Methods and Challenges for Assessing Heterogeneity

A machine learning personalized treatment rule to optimize assignment to psychotherapies for grief among veterans

BACKGROUND

Complex grief patterns are associated with significant suffering, functional impairments, health and mental health problems, and increased healthcare use. This burden may be even more pronounced among veterans. Behavioral Activation and Therapeutic Exposure (BATE-G) and Cognitive Therapy for Grief (CT-G) are two evidence-based interventions for grief. The goal of this study was to use a precision medicine approach to develop a personalized treatment rule to optimize assignment among these psychotherapies.

METHODS

We analyzed data (N = 155) from a randomized clinical trial comparing BATE-G and CT-G. Outcome weighted learning was used to estimate an optimal personalized treatment rule. Baseline characteristics including demographics, social support, variables related to the death, and psychopathology dimensions were used as prescriptive factors of treatment assignment.

RESULTS

The estimated rule assigned 72 veterans to CT-G and 56 to BATE-G. Assigning participants according to this rule was estimated to lead to markedly lower mean grief level following 6 months from treatment compared to assigning everyone to either BATE-G (Vdopt - VBATE-G = -18.57 [95 % CI: -29.41, -7.72]) or CT-G (Vdopt - VBATE-G = -20.89 [95 % CI: -30.7, -11.07]) regardless of their characteristics.

LIMITATIONS

Participants were primarily male veterans, and identified with Black or White race. The estimated rule was not externally validated.

CONCLUSION

The estimated rule used relatively simple, easily accessible, client characteristics to personalize assignment to treatment using a precision medicine approach based on machine learning and causal inference. Upon further validation, such a rule can be easily implemented in clinical practice to prescriptively maximize treatment benefits.

Dimensional Neuroimaging Endophenotypes: Neurobiological Representations of Disease Heterogeneity Through Machine Learning

Machine learning has been increasingly used to obtain individualized neuroimaging signatures for disease diagnosis, prognosis, and response to treatment in neuropsychiatric and neurodegenerative disorders. Therefore, it has contributed to a better understanding of disease heterogeneity by identifying disease subtypes with different brain phenotypic measures. In this review, we first present a systematic literature overview of studies using machine learning and multimodal magnetic resonance imaging to unravel disease heterogeneity in various neuropsychiatric and neurodegenerative disorders, including Alzheimer's disease, schizophrenia, major depressive disorder, autism spectrum disorder, and multiple sclerosis, as well as their potential in a transdiagnostic framework, where neuroanatomical and neurobiological commonalities were assessed across diagnostic boundaries. Subsequently, we summarize relevant machine learning methodologies and their clinical interpretability. We discuss the potential clinical implications of the current findings and envision future research avenues. Finally, we discuss an emerging paradigm called dimensional neuroimaging endophenotypes. Dimensional neuroimaging endophenotypes dissects the neurobiological heterogeneity of neuropsychiatric and neurodegenerative disorders into low-dimensional yet informative, quantitative brain phenotypic representations, serving as robust intermediate phenotypes (i.e., endophenotypes), presumably reflecting the interplay of underlying genetic, lifestyle, and environmental processes associated with disease etiology.

Childhood adversities characterize the heterogeneity in the brain pattern of individuals during neurodevelopment

BACKGROUND

Several factors shape the neurodevelopmental trajectory. A key area of focus in neurodevelopmental research is to estimate the factors that have maximal influence on the brain and can tip the balance from typical to atypical development.

METHODS

Utilizing a dissimilarity maximization algorithm on the dynamic mode decomposition (DMD) of the resting state functional MRI data, we classified subjects from the cVEDA neurodevelopmental cohort (n = 987, aged 6-23 years) into homogeneously patterned DMD (representing typical development in 809 subjects) and heterogeneously patterned DMD (indicative of atypical development in 178 subjects).

RESULTS

Significant DMD differences were primarily identified in the default mode network (DMN) regions across these groups (p < 0.05, Bonferroni corrected). While the groups were comparable in cognitive performance, the atypical group had more frequent exposure to adversities and faced higher abuses (p < 0.05, Bonferroni corrected). Upon evaluating brain-behavior correlations, we found that correlation patterns between adversity and DMN dynamic modes exhibited age-dependent variations for atypical subjects, hinting at differential utilization of the DMN due to chronic adversities.

CONCLUSION

Adversities (particularly abuse) maximally influence the DMN during neurodevelopment and lead to the failure in the development of a coherent DMN system. While DMN's integrity is preserved in typical development, the age-dependent variability in atypically developing individuals is contrasting. The flexibility of DMN might be a compensatory mechanism to protect an individual in an abusive environment. However, such adaptability might deprive the neural system of the faculties of normal functioning and may incur long-term effects on the psyche.

Preliminary Validation of a General Factor Model of Chronic Overlapping Pain Conditions

Chronic overlapping pain conditions (COPCs) by definition, frequently co-occur, perhaps reflecting their shared etiologies. Their overlapping nature presents a methodological challenge, possibly masking associations between COPCs and health outcomes attributable to either general or specific processes. To address this challenge, we used population-based cohort data to evaluate the predictive validity of a bifactor model of 9 self-reported COPCs by assessing its association with incident pain-related clinical diagnoses; pain-relevant pharmacotherapy; and other health outcomes. We obtained data from a 2005 to 2006 study of Swedish adult twins linked with health data from nationwide registers through 2016 (N = 25,418). We then fit a bifactor model comprising a general COPC factor and 2 independent specific factors measuring pain-related somatic symptoms and neck and shoulder pain. Accounting for age, biological sex, and cancer, the general factor was associated with increased risk of all pain-related outcomes (eg, COPC diagnosis adjusted odds ratio [aOR], 1.71; 95% confidence interval [1.62, 1.81]), most mental health-related outcomes (eg, depression aOR, 1.72 [1.60, 1.85]), and overdose and mortality (eg, all-cause mortality aOR, 1.25 [1.09, 1.43]). The somatic symptoms specific factor was associated with pain-relevant pharmacotherapy (eg, prescribed opioids aOR, 1.25 [1.15, 1.36]), most mental health-related outcomes (eg, depression aOR, 1.95 [1.70, 2.23]), and overdose (eg, nonfatal overdose aOR, 1.66 [1.31, 2.10]). The neck and shoulder pain-specific factor was weakly and inconsistently associated with the outcomes. Findings provide initial support for the validity and utility of a general-factor model of COPCs as a tool to strengthen understanding of co-occurrence, etiology, and consequences of chronic pain. PERSPECTIVE: This article presents associations between a novel measurement model of COPCs and various health outcomes. Findings provide support for measuring pain across multiple domains rather than only measuring pain specific to one physical location in both research and clinical contexts.

Mapping Neurophysiological Subtypes of Major Depressive Disorder Using Normative Models of the Functional Connectome

BACKGROUND

Major depressive disorder (MDD) is a highly heterogeneous disorder that typically emerges in adolescence and can occur throughout adulthood. Studies aimed at quantitatively uncovering the heterogeneity of individual functional connectome abnormalities in MDD and identifying reproducibly distinct neurophysiological MDD subtypes across the lifespan, which could provide promising insights for precise diagnosis and treatment prediction, are still lacking.

METHODS

Leveraging resting-state functional magnetic resonance imaging data from 1148 patients with MDD and 1079 healthy control participants (ages 11-93), we conducted the largest multisite analysis to date for neurophysiological MDD subtyping. First, we characterized typical lifespan trajectories of functional connectivity strength based on the normative model and quantitatively mapped the heterogeneous individual deviations among patients with MDD. Then, we identified neurobiological MDD subtypes using an unsupervised clustering algorithm and evaluated intersite reproducibility. Finally, we validated the subtype differences in baseline clinical variables and longitudinal treatment predictive capacity.

RESULTS

Our findings indicated great intersubject heterogeneity in the spatial distribution and severity of functional connectome deviations among patients with MDD, which inspired the identification of 2 reproducible neurophysiological subtypes. Subtype 1 showed severe deviations, with positive deviations in the default mode, limbic, and subcortical areas and negative deviations in the sensorimotor and attention areas. Subtype 2 showed a moderate but converse deviation pattern. More importantly, subtype differences were observed in depressive item scores and the predictive ability of baseline deviations for antidepressant treatment outcomes.

CONCLUSIONS

These findings shed light on our understanding of different neurobiological mechanisms underlying the clinical heterogeneity of MDD and are essential for developing personalized treatments for this disorder.

Data-driven multivariate identification of gyrification patterns in a transdiagnostic patient cohort: A cluster analysis approach

BACKGROUND

Multivariate data-driven statistical approaches offer the opportunity to study multi-dimensional interdependences between a large set of biological parameters, such as high-dimensional brain imaging data. For gyrification, a putative marker of early neurodevelopment, direct comparisons of patterns among multiple psychiatric disorders and investigations of potential heterogeneity of gyrification within one disorder and a transdiagnostic characterization of neuroanatomical features are lacking.

METHODS

In this study we used a data-driven, multivariate statistical approach to analyze cortical gyrification in a large cohort of N = 1028 patients with major psychiatric disorders (Major depressive disorder: n = 783, bipolar disorder: n = 129, schizoaffective disorder: n = 44, schizophrenia: n = 72) to identify cluster patterns of gyrification beyond diagnostic categories.

RESULTS

Cluster analysis applied on gyrification data of 68 brain regions (DK-40 atlas) identified three clusters showing difference in overall (global) gyrification and minor regional variation (regions). Newly, data-driven subgroups are further discriminative in cognition and transdiagnostic disease risk factors.

CONCLUSIONS

Results indicate that gyrification is associated with transdiagnostic risk factors rather than diagnostic categories and further imply a more global role of gyrification related to mental health than a disorder specific one. Our findings support previous studies highlighting the importance of association cortices involved in psychopathology. Explorative, data-driven approaches like ours can help to elucidate if the brain imaging data on hand and its a priori applied grouping actually has the potential to find meaningful effects or if previous hypotheses about the phenotype as well as its grouping have to be revisited.

Mechanisms of cognitive disinhibition explain individual differences in adult attention deficit hyperactivity disorder traits

BACKGROUND

Attention deficit hyperactivity disorder (ADHD) in adults is strongly associated with psychiatric comorbidity and functional impairment. Here, we aimed to use a newly developed online cognitive battery with strong psychometric properties for measuring individual differences in three cognitive mechanisms proposed to underlie ADHD traits in adults: 1) attentional control - the ability to mobilize cognitive resources to stop a prepotent motor response; 2) information sampling/gathering - adequate sampling of information in a stimulus detection task prior to making a decision; and 3) shifting - the ability to adapt behavior in response to positive and negative contingencies.

METHODS

This cross-sectional and correlational study recruited 650 adults (330 males) aged 18-69 years (M = 33.06; MD = 31.00; SD = 10.50), with previously diagnosed ADHD (n = 329) and those from the general community without a history of ADHD (n = 321). Self-report measures of ADHD traits (i.e., inattention/disorganization, impulsivity, hyperactivity) and the cognitive battery were completed online.

RESULTS

Latent class analysis, exploratory structural equation modeling and factor mixture modeling revealed self-reported ADHD traits formed a unidimensional and approximately normally distributed phenotype. Bayesian structural equation modeling demonstrated that all three mechanisms measured by the cognitive battery, explained unique, incremental variance in ADHD traits, with a total of 15.9% explained in the ADHD trait factor.

CONCLUSIONS

Attentional control and shifting, as well as the less researched cognitive process of information gathering, explain individual difference variance in self-reported ADHD traits with potential to yield genetic and neurobiological insights into adult ADHD.

Common and disorder-specific cortical thickness alterations in internalizing, externalizing and thought disorders during early adolescence: an Adolescent Brain and Cognitive Development study

BACKGROUND

A growing body of neuroimaging studies has reported common neural abnormalities among mental disorders in adults. However, it is unclear whether the distinct disorder-specific mechanisms operate during adolescence despite the overlap among disorders.

METHODS

We studied a large cohort of more than 11 000 preadolescent (age 9-10 yr) children from the Adolescent Brain and Cognitive Development cohort. We adopted a regrouping approach to compare cortical thickness (CT) alterations and longitudinal changes between healthy controls (n = 4041) and externalizing (n = 1182), internalizing (n = 1959) and thought disorder (n = 347) groups. Genome-wide association study (GWAS) was performed on regional CT across 4468 unrelated European youth.

RESULTS

Youth with externalizing or internalizing disorders exhibited increased regional CT compared with controls. Externalizing (p = 8 × 10-4, Cohen d = 0.10) and internalizing disorders (p = 2 × 10-3, Cohen d = 0.08) shared thicker CT in the left pars opercularis. The somatosensory and the primary auditory cortex were uniquely affected in externalizing disorders, whereas the primary motor cortex and higher-order visual association areas were uniquely affected in internalizing disorders. Only youth with externalizing disorders showed decelerated cortical thinning from age 10-12 years. The GWAS found 59 genome-wide significant associated genetic variants across these regions. Cortical thickness in common regions was associated with glutamatergic neurons, while internalizing-specific regional CT was associated with astrocytes, oligodendrocyte progenitor cells and GABAergic neurons.

LIMITATIONS

The sample size of the GWAS was relatively small.

CONCLUSION

Our study provides strong evidence for the presence of specificity in CT, developmental trajectories and underlying genetic underpinnings among externalizing and internalizing disorders during early adolescence. Our results support the neurobiological validity of the regrouping approach that could supplement the use of a dimensional approach in future clinical practice.

An overview of clustering methods with guidelines for application in mental health research

Cluster analyzes have been widely used in mental health research to decompose inter-individual heterogeneity by identifying more homogeneous subgroups of individuals. However, despite advances in new algorithms and increasing popularity, there is little guidance on model choice, analytical framework and reporting requirements. In this paper, we aimed to address this gap by introducing the philosophy, design, advantages/disadvantages and implementation of major algorithms that are particularly relevant in mental health research. Extensions of basic models, such as kernel methods, deep learning, semi-supervised clustering, and clustering ensembles are subsequently introduced. How to choose algorithms to address common issues as well as methods for pre-clustering data processing, clustering evaluation and validation are then discussed. Importantly, we also provide general guidance on clustering workflow and reporting requirements. To facilitate the implementation of different algorithms, we provide information on R functions and libraries.

A graph theoretic approach to neurodegeneration: five data-driven neuropsychological subtypes in mild cognitive impairment

Mild cognitive Impairment (MCI) is notoriously heterogenous in terms of clinical presentation, neuroimaging correlates, and subsequent progression. Predicting who will progress to dementia, which type of dementia, and over what timeframe is challenging. Previous work has attempted to identify MCI subtypes using neuropsychological measures in an effort to address this challenge; however, there is no consensus on approach, which may account for some of the variability. Using a hierarchical community detection approach, we examined cognitive subtypes within an MCI sample (from the Alzheimer's Disease Neuroimaging Initiative [ADNI] study). We then examined whether these subtypes were related to biomarkers (e.g., cortical volumes, fluorodeoxyglucose (FDG)-positron emission tomography (PET) hypometabolism) or clinical progression. We identified five communities (i.e., cognitive subtypes) within the MCI sample: 1) predominantly memory impairment, 2) predominantly language impairment, 3) cognitively normal, 4) multidomain, with notable executive dysfunction, 5) multidomain, with notable processing speed impairment. Community membership was significantly associated with 1) cortical volume in the hippocampus, entorhinal cortex, and fusiform cortex; 2) FDG PET hypometabolism in the posterior cingulate, angular gyrus, and inferior/middle temporal gyrus; and 3) conversion to dementia at follow up. Overall, community detection as an approach appears a viable method for identifying unique cognitive subtypes in a neurodegenerative sample that were linked to several meaningful biomarkers and modestly with progression at one year follow up.

Functional connectomics in depression: insights into therapies

Depression is a common mental disorder characterized by heterogeneous cognitive and behavioral symptoms. The emerging research paradigm of functional connectomics has provided a quantitative theoretical framework and analytic tools for parsing variations in the organization and function of brain networks in depression. In this review, we first discuss recent progress in depression-associated functional connectome variations. We then discuss treatment-specific brain network outcomes in depression and propose a hypothetical model highlighting the advantages and uniqueness of each treatment in relation to the modulation of specific brain network connectivity and symptoms of depression. Finally, we look to the future promise of combining multiple treatment types in clinical practice, using multisite datasets and multimodal neuroimaging approaches, and identifying biological depression subtypes.

Specificity of polygenic signatures across symptom dimensions in bipolar disorder: an analysis of UK Bipolar Disorder Research Network data

BACKGROUND

Current definitions and clinical heterogeneity in bipolar disorder are major concerns as they obstruct aetiological research and impede drug development. Therefore, stratification of bipolar disorder is a high priority. To inform stratification, our analysis aimed to examine the patterns and relationships between polygenic liability for bipolar disorder, major depressive disorder (MDD), and schizophrenia with multidimensional symptom representations of bipolar disorder.

METHODS

In this analysis, data from the UK Bipolar Disorder Research Network (BDRN) were assessed with the Operational Checklist for Psychotic Disorders. Individuals with bipolar disorder as defined in DSM-IV, of European ancestry (self-reported), aged 18 years or older at time of interview, living in the UK, and registered with the BDRN were eligible for inclusion. Psychopathological variables obtained via interview by trained research psychologists or psychiatrists and psychiatric case notes were used to identify statistically distinct symptom dimensions, calibrated with exploratory factor analysis and validated with confirmatory factor analysis (CFA). CFA was extended to include three polygenic risk scores (PRSs) indexing liability for bipolar disorder, MDD, and schizophrenia in a multiple indicator multiple cause (MIMIC) structural equation model to estimate PRS relationships with symptom dimensions.

FINDINGS

Of 4198 individuals potentially eligible for inclusion, 4148 (2804 [67·6%] female individuals and 1344 [32·4%] male individuals) with a mean age at interview of 45 years (SD 12·03) were available for analysis. Three reliable dimensions (mania, depression, and psychosis) were identified. The MIMIC model fitted the data well (root mean square error of approximation 0·021, 90% CI 0·019-0·023; comparative fit index 0·99) and suggests statistically distinct symptom dimensions also have distinct polygenic profiles. The PRS for MDD was strongly associated with the depression dimension (standardised β 0·125, 95% CI 0·080-0·171) and the PRS for schizophrenia was strongly associated with the psychosis dimension (0·108, 0·082-0·175). For the mania dimension, the PRS for bipolar disorder was weakly associated (0·050, 0·002-0·097).

INTERPRETATION

Our findings support the hypothesis that genetic heterogeneity underpins clinical heterogeneity, suggesting that different symptom dimensions within bipolar disorder have partly distinct causes. Furthermore, our results suggest that a specific symptom dimension has a similar cause regardless of the primary psychiatric diagnosis, supporting the use of symptom dimensions in precision psychiatry.

FUNDING

Wellcome Trust and UK Medical Research Council.

Functional activity level reported by an informant is an early predictor of Alzheimer's disease

BACKGROUND

Loss of autonomy in day-to-day functioning is one of the feared outcomes of Alzheimer's disease (AD), and relatives may have been worried by subtle behavioral changes in ordinary life situations long before these changes are given medical attention. In the present study, we ask if such subtle changes should be given weight as an early predictor of a future AD diagnosis.

METHODS

Longitudinal data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) were used to define a group of adults with a mild cognitive impairment (MCI) diagnosis remaining stable across several visits (sMCI, n=360; 55-91 years at baseline), and a group of adults who over time converted from having an MCI diagnosis to an AD diagnosis (cAD, n=320; 55-88 years at baseline). Eleven features were used as input in a Random Forest (RF) binary classifier (sMCI vs. cAD) model. This model was tested on an unseen holdout part of the dataset, and further explored by three different permutation-driven importance estimates and a comprehensive post hoc machine learning exploration.

RESULTS

The results consistently showed that measures of daily life functioning, verbal memory function, and a volume measure of hippocampus were the most important predictors of conversion from an MCI to an AD diagnosis. Results from the RF classification model showed a prediction accuracy of around 70% in the test set. Importantly, the post hoc analyses showed that even subtle changes in everyday functioning noticed by a close informant put MCI patients at increased risk for being on a path toward the major cognitive impairment of an AD diagnosis.

CONCLUSION

The results showed that even subtle changes in everyday functioning should be noticed when reported by relatives in a clinical evaluation of patients with MCI. Information of these changes should also be included in future longitudinal studies to investigate different pathways from normal cognitive aging to the cognitive decline characterizing different stages of AD and other neurodegenerative disorders.

Evaluating the predictive performance of subtyping: A criterion for cluster mean-based prediction

Heterogeneity is a frequent issue in population data analyses in medicine, biology, and the social sciences. A common approach for handling heterogeneity is to use a clustering algorithm to group similar samples, considering samples within the same group to be homogeneous. This approach is known as "subtyping" or "subgrouping." Methods for evaluating the validity of subtyping have yet to be fully established. In this study, we propose the cost of cluster mean-based prediction (CCMP) as a metric for evaluating the accuracy of predictions based on subtyping. By selecting the minimum CCMP among several candidate clustering results, the optimal subtype classification in terms of prediction accuracy can be determined. The computational implementation of the CCMP is validated with numerical experiments. We also examine some properties of subtype classification selected by CCMP.

Simulating developmental diversity: Impact of neural stochasticity on atypical flexibility and hierarchy

Introduction

Investigating the pathological mechanisms of developmental disorders is a challenge because the symptoms are a result of complex and dynamic factors such as neural networks, cognitive behavior, environment, and developmental learning. Recently, computational methods have started to provide a unified framework for understanding developmental disorders, enabling us to describe the interactions among those multiple factors underlying symptoms. However, this approach is still limited because most studies to date have focused on cross-sectional task performance and lacked the perspectives of developmental learning. Here, we proposed a new research method for understanding the mechanisms of the acquisition and its failures in hierarchical Bayesian representations using a state-of-the-art computational model, referred to as in silico neurodevelopment framework for atypical representation learning.

Methods

Simple simulation experiments were conducted using the proposed framework to examine whether manipulating the neural stochasticity and noise levels in external environments during the learning process can lead to the altered acquisition of hierarchical Bayesian representation and reduced flexibility.

Results

Networks with normal neural stochasticity acquired hierarchical representations that reflected the underlying probabilistic structures in the environment, including higher-order representation, and exhibited good behavioral and cognitive flexibility. When the neural stochasticity was high during learning, top-down generation using higher-order representation became atypical, although the flexibility did not differ from that of the normal stochasticity settings. However, when the neural stochasticity was low in the learning process, the networks demonstrated reduced flexibility and altered hierarchical representation. Notably, this altered acquisition of higher-order representation and flexibility was ameliorated by increasing the level of noises in external stimuli.

Discussion

These results demonstrated that the proposed method assists in modeling developmental disorders by bridging between multiple factors, such as the inherent characteristics of neural dynamics, acquisitions of hierarchical representation, flexible behavior, and external environment.

Leveraging Machine Learning for Gaining Neurobiological and Nosological Insights in Psychiatric Research

Much attention is currently devoted to developing diagnostic classifiers for mental disorders. Complementing these efforts, we highlight the potential of machine learning to gain biological insights into the psychopathology and nosology of mental disorders. Studies to this end have mainly used brain imaging data, which can be obtained noninvasively from large cohorts and have repeatedly been argued to reveal potentially intermediate phenotypes. This may become particularly relevant in light of recent efforts to identify magnetic resonance imaging-derived biomarkers that yield insight into pathophysiological processes as well as to refine the taxonomy of mental illness. In particular, the accuracy of machine learning models may be used as dependent variables to identify features relevant to pathophysiology. Moreover, such approaches may help disentangle the dimensional (within diagnosis) and often overlapping (across diagnoses) symptomatology of psychiatric illness. We also point out a multiview perspective that combines data from different sources, bridging molecular and system-level information. Finally, we summarize recent efforts toward a data-driven definition of subtypes or disease entities through unsupervised and semisupervised approaches. The latter, blending unsupervised and supervised concepts, may represent a particularly promising avenue toward dissecting heterogeneous categories. Finally, we raise several technical and conceptual aspects related to the reviewed approaches. In particular, we discuss common pitfalls pertaining to flawed input data or analytic procedures that would likely lead to unreliable outputs.

Symptom-Based Profiling and Multimodal Neuroimaging of a Large Preteenage Population Identifies Distinct Obsessive-Compulsive Disorder-like Subtypes With Neurocognitive Differences

BACKGROUND

Obsessive-compulsive disorder (OCD) is characterized by both internalizing (anxiety) and externalizing (compulsivity) symptoms. Currently, little is known about their interrelationships and their relative contributions to disease heterogeneity. Our goal is to resolve affective and cognitive symptom heterogeneity related to internalized and externalized symptom dimensions by determining subtypes of children with OCD symptoms, and to identify any corresponding neural differences.

METHODS

A total of 1269 children with OCD symptoms screened using the Child Behavior Checklist Obsessive-Compulsive Symptom scale and 3987 matched control subjects were obtained from the Adolescent Brain Cognitive Development (ABCD) Study. Consensus hierarchical clustering was used to cluster children with OCD symptoms into distinct subtypes. Ten neurocognitive task scores and 20 Child Behavior Checklist syndrome scales were used to characterize cognitive/behavioral differences. Gray matter volume, fractional anisotropy of major white matter fiber tracts, and functional connectivity among networks were used in case-control studies.

RESULTS

We identified two subgroups with contrasting patterns in internalized and externalized dimensions. Group 1 showed compulsive thoughts and repeated acts but relatively low anxiety symptoms, whereas group 2 exhibited higher anxiety and perfectionism and relatively low repetitive behavior. Only group 1 had significant cognitive impairments and gray matter volume reductions in the bilateral inferior parietal lobe, precentral gyrus, and precuneus gyrus, and had white matter tract fractional anisotropy reductions in the corticostriatal fasciculus.

CONCLUSIONS

Children with OCD symptoms are heterogeneous at the level of symptom clustering and its underlying neural basis. Two subgroups represent distinct patterns of externalizing and internalizing symptoms, suggesting that anxiety is not its major predisposing factor. These results may have implications for the nosology and treatment of preteenage OCD.

Heterogeneity in caregiving-related early adversity: Creating stable dimensions and subtypes

Early psychosocial adversities exist at many levels, including caregiving-related, extrafamilial, and sociodemographic, which despite their high interrelatedness may have unique impacts on development. In this paper, we focus on caregiving-related early adversities (crEAs) and parse the heterogeneity of crEAs via data reduction techniques that identify experiential cooccurrences. Using network science, we characterized crEA cooccurrences to represent the comorbidity of crEA experiences across a sample of school-age children (n = 258; 6-12 years old) with a history of crEAs. crEA dimensions (variable level) and crEA subtypes (subject level) were identified using parallel factor analysis/principal component analysis and graph-based Louvain community detection. Bagging enhancement with cross-validation provided estimates of robustness. These data-driven dimensions/subtypes showed evidence of stability, transcended traditional sociolegally defined groups, were more homogenous than sociolegally defined groups, and reduced statistical correlations with sociodemographic factors. Finally, random forests showed both unique and common predictive importance of the crEA dimensions/subtypes for childhood mental health symptoms and academic skills. These data-driven outcomes provide additional tools and recommendations for crEA data reduction to inform precision medicine efforts in this area.

Regional Radiomics Similarity Networks Reveal Distinct Subtypes and Abnormality Patterns in Mild Cognitive Impairment

Individuals with mild cognitive impairment (MCI) of different subtypes show distinct alterations in network patterns. The first aim of this study is to identify the subtypes of MCI by employing a regional radiomics similarity network (R2SN). The second aim is to characterize the abnormality patterns associated with the clinical manifestations of each subtype. An individual-level R2SN is constructed for N = 605 normal controls (NCs), N = 766 MCI patients, and N = 283 Alzheimer's disease (AD) patients. MCI patients' R2SN profiles are clustered into two subtypes using nonnegative matrix factorization. The patterns of brain alterations, gene expression, and the risk of cognitive decline in each subtype are evaluated. MCI patients are clustered into "similar to the pattern of NCs" (N-CI, N = 252) and "similar to the pattern of AD" (A-CI, N = 514) subgroups. Significant differences are observed between the subtypes with respect to the following: 1) clinical measures; 2) multimodal neuroimaging; 3) the proportion of progression to dementia (61.54% for A-CI and 21.77% for N-CI) within three years; 4) enriched genes for potassium-ion transport and synaptic transmission. Stratification into the two subtypes provides new insight for risk assessment and precise early intervention for MCI patients.

Mental health in the UK Biobank: A roadmap to self-report measures and neuroimaging correlates

The UK Biobank (UKB) is a highly promising dataset for brain biomarker research into population mental health due to its unprecedented sample size and extensive phenotypic, imaging, and biological measurements. In this study, we aimed to provide a shared foundation for UKB neuroimaging research into mental health with a focus on anxiety and depression. We compared UKB self-report measures and revealed important timing effects between scan acquisition and separate online acquisition of some mental health measures. To overcome these timing effects, we introduced and validated the Recent Depressive Symptoms (RDS-4) score which we recommend for state-dependent and longitudinal research in the UKB. We furthermore tested univariate and multivariate associations between brain imaging-derived phenotypes (IDPs) and mental health. Our results showed a significant multivariate relationship between IDPs and mental health, which was replicable. Conversely, effect sizes for individual IDPs were small. Test-retest reliability of IDPs was stronger for measures of brain structure than for measures of brain function. Taken together, these results provide benchmarks and guidelines for future UKB research into brain biomarkers of mental health.

Understanding Anhedonia from a Genomic Perspective

Anhedonia, or the decreased ability to experience pleasure, is a cardinal symptom of major depression that commonly occurs within other forms of psychopathology. Supportive of long-held theory that anhedonia represents a genetically influenced vulnerability marker for depression, evidence from twin studies suggests that it is moderately-largely heritable. However, the genomic sources of this heritability are just beginning to be understood. In this review, we survey what is known about the genomic architecture underlying anhedonia and related constructs. We briefly review twin and initial candidate gene studies before focusing on genome-wide association study (GWAS) and polygenic efforts. As large samples are needed to reliably detect the small effects that typically characterize common genetic variants, the study of anhedonia and related phenotypes conflicts with current genomic research requirements and frameworks that prioritize sample size over precise phenotyping. This has resulted in few and underpowered studies of anhedonia-related constructs that have largely failed to reliably identify individual variants. Nonetheless, the polygenic architecture of anhedonia-related constructs identified in these studies has genetic overlap with depression and schizophrenia as well as related brain structure (e.g., striatal volume), providing important clues to etiology that may usefully guide refinement in nosology. As we await the accumulation of larger samples for more well-powered GWAS of reward-related constructs, novel analytic techniques that leverage GWAS summary statistics (e.g., genomic structural equation modeling) may currently be used to help characterize how the genomic architecture of anhedonia is shared and distinct from that underlying other constructs (e.g., depression, neuroticism, anxiety).

The Monash Autism-ADHD genetics and neurodevelopment (MAGNET) project design and methodologies: a dimensional approach to understanding neurobiological and genetic aetiology

Background

ASD and ADHD are prevalent neurodevelopmental disorders that frequently co-occur and have strong evidence for a degree of shared genetic aetiology. Behavioural and neurocognitive heterogeneity in ASD and ADHD has hampered attempts to map the underlying genetics and neurobiology, predict intervention response, and improve diagnostic accuracy. Moving away from categorical conceptualisations of psychopathology to a dimensional approach is anticipated to facilitate discovery of data-driven clusters and enhance our understanding of the neurobiological and genetic aetiology of these conditions. The Monash Autism-ADHD genetics and neurodevelopment (MAGNET) project is one of the first large-scale, family-based studies to take a truly transdiagnostic approach to ASD and ADHD. Using a comprehensive phenotyping protocol capturing dimensional traits central to ASD and ADHD, the MAGNET project aims to identify data-driven clusters across ADHD-ASD spectra using deep phenotyping of symptoms and behaviours; investigate the degree of familiality for different dimensional ASD-ADHD phenotypes and clusters; and map the neurocognitive, brain imaging, and genetic correlates of these data-driven symptom-based clusters.

Methods

The MAGNET project will recruit 1,200 families with children who are either typically developing, or who display elevated ASD, ADHD, or ASD-ADHD traits, in addition to affected and unaffected biological siblings of probands, and parents. All children will be comprehensively phenotyped for behavioural symptoms, comorbidities, neurocognitive and neuroimaging traits and genetics.

Conclusion

The MAGNET project will be the first large-scale family study to take a transdiagnostic approach to ASD-ADHD, utilising deep phenotyping across behavioural, neurocognitive, brain imaging and genetic measures.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13229-021-00457-3.

Blood metabolic signatures of hikikomori, pathological social withdrawal

Background

A severe form of pathological social withdrawal, ‘hikikomori,’ has been acknowledged in Japan, spreading worldwide, and becoming a global health issue. The pathophysiology of hikikomori has not been clarified, and its biological traits remain unexplored.

Methods

Drug-free patients with hikikomori (n = 42) and healthy controls (n = 41) were recruited. Psychological assessments for the severity of hikikomori and depression were conducted. Blood biochemical tests and plasma metabolome analysis were performed. Based on the integrated information, machine-learning models were created to discriminate cases of hikikomori from healthy controls, predict hikikomori severity, stratify the cases, and identify metabolic signatures that contribute to each model.

Results

Long-chain acylcarnitine levels were remarkably higher in patients with hikikomori; bilirubin, arginine, ornithine, and serum arginase were significantly different in male patients with hikikomori. The discriminative random forest model was highly performant, exhibiting an area under the ROC curve of 0.854 (confidential interval = 0.648–1.000). To predict hikikomori severity, a partial least squares PLS-regression model was successfully created with high linearity and practical accuracy. In addition, blood serum uric acid and plasma cholesterol esters contributed to the stratification of cases.

Conclusions

These findings reveal the blood metabolic signatures of hikikomori, which are key to elucidating the pathophysiology of hikikomori and also useful as an index for monitoring the treatment course for rehabilitation.

Characterizing the impact of adversity, abuse, and neglect on adolescent amygdala resting-state functional connectivity

Characterizing typologies of childhood adversity may inform the development of risk profiles and corresponding interventions aimed at mitigating its lifelong consequences. A neurobiological grounding of these typologies requires systematic comparisons of neural structure and function among individuals with different exposure histories. Using seed-to-whole brain analyses, this study examined associations between childhood adversity and amygdala resting-state functional connectivity (rs-fc) in adolescents aged 11–19 years across three independent studies (N = 223; 127 adversity group) in both general and dimensional models of adversity (comparing abuse and neglect). In a general model, adversity was associated with altered amygdala rs-fc with clusters within the left anterior lateral prefrontal cortex. In a dimensional model, abuse was associated with altered amygdala rs-fc within the orbitofrontal cortex, dorsal precuneus, posterior cingulate cortex, and dorsal anterior cingulate cortex/anterior mid-cingulate cortex, as well as within the dorsal attention, visual, and somatomotor networks. Neglect was associated with altered amygdala rs-fc with the hippocampus, supplementary motor cortex, temporoparietal junction, and regions within the dorsal attention network. Both general and dimensional models revealed unique regions, potentially reflecting pathways by which distinct histories of adversity may influence adolescent behavior, cognition, and psychopathology.

Towards precise resting-state fMRI biomarkers in psychiatry: synthesizing developments in transdiagnostic research, dimensional models of psychopathology, and normative neurodevelopment

Searching for biomarkers has been a chief pursuit of the field of psychiatry. Toward this end, studies have catalogued candidate resting-state biomarkers in nearly all forms of mental disorder. However, it is becoming increasingly clear that these biomarkers lack specificity, limiting their capacity to yield clinical impact. We discuss three avenues of research that are overcoming this limitation: (i) the adoption of transdiagnostic research designs, which involve studying and explicitly comparing multiple disorders from distinct diagnostic axes of psychiatry; (ii) dimensional models of psychopathology that map the full spectrum of symptomatology and that cut across traditional disorder boundaries; and (iii) modeling individuals' unique functional connectomes throughout development. We provide a framework for tying these subfields together that draws on tools from machine learning and network science.