of subcortical regions in adolescents with social anxiety disorder

doi:10.21203/rs.3.rs-1945877/v1

Download PDF

Research Article

Age-dependent alterations in the coordinated development of subcortical regions in adolescents with social anxiety disorder

https://doi.org/10.21203/rs.3.rs-1945877/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Subcortical brain regions play essential roles in the onset of social anxiety disorder (SAD). While adolescence is the peak period of SAD, the relationships between abnormal development of the subcortical regions during this period and SAD are still unclear. This study investigated the age-dependent alterations in structural co-variance among subcortical regions and between subcortical and cortical regions, aiming to reflect aberrant coordination during development in adolescents with SAD. High-resolution T1-weighted images were obtained from 76 adolescents with SAD and 67 healthy controls (HC), ranging from 11.0 to 17.9 years. We found that the SAD group exhibited significantly enhanced structural co-variance among key striatum regions (putamen and caudate). While the co-variance decreased with age in healthy adolescents, the co-variance in SAD adolescents stayed high, leading to more apparent abnormalities in middle adolescence. Moreover, the striatum’s mean structural co-variance with cortical regions decreased with age in HC but increased with age in SAD. The age-dependence of the abnormal coordination indicates atypical developmental trajectories of the striatum and its balance with the cortical regions in adolescents with SAD.

Social anxiety disorder (SAD) is one of the most common mental disorders in children and adolescents [1], characterized by excessive fears of social situations [2], negative emotionality, and social anhedonia [3, 4]. Adolescence is the most prevalent period for the emergence of pathological social anxiety [5, 6]. In this period, neural circuits experience rapid maturation, including synaptic pruning and myelination [7], leading to a tighter association between subcortical and cortical regions [8, 9]. Adolescents' psychology and behavior also undergo corresponding changes with typical brain development, like increased risk-taking, novelty pursuit, and social salience [10]. However, the brain developmental trajectories of adolescents are vulnerable to adverse environmental and life events [9, 11], increasing their susceptibility to SAD.

Most studies have found that the brain mechanisms of social anxiety are related to abnormalities in amygdala-related fear circuitry. Human and animal research has consistently shown that the amygdala is crucially involved in processing threat bias [12, 13]. Individuals with SAD are associated with amygdala hyperreactivity in response to social decisions [14], threat-related stimuli, and symptom provocation [15]. In fear circuitry, the cortex regulates top-down association, attention, and interpretation processes [16–18]. Freitas-Ferrari et al. proposed that individuals with SAD (compared to controls) are characterized by dysfunctional front-limbic (fear) circuitry (including the amygdala, anterior cingulate cortex, ventromedial prefrontal cortex, dorsolateral prefrontal cortex, etc.) [19]. Brühl et al. further added the role of parietal and occipital regions within the circuitry [20].

Recent evidence further suggests that abnormalities in the striatum are strongly associated with social anhedonia. The striatum, encompassing the caudate, putamen, and nucleus accumbens [21], can be assumed to be a critical region of reward circuitry for processing social motivation [22]. Reduced striatal response to social reward [23] and overactive striatal response to social punishment were found in SAD [24]. The disturbance of the reward circuitry was also found in SAD. Recent imaging studies have further supported extensive functional changes in the reward circuitry in adolescents with SAD [25]. Adults with SAD had decreased functional connectivity in the brain-wide network, including frontal, occipital, parietal, and temporal lobes, and increased local efficiency of the right putamen [26]. Thus, both the amygdala-related system and the striatum-related system may play critical roles in the pathophysiology of SAD.

According to the triadic model [27–29], the striatum-related system is responsible for receiving reward cues related to positive valence emotion and the amygdala-related system is responsible for the feedback of aversive cues related to negative emotion. Both systems are subject to the cognitive regulation of the cortex. The coordinated development of these three modules (striatum, amygdala, and cortex) maintains adolescents' typical behavioral and psychological changes. In particular, the balance between the key subcortical regions and the maturation of the cortices’ ability to regulate subcortical regions both play a crucial role in improving social motivation and anxiety regulation [27–29]. However, social skills and the ability to cope with anxiety in adolescents with SAD do not improve along with brain development, suggesting abnormal developmental coordination among the striatum, amygdala, and cortical regions. Therefore, it is necessary to reveal further the altered coordinated development between key subcortical regions and between subcortical and cortical regions in adolescents with SAD.

Structural co-variance between two brain regions reflects synchronized maturational change of brain region pairs, perhaps mediated by axonal connections forming and reforming throughout development [30, 31], implying that brain areas growing together are highly correlated in size [32–34]. A longitudinal study found that brain regions that were structurally co-variant were also correlated with the rate of change in cortical thickness during development [35]. This approach provides evidence for the atypical development of brain regions in other mental disorders [33, 36, 37]. Therefore, we can examine age-specific differences in structural co-variance between key subcortical regions and subcortical and cortical regions in adolescents with SAD using structural magnetic resonance imaging (sMRI).

The present study systematically investigated abnormalities in subcortical areas in adolescents with SAD. We focused on the volumes of seven pairs of subcortical regions, including the amygdala, putamen, caudate nucleus, thalamus, hippocampus, nucleus accumbens, and globus pallidus. We used structural co-variance to examine the developmental coordination within these subcortical regions and between the subcortical and cortical regions. To further reveal abnormal trajectories of coordination in subcortical regions in SAD adolescents, we applied a sliding-age window approach to reflect age-dependent structural co-variance among the subcortical regions and between the subcortical and cortical regions. We hypothesized that the amygdala or striatum (caudate, putamen, and nucleus accumbens) would exhibit age-dependent structural morphology in adolescents with SAD.

Participants

This study included 143 participants comprising 76 adolescents with SAD (SAD group) and 67 healthy controls (HC group). The age range was 11.0-17.9 years (see demographical information in Table 1). The 76 adolescents with SAD were consecutively recruited through advertisements or referred by psychiatrists from the children and adolescent psychiatry clinic in Shanghai Mental Health Center, Shanghai Jiao Tong University School of Medicine. The Ethical Committee of the Shanghai Mental Health Center approved the study design (2018-46), and all participants provided written informed consent.

To validate the diagnosis of SAD, trained clinicians first conducted face-to-face interviews with the patients and at least one of their primary caregivers by using the child version [38] and parent version [39] of the Mini-International Neuropsychiatric Interview for Children and Adolescents (MINI-Kid). A senior pediatric psychiatrist further verified the primary diagnosis according to DSM-5 through a clinical interview. The patients did not receive any systemic medication or psychotherapy before enrollment and did not receive any medication two weeks before the MRI scanning. Patients who completely met the diagnostic criteria of other mental disorders, including post-traumatic stress disorder (PTSD), bipolar disorder, schizophrenia, autism spectrum disorder, mental retardation, tic disorder, anorexia nervosa, bulimia nervosa, and high risk of suicide, were excluded from the SAD group.

Healthy participants were recruited from schools and communities through advertisements and educational programs. The participants and at least one of their primary caregivers were separately assessed with the child and parent versions of the MINI-Kid to exclude past or present mental disorders. For those judged to have no mental illness or have suspected diagnosis in the MINI-Kid interview, a senior psychiatrist further interviewed the participants and their primary caregivers to exclude any DSM-5 diagnoses of mental disorders. Participants with left-handedness, organic brain disorders, a history of severe physical illnesses, or severe suicidal tendencies were excluded from both groups. One participant was excluded from the supplementary analysis aiming to control symptoms because of lost behavioral data.

Imaging data acquisition

High-resolution T1-weighted structural images were acquired on a 3.0T Siemens Trio MRI scanner (Siemens Healthcare, Erlangen, Germany), using a magnetization-prepared rapid gradient echo sequence with the following parameters: repetition time = 2.53 s, echo time = 3.65 ms, inverse time = 1.1 s, flip angle = 7°, field of view = 256*256 mm², matrix = 256*256, 224 coronal slices, and slice thickness = 1.0 mm.

Image preprocessing and quality control

The T1-weighted images were processed using the FreeSurfer (V6.0.0) “recon-all” pipeline [40, 41], and the main steps include bias field correction, skull removal, tissue segmentation, surface reconstruction, and anatomical labeling. As a result, we obtained the gray matter volume estimations of 68 cortical regions based on Desikan Killiany (DK) Atlas [42] and 14 subcortical structures based on Aseg Atlas [43]. In addition, the qualities of skull removal, tissue segmentation, surface reconstruction, and anatomical labeling were visually checked to rule out observable errors.

Assessment of symptom severity

We measured participants’ clinical symptoms by using the Social Anxiety Scale for Children (SASC) with two sub-scales containing Social avoidance and distress (SAD-scale) and Fear of negative evaluation (FNE-scale) [44], and the Depression Self Rating Scale for Children (DSRS-C) [45].

Analysis of demographic information

Age, SASC, and DSRS-C scores were compared between the SAD and HC groups using Welch’s t-tests for continuous variables. Sex distributions were examined using chi-square tests.

Subcortical volume analysis

Associations between SAD and the volume of the thalamus, caudate nucleus, globus pallidus, putamen, amygdala, hippocampus, and nucleus accumbens were examined using stepwise multiple regression analysis. All variables were centered in the regression models. The primary linear model can be presented as:

$$Y=diagnosis+diagnosis\times age+diagnosis\times sex+age+sex+TIV$$

where Y refers to the volume of each subcortical region and TIV means total intracranial volume. We selected the best models by the lowest Akaike information criterion (AIC), and the interaction effects would be dropped from the model if they were non-significant (p > 0.05). The main effects were kept in all formulas with covariates. The type-I error was controlled using the False discovery rate (FDR).

We computed the correlations between the two symptom measures and the volume of subcortical structures in the SAD and HC groups separately because of considerable discrepancy in the symptom severity in the two groups. TIV was first regressed out from the volume of each subcortical region. Then partial correlation coefficients were then computed between the symptom measures (SASC total scores with two sub-scores and DSRS-C total scores) and the residual volume of each subcortical region with covariates including age and sex. Multiple comparison error was corrected using FDR.

Co-variance analysis of subcortical regions

Volume co-variance analyses were performed among the subcortical regions to detect aberrant volume coordination in subcortical-subcortical pairs [35]. First, the partial correlation was used to estimate abnormal volume co-variance for each pair of subcortical regions, with age, sex, and TIV controlled. The analysis was carried out among 91 pairs of subcortical regions for the SAD and HC groups separately. Then the correlation coefficients were transformed into Fisher’s Z values. Finally, the Fisher’s Z values for each pair of subcortical regions were compared between the SAD and HC group using a Z statistic as the following:

$${Z}_{i}=\left({Z}_{i\left(SAD\right)}- {Z}_{i\left(HC\right)}\right)/\sqrt{1/({n}_{SAD}-3)+1/({n}_{HC}-3)}$$

where n_SAD and n_HC refer to sample sizes of the two groups, and Z_i(SAD) and Z_i(HC) indicate the Fisher’s Z values for the SAD and HC groups for the i^th pair of subcortical regions. FDR was used for multiple comparison corrections.

To examine the age effect on volume co-variance of the subcortical region, we divided subjects into early (aged 11.0-14.9 years) and middle (aged 15.0-17.9 years) adolescence [46] to repeat the same volume co-variance analysis as above. We further applied a sliding window approach to reflect the age-dependent group difference on the volume co-variance among pairs of subcortical structures to reveal the latent inflection point. The width of the age windows was three years, and the time step was one year, yielding the following age groups: 11.0-13.9, 12.0-14.9, 13.0-15.9, 14.0-16.9, and 15.0-17.9 years old. Within each age group, the sample sizes of SAD and HC groups were 27 and 27, 40 and 24, 45 and 14, 42 and 38, and 36 and 36, respectively. We conducted the volume co-variance analyses as above within each age group and diagnosis group.

Co-variance analysis between subcortical regions and cortical regions

We examined the volume co-variance between subcortical and cortical regions in each hemisphere. The analysis was repeated in early (aged 11.0-14.9 years) and middle (aged 15.0-17.9 years) adolescence, respectively. First, the partial correlation was used to estimate co-variance for each pair of regions (14 subcortical regions by 68 cortical regions), controlling age, sex, and TIV. Then, the co-variance was calculated in the SAD and HC groups separately, transformed into Fisher’s Z, and statistically compared using the Z statistic as described above. Multiple comparison error was corrected using the false discovery rate (FDR) method.

We considered the coordinated development between the subcortical and cortical regions as a network and characterized the roles of the subcortical regions using centrality. Specifically, we averaged the co-variance measures between each subcortical region and all cortical regions. The volume co-variance centrality measures were compared between the SAD and HC groups. We further applied the sliding age-window approach to the volume co-variance centrality of the subcortical regions. The significance of the group differences was evaluated using permutation tests that randomly shuffled the group labels 10,000 times to generate null distributions of the co-variance centrality measures.

To rule out the possibility that the potential difference in structural co-variance could be due to the symptom severity, we repeated the structural co-variance analysis by using the symptom severity measure (SASC or DSRS-C total scores) as a covariate among subcortical regions and between subcortical and cortical regions in early and middle adolescence.

Demographics and clinical data

As shown in Table 1, there was no significant difference in sex ratio (p = 0.07) or age (p = 0.65) between the SAD and HC groups. The SAD group showed significantly higher scores in the SASC (p < 0.01) and its subscales-SAD (p < 0.01) and subscales-FNE (p < 0.01), and the DSRS-C (p < 0.01).

------------------------------------------------------- Table 1 -------------------------------------------------------

	SAD (n = 76)	HC (n = 67)	Test Statistic	P value
Female	54	37	Χ = 3.20, df = 1	0.07
Age, mean±SD, years	14.91±1.60	14.73±2.04	t = - .45, df = 123.41	0.65
SASC, mean±SD	14.25±4.04	5.76±3.92	t = -12.70, df = 138.99	<0.01***
SAD-scale, mean±SD	5.65±1.74	1.64±1.72	t = -9.26, df = 138.54	<0.01***
FNE-scale, mean±SD	8.60±2.89	4.12±2.86	t = - 13.81, df = 138.48	<0.01***
DSRS-C, mean±SD	19.67±6.51	7.48±5.65	t = - 11.94, df = 139.88	<0.01***

Abbreviations: SAD, social anxiety disorder; SASC, the Social Anxiety Scale for Children; SAD-scale, Social avoidance and distress; FNE-scale, Fear of negative evaluation; DSRS-C, Depression Self Rating Scale for Children

* P value is less than 0.05; ** P value is less than 0.01; *** P value is less than 0.001;

Subcortical volume comparison and associations with symptoms

As shown in Table 2, there was no main effect of diagnosis on the 14 subcortical regions. The interactions including diagnosis*age and diagnosis*sex were far away from the significance level so that it was dropped from the final models by the step-wise regression algorithm. Within the SAD group, the social avoidance and distress score was negatively correlated with the volume of the left putamen (r = -0.24; FDR-corrected p = 0.02), left hippocampus (r = -0.28; FDR-corrected p < 0.01), right thalamus (r = -0.23; FDR-corrected p = 0.03). The total score of DSRS-C had positive correlation with the left amygdala (r = 0.27; FDR-corrected p = 0.01) and right accumbens area (r = 0.27; FDR-corrected p = 0.01) in HC group. Figure 1 shows scatter plots representing the associations between the symptoms and the subcortical volumes.

------------------------------------------------------- Table 2 -------------------------------------------------------

Region	Diagnosis		Sex		Age		TIV
Region	β	P value	β	P value	β	P value	β	P value
Thalamus (left)	-0.12	0.34	0.40	0.01*	0.01	0.88	0.55	<0.01***
Thalamus (right)	0.11	0.34	0.28	0.06	-0.03	0.38	0.66	<0.01***
Caudate (left)	0.05	0.69	0.35	0.04	-0.10	0.01*	0.50	<0.01***
Caudate (right)	0.03	0.79	0.18	0.28	-0.10	0.01*	0.59	<0.01***
Putamen (left)	-0.12	0.35	0.80	<0.01***	-0.06	0.13	0.34	<0.01***
Putamen (right)	-0.17	0.19	0.81	<0.01***	-0.04	0.24	0.34	<0.01***
Globus pallidus (left)	-0.20	0.17	0.12	0.54	-0.06	0.12	0.50	<0.01***
Globus pallidus (right)	-0.27	0.06	0.19	0.30	-0.07	0.07	0.52	<0.01***
Hippocampus (left)	-0.05	0.69	0.40	0.02*	0.03	0.51	0.48	<0.01***
Hippocampus (right)	0.03	0.80	0.23	0.18	0.02	0.52	0.56	<0.01***
Amygdala (left)	0.07	0.61	0.60	<0.01***	-0.0006	0.99	0.44	<0.01***
Amygdala (right)	-0.07	0.61	0.47	0.01*	-0.045	0.24	0.47	<0.01***
Nucleus accumbens (left)	-0.01	0.95	0.48	0.01*	-0.13	<0.01**	0.34	<0.01***
Nucleus accumbens (right)	-0.15	0.29	0.26	0.16	-0.11	0.01*	0.46	<0.01***

Abbreviations: TIV, total intracranial volume

Multiple comparison correction is labeled by bold and italic letters.

* P value is less than 0.05; ** P value is less than 0.01; *** P value is less than 0.001.

------------------------------------------------------- Figure 1 -----------------------------------------------------

Structural co-variance in the subcortical region

Figures 2A presents the volume co-variance among the subcortical regions in the SAD and HC groups and the co-variance differences between the two groups. Compared to the HC group, the SAD group showed significantly higher co-variance between the left caudate and right putamen (SAD: r = 0.46; HC: r = -0.10; difference: Z = -3.46, p < 0.05; all results were FDR-corrected, same below). These regions are key regions of the striatum.

--------------------------------------------------------- Figure 2 ---------------------------------------------------

As the brain rapidly develops in adolescence, the co-variance within subcortical regions in different periods of the adolescence may be different. We then examined the difference between the SAD and HC groups in early (11.0-14.9 years) and middle adolescence (15.0-17.9 years) separately. As shown in Figure 2B, there was no significant difference in the subcortical volume co-variance between the two groups in early adolescence. In contrast, as shown in Figure 2C, the SAD group exhibited significantly enhanced volume co-variance in four pairs of subcortical regions in the middle adolescence, including the left caudate-right putamen (SAD: r = 0.52; HC: r = -0.39; difference: Z = -4.00, p < 0.01), left caudate- Left putamen (SAD: r = 0.45; HC: r = -0.36; difference: Z = -3.50, p = 0.02), right caudate-right putamen (SAD: r = 0.37; HC: r = -0.36; difference: Z = -3.24, p = 0.04), and left putamen-right caudate (SAD: r = 0.40; HC: r = -0.36; difference: Z = -3.09, p < 0.05). Figure 3 presents scatter plots that reflect details of the above findings. To quantify the differences between age stages, Figure 2D presents comparisons of the co-variance in these 4 pairs of subcortical regions in different age stages.

--------------------------------------------------------- Figure 3 ---------------------------------------------------

The above findings support that the abnormality of structural co-variance in SAD was modulated by age in adolescence. To further characterize the age-dependent co-variance differences, we compared the structural co-variance trajectories in 3-year bins. As shown in Figure 4, we found significant increases in volume co-variance between the left caudate and right putamen (SAD: r = 0.34; HC: r = -0.70; difference: Z = -3.63, p = 0.01), left caudate and left putamen (SAD: r = 0.30; HC: r = -0.56; difference: Z = -2.78, p < 0.10), left putamen and right caudate (SAD: r = 0.33; HC: r = -0.62; difference: Z = -3.12, p = 0.05), and right caudate and right putamen (SAD: r = 0.36; HC: r = -0.70; difference: Z =-3.65, p = 0.01) in the SAD group during 13.0-15.9 years.

In the 14.0-16.9 years bin, there were significant increases in volume co-variance in SAD between the left caudate and right putamen (SAD: r = 0.42; HC: r = -0.38; difference: Z = -3.62, p = 0.03), left caudate and left putamen (SAD: r = 0.37; HC: r = -0.34; difference: Z = -3.18, p = 0.04), left putamen and right caudate (SAD: r = 0.33; HC: r = -0.33; difference: Z = -2.94, p = 0.08), and right caudate and right putamen (SAD: r = 0.37; HC: r = -0.35; difference: Z = -3.22, p = 0.04).

For the 15.0-17.9 years old period, we observed significantly larger co-variance in SAD between the left caudate and right putamen (SAD: r = 0.52; HC: r = -0.39; difference: Z = -4.00, p = 0.02), left caudate and left putamen (SAD: r = 0.45; HC: r = -0.36; difference: Z = -3.50, p = 0.02), left putamen and right caudate (SAD: r = 0.37; HC: r = -0.35; difference: Z = -3.09, p < 0.05), and right caudate and right putamen (SAD: r = 0.40; HC: r = -0.36; difference: Z = -3.24, p = 0.04). The structural co-variance of these four pairs of subcortical regions started to exhibit abnormality from 13.0-15.9 years, revealing a sensitive period for the emergence of SAD-related subcortical structural alteration.

--------------------------------------------------------- Figure 4 ---------------------------------------------------

Subcortical-cortical co-variance

We did not observe any group difference in co-variance between specific subcortical and specific cortical regions. Nonetheless, as shown in Figure 5A, the co-variance centrality of a few subcortical regions exhibited significant difference between the HC and SAD groups in early and middle adolescence respectively. Specifically, we found that the volume co-variance centrality measures of the left globus pallidus (SAD: mean r = -0.09; HC: mean r = 0.12; p = 0.03), right nucleus accumbens (SAD: mean r = -0.08; HC: mean r = 0.21; p = 0.01), and right putamen (SAD: mean r = -0.08; HC: mean r = 0.12; p = 0.04) were significantly higher in HC in early adolescence. Besides, the left nucleus accumbens exhibited higher volume co-variance centrality in HC than in SAD in early adolescence (SAD: mean r = 0.01.; HC: mean r = 0.19; p = 0.08). In contrast, the volume co-variance centrality of the left putamen (SAD: mean r = 0.13; HC: mean r = -0.02; p = 0.03), right caudate nucleus (SAD: mean r = 0.13; HC: mean r = 0.01; p = 0.04), and right putamen (SAD: mean r = 0.15; HC: mean r = 0.01; p = 0.03) were significantly higher in the SAD group in middle adolescence.

--------------------------------------------------------- Figure 5 ---------------------------------------------------

A further sliding age-window analysis showed that subcortical-cortical co-variance developed in different patterns in HC and SAD, as depicted in Figure 5B. Overall, the HC group exhibited a trajectory of high-to-low cortical-subcortical co-variance, while the SAD group showed an inversed trajectory.

We observed that the co-variance centrality measures of the right putamen (SAD: mean r = -0.11; HC: mean r = 0.17; p = 0.03) and right nucleus accumbens (SAD: mean r = -0.11; HC: mean r = 0.22; p = 0.02) were significantly lower in the SAD group in 11.0-13.9 years. The right amygdala showed the same tendency (SAD: mean r = 0.06; HC: mean r = 0.27; p = 0.05).

For the 12.0-14.9 years bin, the co-variance centrality measures of the left globus pallidus (SAD: mean r = -0.09; HC: mean r = 0.19; p = 0.01), right putamen (SAD: mean r = -0.08; HC: mean r = 0.16; p = 0.01) and right nucleus accumbens (SAD: mean r = -0.08; HC: mean r = 0.26; p < .01) were significantly lower in the SAD group. The same trend was marginally significant in the left putamen (SAD: mean r = -0.02; HC: mean r = 0.13; p = 0.09) and the left nucleus accumbens (SAD: mean r = 0.01; HC: mean r = 0.19; p = 0.07).

Among 13.0-15.9 years old, adolescents with SAD had significantly reduced subcortical-cortical co-variance in left globus pallidus (SAD: mean r = -0.11; HC: mean r = 0.18; p < .01), left nucleus accumbens (SAD: mean r = 0.04; HC: mean r = 0.27; p = 0.04), right nucleus accumbens (SAD: mean r = 0.01; HC: mean r = 0.24; p = 0.03), left amygdala (SAD: mean r = 0.05; HC: mean r = 0.30; p = 0.01), right amygdala (SAD: mean r = 0.06; HC: mean r = 0.31; p < 0.01), and right hippocampus (SAD: mean r = 0.12; HC: mean r = 0.30; p = 0.04) .

In the 14.0-16.9 bin, the co-variance centrality measures of the right caudate nucleus (SAD: mean r = 0.12; HC: mean r = 0.02; p = 0.05) was higher in SAD group. The left putamen (SAD: mean r = 0.13; HC: mean r = -0.02; p = 0.03), right caudate (SAD: mean r = 0.13; HC: mean r = 0.01; p = 0.04), and right putamen (SAD: mean r = 0.15; HC: mean r = 0.01; p = 0.03) was higher in the SAD group for the 15.0-17.9 bin. Other subcortical regions showed no group difference in the 14.0-17.9 bin. These observations reflected abnormal developmental trajectories of subcortical-cortical co-variance in adolescence with SAD.

After controlling the influence of symptom severity (SASC and DSRS-C) respectively, the structural co-variance between subcortical regions and subcortical and cortical regions within each group and the difference between groups were similar to the above results above in early and middle adolescence (Table S1~8).

The present study systematically examined the volumes of subcortical areas in adolescents diagnosed with SAD. We found that adolescents with SAD showed significantly enhanced structural co-variance in striatal regions (caudate and putamen). These abnormalities become apparent in SAD from the age of 14 years. Furthermore, we found reversed developmental trajectories of striatal-cortical co-variance tendencies in HC and SAD adolescents. The HC group showed a decrease in striatal-cortical co-variance centrality with age. In contrast, the SAD group showed scant striatal-cortical co-variance centrality in early adolescence but exaggerated co-variance in mid-adolescence. These findings provide new insights into the abnormal brain development in adolescents with SAD.

The developmental transition of the brain during adolescence is characterized by a global structural co-variance that peaks in early adolescence and then declines in mid-adolescence [47, 48]. The decline of structural co-variance in brain networks during adolescence, as shown in the HC group in our study, has significant implications. As the extent of myelination was more pronounced in brain regions with significantly decreased structural co-variance [47], the reduction of structural co-variance within the striatal system during adolescence may be the result of progressvie brain maturation. The typical development of the striatal system in healthy adolescents allows them to respond flexibly to various social stimuli [49, 50]. During adolescence, the dorsal and ventral striatum showed age-related increases and decreases in neural activity, respectively [51], allowing the striatum to confer hypersensitivity to reward and maintains social motivation in healthy adolescents [49-52].

However, in adolescents with SAD, structural co-variance in the key regions in the striatum failed to decrease as it did in HC (Figure 4). Since structural co-variance reflects the developmental coordination of different brain regions [32, 35], the persistent high structural co-variance in the striatum suggests that striatum’s development is hindered in adolescents with SAD. This argument echoes longitudinal study reflecting adolescent’s striatal activity diminished under longtime social stress, associated with higher anhedonia [53]. Furthermore, the striatum was deactivated for social reward in adolescents with SAD compared with HC [23, 54]. Thus, the stunted striatum may render adolescents with SAD to exhibit markedly reduced social motivation.

Our study further revealed that the structural co-variances between the key regions of the striatum and cortex were significantly deficient in early adolescence and significantly enhanced in mid-adolescence in SAD (Figure 5). This developmental trajectory may suggest a delayed cortical development in adolescents with SAD compared with healthy adolescents. In healthy adolescents, the cortex regulates the striatum for cognitive remodeling [55], and this kind of ability increases with age [56]. However, most adolescents with SAD lack the ability of cognitive reappraisal towards social stimuli. This ability relies on how the cerebral cortex dynamically modulates the striatum [55, 56]. Therefore, the delayed coordination between the striatum and the cortex can explain the weakened regulation of the cortex in SAD.

The results of structural co-variance did not change after separately controlling for the effects of symptom severity (SASC and DSRS-C). Long-term interaction between heredity and the environment influences structural co-variance between brain regions [32]. Thus, the abnormally developmental coordination in the striatum-related system in SAD reflected long-term alterations in the brain system in SAD. According to the triadic model [27-29], the abnormal development of the striatal system may break the balance between the amygdala-related system and the striatum-related system during adolescence, thereby predisposing adolescents with SAD to exhibit social phobia and social anhedonia.

Research on age-dependent differences in subcortical volume between typically developing adolescents and socially anxious adolescents is sparse. Adults with SAD were reported to have volumetric alteration in those subcortical regions [57]. However, we did not find that the volume of the amygdala and striatum varied across developmental stages between SAD and HC during adolescence. Subcortical regions are still developing in this crucial period ranging from 11.0 to 17.9 years, in which brain regions undergo significant structural change with age [58, 59]. The volume of the amygdala and striatum in SAD might be influenced by the progression of the disease and show neural plasticity during puberty, which may help explain why no difference was found between SAD and HC during adolescence.

According to results of the correlation analysis (Figure 1), we believe that alterations in the volume of subcortical regions in adolescents are associated with increased vulnerability to anxiety and depression. Consistent with the previous study, in the HC group, we found that a larger amygdalar structure was tied to general depression independent of age and sex [60, 61], and an increased nucleus accumbens volume contributed to adolescent depression [62]. Moreover, we were intrigued by the finding that the volumes of left putamen, right thalamus, and left hippocampus were negatively correlated with social avoidance and distress sex in adolescents with SAD, controlling for age and sex. These results suggest the important role of the putamen, thalamus, and hippocampus in responding to social interaction and imply their potential effect on social avoidance symptoms. The putamen and thalamus are part of the striatum-thalamus-cortical network important for learning actions based on their predicted outcomes and regulating cognitive or emotional behaviors [21, 63, 64]. The constant encoding, updating, and flexible expression of memory representations to modulate social behavior flexibly depend upon the hippocampal-dependent memory system [65, 66].

Our study highlights that the striatum-related system of SAD patients is affected from an early age and show abnormal developmental trajectories. However, current psychological and drug treatments for social anxiety have focused on reducing excessive negative emotions, acting mainly on the negative emotion system associated with the amygdala [67, 68], rather than improving deficits in positive emotions [69, 70]. This situation leaves many young people with SAD still facing ineffective treatment [70]. Since anhedonia is associated with deficits in striatal-related systems [50, 71], our findings support the introduction of therapies with the primary goal of improving positive affect, such as positive thinking-based cognitive therapy (Mindfulness-Based Cognitive Therapy, MBCT) [72], into interventions for adolescents with SAD.

A limitation of this study is that the cross-sectional nature cannot infer the causal relationship between the abnormalities in structural co-variance and the symptoms of social anxiety. Therefore, a longitudinal design is required to clarify whether the abnormal structural co-variance represents vulnerability or consequences of social anxiety. In addition, we found that the coordinated development of the striatal system in adolescents with SAD was abnormal, mainly involving reward motivation circuits. However, more direct links between the reported abnormality and social anhedonia need further validation. Finally, the structural co-variance characteristics found in this study should be examined in other types of anxiety disorders to clarify the generalizability and specificity among related mental disorders.

In adolescents with SAD, we found age-dependent abnormalities in the volume co-variance between striatal components and between striatum and cortex. The co-variance increases abnormally from early to middle adolescence and maintains a high level in middle adolescence. These findings suggest that adolescents with SAD may have abnormal striatum development.

Competing Interests: The authors declare no conflict of interest.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (81971682, 81571756); Shanghai Science and Technology Commission (20Y11906700); Project of Shanghai Children’s Health Service Capacity Construction (GDEK201702); Clinical Research Project of Shanghai Mental Health Center (CRC2018DSJ01-5, CRC2019ZD04); Natural Science Foundation of Shanghai (20ZR1472800); Shanghai Municipal Commission of Education (Gaofeng Clinical Medicine-20171929); Shanghai Science and Technology Commission (18JC1420305); Shanghai Municipal Health Commission (2019ZB0201, 2018BR17); Shanghai Clinical Research Center for Mental Health (19MC1911100); and Research Funds from Shanghai Mental Health Center (13dz2260500, 2018-YJ-02, 2018-YJ-05).

Mathew SJ, Coplan JD, Gorman JM. Neurobiological mechanisms of social anxiety disorder. Am J Psychiatry 2001; 158: 1558–1567.
Stein MB, Stein DJ. Social anxiety disorder. Lancet Lond Engl 2008; 371: 1115–1125.
Richey JA, Rittenberg A, Hughes L, Damiano CR, Sabatino A, Miller S et al. Common and distinct neural features of social and non-social reward processing in autism and social anxiety disorder. Soc Cogn Affect Neurosci 2014; 9: 367–377.
Tb K. Social anxiety spectrum and diminished positive experiences: theoretical synthesis and meta-analysis. Clin Psychol Rev 2007; 27. doi:10.1016/j.cpr.2006.12.003.
Beesdo K, Knappe S, Pine DS. Anxiety and Anxiety Disorders in Children and Adolescents: Developmental Issues and Implications for DSM-V. Psychiatr Clin North Am 2009; 32: 483–524.
S X, X Z, W C, Z Y. Adolescent anxiety disorders and the developing brain: comparing neuroimaging findings in adolescents and adults. Gen Psychiatry 2021; 34. doi:10.1136/gpsych-2020-100411.
Natu VS, Gomez J, Barnett M, Jeska B, Kirilina E, Jaeger C et al. Apparent thinning of human visual cortex during childhood is associated with myelination. Proc Natl Acad Sci U S A 2019; 116: 20750–20759.
Heller AS, Cohen AO, Dreyfuss MFW, Casey BJ. Changes in cortico-subcortical and subcortico-subcortical connectivity impact cognitive control to emotional cues across development. Soc Cogn Affect Neurosci 2016; 11: 1910–1918.
Tottenham N, Galván A. Stress and the Adolescent Brain: Amygdala-Prefrontal Cortex Circuitry and Ventral Striatum as Developmental Targets. Neurosci Biobehav Rev 2016; 70: 217–227.
Spear LP. The adolescent brain and age-related behavioral manifestations. Neurosci Biobehav Rev 2000; 24: 417–463.
Caouette JD, Guyer AE. Gaining insight into adolescent vulnerability for social anxiety from developmental cognitive neuroscience. Dev Cogn Neurosci 2014; 8: 65–76.
Etkin A, Wager TD. Functional Neuroimaging of Anxiety: A Meta-Analysis of Emotional Processing in PTSD, Social Anxiety Disorder, and Specific Phobia. Am J Psychiatry 2007; 164: 1476–1488.
Phelps EA, LeDoux JE. Contributions of the Amygdala to Emotion Processing: From Animal Models to Human Behavior. Neuron 2005; 48: 175–187.
Schultz J, Willems T, Gädeke M, Chakkour G, Franke A, Weber B et al. A human subcortical network underlying social avoidance revealed by risky economic choices. eLife 2019; 8: e45249.
Miskovic V, Schmidt LA. Social fearfulness in the human brain. Neurosci Biobehav Rev 2012; 36: 459–478.
de Carvalho MR, Rozenthal M, Nardi AE. The fear circuitry in panic disorder and its modulation by cognitive-behaviour therapy interventions. World J Biol Psychiatry Off J World Fed Soc Biol Psychiatry 2010; 11: 188–198.
Shin LM, Liberzon I. The Neurocircuitry of Fear, Stress, and Anxiety Disorders. Neuropsychopharmacology 2010; 35: 169–191.
Tovote P, Fadok JP, Lüthi A. Neuronal circuits for fear and anxiety. Nat Rev Neurosci 2015; 16: 317–331.
Freitas-Ferrari MC, Hallak JEC, Trzesniak C, Filho AS, Machado-de-Sousa JP, Chagas MHN et al. Neuroimaging in social anxiety disorder: A systematic review of the literature. Prog Neuropsychopharmacol Biol Psychiatry 2010; 34: 565–580.
Brühl AB, Delsignore A, Komossa K, Weidt S. Neuroimaging in social anxiety disorder—A meta-analytic review resulting in a new neurofunctional model. Neurosci Biobehav Rev 2014; 47: 260–280.
Marchand WR. Cortico-basal ganglia circuitry: a review of key research and implications for functional connectivity studies of mood and anxiety disorders. Brain Struct Funct 2010; 215: 73–96.
Ruff CC, Fehr E. The neurobiology of rewards and values in social decision making. Nat Rev Neurosci 2014; 15: 549–562.
Becker MPI, Simon D, Miltner WHR, Straube T. Altered activation of the ventral striatum under performance-related observation in social anxiety disorder. Psychol Med 2017; 47: 2502–2512.
Crane NA, Chang F, Kinney KL, Klumpp H. Individual differences in striatal and amygdala response to emotional faces are related to symptom severity in social anxiety disorder. NeuroImage Clin 2021; 30: 102615.
Arnold Anteraper S, Triantafyllou C, Sawyer AT, Hofmann SG, Gabrieli JD, Whitfield-Gabrieli S. Hyper-connectivity of subcortical resting-state networks in social anxiety disorder. Brain Connect 2014; 4: 81–90.
Zhu H, Qiu C, Meng Y, Yuan M, Zhang Y, Ren Z et al. Altered Topological Properties of Brain Networks in Social Anxiety Disorder: A Resting-state Functional MRI Study. Sci Rep 2017; 7: 43089.
Ernst M. The triadic model perspective for the study of adolescent motivated behavior. Brain Cogn 2014; 89: 104–111.
ERNST M, PINE DS, HARDIN M. Triadic model of the neurobiology of motivated behavior in adolescence. Psychol Med 2006; 36: 299–312.
Ernst M, Fudge JL. A developmental neurobiological model of motivated behavior: anatomy, connectivity and ontogeny of the triadic nodes. Neurosci Biobehav Rev 2009; 33: 367–382.
Bullmore ET, Frangou S, Murray RM. The dysplastic net hypothesis: an integration of developmental and dysconnectivity theories of schizophrenia. Schizophr Res 1997; 28: 143–156.
Burgoyne RD, Graham ME, Cambray-Deakin M. Neurotrophic effects of NMDA receptor activation on developing cerebellar granule cells. J Neurocytol 1993; 22: 689–695.
Alexander-Bloch A, Giedd JN, Bullmore E. Imaging structural co-variance between human brain regions. Nat Rev Neurosci 2013; 14: 322–336.
Liu Z, Hu Y, Zhang Y, Liu W, Zhang L, Wang Y et al. Altered gray matter volume and structural co-variance in adolescents with social anxiety disorder: evidence for a delayed and unsynchronized development of the fronto-limbic system. Psychol Med 2021; 51: 1742–1751.
Mechelli A, Friston KJ, Frackowiak RS, Price CJ. Structural Covariance in the Human Cortex. J Neurosci 2005; 25: 8303–8310.
Alexander-Bloch A, Raznahan A, Bullmore E, Giedd J. The Convergence of Maturational Change and Structural Covariance in Human Cortical Networks. J Neurosci 2013; 33: 2889–2899.
Duan X, Wang R, Xiao J, Li Y, Huang X, Guo X et al. Subcortical structural covariance in young children with autism spectrum disorder. Prog Neuropsychopharmacol Biol Psychiatry 2020; 99: 109874.
Wannan CMJ, Cropley VL, Chakravarty MM, Bousman C, Ganella EP, Bruggemann JM et al. Evidence for Network-Based Cortical Thickness Reductions in Schizophrenia. Am J Psychiatry 2019; 176: 552–563.
Liu Y-X, Liu J, Wang Y-F. Reliability and validity of Chinese version of the Mini International Neuropsychiatric Interview for Children and Adolescents (child version). Chin Ment Health J 2011; 25: 8–13.
Yu L. Reliability and validity of Chinese version of the Mini International Neuropsychiatric Interview for Children and Adolescents(Parent Version). undefined 2010.https://www.semanticscholar.org/paper/Reliability-and-validity-of-Chinese-version-of-the-Yu/bfb11cb1dc358f7aa8c56b8b13ae69aa5fa7958b (accessed 20 May2022).
Dale AM, Fischl B, Sereno MI. Cortical Surface-Based Analysis: I. Segmentation and Surface Reconstruction. NeuroImage 1999; 9: 179–194.
Fischl B, Sereno MI, Dale AM. Cortical surface-based analysis. II: Inflation, flattening, and a surface-based coordinate system. NeuroImage 1999; 9: 195–207.
Desikan RS, Ségonne F, Fischl B, Quinn BT, Dickerson BC, Blacker D et al. An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. NeuroImage 2006; 31: 968–980.
Fischl B, Salat DH, Busa E, Albert M, Dieterich M, Haselgrove C et al. Whole Brain Segmentation: Automated Labeling of Neuroanatomical Structures in the Human Brain. Neuron 2002; 33: 341–355.
La Greca AM, Stone WL. Social Anxiety Scale for Children-Revised: Factor Structure and Concurrent Validity. J Clin Child Psychol 1993; 22: 17–27.
Birleson P, Hudson I, Buchanan DG, Wolff S. Clinical Evaluation of a Self-Rating Scale for Depressive Disorder in Childhood (depression Self-Rating Scale). J Child Psychol Psychiatry 1987; 28: 43–60.
Hardin AP, Hackell JM, COMMITTEE ON PRACTICE AND AMBULATORY MEDICINE. Age Limit of Pediatrics. Pediatrics 2017; 140: e20172151.
Váša F, Seidlitz J, Romero-Garcia R, Whitaker KJ, Rosenthal G, Vértes PE et al. Adolescent Tuning of Association Cortex in Human Structural Brain Networks. Cereb Cortex 2018; 28: 281–294.
Vijayakumar N, Ball G, Seal ML, Mundy L, Whittle S, Silk T. The development of structural covariance networks during the transition from childhood to adolescence. Sci Rep 2021; 11: 9451.
Raznahan A, Shaw PW, Lerch JP, Clasen LS, Greenstein D, Berman R et al. Longitudinal four-dimensional mapping of subcortical anatomy in human development. Proc Natl Acad Sci 2014; 111: 1592–1597.
Telzer EH. Dopaminergic reward sensitivity can promote adolescent health: A new perspective on the mechanism of ventral striatum activation. Dev Cogn Neurosci 2016; 17: 57–67.
Larsen B, Luna B. In vivo evidence of neurophysiological maturation of the human adolescent striatum. Dev Cogn Neurosci 2015; 12: 74–85.
Bhanji JP, Delgado MR. The Social Brain and Reward: Social Information Processing in the Human Striatum. Wiley Interdiscip Rev Cogn Sci 2014; 5: 61–73.
Hanson JL, Hariri AR, Williamson DE. Blunted Ventral Striatum Development in Adolescence Reflects Emotional Neglect and Predicts Depressive Symptoms. Biol Psychiatry 2015; 78: 598–605.
Zhang Y, Zhu C, Chen H, Duan X, Lu F, Li M et al. Frequency-dependent alterations in the amplitude of low-frequency fluctuations in social anxiety disorder. J Affect Disord 2015; 174: 329–335.
Wager TD, Davidson ML, Hughes BL, Lindquist MA, Ochsner KN. Prefrontal-Subcortical Pathways Mediating Successful Emotion Regulation. Neuron 2008; 59: 1037–1050.
van den Bos W, Cohen MX, Kahnt T, Crone EA. Striatum–Medial Prefrontal Cortex Connectivity Predicts Developmental Changes in Reinforcement Learning. Cereb Cortex 2012; 22: 1247–1255.
Bas-Hoogendam JM, van Steenbergen H, Nienke Pannekoek J, Fouche J-P, Lochner C, Hattingh CJ et al. Voxel-based morphometry multi-center mega-analysis of brain structure in social anxiety disorder. NeuroImage Clin 2017; 16: 678–688.
Low LK, Cheng H-J. Axon pruning: an essential step underlying the developmental plasticity of neuronal connections. Philos Trans R Soc B Biol Sci 2006; 361: 1531–1544.
Østby Y, Tamnes CK, Fjell AM, Westlye LT, Due-Tønnessen P, Walhovd KB. Heterogeneity in Subcortical Brain Development: A Structural Magnetic Resonance Imaging Study of Brain Maturation from 8 to 30 Years. J Neurosci 2009; 29: 11772–11782.
Albaugh MD, Nguyen T-V, Ducharme S, Collins DL, Botteron KN, D’Alberto N et al. Age-related volumetric change of limbic structures and subclinical anxious/depressed symptomatology in typically developing children and adolescents. Biol Psychol 2017; 124: 133–140.
Foell J, Palumbo IM, Yancey JR, Vizueta N, Demirakca T, Patrick CJ. BIOBEHAVIORAL THREAT SENSITIVITY AND AMYGDALA VOLUME: A TWIN NEUROIMAGING STUDY. NeuroImage 2019; 186: 14–21.
Lee KH, Yoo JH, Lee J, Kim SH, Han JY, Hong S-B et al. The indirect effect of peer problems on adolescent depression through nucleus accumbens volume alteration. Sci Rep 2020; 10: 12870.
Boyes A, McLoughlin LT, Anderson H, Schwenn P, Shan Z, Gatt JM et al. Basal ganglia correlates of wellbeing in early adolescence. Brain Res 2022; 1774: 147710.
Zhang Y, Liu W, Lebowitz ER, Zhang F, Hu Y, Liu Z et al. Abnormal asymmetry of thalamic volume moderates stress from parents and anxiety symptoms in children and adolescents with social anxiety disorder. Neuropharmacology 2020; 180: 108301.
Felix-Ortiz AC, Tye KM. Amygdala Inputs to the Ventral Hippocampus Bidirectionally Modulate Social Behavior. J Neurosci 2014; 34: 586–595.
Rubin RD, Watson PD, Duff MC, Cohen NJ. The role of the hippocampus in flexible cognition and social behavior. Front Hum Neurosci 2014; 8.https://www.frontiersin.org/article/10.3389/fnhum.2014.00742 (accessed 12 Feb2022).
Ashar YK, Clark J, Gunning FM, Goldin P, Gross JJ, Wager TD. Brain markers predicting response to cognitive-behavioral therapy for social anxiety disorder: an independent replication of Whitfield-Gabrieli et al. 2015. Transl Psychiatry 2021; 11: 1–9.
Whitfield-Gabrieli S, Ghosh SS, Nieto-Castanon A, Saygin Z, Doehrmann O, Chai XJ et al. Brain connectomics predict response to treatment in social anxiety disorder. Mol Psychiatry 2016; 21: 680–685.
Bystritsky A, Khalsa SS, Cameron ME, Schiffman J. Current Diagnosis and Treatment of Anxiety Disorders. Pharm Ther 2013; 38: 30–57.
Craske MG, Meuret AE, Ritz T, Treanor M, Dour HJ. Treatment for Anhedonia: A Neuroscience Driven Approach. Depress Anxiety 2016; 33: 927–938.
Gorwood P. Neurobiological mechanisms of anhedonia. Dialogues Clin Neurosci 2008; 10: 291–299.
Strege MV, Swain D, Bochicchio L, Valdespino A, Richey JA. A Pilot Study of the Effects of Mindfulness-Based Cognitive Therapy on Positive Affect and Social Anxiety Symptoms. Front Psychol 2018; 9.https://www.frontiersin.org/article/10.3389/fpsyg.2018.00866 (accessed 5 Jun2022).

No competing interests reported.

Supplementary.docx

Download PDF

Editorial decision: Major revision
19 Oct, 2022
Reviews received at journal
10 Oct, 2022
Reviewers agreed at journal
21 Sep, 2022
Reviews received at journal
19 Sep, 2022
Reviewers agreed at journal
05 Sep, 2022
Reviewers invited by journal
22 Aug, 2022
Editor assigned by journal
10 Aug, 2022
Submission checks completed at journal
10 Aug, 2022
First submitted to journal
09 Aug, 2022

You are reading this latest preprint version

Age-dependent alterations in the coordinated development of subcortical regions in adolescents with social anxiety disorder

Status:

Version 1

Abstract

Figures

Introduction

Methods

Participants

Imaging data acquisition

Image preprocessing and quality control

Assessment of symptom severity

Analysis of demographic information

Subcortical volume analysis

Co-variance analysis of subcortical regions

Co-variance analysis between subcortical regions and cortical regions

Results

Discussion

Conclusions

Statements And Declarations

Acknowledgments

References

Additional Declarations

Supplementary Files

Status:

Version 1