Multidimensional prediction of treatment response to antidepressants with cognitive control and functional MRI

Neuroimaging for precision medicine in psychiatry

Although the lifetime burden due to mental disorders is increasing, we lack tools for more precise diagnosing and treating prevalent and disabling disorders such as major depressive disorder. We lack strategies for selecting among available treatments or expediting access to new treatment options. This critical review concentrates on functional neuroimaging as a modality of measurement for precision psychiatry, focusing on major depressive and anxiety disorders. We begin by outlining evidence for the use of functional neuroimaging to stratify the heterogeneity of these disorders, based on underlying circuit dysfunction. We then review the current landscape of how functional neuroimaging-derived circuit predictors can predict treatment outcomes and clinical trajectories in depression and anxiety. Future directions for advancing clinically appliable neuroimaging measures are considered. We conclude by considering the opportunities and challenges of translating neuroimaging measures into practice. As an illustration, we highlight one approach for quantifying brain circuit function at an individual level, which could serve as a model for clinical translation.

Breakthroughs and challenges for generating brain network-based biomarkers of treatment response in depression

Treatment outcomes widely vary for individuals diagnosed with major depressive disorder, implicating a need for deeper understanding of the biological mechanisms conferring a greater likelihood of response to a particular treatment. Our improved understanding of intrinsic brain networks underlying depression psychopathology via magnetic resonance imaging and other neuroimaging modalities has helped reveal novel and potentially clinically meaningful biological markers of response. And while we have made considerable progress in identifying such biomarkers over the last decade, particularly with larger, multisite trials, there are significant methodological and practical obstacles that need to be overcome to translate these markers into the clinic. The aim of this review is to review current literature on brain network structural and functional biomarkers of treatment response or selection in depression, with a specific focus on recent large, multisite trials reporting predictive accuracy of candidate biomarkers. Regarding pharmaco- and psychotherapy, we discuss candidate biomarkers, reporting that while we have identified candidate biomarkers of response to a single intervention, we need more trials that distinguish biomarkers between first-line treatments. Further, we discuss the ways prognostic neuroimaging may help to improve treatment outcomes to neuromodulation-based therapies, such as transcranial magnetic stimulation and deep brain stimulation. Lastly, we highlight obstacles and technical developments that may help to address the knowledge gaps in this area of research. Ultimately, integrating neuroimaging-derived biomarkers into clinical practice holds promise for enhancing treatment outcomes and advancing precision psychiatry strategies for depression management. By elucidating the neural predictors of treatment response and selection, we can move towards more individualized and effective depression interventions, ultimately improving patient outcomes and quality of life.

Accelerated Theta Burst Transcranial Magnetic Stimulation for Refractory Depression in Autism Spectrum Disorder

PURPOSE

Major depressive disorder (MDD) disproportionately affects those living with autism spectrum disorder (ASD) and is associated with significant impairment and treatment recidivism.

METHODS

We studied the use of accelerated theta burst stimulation (ATBS) for the treatment of refractory MDD in ASD (3 treatments daily x 10 days). This prospective open-label 12-week trial included 10 subjects with a mean age of 21.5 years, randomized to receive unilateral or bilateral stimulation of the dorsolateral prefrontal cortex.

RESULTS

One participant dropped out of the study due to intolerability. In both treatment arms, depressive symptoms, scored on the Hamilton Depression Rating Scale scores, diminished substantially. At 12 weeks post-treatment, full remission was sustained in 5 subjects and partial remission in 3 subjects. Treatment with ATBS, regardless of the site of stimulation, was associated with a significant, substantial, and sustained improvement in depressive symptomatology via the primary outcome measure, the Hamilton Depression Rating Scale. Additional secondary measures, including self-report depression scales, fluid cognition, and sleep quality, also showed significant improvement. No serious adverse events occurred during the study. Mild transient headaches were infrequently reported, which are expected side effects of ATBS.

CONCLUSION

Overall, ATBS treatment was highly effective and well-tolerated in individuals with ASD and co-occurring MDD. The findings support the need for a larger, sham-controlled randomized controlled trial to further evaluate efficacy of ATBS in this population.

Brain Circuit-Derived Biotypes for Treatment Selection in Mood Disorders: A Critical Review and Illustration of a Functional Neuroimaging Tool for Clinical Translation

Although the lifetime burden due to major depressive disorder is increasing, we lack tools for selecting the most effective treatments for each patient. One-third to one-half of patients with major depressive disorder do not respond to treatment, and we lack strategies for selecting among available treatments or expediting access to new treatment options. This critical review concentrates on functional neuroimaging as a modality of measurement for precision psychiatry. We begin by summarizing the current landscape of how functional neuroimaging-derived circuit predictors can forecast treatment outcomes in depression. Then, we outline the opportunities and challenges in integrating circuit predictors into clinical practice. We highlight one standardized and reproducible approach for quantifying brain circuit function at an individual level, which could serve as a model for clinical translation. We conclude by evaluating the prospects and practicality of employing neuroimaging tools, such as the one that we propose, in routine clinical practice.

Cortical glutamate, Glx, and total N-acetylaspartate: potential biomarkers of repetitive transcranial magnetic stimulation treatment response and outcomes in major depression

Repetitive transcranial magnetic stimulation (rTMS) is an effective treatment for individuals with major depressive disorder (MDD) who have not improved with standard therapies. However, only 30-45% of patients respond to rTMS. Predicting response to rTMS will benefit both patients and providers in terms of prescribing and targeting treatment for maximum efficacy and directing resources, as individuals with lower likelihood of response could be redirected to more suitable treatment alternatives. In this exploratory study, our goal was to use proton magnetic resonance spectroscopy to examine how glutamate (Glu), Glx, and total N-acetylaspartate (tNAA) predict post-rTMS changes in overall MDD severity and symptoms, and treatment response. Metabolites were measured in a right dorsal anterior cingulate cortex voxel prior to a standard course of 10 Hz rTMS to the left DLPFC in 25 individuals with MDD. MDD severity and symptoms were evaluated via the Inventory of Depression Symptomatology Self-Report (IDS-SR). rTMS response was defined as ≥50% change in full-scale IDS-SR scores post treatment. Percent change in IDS-SR symptom domains were evaluated using principal component analysis and established subscales. Generalized linear and logistic regression models were used to evaluate the relationship between baseline Glu, Glx, and tNAA and outcomes while controlling for age and sex. Participants with baseline Glu and Glx levels in the lower range had greater percent change in full scale IDS-SR scores post-treatment (p < 0.001), as did tNAA (p = 0.007). Low glutamatergic metabolite levels also predicted greater percent change in mood/cognition symptoms (p ≤ 0.001). Low-range Glu, Glx, and tNAA were associated with greater improvement on the immuno-metabolic subscale (p ≤ 0.003). Baseline Glu predicted rTMS responder status (p = 0.025) and had an area under the receiving operating characteristic curve of 0.81 (p = 0.009), demonstrating excellent discriminative ability. Baseline Glu, Glx, and tNAA significantly predicted MDD improvement after rTMS; preliminary evidence also demonstrates metabolite association with symptom subdomain improvement post-rTMS. This work provides feasibility for a personalized medicine approach to rTMS treatment selection, with individuals with Glu levels in the lower range potentially being the best candidates.

Trait- and state-like co-activation pattern dynamics in current and remitted major depressive disorder

BACKGROUND

Distinguishing between trait- and state-like neural alternations in major depressive disorder (MDD) may advance our understanding of this recurring disorder. We aimed to investigate dynamic functional connectivity alternations in unmedicated individuals with current or past MDD using co-activation pattern analyses.

METHODS

Resting-state functional magnetic resonance imaging data were acquired from individuals with first-episode current MDD (cMDD, n = 50), remitted MDD (rMDD, n = 44), and healthy controls (HCs, n = 64). Using a data-driven consensus clustering technique, four whole-brain states of spatial co-activation were identified and associated metrics (dominance, entries, transition frequency) were analyzed with respect to clinical characteristics.

RESULTS

Relative to rMDD and HC, cMDD showed increased dominance and entries of state 1 (primarily involving default mode network (DMN)), and decreased dominance of state 4 (mostly involving frontal-parietal network (FPN)). Among cMDD, state 1 entries correlated positively with trait rumination. Conversely, relative to cMDD and HC, individuals with rMDD were characterized by increased state 4 entries. Relative to HC, both MDD groups showed increased state 4-to-1 (FPN to DMN) transition frequency but reduction in state 3 (spanning visual attention, somatosensory, limbic networks), with the former metric specifically related to trait rumination.

LIMITATIONS

Further confirmation with longitudinal studies are required.

CONCLUSIONS

Regardless of symptoms, MDD was characterized by increased FPN-to-DMN transitions and reduced dominance of a hybrid network. State-related effect emerged in regions critically implicated in repetitive introspection and cognitive control. Asymptomatic individuals with past MDD were uniquely linked to increased FPN entries. Our findings identify trait-like brain network dynamics that might increase vulnerability to future MDD.

AI-assisted prediction of differential response to antidepressant classes using electronic health records

Antidepressant selection is largely a trial-and-error process. We used electronic health record (EHR) data and artificial intelligence (AI) to predict response to four antidepressants classes (SSRI, SNRI, bupropion, and mirtazapine) 4 to 12 weeks after antidepressant initiation. The final data set comprised 17,556 patients. Predictors were derived from both structured and unstructured EHR data and models accounted for features predictive of treatment selection to minimize confounding by indication. Outcome labels were derived through expert chart review and AI-automated imputation. Regularized generalized linear model (GLM), random forest, gradient boosting machine (GBM), and deep neural network (DNN) models were trained and their performance compared. Predictor importance scores were derived using SHapley Additive exPlanations (SHAP). All models demonstrated similarly good prediction performance (AUROCs ≥ 0.70, AUPRCs ≥ 0.68). The models can estimate differential treatment response probabilities both between patients and between antidepressant classes for the same patient. In addition, patient-specific factors driving response probabilities for each antidepressant class can be generated. We show that antidepressant response can be accurately predicted from real-world EHR data with AI modeling, and our approach could inform further development of clinical decision support systems for more effective treatment selection.

Using Machine Learning to Predict Antidepressant Treatment Outcome From Electronic Health Records

Objective

To evaluate if a machine learning approach can accurately predict antidepressant treatment outcome using electronic health records (EHRs) from patients with depression.

Method

This study examined 808 patients with depression at a New York City-based outpatient mental health clinic between June 13, 2016 and June 22, 2020. Antidepressant treatment outcome was defined based on trend in depression symptom severity over time and was categorized as either "Recovering" or "Worsening" (i.e., non-Recovering), measured by the slope of individual-level Patient Health Questionnaire-9 (PHQ-9) score trajectory spanning 6 months following treatment initiation. A patient was designated as "Recovering" if the slope is less than 0 and as "Worsening" if the slope was no less than 0. Multiple machine learning (ML) models including L2 norm regularized Logistic Regression, Naive Bayes, Random Forest, and Gradient Boosting Decision Tree (GBDT) were used to predict treatment outcome based on additional data from EHRs, including demographics and diagnoses. Shapley Additive Explanations were applied to identify the most important predictors.

Results

The GBDT achieved the best results of predicting "Recovering" (AUC: 0.7654 ± 0.0227; precision: 0.6002 ± 0.0215; recall: 0.5131 ± 0.0336). When excluding patients with low PHQ-9 scores (<10) at baseline, the results of predicting "Recovering" (AUC: 0.7254 ± 0.0218; precision: 0.5392 ± 0.0437; recall: 0.4431 ± 0.0513) were obtained. Prior diagnosis of anxiety, psychotherapy, recurrent depression, and baseline depression symptom severity were strong predictors.

Conclusions

The results demonstrate the potential utility of using ML in longitudinal EHRs to predict antidepressant treatment outcome. Our predictive tool holds the promise to accelerate personalized medical management in patients with psychiatric illnesses.

Response Inhibition and Predicting Response to Pharmacological and Cognitive Behavioral Therapy Treatments for Major Depressive Disorder: A Canadian Biomarker Integration Network for Depression Study

BACKGROUND

Major depressive disorder (MDD) is associated with various cognitive impairments, including response inhibition. Deficits in response inhibition may also underlie poor antidepressant treatment response. Recent studies revealed that the neurobiological correlates of response inhibition can predict response to pharmacological treatments. However, the generalizability of this finding to first-line nonpharmacological treatments, particularly cognitive behavioral therapy, remains to be investigated.

METHODS

Data from two independent treatment protocols were combined, one in which 65 patients with MDD underwent treatment with escitalopram, and the other in which 41 patients with MDD underwent a course of cognitive behavioral therapy. A total of 25 healthy control subjects were also recruited. Neural correlates of response inhibition were captured by participants completing a Go/NoGo task during electroencephalography recording. Response inhibition-related measures of interest included the amplitudes of the N2 and P3 event-related potentials.

RESULTS

Pretreatment P3 amplitude, which has been linked to both the motor and cognitive aspects of response inhibition, was a significant predictor of change in depressive symptoms following escitalopram and cognitive behavioral therapy treatment. A greater pretreatment P3 amplitude was associated with a greater reduction in depressive severity. In addition, the pretreatment P3 amplitude was found to be significantly greater at baseline in remitters than in nonremitters and healthy control subjects.

CONCLUSIONS

The integrity of response inhibition may be critical for a successful course of pharmacological or psychological treatment for MDD. Electrophysiological correlates of response inhibition may have utility as a general prognostic marker of treatment response in MDD. Future studies may investigate the benefit of preceding first-line treatments with interventions that improve response inhibition in MDD.

Computational approaches to treatment response prediction in major depression using brain activity and behavioral data: A systematic review

Major depressive disorder is a heterogeneous diagnostic category with multiple available treatments. With the goal of optimizing treatment selection, researchers are developing computational models that attempt to predict treatment response based on various pretreatment measures. In this paper, we review studies that use brain activity data to predict treatment response. Our aim is to highlight and clarify important methodological differences between various studies that relate to the incorporation of domain knowledge, specifically within two approaches delineated as data-driven and theory-driven. We argue that theory-driven generative modeling, which explicitly models information processing in the brain and thus can capture disease mechanisms, is a promising emerging approach that is only beginning to be utilized in treatment response prediction. The predictors extracted via such models could improve interpretability, which is critical for clinical decision-making. We also identify several methodological limitations across the reviewed studies and provide suggestions for addressing them. Namely, we consider problems with dichotomizing treatment outcomes, the importance of investigating more than one treatment in a given study for differential treatment response predictions, the need for a patient-centered approach for defining treatment outcomes, and finally, the use of internal and external validation methods for improving model generalizability.

MRI predictors of pharmacotherapy response in major depressive disorder

Major depressive disorder is among the most prevalent psychiatric disorders, exacting a substantial personal, social, and economic toll. Antidepressant treatment typically involves an individualized trial and error approach with an inconsistent success rate. Despite a pressing need, no reliable biomarkers for predicting treatment outcome have yet been discovered. Brain MRI measures hold promise in this regard, though clinical translation remains elusive. In this review, we summarize structural MRI and functional MRI (fMRI) measures that have been investigated as predictors of treatment outcome. We broadly divide these into five categories including three structural measures: volumetric, white matter burden, and white matter integrity; and two functional measures: resting state fMRI and task fMRI. Currently, larger hippocampal volume is the most widely replicated predictor of successful treatment. Lower white matter hyperintensity burden has shown robustness in late life depression. However, both have modest discriminative power. Higher fractional anisotropy of the cingulum bundle and frontal white matter, amygdala hypoactivation and anterior cingulate cortex hyperactivation in response to negative emotional stimuli, and hyperconnectivity within the default mode network (DMN) and between the DMN and executive control network also show promise as predictors of successful treatment. Such network-focused measures may ultimately provide a higher-dimensional measure of treatment response with closer ties to the underlying neurobiology.

Predicting Patient Wait Times by Using Highly Deidentified Data in Mental Health Care: Enhanced Machine Learning Approach

BACKGROUND

Wait times impact patient satisfaction, treatment effectiveness, and the efficiency of care that the patients receive. Wait time prediction in mental health is a complex task and is affected by the difficulty in predicting the required number of treatment sessions for outpatients, high no-show rates, and the possibility of using group treatment sessions. The task of wait time analysis becomes even more challenging if the input data has low utility, which happens when the data is highly deidentified by removing both direct and quasi identifiers.

OBJECTIVE

The first aim of this study was to develop machine learning models to predict the wait time from referral to the first appointment for psychiatric outpatients by using real-time data. The second aim was to enhance the performance of these predictive models by utilizing the system's knowledge while the input data were highly deidentified. The third aim was to identify the factors that drove long wait times, and the fourth aim was to build these models such that they were practical and easy-to-implement (and therefore, attractive to care providers).

METHODS

We analyzed retrospective highly deidentified administrative data from 8 outpatient clinics at Ontario Shores Centre for Mental Health Sciences in Canada by using 6 machine learning methods to predict the first appointment wait time for new outpatients. We used the system's knowledge to mitigate the low utility of our data. The data included 4187 patients who received care through 30,342 appointments.

RESULTS

The average wait time varied widely between different types of mental health clinics. For more than half of the clinics, the average wait time was longer than 3 months. The number of scheduled appointments and the rate of no-shows varied widely among clinics. Despite these variations, the random forest method provided the minimum root mean square error values for 4 of the 8 clinics, and the second minimum root mean square error for the other 4 clinics. Utilizing the system's knowledge increased the utility of our highly deidentified data and improved the predictive power of the models.

CONCLUSIONS

The random forest method, enhanced with the system's knowledge, provided reliable wait time predictions for new outpatients, regardless of low utility of the highly deidentified input data and the high variation in wait times across different clinics and patient types. The priority system was identified as a factor that contributed to long wait times, and a fast-track system was suggested as a potential solution.

Neural correlates of affective control regions induced by common therapeutic strategies in major depressive disorders: An activation likelihood estimation meta-analysis study

In major depressive disorder (MDD), not only the pathophysiology of this disease is unknown but also the mechanisms of clinical efficacy across its therapeutic strategies are unclear. Although neuroimaging studies adopted activation likelihood estimation (ALE) approach to identify the convergent abnormalities of human brain in the MDD patients, the common alterations after antidepressant therapies were not summarized. Thus, we extracted the coordinates of brain regions in the MDD patients that showed differences in resting-state function, gray matter morphometry, and task-evoked neuronal responses after therapies. The ALE algorithm (GingerALE2.0.3) was employed in all 53 studies (64 experiments with 1406 MDD patients). Consistent results across treatment therapies were reported in the affective control network, including the bilateral thalamus, bilateral amygdala/parahippocampal gyrus, right anterior cingulate cortex/middle frontal gyrus, and right insular cortex/claustrum. Only electroconvulsive therapy partially replicated above findings. Our results indicate the antidepressant therapies efficiently influence core structures of the affective control network, which might be the underlying mechanism of remission in depression and provides potential targets for further treatment strategies.

Using Network Parcels and Resting-State Networks to Estimate Correlates of Mood Disorder and Related Research Domain Criteria Constructs of Reward Responsiveness and Inhibitory Control

BACKGROUND

Resting-state graph-based network edges can be powerful tools for identification of mood disorders. We address whether these edges can be integrated with Research Domain Criteria (RDoC) constructs for accurate identification of mood disorder-related markers, while minimizing active symptoms of disease.

METHODS

We compared 132 individuals with currently remitted or euthymic mood disorder with 65 healthy comparison participants, ages 18-30 years. Subsets of smaller brain parcels, combined into three prominent networks and one network of parcels overlapping across these networks, were used to compare edge differences between groups. Consistent with the RDoC framework, we evaluated individual differences with performance measure regressors of inhibitory control and reward responsivity. Within an omnibus regression model, we predicted edges related to diagnostic group membership, performance within both RDoC domains, and relevant interactions.

RESULTS

There were several edges of mood disorder group, predominantly of greater connectivity across networks, different than those related to individual differences in inhibitory control and reward responsivity. Edges related to diagnosis and inhibitory control did not align well with prior literature, whereas edges in relation to reward responsivity constructs showed greater alignment with prior literature. Those edges in interaction between RDoC constructs and diagnosis showed a divergence for inhibitory control (negative interactions in default mode) relative to reward (positive interactions with salience and emotion network).

CONCLUSIONS

In conclusion, there is evidence that prior simple network models of mood disorders are currently of insufficient biological or diagnostic clarity or that parcel-based edges may be insufficiently sensitive for these purposes.

Machine learning in the prediction of depression treatment outcomes: a systematic review and meta-analysis

BACKGROUND

Multiple treatments are effective for major depressive disorder (MDD), but the outcomes of each treatment vary broadly among individuals. Accurate prediction of outcomes is needed to help select a treatment that is likely to work for a given person. We aim to examine the performance of machine learning methods in delivering replicable predictions of treatment outcomes.

METHODS

Of 7732 non-duplicate records identified through literature search, we retained 59 eligible reports and extracted data on sample, treatment, predictors, machine learning method, and treatment outcome prediction. A minimum sample size of 100 and an adequate validation method were used to identify adequate-quality studies. The effects of study features on prediction accuracy were tested with mixed-effects models. Fifty-four of the studies provided accuracy estimates or other estimates that allowed calculation of balanced accuracy of predicting outcomes of treatment.

RESULTS

Eight adequate-quality studies reported a mean accuracy of 0.63 [95% confidence interval (CI) 0.56-0.71], which was significantly lower than a mean accuracy of 0.75 (95% CI 0.72-0.78) in the other 46 studies. Among the adequate-quality studies, accuracies were higher when predicting treatment resistance (0.69) and lower when predicting remission (0.60) or response (0.56). The choice of machine learning method, feature selection, and the ratio of features to individuals were not associated with reported accuracy.

CONCLUSIONS

The negative relationship between study quality and prediction accuracy, combined with a lack of independent replication, invites caution when evaluating the potential of machine learning applications for personalizing the treatment of depression.

Resting-state functional connectivity predictors of treatment response in schizophrenia - A systematic review and meta-analysis

We aimed to systematically synthesize and quantify the utility of pre-treatment resting-state functional magnetic resonance imaging (rs-fMRI) in predicting antipsychotic response in schizophrenia. We searched the PubMed/MEDLINE database for studies that examined the magnitude of association between baseline rs-fMRI assessment and subsequent response to antipsychotic treatment in persons with schizophrenia. We also performed meta-analyses for quantifying the magnitude and accuracy of predicting response defined continuously and categorically. Data from 22 datasets examining 1280 individuals identified striatal and default mode network functional segregation and integration metrics as consistent determinants of treatment response. The pooled correlation coefficient for predicting improvement in total symptoms measured continuously was ~0.47 (12 datasets; 95% CI: 0.35 to 0.59). The pooled odds ratio of predicting categorically defined treatment response was 12.66 (nine datasets; 95% CI: 7.91-20.29), with 81% sensitivity and 76% specificity. rs-fMRI holds promise as a predictive biomarker of antipsychotic treatment response in schizophrenia. Future efforts need to focus on refining feature characterization to improve prediction accuracy, validate prediction models, and evaluate their implementation in clinical practice.

Abnormal Cerebellar Volume in Patients with Remitted Major Depression with Persistent Cognitive Deficits

Cerebellar involvement in major depressive disorder (MDD) has been demonstrated by a growing number of studies, but it is unknown whether cognitive functioning in depressed individuals is related to cerebellar gray matter volume (GMV) abnormalities. Impaired attention and executive dysfunction are characteristic cognitive deficits in MDD, and critically, they often persist despite remission of mood symptoms. In this study, we investigated cerebellar GMV in patients with remitted MDD (rMDD) that showed persistent cognitive impairment. We applied cerebellum-optimized voxel-based morphometry in 37 patients with rMDD and with cognitive deficits, in 12 patients with rMDD and without cognitive deficits, and in 36 healthy controls (HC). Compared with HC, rMDD patients with cognitive deficits had lower GMV in left area VIIA, crus II, and in vermal area VIIB. In patients with rMDD, regression analyses demonstrated significant associations between GMV reductions in both regions and impaired attention and executive dysfunction. Compared with HC, patients without cognitive deficits showed increased GMV in bilateral area VIIIB. This study supports cerebellar contributions to the cognitive dimension of MDD. The data also point towards cerebellar area VII as a potential target for non-invasive brain stimulation to treat cognitive deficits related to MDD.

Early post-treatment blood oxygenation level-dependent responses to emotion processing associated with clinical response to pharmacological treatment in major depressive disorder

INTRODUCTION

Pre-treatment blood oxygenation level-dependent (BOLD) functional magnetic resonance imaging (fMRI) has been used for the early identification of patients with major depressive disorder (MDD) who later respond or fail to respond to medication. However, BOLD responses early after treatment initiation may offer insight into early neural changes associated with later clinical response. The present study evaluated both pre-treatment and early post-treatment fMRI responses to an emotion processing task, to further our understanding of neural changes associated with a successful response to pharmacological intervention.

METHODS

MDD patients who responded (n = 22) and failed to respond (n = 12) after 8 weeks of treatment with either citalopram or quetiapine extended release, and healthy controls (n = 18) underwent two fMRI scans, baseline (pre-treatment), and early post-treatment (one week after treatment commencement). Participants completed an emotional face matching task at both scans.

RESULTS

Using threshold-free cluster enhancement (TFCE) and non-parametric permutation testing, fMRI activation maps showed that after one week of treatment, responders demonstrated increased activation in the left parietal lobule, precentral gyrus, and bilateral insula (all P < 0.05 threshold-free cluster enhancement (TFCE) family-wise error-corrected) to negative facial expressions. Non-responders showed some small increases in the precentral gyrus, while controls showed no differences between scans. Compared to non-responders, responders showed some increased activation in the superior parietal lobule and middle temporal gyrus at the post-treatment scan. There were no group differences between responders, non-responders, and controls at baseline.

CONCLUSIONS

One week after treatment commencement, BOLD signal changes in the parietal lobules, insula, and middle temporal gyrus were related to clinical response to pharmacological treatment.

The promise of machine learning in predicting treatment outcomes in psychiatry

For many years, psychiatrists have tried to understand factors involved in response to medications or psychotherapies, in order to personalize their treatment choices. There is now a broad and growing interest in the idea that we can develop models to personalize treatment decisions using new statistical approaches from the field of machine learning and applying them to larger volumes of data. In this pursuit, there has been a paradigm shift away from experimental studies to confirm or refute specific hypotheses towards a focus on the overall explanatory power of a predictive model when tested on new, unseen datasets. In this paper, we review key studies using machine learning to predict treatment outcomes in psychiatry, ranging from medications and psychotherapies to digital interventions and neurobiological treatments. Next, we focus on some new sources of data that are being used for the development of predictive models based on machine learning, such as electronic health records, smartphone and social media data, and on the potential utility of data from genetics, electrophysiology, neuroimaging and cognitive testing. Finally, we discuss how far the field has come towards implementing prediction tools in real-world clinical practice. Relatively few retrospective studies to-date include appropriate external validation procedures, and there are even fewer prospective studies testing the clinical feasibility and effectiveness of predictive models. Applications of machine learning in psychiatry face some of the same ethical challenges posed by these techniques in other areas of medicine or computer science, which we discuss here. In short, machine learning is a nascent but important approach to improve the effectiveness of mental health care, and several prospective clinical studies suggest that it may be working already.

Applications of artificial intelligence to improve patient flow on mental health inpatient units - Narrative literature review

BACKGROUND

Despite a growing body of research into both Artificial intelligence and mental health inpatient flow issues, few studies adequately combine the two. This review summarises findings in the fields of AI in psychiatry and patient flow from the past 5 years, finds links and identifies gaps for future research.

METHODS

The OVID database was used to access Embase and Medline. Top journals such as JAMA, Nature and The Lancet were screened for other relevant studies. Selection bias was limited by strict inclusion and exclusion criteria.

RESEARCH

3,675 papers were identified in March 2020, of which a limited number focused on AI for mental health unit patient flow. After initial screening, 323 were selected and 83 were subsequently analysed. The literature review revealed a wide range of applications with three main themes: diagnosis (33%), prognosis (39%) and treatment (28%). The main themes that emerged from AI in patient flow studies were: readmissions (41%), resource allocation (44%) and limitations (91%). The review extrapolates those solutions and suggests how they could potentially improve patient flow on mental health units, along with challenges and limitations they could face.

CONCLUSION

Research widely addresses potential uses of AI in mental health, with some focused on its applicability in psychiatric inpatients units, however research rarely discusses improvements in patient flow. Studies investigated various uses of AI to improve patient flow across specialities. This review highlights a gap in research and the unique research opportunity it presents.

Resting State Functional Connectivity of Brain With Electroconvulsive Therapy in Depression: Meta-Analysis to Understand Its Mechanisms

Introduction: Electroconvulsive therapy (ECT) is a commonly used brain stimulation treatment for treatment-resistant or severe depression. This study was planned to find the effects of ECT on brain connectivity by conducting a systematic review and coordinate-based meta-analysis of the studies performing resting state fMRI (rsfMRI) in patients with depression receiving ECT. Methods: We systematically searched the databases published up to July 31, 2020, for studies in patients having depression that compared resting-state functional connectivity (rsFC) before and after a course of pulse wave ECT. Meta-analysis was performed using the activation likelihood estimation method after extracting details about coordinates, voxel size, and method for correction of multiple comparisons corresponding to the significant clusters and the respective rsFC analysis measure with its method of extraction. Results: Among 41 articles selected for full-text review, 31 articles were included in the systematic review. Among them, 13 articles were included in the meta-analysis, and a total of 73 foci of 21 experiments were examined using activation likelihood estimation in 10 sets. Using the cluster-level interference method, one voxel-wise analysis with the measure of amplitude of low frequency fluctuations and one seed-voxel analysis with the right hippocampus showed a significant reduction (p < 0.0001) in the left cingulate gyrus (dorsal anterior cingulate cortex) and a significant increase (p < 0.0001) in the right hippocampus with the right parahippocampal gyrus, respectively. Another analysis with the studies implementing network-wise (posterior default mode network: dorsomedial prefrontal cortex) resting state functional connectivity showed a significant increase (p < 0.001) in bilateral posterior cingulate cortex. There was considerable variability as well as a few key deficits in the preprocessing and analysis of the neuroimages and the reporting of results in the included studies. Due to lesser studies, we could not do further analysis to address the neuroimaging variability and subject-related differences. Conclusion: The brain regions noted in this meta-analysis are reasonably specific and distinguished, and they had significant changes in resting state functional connectivity after a course of ECT for depression. More studies with better neuroimaging standards should be conducted in the future to confirm these results in different subgroups of depression and with varied aspects of ECT.

A Lifespan Model of Interference Resolution and Inhibitory Control: Risk for Depression and Changes with Illness Progression

The cognitive processes involved in inhibitory control accuracy (IC) and interference resolution speed (IR) or broadly - inhibition - are discussed in this review, and both are described within the context of a lifespan model of mood disorders. Inhibitory control (IC) is a binary outcome (success or no for response selection and inhibition of unwanted responses) for any given event that is influenced to an extent by IR. IR refers to the process of inhibition, which can be manipulated by task design in earlier and later stages through use of distractors and timing, and manipulation of individual differences in response proclivity. We describe the development of these two processes across the lifespan, noting factors that influence this development (e.g., environment, adversity and stress) as well as inherent difficulties in assessing IC/IR prior to adulthood (e.g., cross-informant reports). We use mood disorders as an illustrative example of how this multidimensional construct can be informative to state, trait, vulnerability and neuroprogression of disease. We present aggregated data across numerous studies and methodologies to examine the lifelong development and degradation of this subconstruct of executive function, particularly in mood disorders. We highlight the challenges in identifying and measuring IC/IR in late life, including specificity to complex, comorbid disease processes. Finally, we discuss some potential avenues for treatment and accommodation of these difficulties across the lifespan, including newer treatments using cognitive remediation training and neuromodulation.

Predicting response to psychiatric surgery: a systematic review of neuroimaging findings

BACKGROUND

Psychiatric surgery, including deep brain stimulation and stereotactic ablation, is an important treatment option in severe refractory psychiatric illness. Several large trials have demonstrated response rates of approximately 50%, underscoring the need to identify and select responders preoperatively. Recent advances in neuroimaging have brought this possibility into focus. We systematically reviewed the psychiatric surgery neuroimaging literature to assess the current state of evidence for preoperative imaging predictors of response.

METHODS

We performed this study in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and Meta-analysis of Observational Studies in Epidemiology (MOOSE) frameworks, and preregistered it using PROSPERO. We systematically searched the Medline, Embase and Cochrane databases for studies reporting preoperative neuroimaging analyses correlated with clinical outcomes in patients who underwent psychiatric surgery. We recorded and synthesized the methodological details, imaging results and clinical correlations from these studies.

RESULTS

After removing duplicates, the search yielded 8388 unique articles, of which 7 met the inclusion criteria. The included articles were published between 2001 and 2018 and reported on the outcomes of 101 unique patients. Of the 6 studies that reported significant findings, all identified clusters of hypermetabolism, hyperconnectivity or increased size in the frontostriatal limbic circuitry.

LIMITATIONS

The included studies were few and highly varied, spanning 2 decades.

CONCLUSION

Although few studies have analyzed preoperative imaging for predictors of response to psychiatric surgery, we found consistency among the reported results: most studies implicated overactivity in the frontostriatal limbic network as being correlated with clinical response. Larger prospective studies are needed.

REGISTRATION

www.crd.york.ac.uk/prospero/display_record.php?RecordID=131151.

Connectivity of the Cognitive Control Network During Response Inhibition as a Predictive and Response Biomarker in Major Depression: Evidence From a Randomized Clinical Trial

BACKGROUND

In treating major depressive disorder, we lack tests anchored in neurobiology that predict antidepressant efficacy. Cognitive impairments are a particularly disabling feature of major depressive disorder. We tested whether functional connectivity during a response-inhibition task can predict response to antidepressants and whether its changes over time are correlated to symptom changes.

METHODS

We analyzed data from outpatients with major depressive disorder (n = 124) randomized to receive escitalopram, sertraline, or venlafaxine (8 weeks) and healthy control subjects (n = 59; age 18-65 years). Before and after treatment, participants were interviewed and scanned using functional magnetic resonance imaging, and functional connectivity was measured using generalized psychophysiological interaction during response inhibition (Go-NoGo task). We investigated the interaction between treatment type and response (≥50% reduction on self-reported symptoms), coupling differences between responders and nonresponders at baseline, their correlation with symptom improvement, and their changes in time.

RESULTS

During response inhibition, connectivity between the dorsolateral prefrontal cortex/supramarginal gyrus and supramarginal gyrus/middle temporal gyrus was associated with response to sertraline and venlafaxine, but not escitalopram. Sertraline responders had higher functional connectivity between these regions compared with nonresponders, whereas venlafaxine responders had lower functional connectivity. For sertraline, attenuation of connectivity in the precentral and superior temporal gyri correlated with posttreatment symptom improvement. For venlafaxine, enhancement of connectivity between the orbitofrontal cortex and subcortical regions correlated with symptom improvement.

CONCLUSIONS

Connectivity of the cognitive control circuit during response inhibition selectively and differentially predicts antidepressant treatment response and correlates with symptom improvement. These quantitative markers tied to the neurobiology of cognitive features of depression could be used translationally to predict and evaluate treatment response.

Neuroimaging as a Tool for Individualized Treatment Choice in Depression: the Past, the Present and the Future

Purpose of Review

This paper aims to review the findings on neuroimaging as a tool for facilitating individualized treatment choice in depression.

Recent Findings

Neuroimaging has allowed the exploration of neural candidates for response biomarkers. In less than two decades, the field has expanded from small single drug studies to large multisite initiatives testing multiple interventions; from simple analytical methods to employing artificial intelligence, with an aim of establishing models based on a variety of data, such as neuroimaging, biological, psychological and clinical measures.

Summary

Neural biomarkers of response may play an important role in treatment response prediction. It seems likely that they will need to be considered together with other types of data in complex models in order to achieve the high accuracy and generalizability of results necessary for clinical use.

Connectivity patterns of cognitive control network in first episode medication-naive depression and remitted depression

BACKGROUND

Cognitive dysfunctions, such as impaired cognitive control, are frequently observed in patients with major depressive disorder (MDD). Although the cognitive control network (CCN) is widely considered a core feature of major depressive disorder (MDD), the relationship between cognitive dysfunction and symptom dimensions remains unclear. This study investigated differences in resting-state functional connectivity of the cognitive control network (CCN) between first-episode medication-naive MDD patients and remitted MDD.

METHODS

We collected resting-state functional MRI (rs-fMRI) data from 22 first-episode medication-naive major depressive disorder (fMDD) patients, 20 patients previously diagnosed with MDD in the remitted phase of depression (rMDD), and 20 healthy controls (HC). The CCN was derived from fMRI images using independent component analysis (ICA), a data-driven image analysis method.

RESULTS

Changes in functional connectivity (FC) within the CCN was mainly attenuated in the right dorsolateral prefrontal cortex and the left inferior parietal lobule, while strengthened in the right dorsal anterior cingulate cortex and the right insula in both fMDD and rMDD groups. Compared with the fMDD group, the rMDD group had decreased FC in the bilateral insula and the right dorsolateral prefrontal cortex. Further analysis explored that the FC in the bilateral insula, the right dorsal anterior cingulate cortex and the right inferior parietal lobule were correlated positively cognitive disturbance factor scores in both patients groups.

CONCLUSIONS

These findings are in agreement with the previous findings that the cognitive control network are impaired in MDD. Furthermore, our results suggest that the alteration of CCN might underpin the cognitive disturbance and the distinct patterns of the CCN between fMDD and rMDD patients may be an important target for effective cognitive remediation in MDD.

Predicting escitalopram monotherapy response in depression: The role of anterior cingulate cortex

Neuroimaging biomarkers of treatment efficacy can be used to guide personalized treatment in major depressive disorder (MDD). Escitalopram is recommended as first-line therapy for MDD and severe depression. An interesting hypothesis suggests that the reconfiguration of dynamic brain networks might provide important insights into antidepressant mechanisms. The present study assesses whether the spatiotemporal modulation across functional brain networks could serve as a predictor of effective antidepressant treatment with escitalopram. A total of 106 first-episode, drug-naïve patients and 109 healthy controls from three different multicenters underwent resting-state functional magnetic resonance imaging. Patients were considered as responders if they had a reduction of at least 50% in Hamilton Rating Scale for Depression scores at endpoint (>2 weeks). Multilayer modularity framework was applied on the whole brain to construct features in relation to network dynamic characters that were used for multivariate pattern analysis. Linear soft-threshold support vector machine models were used to separate responders from nonresponders. The permutation tests demonstrated the robustness of discrimination performances. The discriminative regions formed a spatially distributed pattern with anterior cingulate cortex (ACC) as the hub in the default mode subnetwork. Interestingly, a significantly larger module allegiance of ACC was also found in treatment responders compared to nonresponders, suggesting high interactivities of ACC to other regions may be beneficial for the recovery after treatment. Consistent results across multicenters confirmed that ACC could serve as a predictor of escitalopram monotherapy treatment outcome, implying strong likelihood of replication in the future.

The Urge to Decide and Act: Implications for Brain Function and Dysfunction

Humans and other animals are motivated to act so as to maximize their subjective reward rate. Here, we propose that reward rate maximization is accomplished by adjusting a context-dependent “urgency signal,” which influences both the commitment to a developing action choice and the vigor with which the ensuing action is performed. We review behavioral and neurophysiological data suggesting that urgency is controlled by projections from the basal ganglia to cerebral cortical regions, influencing neural activity related to decision making as well as activity related to action execution. We also review evidence suggesting that different individuals possess specific policies for adjusting their urgency signal to particular contextual variables, such that urgency constitutes an individual trait which jointly influences a wide range of behavioral measures commonly related to the overall quality and hastiness of one’s decisions and actions. Consequently, we argue that a central mechanism for reward rate maximization provides a potential link between personality traits such as impulsivity, as well as some of the motivation-related symptomology of clinical disorders such as depression and Parkinson’s disease.

The Translational Potential of Neuroimaging Genomic Analyses To Diagnosis And Treatment In The Mental Disorders

Imaging genomics focuses on characterizing genomic influence on the variation of neurobiological traits, holding promise for illuminating the pathogenesis, reforming the diagnostic system, and precision medicine of mental disorders. This paper aims to provide an overall picture of the current status of neuroimaging-genomic analyses in mental disorders, and how we can increase their translational potential into clinical practice. The review is organized around three perspectives. (a) Towards reliability, generalizability and interpretability, where we summarize the multivariate models and discuss the considerations and trade-offs of using these methods and how reliable findings may be reached, to serve as ground for further delineation. (b) Towards improved diagnosis, where we outline the advantages and challenges of constructing a dimensional transdiagnostic model and how imaging genomic analyses map into this framework to aid in deconstructing heterogeneity and achieving an optimal stratification of patients that better inform treatment planning. (c) Towards improved treatment. Here we highlight recent efforts and progress in elucidating the functional annotations that bridge between genomic risk and neurobiological abnormalities, in detecting genomic predisposition and prodromal neurodevelopmental changes, as well as in identifying imaging genomic biomarkers for predicting treatment response. Providing an overview of the challenges and promises, this review hopefully motivates imaging genomic studies with multivariate, dimensional and transdiagnostic designs for generalizable and interpretable findings that facilitate development of personalized treatment.

The pulvinar nucleus and antidepressant treatment: dynamic modeling of antidepressant response and remission with ultra-high field functional MRI

Functional magnetic resonance imaging (fMRI) successfully disentangled neuronal pathophysiology of major depression (MD), but only a few fMRI studies have investigated correlates and predictors of remission. Moreover, most studies have used clinical outcome parameters from two time points, which do not optimally depict differential response times. Therefore, we aimed to detect neuronal correlates of response and remission in an antidepressant treatment study with 7 T fMRI, potentially harnessing advances in detection power and spatial specificity. Moreover, we modeled outcome parameters from multiple study visits during a 12-week antidepressant fMRI study in 26 acute (aMD) patients compared to 36 stable remitted (rMD) patients and 33 healthy control subjects (HC). During an electrical painful stimulation task, significantly higher baseline activity in aMD compared to HC and rMD in the medial thalamic nuclei of the pulvinar was detected (p = 0.004, FWE-corrected), which was reduced by treatment. Moreover, clinical response followed a sigmoid function with a plateau phase in the beginning, a rapid decline and a further plateau at treatment end. By modeling the dynamic speed of response with fMRI-data, perigenual anterior cingulate activity after treatment was significantly associated with antidepressant response (p < 0.001, FWE-corrected). Temporoparietal junction (TPJ) baseline activity significantly predicted non-remission after 2 antidepressant trials (p = 0.005, FWE-corrected). The results underline the importance of the medial thalamus, attention networks in MD and antidepressant treatment. Moreover, by using a sigmoid model, this study provides a novel method to analyze the dynamic nature of response and remission for future trials.

Multidimensional imaging techniques for prediction of treatment response in major depressive disorder

A large number of studies have attempted to use neuroimaging tools to aid in treatment prediction models for major depressive disorder (MDD). Most such studies have reported on only one dimension of function and prediction at a time. In this study, we used three different tasks across domains of function (emotion processing, reward anticipation, and cognitive control, plus resting state connectivity completed prior to start of medication to predict treatment response in 13-36 adults with MDD. For each experiment, adults with MDD were prescribed only label duloxetine (all experiments), whereas another subset were prescribed escitalopram. We used a KeyNet (both Task derived masks and Key intrinsic Network derived masks) approach to targeting brain systems in a specific match to tasks. The most robust predictors were (Dichter et al., 2010) positive response to anger and (Gong et al., 2011) negative response to fear within relevant anger and fear TaskNets and Salience and Emotion KeyNet (Langenecker et al., 2018) cognitive control (correct rejections) within Inhibition TaskNet (negative) and Cognitive Control KeyNet (positive). Resting state analyses were most robust for Cognitive control Network (positive) and Salience and Emotion Network (negative). Results differed by whether an -fwhm or -acf (more conservative) adjustment for multiple comparisons was used. Together, these results implicate the importance of future studies with larger sample sizes, multidimensional predictive models, and the importance of using empirically derived masks for search areas.

Cognitive control and network disruption in remitted depression: a correlate of childhood adversity

Individuals in a major depressive episode often display impairment in cognitive control, and this impairment exists outside of the acute phase of illness. Impairment in cognitive control also has been associated with exposure to childhood adversity (CA). The current study examined whether exposure to CA can explain variance in a component of cognitive control—inhibitory control—independent of diagnostic status in young adults with and without a history of depression. Healthy control individuals (n = 40) and individuals with remitted major depressive disorder (n = 53) completed a task measuring inhibitory control, reported level of CA and completed a scanning session to assess gray matter volume and resting state connectivity in regions associated with cognitive control. The results demonstrate that higher levels of CA were associated with poorer inhibitory control, reduced right middle frontal gyrus gray matter, decreased connectivity of salience and emotion networks and increased connectivity in cognitive control networks, even after controlling for diagnostic status, residual depression symptoms and current stressors. Together, the results suggest that inhibitory control impairment and intrinsic connectivity changes may be characterized as developmental sequelae of early stress exposure.

Cognitive Control as a 5-HT1A-Based Domain That Is Disrupted in Major Depressive Disorder

Heterogeneity within Major Depressive Disorder (MDD) has hampered identification of biological markers (e.g., intermediate phenotypes, IPs) that might increase risk for the disorder or reflect closer links to the genes underlying the disease process. The newer characterizations of dimensions of MDD within Research Domain Criteria (RDoC) domains may align well with the goal of defining IPs. We compare a sample of 25 individuals with MDD compared to 29 age and education matched controls in multimodal assessment. The multimodal RDoC assessment included the primary IP biomarker, positron emission tomography (PET) with a selective radiotracer for 5-HT1A [(11C)WAY-100635], as well as event-related functional MRI with a Go/No-go task targeting the Cognitive Control network, neuropsychological assessment of affective perception, negative memory bias and Cognitive Control domains. There was also an exploratory genetic analysis with the serotonin transporter (5-HTTLPR) and monamine oxidase A (MAO-A) genes. In regression analyses, lower 5-HT1A binding potential (BP) in the MDD group was related to diminished engagement of the Cognitive Control network, slowed resolution of interfering cognitive stimuli, one element of Cognitive Control. In contrast, higher/normative levels of 5-HT1A BP in MDD (only) was related to a substantial memory bias toward negative information, but intact resolution of interfering cognitive stimuli and greater engagement of Cognitive Control circuitry. The serotonin transporter risk allele was associated with lower 1a BP and the corresponding imaging and cognitive IPs in MDD. Lowered 5HT 1a BP was present in half of the MDD group relative to the control group. Lowered 5HT 1a BP may represent a subtype including decreased engagement of Cognitive Control network and impaired resolution of interfering cognitive stimuli. Future investigations might link lowered 1a BP to neurobiological pathways and markers, as well as probing subtype-specific treatment targets.

Developmental changes in resting-state functional networks among individuals with and without internalizing psychopathologies

BACKGROUND

Three well-established intrinsic connectivity networks (ICNs) involved in cognitive-affective processing include the cognitive control network (CCN), default mode network (DMN), and salience and emotional network (SEN). Despite recent advances in understanding developmental changes in these ICNs, the majority of research has focused on single seeds or networks in isolation with limited age ranges. Additionally, although internalizing psychopathologies (IPs), such as anxiety and depression, are often characterized by maladaptive cognitive-affective processing styles, it is not clear how IP history influences age-related changes in brain networks.

METHOD

The current study aimed to characterize the normative development of the CCN, DMN, and SEN across a large age-span (7-29 year olds) of typically developing (TD) individuals (n = 97). We also explore how age may impact differences in network connectivity between TD individuals and patients with IPs (n = 136).

RESULTS

Among TD individuals, DMN and CCN connectivity strengthened with age, whereas connectivity between the SEN and ventromedial prefrontal cortex weakened across development. When exploring group (IP vs. TD) differences, the IP group was characterized by greater connectivity between the CCN and cerebellum and between the SEN and caudate from childhood to early adulthood, relative to TD individuals. In addition, patients with IPs, versus TD individuals, exhibited reduced connectivity between the SEN and medial frontal gyrus from adolescence to adulthood.

CONCLUSIONS

The current findings shed light on differential age-related changes in brain network patterns among psychiatrically free, TD individuals and those with internalizing disorders, and may provide plausible targets for novel mechanism-based treatments that differ based on developmental stage.

Intrinsic Brain Network Biomarkers of Antidepressant Response: a Review

PURPOSE OF REVIEW

Poor treatment response is a hallmark of major depressive disorder. To tackle this problem, recent neuroimaging studies have sought to characterize antidepressant response in terms of pretreatment differences in intrinsic functional brain networks. Our aim is to review recent studies that predict antidepressant response using intrinsic network connectivity. We discuss current methodological limitations and directions for future antidepressant biomarker studies.

RECENT FINDINGS

Functional connectivity stemming from the subgenual and rostral anterior cingulate has shown particular consistency in predicting antidepressant response. Differences in this connectivity may prove fruitful in differentiating treatment responders to many antidepressant interventions. Future biomarker studies should integrate biological MDD subtypes to address the disorder's inherent clinical heterogeneity. These clinical and scientific advancements have the potential to address this population marked by limited treatment response. Methodological considerations, including patient selection, response criteria, and model overfitting, will require future investigation to ensure that biomarkers generalize for prospective prediction of treatment response.

Development of Neuroimaging-Based Biomarkers in Psychiatry

This chapter presents an overview of accumulating neuroimaging data with emphasis on translational potential. The subject will be described in the context of three disease states, i.e., schizophrenia, bipolar disorder, and major depressive disorder, and for three clinical goals, i.e., disease risk assessment, subtyping, and treatment decision.

Classifying Treated vs. Untreated MDD Adolescents from Anatomical Connectivity using Nonlinear SVM

Identification of the treatment-related responders for adolescent Major Depressive Disorder (MDD) is urgently needed to develop effective treatments. In this paper, machine learning based classifiers are used to reveal anatomical features as responders for distinguishing MDD patients who have received treatment from those who never received any treatment. The features are drawn from two sets of measurements: 1) anatomical connectivity defined by diffusion tensor imaging measurements between a pair of brain regions, and 2) topological measurements from anatomical networks. Feature selection was performed based on p-value and minimum redundancy maximum relevance (mRMR) method to achieve improved classification accuracy. The classification performance is evaluated with a leave-one-out cross-validation method using 37 treated and 15 untreated subjects. The proposed methodology achieves 73% accuracy, 100% specificity, and 100% precision for 52 subjects. The most distinguishing features are the strength of the right hippocampus of the mean diffusivity (MD) network at 18% density and of the track-count (TR) network, the participation coefficient of the left middle temporal gyrus of the radial diffusivity (RD) network at 20% density, the axial diffusivity (AD) connectivity between right middle temporal gyrus and right supramarginal gyrus, the betweenness centrality of the right hippocampus of the TR network at 11% density, the apparent diffusion coefficient (ADC) connectivity between the left pars opercularis and the left rostral anterior cingulate cortex, the clustering coefficient of the middle anterior corpus callosum of the TR network at 11% density, and the AD connectivity between the left pars opercularis and the left rostral anterior cingulate cortex.

Predicting Treatment Response in Depression: The Role of Anterior Cingulate Cortex

Background

Identification of biomarkers predicting therapeutic outcome of antidepressant treatment is one of the most important tasks in current research because it may transform the lengthy process of finding the right treatment for a given individual with depression. In the current study, we explored the potential of pretreatment pregenual anterior cingulate cortex activity as a putative biomarker of treatment response.

Methods

Thirty-two medication-free patients with depression were treated for 6 weeks with a selective serotonin reuptake inhibitor, escitalopram. Before treatment began, patients underwent an fMRI scan testing response to brief, masked, presentations of facial expression depicting sadness and happiness.

Results

After 6 weeks of treatment, there were 20 selective serotonin reuptake inhibitor responders and 12 nonresponders. Increased pretreatment pregenual anterior cingulate cortex activity to sad vs happy faces was observed in responders relative to nonresponders. A leave-one-out analysis suggested that activity in the anterior cingulate cortex was able to predict response status at the level of the individual participant.

Conclusions

The study supports the notion of pregenual anterior cingulate cortex as a promising predictor of antidepressant response.

Cognitive control neuroimaging measures differentiate between those with and without future recurrence of depression

Background

Major Depressive Disorder (MDD) is a prevalent, disruptive illness. A majority of those with MDD are at high risk for recurrence and increased risk for morbidity and mortality. This study examined whether multimodal baseline (and retest) Cognitive Control performance and neuroimaging markers (task activation and neural connectivity between key brain nodes) could differentiate between those with and without future recurrence of a major depressive (MD) episode within one year. We hypothesized that performance and neuroimaging measures of Cognitive Control would identify markers that differ between these two groups.

Methods

A prospective cohort study of young adults (ages 18–23) with history (h) of early-onset MDD (N = 60), now remitted, and healthy young adults (N = 49). Baseline Cognitive Control measures of performance, task fMRI and resting state connectivity (and reliability retest 4–12 weeks later) were used to compare those with future recurrence of MDD (N = 21) relative to those without future recurrence of MDD (N = 34 with resilience). The measures tested were (1) Parametric Go/No-Go (PGNG) performance, and task activation for (2) PGNG Correct Rejections, (3) PGNG Commission errors, and (4 & 5), resting state connectivity analyses of Cognitive Control Network to and from subgenual anterior cingulate.

Results

Relative to other groups at baseline, the group with MDD Recurrence had less bilateral middle frontal gyrus activation during commission errors. MDD Recurrence exhibited greater connectivity of right middle frontal gyrus to subgenual anterior cingulate (SGAC). SGAC connectivity was also elevated in this group to numerous regions in the Cognitive Control Network. Moderate to strong ICCs were present from test to retest, and highest for rs-fMRI markers. There were modest, significant correlations between task, connectivity and behavioral markers that distinguished between groups.

Conclusion

Markers of Cognitive Control function could identify those with early course MD who are at risk for depression recurrence. Those at high risk for recurrence would benefit from maintenance or preventative treatments. Future studies could test and validate these markers as potential predictors, accounting for sample selection and bias in feature detection.

•

Tools are needed to increase identification of MDD recurrence

•

Cognitive control behavior and depression symptoms have been predictive of recurrence in prior studies, but with low accuracy

•

In remitted Major Depressive Disorder, those who will go on to have future depressive episodes differed in cognitive control activation and connectivity

•

Symptoms, performance, task activation, and seed-based connectivity can contribute to identification of risk for recurrence

Progress in Elucidating Biomarkers of Antidepressant Pharmacological Treatment Response: A Systematic Review and Meta-analysis of the Last 15 Years

BACKGROUND

Antidepressant drugs are widely prescribed, but response rates after 3 months are only around one-third, explaining the importance of the search of objectively measurable markers predicting positive treatment response. These markers are being developed in different fields, with different techniques, sample sizes, costs, and efficiency. It is therefore difficult to know which ones are the most promising.

OBJECTIVE

Our purpose was to compute comparable (i.e., standardized) effect sizes, at study level but also at marker level, in order to conclude on the efficacy of each technique used and all analyzed markers.

METHODS

We conducted a systematic search on the PubMed database to gather all articles published since 2000 using objectively measurable markers to predict antidepressant response from five domains, namely cognition, electrophysiology, imaging, genetics, and transcriptomics/proteomics/epigenetics. A manual screening of the abstracts and the reference lists of these articles completed the search process.

RESULTS

Executive functioning, theta activity in the rostral Anterior Cingular Cortex (rACC), and polysomnographic sleep measures could be considered as belonging to the best objectively measured markers, with a combined d around 1 and at least four positive studies. For inter-category comparisons, the approaches that showed the highest effect sizes are, in descending order, imaging (combined d between 0.703 and 1.353), electrophysiology (0.294-1.138), cognition (0.929-1.022), proteins/nucleotides (0.520-1.18), and genetics (0.021-0.515).

CONCLUSION

Markers of antidepressant treatment outcome are numerous, but with a discrepant level of accuracy. Many biomarkers and cognitions have sufficient predictive value (d ≥ 1) to be potentially useful for clinicians to predict outcome and personalize antidepressant treatment.

Pathways to Neuroprediction: Opportunities and challenges to prediction of treatment response in depression

PURPOSE OF REVIEW

We set out to review the current state of science in neuroprediction, using biological measures of brain function, with task based fMRI to prospectively predict response to a variety of treatments.

RECENT FINDINGS

Task-based fMRI neuroprediction studies are balanced between whole brain and ROI specific analyses. The predominant tasks are emotion processing, with ROIs based upon amygdala and subgenual anterior cingulate gyrus, both within the salience and emotion network. A rapidly emerging new area of neuroprediction is of disease course and illness recurrence. Concerns include use of open-label and single arm studies, lack of consideration of placebo effects, unbalanced adjustments for multiple comparisons (over focus on type I error), small sample sizes, unreported effect sizes, overreliance on ROI studies.

SUMMARY

There is a need to adjust neuroprediction study reporting so that greater coherence can facilitate meta analyses, and increased funding for more multiarm studies in neuroprediction.

Differences in the cognitive profile of depression between racial groups

BACKGROUND

There is little research concerning whether race is associated with different clinical presentations of psychiatric disorders. Understanding the racial aspects of cognitive differences among depressed adults could help identify specific targets for depression treatment.

METHODS

We recruited 59 participants (52.5% white, 47.5% African American), age 18 to 60, with a current diagnosis of major depressive disorder (MDD). All participants underwent a cognitive battery that included tasks assessing decision-making, response inhibition, working memory, and executive planning. Differences between white and African American groups were analyzed using analysis of variance or equivalent nonparametric tests.

RESULTS

There were no differences in demographic variables of age, sex, and education, but white individuals reported a significantly earlier age of first depressive episode compared with African American individuals. The African American group, however, showed significantly lower quality of decision-making and less risk adjustment on the Cambridge Gambling Task, and made more errors on the Spatial Working Memory task.

CONCLUSIONS

Our finding that specific cognitive domains differ as a function of racial differences in MDD might have differential prognostic and treatment implications.

Disrupted engagement of networks supporting hot and cold cognition in remitted major depressive disorder

BACKGROUND

Major depressive disorder (MDD) is characterized by dysfunction in cognitive and emotional systems. However, the neural network correlates of cognitive control (cold cognition) and emotion processing (hot cognition) during the remitted state of MDD (rMDD) remain unclear and not fully probed, which has important implications for identifying intermediate phenotypes of depression risk.

METHODS

43 young adults with rMDD and 33 healthy controls (HCs) underwent fMRI while completing separate tasks of cold cognition (Parametric Go/No-Go test) and hot cognition (Facial Emotion Processing Test). Two 2 group (rMDD, HC) × 2 event (sad/fearful faces, correct rejections) factorial models of activation were calculated in SPM8. Functional activation was evaluated in the salience and emotional network (SEN) and the cognitive control network (CCN), including hypothesized interaction between group and task within the CCN.

RESULTS

Individuals with rMDD demonstrated greater spatial extent of suprathreshold activation within the SEN during sad faces relative to HCs. There were several regions within the CCN in which HCs showed greater activation than rMDD during correct rejections of lures, whereas individuals with rMDD showed greater activation than HCs during sad or fearful faces.

LIMITATIONS

Results were not directly compared with active MDD.

CONCLUSIONS

These results provide evidence of deficient CCN engagement during cognitive control in rMDD (dysfunctional cold cognition). Elevated SEN activation during sad faces could represent heightened salience of negative emotional faces in rMDD; elevated CCN activation during emotional faces in rMDD could represent compensatory regulatory control. These group differences may represent vulnerability factors, scars of prior depressive episodes, or processes maintaining wellness.

Neural and Behavioral Predictors of Treatment Efficacy on Mood Symptoms and Cognition in Mood Disorders: A Systematic Review

Background: The clinical and etiological heterogeneity of mood disorders impede identification of effective treatments for the individual patient. This highlights a need for early neuronal and behavioral biomarkers for treatment efficacy, which can provide a basis for more personalized treatments. The present systematic review aimed to identify the most consistent neuronal and behavioral predictors of treatment efficacy on mood symptoms and cognitive impairment in mood disorders. Methods: We identified and included 60 original peer-reviewed studies investigating neuroimaging and behavioral predictors of treatment efficacy within the domains of emotional and non-emotional cognition, structural neuroimaging, and resting state functional connectivity in patients with unipolar or bipolar disorder. Results: Lower baseline responsivity in limbic regions coupled with heightened medial and dorsal prefrontal responses to emotional stimuli were the most consistent predictors of response to pharmacotherapy for depression. In contrast, heightened limbic and ventral prefrontal reactivity to emotional stimuli seemed to predict efficacy of psychological interventions. Early modulation of fronto-limbic activity and reduction in negative bias were also associated with treatment response. Better performance on non-emotional tests at baseline was relatively consistently associated with efficacy on mood symptoms, whereas the association between neural activity during non-emotional tests and treatment response was less clear. Other baseline factors associated with treatment response were greater white matter integrity, resting state functional connectivity, more prefrontal gray matter volume as well as an early increase following short administered treatment. Finally, emerging evidence indicates that baseline cognitive deficits are associated with greater chances of achieving treatment efficacy on cognition. Conclusions: Patients' profile of emotional and non-emotional cognition and neural activity-and the early treatment-associated changes in neural and cognitive function-may be useful for guiding treatments for depression. While cognitive deficits at baseline seem to improve chances of treatment efficacy on cognition, more studies of this association are urgently needed.

A Systematic Review of Cognitive Predictors of Treatment Outcome in Major Depression

Background: Research suggests that only 50% of patients with major depression respond to psychotherapy or pharmacological treatment, and relapse is common. Therefore, there is interest in elucidating factors that help predict clinical response. Cognitive impairment is a key feature of depression, which often persists beyond remission; thus, the aim of this systematic review was to determine whether baseline cognitive functioning can predict treatment outcomes in individuals with depression. Method: Studies examining cognitive predictors of treatment response in depression were identified using Pub Med and Web of Science databases. Given the heterogeneity of outcome measures, the variety of treatment protocols, and the differing ways in which data was presented and analyzed, a narrative rather than meta-analytic review technique was used. Results: 39 studies met inclusion criteria. Findings in younger adult samples were inconclusive. There was some evidence for a predictive effect of executive function and to a lesser extent, psychomotor speed, on treatment response. There was no evidence of learning or memory being associated with treatment response. In older-aged samples, the evidence was much more consistent, suggesting that poor executive function predicts poor response to SSRIs. Conclusions: Findings from the present review suggest that certain aspects of cognitive functioning, particularly executive function, may be useful in predicting treatment response in depression. This is certainly the case in elderly samples, with evidence suggesting that poor executive functioning predicts poor response to SSRIs. With further research, baseline cognitive functioning may serve as a factor which helps guide clinical decision making. Moreover, cognitive deficits may become targets for specific pharmacological or psychological treatments, with the hope of improving overall outcome.

Neuroimaging Advances for Depression

Depression is one of the world's most prevalent mental health problems, with as many as 350 million sufferers worldwide and close to 20 million sufferers in the US. While neuroimaging applications for identifying various types of depression have made enormous strides in recent years, no findings have been sufficiently replicated or considered significant enough to warrant application in clinical settings. Our authors are well equipped to tell us what the future may bring.

Brain-Based Biomarkers for the Treatment of Depression: Evolution of an Idea

An ambition of depression biomarker research is to augment psychometric and cognitive assessment of clinically relevant phenomena with neural measures. Although such applications have been slow to arrive, we observe a steady evolution of the idea and anticipate emerging technologies with some optimism. To highlight critical themes and innovations in depression biomarker research, we take as our point of reference a specific research narrative. We begin with an early model of frontal-limbic dysfunction, which represents a conceptual shift from localized pathology to understanding symptoms as an emergent property of distributed networks. Over the decades, this model accommodates perspectives from neurology, psychiatry, clinical, and cognitive neuroscience, and preserves past insight as more complex methods become available. We also track the expanding mission of brain biomarker research: from the development of diagnostic tools to treatment selection algorithms, measures of neurocognitive functioning and novel targets for neuromodulation. To conclude, we draw from this particular research narrative future directions for biomarker research. We emphasize integration of measurement modalities to describe dynamic change in domain-general networks, and we speculate that a brain-based framework for psychiatric problems may dissolve classical diagnostic and disciplinary boundaries. (JINS, 2017, 23, 870-880).