Search

We currently observe a disconcerting phenomenon in machine learning studies in psychiatry: While we would expect larger samples to yield better results due to the availability of more data, larger machine learning studies consistently show much weaker performance than the numerous small-scale studies. Here, we systematically investigated this effect focusing on one of the most heavily studied questions in the field, namely the classification of patients suffering from Major Depressive Disorder (MDD) and healthy controls based on neuroimaging data. Drawing upon structural MRI data from a balanced sample of N = 1868 MDD patients and healthy controls from our recent international Predictive Analytics Competition (PAC), we first trained and tested a classification model on the full dataset which yielded an accuracy of 61%. Next, we mimicked the process by which researchers would draw samples of various sizes (N = 4 to N = 150) from the population and showed a strong risk of misestimation. Specifically, for small sample sizes (N = 20), we observe accuracies of up to 95%. For medium sample sizes (N = 100) accuracies up to 75% were found. Importantly, further investigation showed that sufficiently large test sets effectively protect against performance misestimation whereas larger datasets per se do not. While these results question the validity of a substantial part of the current literature, we outline the relatively low-cost remedy of larger test sets, which is readily available in most cases.

Preclinical evidence indicates that the endocannabinoid system is involved in neural responses to reward. This study aimed to investigate associations between basal serum concentrations of the endocannabinoids anandamide (AEA) and 2-arachidonylglycerol (2-AG) with brain functional reward processing. Additionally, a personality measure of reward dependence was obtained. Brain functional data were obtained of 30 right-handed adults by conducting fMRI at 3 Tesla using a reward paradigm. Reward dependence was obtained using the subscale reward dependence of the Tridimensional Personality Questionnaire (TPQ). Basal concentrations of AEA and 2-AG were determined in serum. Analyzing the fMRI data, for AEA and 2-AG ANCOVAs were calculated using a full factorial model, with condition (reward > control, loss > control) and concentrations for AEA and 2-AG as factors. Regression analyses were conducted for AEA and 2-AG on TPQ-RD scores. A whole-brain analysis showed a significant interaction effect of AEA concentration by condition (positive vs. negative) within the putamen (x = 26, y = 16, z = −8, F13.51, TFCE((1, 54)) = 771.68, k = 70, P(FWE) = 0.044) resulting from a positive association of basal AEA concentrations and putamen activity to rewarding stimuli, while this association was absent in the loss condition. AEA concentrations were significantly negatively correlated with TPQ reward dependence scores (r(spearman) = −0.56, P = 0.001). These results show that circulating AEA may modulate brain activation during reward feedback and that the personality measure reward dependence is correlated with AEA concentrations in healthy human volunteers. Future research is needed to further characterize the nature of the lipids’ influence on reward processing, the impact on reward anticipation and outcome, and on vulnerability for psychiatric disorders.

Transgender individuals (TIs) show brain structural alterations that differ from their biological sex as well as their perceived gender. To substantiate evidence that the brain structure of TIs differs from male and female, we use a combined multivariate and univariate approach. Gray matter segments resulting from voxel-based morphometry preprocessing of $N = 1753$ cisgender (CG) healthy participants were used to train ($N=1402$) and validate (20 % hold-out; $N = 351$) a support-vector machine classifying the biological sex. As a second validation, we classified $N = 1104$ patients with depression. A third validation was performed using the matched CG sample of the transgender women (TWs) application-sample. Subsequently, the classifier was applied to $N = 26$ TWs. Finally, we compared brain volumes of CG-men, women and TW-pre/post treatment (cross-sex hormone treatment) in a univariate analysis controlling for sexual orientation, age and total brain volume. The application of our biological sex classifier to the transgender sample resulted in a significantly lower true positive rate (TPR) (TPR-male = 56.0 %). The TPR did not differ between CG-individuals with (TPR-male = 86.9 %) and without depression (TPR-male = 88.5 %). The univariate analysis of the transgender application-sample revealed that TW-pre/post treatment show brain structural differences from CG-women and CG-men in the putamen and insula, as well as the whole-brain analysis. Our results support the hypothesis that brain structure in TW differs from brain structure of their biological sex (male) as well as their perceived gender (female). This finding substantiates evidence that TIs show specific brain structural alterations leading to a different pattern of brain structure than CG-individuals., Content adapted to the publication at Neuropsychopharmacology

Transgender individuals (TIs) show brain-structural alterations that differ from their biological sex as well as their perceived gender. To substantiate evidence that the brain structure of TIs differs from male and female, we use a combined multivariate and univariate approach. Gray matter segments resulting from voxel-based morphometry preprocessing of N = 1753 cisgender (CG) healthy participants were used to train (N = 1402) and validate (20% holdout N = 351) a support-vector machine classifying the biological sex. As a second validation, we classified N = 1104 patients with depression. A third validation was performed using the matched CG sample of the transgender women (TW) application sample. Subsequently, the classifier was applied to N = 26 TW. Finally, we compared brain volumes of CG-men, women, and TW-pre/post treatment cross-sex hormone treatment (CHT) in a univariate analysis controlling for sexual orientation, age, and total brain volume. The application of our biological sex classifier to the transgender sample resulted in a significantly lower true positive rate (TPR-male = 56.0%). The TPR did not differ between CG-individuals with (TPR-male = 86.9%) and without depression (TPR-male = 88.5%). The univariate analysis of the transgender application-sample revealed that TW-pre/post treatment show brain-structural differences from CG-women and CG-men in the putamen and insula, as well as the whole-brain analysis. Our results support the hypothesis that brain structure in TW differs from brain structure of their biological sex (male) as well as their perceived gender (female). This finding substantiates evidence that TIs show specific brain-structural alterations leading to a different pattern of brain structure than CG-individuals.

DNA methylation profiles of the serotonin transporter gene (SLC6A4) have been shown to alter SLC6A4 expression, drive antidepressant treatment response and modify brain functions. This study investigated whether methylation of an AluJb element in the SLC6A4 promotor was associated with major depressive disorder (MDD), amygdala reactivity to emotional faces, 5-HTTLPR/rs25531 polymorphism, and recent stress. MDD patients (n=122) and healthy controls (HC, n=176) underwent fMRI during an emotional face-matching task. Individual SLC6A4 AluJb methylation profiles were ascertained and associated with MDD, amygdala reactivity, 5-HTTLPR/rs25531, and stress. SLC6A4 AluJb methylation was significantly lower in MDD compared to HC and in stressed compared to less stressed participants. Lower AluJb methylation was particularly found in 5-HTTLPR/rs25531 risk allele carriers under stress and correlated with less depressive episodes. fMRI analysis revealed a significant interaction of AluJb methylation and diagnosis in the amygdala, with MDD patients showing lower AluJb methylation associated with decreased amygdala reactivity. While no joint effect of AluJb methylation and 5-HTTLPR/rs25531 existed, risk allele carriers showed significantly increased bilateral amygdala activation. These findings suggest a role of SLC6A4 AluJb methylation in MDD, amygdala reactivity, and stress reaction, partly interwoven with 5-HTTLPR/rs25531 effects. Patients with low methylation in conjunction with a shorter MDD history and decreased amygdala reactivity might feature a more stress-adaptive epigenetic process, maybe via theoretically possible endogenous antidepressant-like effects. In contrast, patients with higher methylation might possibly suffer from impaired epigenetic adaption to chronic stress. Further, the 5-HTTLPR/rs25531 association with amygdala activation was confirmed in our large sample.

Adolescent-onset major depressive disorder (MDD) is associated with an increased risk of recurrent depressive episodes, suicidal behaviors, and psychiatric morbidity throughout the lifespan. The objective of the present study was to investigate brain structural and functional changes in adolescent patients with MDD. Furthermore, we aimed to clarify the influence of early-life stress on brain function and structure. The study investigated adolescent patients with severe MDD (n=20, mean age=16.0, range=15-18 years) and a control sample of matched healthy adolescents (n=21, mean age=16.6, range=15-18 years). Functional MRI data were obtained using a face-matching paradigm to investigate emotion processing. Structural MRI data were analyzed using voxel-based morphometry (VBM). In line with previous studies on adult MDD, adolescent patients showed elevated amygdala activity to negative and reduced amygdala activity to positive emotional stimuli. Furthermore, MDD patients showed smaller hippocampal volumes compared to healthy adolescents. Higher levels of childhood maltreatment were associated with smaller hippocampal volumes in both depressed patients and healthy controls, whereby no associations between amygdala reactivity and childhood maltreatment were found. Our results suggest that hippocampal alterations in youth MDD patients may at least partly be traced back to higher occurrence of early-life adverse experiences. Regarding the strong morphometric impact of childhood maltreatment and its distinctly elevated prevalence in MDD populations, this study provides an alternative explanation for frequently observed limbic structural abnormalities in depressed patients.

Distinguishing bipolar disorder from major depressive disorder is a major challenge in psychiatric treatment. Consequently, there has been growing interest in identifying neuronal biomarkers of disorder-specific pathophysiological processes to differentiate affective disorders. Thirty-six depressed bipolar patients, 36 depressed unipolar patients, and 36 matched healthy controls (HCs) participated in an fMRI experiment. Emotional faces served as stimuli in a matching task. We investigated neural activation towards angry, fearful, and happy faces focusing on prototypical regions related to emotion processing, ie, the amygdala and the anterior cingulate gyrus (ACG). Furthermore, we employed a whole-brain and a multivariate pattern classification analysis. Unipolar patients showed abnormally reduced ACG activation toward happy and fearful faces compared with bipolar patients and HCs respectively. Furthermore, the whole-brain analysis revealed significantly increased activation in bipolar patients compared with unipolar patients in the fearful condition in the right frontal and parietal cortex. Moreover, the multivariate pattern classification analysis yielded significant classification rates of up to 72% based on ACG activation elicited by fearful faces. Our results question the rather ‘amygdalocentric' neurobiological models of mood disorders. We observed patterns of abnormally reduced ventral and supragenual ACG activation, potentially indicating impaired bottom-up emotion processing and automatic emotion regulation specifically in unipolar but not in bipolar individuals.

Social anxiety disorder (SAD) is a commonly occurring and highly disabling disorder. The neuropeptide oxytocin and its receptor (OXTR) have been implicated in social cognition and behavior. This study-for the first time applying a multilevel epigenetic approach-investigates the role of OXTR gene methylation in categorical, dimensional, and intermediate neuroendocrinological/neural network phenotypes of social anxiety. A total of 110 unmedicated patients with SAD and matched 110 controls were analyzed for OXTR methylation by direct sequencing of sodium bisulfite-converted DNA extracted from whole blood. Furthermore, OXTR methylation was investigated regarding SAD-related traits (Social Phobia Scale (SPS) and Social Interaction Anxiety Scale (SIAS)), salivary cortisol response during the Trier social stress test (TSST), and amygdala responsiveness to social phobia related verbal stimuli using fMRI. Significantly decreased OXTR methylation particularly at CpG Chr3: 8 809 437 was associated with (1) the categorical phenotype of SAD (p

Adenosine A1 receptors (A1ARs) and the interacting adenosine A2A receptors are implicated in neurological and psychiatric disorders. Variants within the corresponding genes ADORA1 and ADORA2A were shown associated with pathophysiologic alterations, particularly increased anxiety. It is unknown so far, if these variants might modulate the A1AR distribution and availability in different brain regions. In this pilot study, the influence of ADORA1 and ADORA2A variants on in vivo A1AR binding was assessed with the A1AR-selective positron emission tomography (PET) radioligand [(18)F]CPFPX in brains of healthy humans. Twenty-eight normal control subjects underwent PET procedures to calculate the binding potential BPND of [(18)F]CPFPX in cerebral regions and to assess ADORA1 and ADORA2A single nucleotide polymorphism (SNP) effects on regional BPND data. Our results revealed SNPs of both genes associated with [(18)F]CPFPX binding to the A1AR. The strongest effects that withstood even Bonferroni correction of multiple SNP testing were found in non-smoking subjects (N=22) for ADORA2A SNPs rs2236624 and rs5751876 (corr. Pall

Reduced hippocampal volumes are probably the most frequently reported structural neuroimaging finding associated with major depressive disorder (MDD). However, it remains unclear whether altered hippocampal structure represents a risk factor for or a consequence of MDD. Reduced hippocampal volumes were consistently reported in subjects affected by childhood maltreatment. As the prevalence of childhood maltreatment is highly elevated in MDD populations, previous morphometric findings regarding hippocampal atrophy in MDD therefore might have been confounded by maltreatment experiences. The aim of this study was to differentiate the impact of childhood maltreatment from the influence of MDD diagnosis on hippocampal morphometry. Depressed patients (85) as well as 85 age- and sex-matched healthy controls underwent structural MRI. The Childhood Trauma Questionnaire was administered to estimate experiences of childhood maltreatment. Hippocampal volume and surface structure was examined by the use of two independent methods, automated segmentation (FSL-FIRST) and voxel-based morphometry (VBM8). In line with existing studies, MDD patients showed reduced hippocampal volumes, and childhood maltreatment was consistently associated with hippocampal volume loss in both, patients and healthy controls. However, no analysis revealed significant morphological differences between patients and controls if maltreatment experience was regressed out. Our results suggest that hippocampal alterations in MDD patients may at least partly be traced back to higher occurrence of early-life adverse experiences. Regarding the strong morphometric impact of childhood maltreatment and its distinctly elevated prevalence in MDD populations, this study provides an alternative explanation for frequently observed limbic structural abnormalities in depressed patients.

According to preclinical studies, glutamate has been implicated in the pathogenesis of anxiety. In order to elucidate the role of glutamate in anxiety and panic in humans, brain glutamate+glutamine (Glx) levels were measured during cholecystokinin-tetrapeptide (CCK-4)-induced panic using magnetic resonance spectroscopy (MRS). Eighteen healthy subjects underwent a CCK-4 challenge. MR spectra were obtained from the anterior cingulate cortex (ACC) using a single voxel point-resolved spectroscopy method and analyzed using LCModel. A combined fitting of Glx was performed. Panic was assessed using the Acute Panic Inventory (API) and Panic Symptom Scale (PSS) scores. Moreover, hypothalamic–pituitary–adrenal axis stimulation was monitored throughout the challenge. There was a significant panic response following CCK-4 as revealed by a marked increase in both the panic scores (API: F(1,17)=149.41; p

Emergence of suicidal ideation (TESI) during treatment with antidepressants in major depression led to a black box warning. We performed a genome-wide association study to identify genetic markers, which increase the risk for this serious side effect. TESI was evaluated in depressed in-patients (N=397) and defined by an emergence of suicidal thoughts during hospitalization without suicidal thoughts at admission using the suicide item (3) of the Hamilton Depression Rating Scale. Genotype distribution of 405.383 single-nucleotide polymorphisms (SNPs) in patients with TESI (N=32/8.1%) was compared to patients without increase in suicidal ideation (N=329/82.9%) and to a subgroup never reported suicidal ideation (N=79/19.9%). Top results were analyzed in an independent sample (N=501). None variant reached genome-wide significance, the best associated SNP was rs1630535 (p-value=1.3 × 10(-7)). The top 79 SNPs could be analyzed in an independent sample, and 14 variants showed nominal significant association with the same risk allele in the replication sample. A discriminant analysis classifying patients using these 79 SNPs revealed a 91% probability to classify TESI vs non-TESI cases correctly in the replication sample. Although our data need to be interpreted carefully owing to the small numbers in both cohorts, they suggest that a combination of genetic markers might indeed be used to identify patients at risk for TESI.

Recently, several genome-wide association studies (GWAS) on bipolar disorder (BPD) suggested novel risk genes. However, only few of them were followed up and further, the specificity of these genes is even more elusive. To address these issues, we genotyped SNPs in ANK3, CACNA1C, CMTM8, DGKH, EGFR, and NPAS3, which were significantly associated with BPD in previous GWAS, in a sample of 380 BPD patients. Replicated SNPs were then followed up in patients suffering from unipolar depression (UPD; n=387) or adult attention-deficit/hyperactivity disorder (aADHD; n=535). While we could not confirm an association of ANK3, CACNA1C, and EGFR with BPD, 10 SNPs in DGKH, CMTM8, and NPAS3 were nominally associated with disease, with two DGKH markers surviving correction for multiple testing. When these were followed up in UPD and aADHD, seven DGKH SNPs were also associated with UPD, while one SNP each in NPAS3 and CMTM8 and four in DGKH were linked to aADHD. Furthermore, a DGKH haplotype consisting of rs994856/rs9525580/rs9525584 GAT was associated with all disorders tested, while the complementary AGC haplotype was protective. The corresponding haploblock spans a 27-kb region covering exons coding for amino acids 65-243, and thus might include functional variants yet to be identified. We demonstrate an association of DGKH with BPD, UPD, and aADHD by applying a two-stage design. These disorders share the feature of mood instability, so that this phenotype might be associated with genetic variation in DGKH.Neuropsychopharmacology advance online publication, 8 June 2011; doi:10.1038/npp.2011.98.

Recent studies point to a role of neuropeptide-S (NPS) in the etiology of anxiety disorders. In animal models, NPS and its receptor (NPSR) were shown to be highly expressed in the amygdala, a central structure in the fear circuit, also known to be hyper-responsive in anxiety disorders. Recently, a functional polymorphism in the NPSR gene (rs324981 A/T) has been associated with panic disorder and anxiety sensitivity. However, the role of NPSR gene variation in the modulation of fear-related amygdala responsiveness remains to be clarified. In 79 healthy subjects genotyped for NPSR rs324981, amygdala responses were assessed by means of fMRI. The participants were presented with fear-relevant faces in a robust emotion-processing paradigm frequently used to study amygdala responsiveness. We observed a strong association of NPSR T-alleles with right amygdala responsiveness to fear-relevant faces. The association peak was located in the BLA. Furthermore, responsiveness to aversive stimuli within this BLA cluster predicted a participant's self-reported harm avoidance but not depression level. We conclude that NPSR genotype is associated with increased amygdala responsiveness to fear-relevant stimuli. Thereby, NPSR rs324981 apparently causes an indirect effect on anxiety-related traits and potentially contributes to the pathogenesis of anxiety disorders by shaping fear-related limbic activity.

Negatively balanced neurotrophic factors may be important in precipitating clinical depression. Recently, it has been reported that antidepressant therapy may exert positive neurotrophic effects. The aim of this study was to detect probable neurotrophic changes during electroconvulsive therapy (ECT). For this purpose, N-acetylaspartate (NAA), an amino acid exclusively located in neurons, and other brain metabolites such as glutamine/glutamate (Glx), choline (Cho), and creatine (Cr) were measured in patients by localized proton magnetic resonance spectroscopy. A total of 28 severely depressed patients (DSM-IV) were enrolled, and the left amygdalar region was investigated by proton STEAM spectroscopy before and after unilateral ECT. The results were compared with 28 age- and gender-matched controls using nonparametric paired and unpaired tests. A significant increase in NAA was observed only in ECT responders (n=14; p=0.019). Five out of 14 nonresponders to ECT monotherapy were remeasured following a clinical improvement after continued ECT combined with antidepressants and were then found also to present a significant increase in NAA. In all successfully treated patients, parallel observations, that is, increased levels, were made for Glx, whereas Cho and Cr were unchanged. In conclusion, our preliminary finding of increased NAA concentrations after successful ECT may indicate a probable neurotrophic effect of ECT.