Accounting for Dynamic Fluctuations across Time when Examining fMRI Test-Retest Reliability: Analysis of a Reward Paradigm in the EMBARC Study

Treatment Response Prediction in Major Depressive Disorder Using Multimodal MRI and Clinical Data: Secondary Analysis of a Randomized Clinical Trial

OBJECTIVE

Response to antidepressant treatment in major depressive disorder varies substantially between individuals, which lengthens the process of finding effective treatment. The authors sought to determine whether a multimodal machine learning approach could predict early sertraline response in patients with major depressive disorder. They assessed the predictive contribution of MR neuroimaging and clinical assessments at baseline and after 1 week of treatment.

METHODS

This was a preregistered secondary analysis of data from the Establishing Moderators and Biosignatures of Antidepressant Response in Clinical Care (EMBARC) study, a multisite double-blind, placebo-controlled randomized clinical trial that included 296 adult outpatients with unmedicated recurrent or chronic major depressive disorder. MR neuroimaging and clinical data were collected before and after 1 week of treatment. Performance in predicting response and remission, collected after 8 weeks, was quantified using balanced accuracy (bAcc) and area under the receiver operating characteristic curve (AUROC) scores.

RESULTS

A total of 229 patients were included in the analyses (mean age, 38 years [SD=13]; 66% female). Internal cross-validation performance in predicting response to sertraline (bAcc=68% [SD=10], AUROC=0.73 [SD=0.03]) was significantly better than chance. External cross-validation on data from placebo nonresponders (bAcc=62%, AUROC=0.66) and placebo nonresponders who were switched to sertraline (bAcc=65%, AUROC=0.68) resulted in differences that suggest specificity for sertraline treatment compared with placebo treatment. Finally, multimodal models outperformed unimodal models.

CONCLUSIONS

The study results confirm that early sertraline treatment response can be predicted; that the models are sertraline specific compared with placebo; that prediction benefits from integrating multimodal MRI data with clinical data; and that perfusion imaging contributes most to these predictions. Using this approach, a lean and effective protocol could individualize sertraline treatment planning to improve psychiatric care.

A novel technique for delineating the effect of variation in the learning rate on the neural correlates of reward prediction errors in model-based fMRI

Introduction

Computational models play an increasingly important role in describing variation in neural activation in human neuroimaging experiments, including evaluating individual differences in the context of psychiatric neuroimaging. In particular, reinforcement learning (RL) techniques have been widely adopted to examine neural responses to reward prediction errors and stimulus or action values, and how these might vary as a function of clinical status. However, there is a lack of consensus around the importance of the precision of free parameter estimation for these methods, particularly with regard to the learning rate. In the present study, I introduce a novel technique which may be used within a general linear model (GLM) to model the effect of mis-estimation of the learning rate on reward prediction error (RPE)-related neural responses.

Methods

Simulations employed a simple RL algorithm, which was used to generate hypothetical neural activations that would be expected to be observed in functional magnetic resonance imaging (fMRI) studies of RL. Similar RL models were incorporated within a GLM-based analysis method including derivatives, with individual differences in the resulting GLM-derived beta parameters being evaluated with respect to the free parameters of the RL model or being submitted to other validation analyses.

Results

Initial simulations demonstrated that the conventional approach to fitting RL models to RPE responses is more likely to reflect individual differences in a reinforcement efficacy construct (lambda) rather than learning rate (alpha). The proposed method, adding a derivative regressor to the GLM, provides a second regressor which reflects the learning rate. Validation analyses were performed including examining another comparable method which yielded highly similar results, and a demonstration of sensitivity of the method in presence of fMRI-like noise.

Conclusion

Overall, the findings underscore the importance of the lambda parameter for interpreting individual differences in RPE-coupled neural activity, and validate a novel neural metric of the modulation of such activity by individual differences in the learning rate. The method is expected to find application in understanding aberrant reinforcement learning across different psychiatric patient groups including major depression and substance use disorder.

Structural brain measures linked to clinical phenotypes in major depression replicate across clinical centres

Abnormalities in brain structural measures, such as cortical thickness and subcortical volumes, are observed in patients with major depressive disorder (MDD) who also often show heterogeneous clinical features. This study seeks to identify the multivariate associations between structural phenotypes and specific clinical symptoms, a novel area of investigation. T1-weighted magnetic resonance imaging measures were obtained using 3 T scanners for 178 unmedicated depressed patients at four academic medical centres. Cortical thickness and subcortical volumes were determined for the depressed patients and patients' clinical presentation was characterized by 213 item-level clinical measures, which were grouped into several large, homogeneous categories by K-means clustering. The multivariate correlations between structural and cluster-level clinical-feature measures were examined using canonical correlation analysis (CCA) and confirmed with both 5-fold and leave-one-site-out cross-validation. Four broad types of clinical measures were detected based on clustering: an anxious misery composite (composed of item-level depression, anxiety, anhedonia, neuroticism and suicidality scores); positive personality traits (extraversion, openness, agreeableness and conscientiousness); reported history of physical/emotional trauma; and a reported history of sexual abuse. Responses on the item-level anxious misery measures were negatively associated with cortical thickness/subcortical volumes in the limbic system and frontal lobe; reported childhood history of physical/emotional trauma and sexual abuse measures were negatively correlated with entorhinal thickness and left hippocampal volume, respectively. In contrast, the positive traits measures were positively associated with hippocampal and amygdala volumes and cortical thickness of the highly-connected precuneus and cingulate cortex. Our findings suggest that structural brain measures may reflect neurobiological mechanisms underlying MDD features.

Great Expectations: A Critical Review of and Suggestions for the Study of Reward Processing as a Cause and Predictor of Depression

Both human and animal studies support the relationship between depression and reward processing abnormalities, giving rise to the expectation that neural signals of these processes may serve as biomarkers or mechanistic treatment targets. Given the great promise of this research line, we scrutinized those findings and the theoretical claims that underlie them. To achieve this, we applied the framework provided by classical work on causality as well as contemporary approaches to prediction. We identified a number of conceptual, practical, and analytical challenges to this line of research and used a preregistered meta-analysis to quantify the longitudinal associations between reward processing abnormalities and depression. We also investigated the impact of measurement error on reported data. We found that reward processing abnormalities do not reach levels that would be useful for clinical prediction, yet the available evidence does not preclude a possible causal role in depression.

Translating big data to better treatment in bipolar disorder - a manifesto for coordinated action

Bipolar disorder (BD) is a major healthcare and socio-economic challenge. Despite its substantial burden on society, the research activity in BD is much smaller than its economic impact appears to demand. There is a consensus that the accurate identification of the underlying pathophysiology for BD is fundamental to realize major health benefits through better treatment and preventive regimens. However, to achieve these goals requires coordinated action and innovative approaches to boost the discovery of the neurobiological underpinnings of BD, and rapid translation of research findings into development and testing of better and more specific treatments. To this end, we here propose that only a large-scale coordinated action can be successful in integrating international big-data approaches with real-world clinical interventions. This could be achieved through the creation of a Global Bipolar Disorder Foundation, which could bring government, industry and philanthropy together in common cause. A global initiative for BD research would come at a highly opportune time given the seminal advances promised for our understanding of the genetic and brain basis of the disease and the obvious areas of unmet clinical need. Such an endeavour would embrace the principles of open science and see the strong involvement of user groups and integration of dissemination and public involvement with the research programs. We believe the time is right for a step change in our approach to understanding, treating and even preventing BD effectively.

Reward related ventral striatal activity and differential response to sertraline versus placebo in depressed individuals

Medications to treat major depressive disorder (MDD) are not equally effective across patients. Given that neural response to rewards is altered in MDD and given that reward-related circuitry is modulated by dopamine and serotonin, we examined, for the first time, whether reward-related neural activity moderated response to sertraline, an antidepressant medication that targets these neurotransmitters. A total of 222 unmedicated adults with MDD randomized to receive sertraline (n = 110) or placebo (n = 112) in the Establishing Moderators and Biosignatures of Antidepressant Response in Clinical Care (EMBARC) study completed demographic and clinical assessments, and pretreatment functional magnetic resonance imaging while performing a reward task. We tested whether an index of reward system function in the ventral striatum (VS), a key reward circuitry region, moderated differential response to sertraline versus placebo, assessed with the Hamilton Rating Scale for Depression (HSRD) over 8 weeks. We observed a significant moderation effect of the reward index, reflecting the temporal dynamics of VS activity, on week-8 depression levels (Fs ≥ 9.67, ps ≤ 0.002). Specifically, VS responses that were abnormal with respect to predictions from reinforcement learning theory were associated with lower week-8 depression symptoms in the sertraline versus placebo arms. Thus, a more abnormal pattern of pretreatment VS dynamic response to reward expectancy (expected outcome value) and prediction error (difference between expected and actual outcome), likely reflecting serotonergic and dopaminergic deficits, was associated with better response to sertraline than placebo. Pretreatment measures of reward-related VS activity may serve as objective neural markers to advance efforts to personalize interventions by guiding individual-level choice of antidepressant treatment.

Reliability of a functional magnetic resonance imaging task of emotional conflict in healthy participants

Task-based functional neuroimaging methods are increasingly being used to identify biomarkers of treatment response in psychiatric disorders. To facilitate meaningful interpretation of neural correlates of tasks and their potential changes with treatment over time, understanding the reliability of the blood-oxygen-level dependent (BOLD) signal of such tasks is essential. We assessed test-retest reliability of an emotional conflict task in healthy participants collected as part of the Canadian Biomarker Integration Network in Depression. Data for 36 participants, scanned at three time points (weeks 0, 2, and 8) were analyzed, and intra-class correlation coefficients (ICC) were used to quantify reliability. We observed moderate reliability (median ICC values between 0.5 and 0.6), within occipital, parietal, and temporal regions, specifically for conditions of lower cognitive complexity, that is, face, congruent or incongruent trials. For these conditions, activation was also observed within frontal and sub-cortical regions, however, their reliability was poor (median ICC < 0.2). Clinically relevant prognostic markers based on task-based fMRI require high predictive accuracy at an individual level. For this to be achieved, reliability of BOLD responses needs to be high. We have shown that reliability of the BOLD response to an emotional conflict task in healthy individuals is moderate. Implications of these findings to further inform studies of treatment effects and biomarker discovery are discussed.

Test-retest reliability of brain responses to risk-taking during the balloon analogue risk task

The Balloon Analogue Risk Task (BART) provides a reliable and ecologically valid model for the assessment of individual risk-taking propensity and is frequently used in neuroimaging and developmental research. Although the test-retest reliability of risk-taking behavior during the BART is well established, the reliability of brain activation patterns in response to risk-taking during the BART remains elusive. In this study, we used functional magnetic resonance imaging (fMRI) and evaluated the test-retest reliability of brain responses in 34 healthy adults during a modified BART by calculating the intraclass correlation coefficients (ICC) and Dice’s similarity coefficients (DSC). Analyses revealed that risk-induced brain activation patterns showed good test-retest reliability (median ICC = 0.62) and moderate to high spatial consistency, while brain activation patterns associated with win or loss outcomes only had poor to fair reliability (median ICC = 0.33 for win and 0.42 for loss). These findings have important implications for future utility of the BART in fMRI to examine brain responses to risk-taking and decision-making.

Is the encoding of Reward Prediction Error reliable during development?

Reward Prediction Errors (RPEs), defined as the difference between the expected and received outcomes, are integral to reinforcement learning models and play an important role in development and psychopathology. In humans, RPE encoding can be estimated using fMRI recordings, however, a basic measurement property of RPE signals, their test-retest reliability across different time scales, remains an open question. In this paper, we examine the 3-month and 3-year reliability of RPE encoding in youth (mean age at baseline = 10.6 ± 0.3 years), a period of developmental transitions in reward processing. We show that RPE encoding is differentially distributed between the positive values being encoded predominantly in the striatum and negative RPEs primarily encoded in the insula. The encoding of negative RPE values is highly reliable in the right insula, across both the long and the short time intervals. Insula reliability for RPE encoding is the most robust finding, while other regions, such as the striatum, are less consistent. Striatal reliability appeared significant as well once covarying for factors, which were possibly confounding the signal to noise ratio. By contrast, task activation during feedback in the striatum is highly reliable across both time intervals. These results demonstrate the valence-dependent differential encoding of RPE signals between the insula and striatum, and the consistency of RPE signals or lack thereof, during childhood and into adolescence. Characterizing the regions where the RPE signal in BOLD fMRI is a reliable marker is key for estimating reward-processing alterations in longitudinal designs, such as developmental or treatment studies.

Predictive modeling of treatment resistant depression using data from STAR*D and an independent clinical study

Identification of risk factors of treatment resistance may be useful to guide treatment selection, avoid inefficient trial-and-error, and improve major depressive disorder (MDD) care. We extended the work in predictive modeling of treatment resistant depression (TRD) via partition of the data from the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) cohort into a training and a testing dataset. We also included data from a small yet completely independent cohort RIS-INT-93 as an external test dataset. We used features from enrollment and level 1 treatment (up to week 2 response only) of STAR*D to explore the feature space comprehensively and applied machine learning methods to model TRD outcome at level 2. For TRD defined using QIDS-C₁₆ remission criteria, multiple machine learning models were internally cross-validated in the STAR*D training dataset and externally validated in both the STAR*D testing dataset and RIS-INT-93 independent dataset with an area under the receiver operating characteristic curve (AUC) of 0.70–0.78 and 0.72–0.77, respectively. The upper bound for the AUC achievable with the full set of features could be as high as 0.78 in the STAR*D testing dataset. Model developed using top 30 features identified using feature selection technique (k-means clustering followed by χ² test) achieved an AUC of 0.77 in the STAR*D testing dataset. In addition, the model developed using overlapping features between STAR*D and RIS-INT-93, achieved an AUC of > 0.70 in both the STAR*D testing and RIS-INT-93 datasets. Among all the features explored in STAR*D and RIS-INT-93 datasets, the most important feature was early or initial treatment response or symptom severity at week 2. These results indicate that prediction of TRD prior to undergoing a second round of antidepressant treatment could be feasible even in the absence of biomarker data.

Peripheral biomarkers of major depression and antidepressant treatment response: Current knowledge and future outlooks

BACKGROUND

In recent years, we have accomplished a deeper understanding about the pathophysiology of major depressive disorder (MDD). Nevertheless, this improved comprehension has not translated to improved treatment outcome, as identification of specific biologic markers of disease may still be crucial to facilitate a more rapid, successful treatment. Ongoing research explores the importance of screening biomarkers using neuroimaging, neurophysiology, genomics, proteomics, and metabolomics measures.

RESULTS

In the present review, we highlight the biomarkers that are differentially expressed in MDD and treatment response and place a particular emphasis on the most recent progress in advancing technology which will continue the search for blood-based biomarkers.

LIMITATIONS

Due to space constraints, we are unable to detail all biomarker platforms, such as neurophysiological and neuroimaging markers, although their contributions are certainly applicable to a biomarker review and valuable to the field.

CONCLUSIONS

Although the search for reliable biomarkers of depression and/or treatment outcome is ongoing, the rapidly-expanding field of research along with promising new technologies may provide the foundation for identifying key factors which will ultimately help direct patients toward a quicker and more effective treatment for MDD.

Test-retest reliability of cerebral blood flow in healthy individuals using arterial spin labeling: Findings from the EMBARC study

INTRODUCTION

Previous investigations of test-retest reliability of cerebral blood flow (CBF) at rest measured with pseudo-continuous Arterial Spin Labeling (pCASL) demonstrated good reliability, but are limited by the use of similar scanner platforms. In the present study we examined test-retest reliability of CBF in regions implicated in emotion and the default mode network.

MATERIAL AND METHODS

We measured absolute and relative CBF at rest in thirty-one healthy subjects in two scan sessions, one week apart, at four different sites and three different scan platforms. We derived CBF from pCASL images with an automated algorithm and calculated intra-class correlation coefficients (ICCs) across sessions for regions of interest. In addition, we investigated site effects.

RESULTS

For both absolute and relative CBF measures, ICCs were good to excellent (i.e. >0.6) in most brain regions, with highest values observed for the subgenual anterior cingulate cortex and ventral striatum. A leave-one-site-out cross validation analysis did not show a significant effect for site on whole brain CBF and there was no proportional bias across sites. However, a significant site effect was present in the repeated measures ANOVA.

CONCLUSIONS

The high test-retest reliability of CBF measured with pCASL in a range of brain regions implicated in emotion and salience processing, emotion regulation, and the default mode network, which have been previously linked to depression symptomatology supports its use in studies that aim to identify neuroimaging biomarkers of treatment response.

Reliability in adolescent fMRI within two years - a comparison of three tasks

Longitudinal developmental fMRI studies just recently began to focus on within-subject reliability using the intraclass coefficient (ICC). It remains largely unclear which degree of reliability can be achieved in developmental studies and whether this depends on the type of task used. Therefore, we aimed to systematically investigate the reliability of three well-classified tasks: an emotional attention, a cognitive control, and an intertemporal choice paradigm. We hypothesized to find higher reliability in the cognitive task than in the emotional or reward-related task. 104 healthy mid-adolescents were scanned at age 14 and again at age 16 within M = 1.8 years using the same paradigms, scanner, and scanning protocols. Overall, we found both variability and stability (i.e. poor to excellent ICCs) depending largely on the region of interest (ROI) and task. Contrary to our hypothesis, whole brain reliability was fair for the cognitive control task but good for the emotional attention and intertemporal choice task. Subcortical ROIs (ventral striatum, amygdala) resulted in lower ICCs than visual ROIs. Current results add to the yet sparse overall ICC literature in both developing samples and adults. This study shows that analyses of stability, i.e. reliability, are helpful benchmarks for longitudinal studies and their implications for adolescent development.

A pathway linking reward circuitry, impulsive sensation-seeking and risky decision-making in young adults: identifying neural markers for new interventions

High trait impulsive sensation seeking (ISS) is common in 18-25-year olds, and is associated with risky decision-making and deleterious outcomes. We examined relationships among: activity in reward regions previously associated with ISS during an ISS-relevant context, uncertain reward expectancy (RE), using fMRI; ISS impulsivity and sensation-seeking subcomponents; and risky decision-making in 100, transdiagnostically recruited 18-25-year olds. ISS, anhedonia, anxiety, depression and mania were measured using self-report scales; clinician-administered scales also assessed the latter four. A post-scan risky decision-making task measured 'risky' (possible win/loss/mixed/neutral) fMRI-task versus 'sure thing' stimuli. 'Bias' reflected risky over safe choices. Uncertain RE-related activity in left ventrolateral prefrontal cortex and bilateral ventral striatum was positively associated with an ISS composite score, comprising impulsivity and sensation-seeking-fun-seeking subcomponents (ISSc; P⩽0.001). Bias positively associated with sensation seeking-experience seeking (ES; P=0.003). This relationship was moderated by ISSc (P=0.009): it was evident only in high ISSc individuals. Whole-brain analyses showed a positive relationship between: uncertain RE-related left ventrolateral prefrontal cortical activity and ISSc; uncertain RE-related visual attention and motor preparation neural network activity and ES; and uncertain RE-related dorsal anterior cingulate cortical activity and bias, specifically in high ISSc participants (all ps<0.05, peak-level, family-wise error corrected). We identify an indirect pathway linking greater levels of uncertain RE-related activity in reward, visual attention and motor networks with greater risky decision-making, via positive relationships with impulsivity, fun seeking and ES. These objective neural markers of high ISS can guide new treatment developments for young adults with high levels of this debilitating personality trait.

Clashing Diagnostic Approaches: DSM-ICD Versus RDoC

Since at least the middle of the past century, one overarching model of psychiatric classification has reigned supreme, namely, that of the Diagnostic and Statistical Manual of Mental Disorders and the International Statistical Classification of Diseases and Related Health Problems (herein referred to as DSM-ICD). This DSM-ICD approach embraces an Aristotelian view of mental disorders as largely discrete entities that are characterized by distinctive signs, symptoms, and natural histories. Over the past several years, however, a competing vision, namely, the Research Domain Criteria (RDoC) initiative launched by the National Institute of Mental Health, has emerged in response to accumulating anomalies within the DSM-ICD system. In contrast to DSM-ICD, RDoC embraces a Galilean view of psychopathology as the product of dysfunctions in neural circuitry. RDoC appears to be a valuable endeavor that holds out the long-term promise of an alternative system of mental illness classification. We delineate three sets of pressing challenges--conceptual, methodological, and logistical/pragmatic--that must be addressed for RDoC to realize its scientific potential. We conclude with a call for further research, including investigation of a rapprochement between Aristotelian and Galilean approaches to psychiatric classification.

Impact of the glucocorticoid receptor BclI polymorphism on reward expectancy and prediction error related ventral striatal reactivity in depressed and healthy individuals

There is evidence that reward-related neural reactivity is altered in depressive disorders. Glucocorticoids influence dopaminergic transmission, which is widely implicated in reward processing. However, no studies have examined the effect of glucocorticoid receptor gene polymorphisms on reward-related neural reactivity in depressed or healthy individuals. Fifty-nine depressed individuals with major depressive disorder (n=33) or bipolar disorder (n=26), and 32 healthy individuals were genotyped for the glucocorticoid receptor BclI G/C polymorphism, and underwent functional magnetic resonance imaging during a monetary reward task. We examined the effect of the glucocorticoid receptor BclI G/C polymorphism on reward expectancy (RE; expected outcome value) and prediction error (PE; discrepancy between expected and actual outcome) related ventral striatal reactivity. There was a significant interaction between reward condition and BclI genotype (p=0.007). C-allele carriers showed higher PE than RE-related right ventral striatal reactivity (p<0.001), whereas no such difference was observed in G/G homozygotes. Accordingly, C-allele carriers showed a greater difference between PE and RE-related right ventral striatal reactivity than G/G homozygotes (p<0.005), and also showed lower RE-related right ventral striatal reactivity than G/G homozygotes (p=0.011). These findings suggest a slowed transfer from PE to RE-related ventral striatal responses during reinforcement learning in C-allele carriers, regardless of diagnosis, possibly due to altered dopamine release associated with increased sensitivity to glucocorticoids.