Ellen Leibenluft - US grants

Considerable public attention has focused recently on the question of why there has been a rather dramatic upsurge recently in the rate at which the diagnosis of bipolar disorder (BD) is being assigned to children. In large part the upsurge appears to be due to the fact that children with extremely severe irritability, but without distinct manic episodes, are receiving the diagnosis of BD. At the inception of this project, we defined criteria that would allow us to identify reliably children in this controversial diagnostic group. We called the syndrome severe mood and behavioral dysregulation (SMD), and defined it in terms of impairing symptoms that include abnormal baseline mood (i.e. irritability, anger, and/or sadness), hyperarousal (e.g. insomnia, agitation, distractibility), and increased reactivity to negative emotional stimuli. The specific goals of this project are 1) to identify reliably a group of children with severe mood and behavioral dysregulation in order to characterize them clinically and behaviorally, and follow them longitudinally, 2) to compare the brain function of SMD children (assessed with functional MRI and standardized behvioral testing) to that of children with unequivocal BPD; 3) to test appropriate treatments for SMD. The latter is particularly important, since if youth with SMD are found to have a form of BD that would dictate one course of treatment, whereas if they are found to have (for example) a variant of depression, anxiety, and/or attention deficit hyperactivity disorder, this would dictate a rather different plan for treatment.[unreadable] Since the inception of this project, approximately 200 youth have been recruited into the project, approximately 100 of whom are patients with SMD. Approximately 20 new patients were recruited this year. It is very important to note that these youth with SMD suffer very severe psychiatric impairment, and indeed are as ill as are youth with BD in terms of number of medications prescribed, number of psychiatric hospitalizations, and standardized measures of function. In past years, we demonstrated that youth with SMD are at risk for major depression, rather than necessarily BD, in early adulthood, and that they are less likely than youth with BD to have a parent with BD. These data would suggest that youth with SMD should not be considered to be suffering from a form of BD. On the other hand, youth with SMD and BD differ from youth with depression, attention deficit hyperactivity disorder (ADHD), or anxiety in that only those with BD or SMD have deficits in face emotion identification. This year we completed neuroimaging studies comparing these different populations in terms of neural activation while viewing faces and rating the emotion on the face and their own response to the face. We found that youth with BD and those with SMD differed in amygdala activation while viewing such faces. We also finished a study comparing brain activation in these groups while the subjects perform a task requiring motor inhibition or response flexibility. These data are being analyzed.[unreadable] Finally this year we completed a double-blind placebo-controlled study of lithium, with accompanying magnetic resonance spectroscopy (MRS). The results of this study indicate that lithium did not provide more clinical benefit than placebo, nor were effects on brain chemistry discerned with MRS. A new treatment trial, using different pharmacologic agents, has been submitted for approval.

Bipolar disorder (BD) is a highly impairing illness; when it occurs in children and adolescents, it tends to run a particularly severe course. Children with BD typically require extensive treatment with medication and, often, hospitalization. They have high rates of suicide attempts and have great difficulty navigating normal developmental milestones. Pediatric BD takes a very significant toll on children afflicted with the disease and their families.[unreadable] This study is designed to elucidate the brain mechanisms that mediate bipolar disorder (BD) in children and adolescents, and that confer risk of the illness in youth who have a parent or sibling with BD. We are testing the hypotheses that both youth with BD and those at risk for the illness by virtue of having a parent or sibling with BD will have deficits in processing of emotional stimuli, such as reward and punishment, or faces displaying emotion. These hypotheses are operationalized as follows: 1) children with bipolar disorder, or at familial risk for the illness, will have deficits labeling face emotion; 2) children with bipolar disorder, or at familial risk for the disorder, have difficulty adapting their behavior in response to changes in emotional stimuli in their environment; 3) these deficits will be mediated by dysfunction in ventral prefrontal-amygdala-striatal circuitry. Since the beginning of the project we have enrolled approximately 500 subjects, including youth with BD (approximately 100), youth at risk for BD (approximately 60), control subjects, and adults with BD. This year, approximately 75 new subjects were enrolled.[unreadable] Previous work on this project had identified deficits in face emotion identification in youth with BD, as well as deficits in response flexibility. Work this year extended these findings to youth at risk. Specifically, we demonstrated that youth at risk for BD have deficits in face emotion identification comparable to those seen in youth with the illness. We are currently analyzing data to test whether adults with BD have similar dysfunction. In addition, this year we performed fMRI neuroimaging studies to identify how brain mechanisms mediating this deficit differentiate youth with BD from those with severe irritability (see MH002786-07) and from those with attention deficit hyperactivity disorder (ADHD). When viewing faces and rating their emotional response to them, and to the emotion on the face, we found differences between irritable youth and those with BD in amygdala activation. We are currently performing fMRI studies with youth at risk for BD, and with adults with BD, to determine whether this amygdala dysfunction is present in children at risk, and whether it is unique to children with the disorder or present across the developmental spectrum.[unreadable] With regard to response flexibility, this year we obtained new findings both in youth at risk for BD and in youth with the illness. Although we have not, at this point, identified response flexibility deficits per se in youth at risk for BD, we have identified that they, like youth with the illness, have abnormally high variability in response time. This indicates difficulty sustaining attention to a task and, while it has been identified in other illness, such as ADHD, in our at risk population this deficit was independent of ADHD symptoms. More refined behavioral testing is now being conducted to determine the precise nature of this deficit. In youth with BD, we are using the new imaging modality magnetoencephalography (MEG) to identify deficits in response flexibility, especially those that occur in response to frustration. We have found that, in youth with BD, limbic, parietal, and prefrontal brain regions are particularly responsive to negative feedback, while in control subjects there is increased responsiveness to positive feedback. Planned studies will now compare these findings to youth with severe irritability, using both fMRI and MEG techniques.

As noted above, in response to the concern about the appropriate diagnosis for children with chronic, severe irritability, we defined a group of children whom we described as having severe mood dysregulation (SMD). These children have extremely severe irritability and symptoms of hyperarousal (the latter being similar to those seen in attention deficit hyperactivity disorder (ADHD)) and, although they frequently receive the diagnosis of bipolar disorder (BD), they do not indeed meet diagnostic criteria for BD. Our SMD phenotype formed the basis for the new pediatric diagnosis of mood dysregulation disorder with dysphoria in DSM-5. Since the inception of this project, approximately 200 youth with SMD have been recruited into the project. Approximately 20 new patients were recruited this year. It is very important to note that these youth with SMD suffer very severe psychiatric impairment, and indeed are as ill as are youth with BD in terms of number of medications prescribed, number of psychiatric hospitalizations, and standardized measures of function. Also as noted above, in previous years we demonstrated differences between youth with SMD and those with BD in terms of clinical course, family history, and the brain mechanisms associated with behavioral problems. Since irritable youth have decreased ability to tolerate frustration, this year we continued our work using frustrating tasks inside the fMRI scanner to define the neural circuitry mediating irritability. In a paper in press, we found that, when frustrated, youth with SMD had more difficulty than healthy subjects shifting their attention in order to meet task demands. In addition, when frustrated and receiving negative feedback, compared to healthy subjects, those with SMD exhibited decreased activation in the amygdala, striatum, and parietal regions. On the other hand, when subjects were frustrated and received positive feedback, there was no difference in brain activity between healthy subjects and those with SMD. These findings indicate that in response to negative feedback received in the context of frustration, youth with severe, chronic irritability show abnormally reduced activation in regions implicated in emotion, attention, and reward processing. We will now extend this work by adding comparison groups of youth with BD, anxiety, and attention deficit hyperactivity. In a hypothesis consistent with a Research Domain Criteria (RDoC) formulation, we posit that, across all of these groups, increased irritability will be associated with decreased attentional control and amygdala-parietal-striatal deactivation during frustration. In addition, in a collaboration with extramurally-funded investigators at Virginia Commonwealth University, we will be scanning monozygotic and dizygotic twins while they perform this frustrating task in order to quantify environmental and genetic contributions to the neural mechanisms mediating frustration in healthy youth. In this work, Importantly, our work conceptualizing irritability as a dimensional trait present in healthy youth and across different psychopathologies was greatly facilitated by our development of a brief but valid self- and parent-report instrument that could be used to measure irritability in youth. A variant of this scale was used in the field trials for DSM-5. Our scanning study in twins complements our published and ongoing work on the genetic epidemiology of irritability. In this work, we demonstrated that the heritability of irritability is approximately 0.4 (which is moderate for psychiatric illnesses or traits). Ongoing work is designed to describe the developmental trajectory of genetic influences on irritability i.e., the extent to which different genetic mechanisms mediating irritability come on-line as youth develop. Consistent with our finding on the frustration task, we also found attentional dyscontrol in irritable subjects on a task designed to ascertain abnormal attentional bias toward threatening faces. Thus, in both the frustration and the attentional bias tasks, we found attentional dyscontrol in irritable youth in negative emotional contexts. The abnormal attentional bias toward threat that we saw in irritable children is consistent with the bias seen in youth with anxiety, and adds to emerging data about mechanistic and clinical links between anxiety and irritability. This link has considerable public health importance, since anxiety is a frequent but often unrecognized cause of irritability in youth presenting for psychiatric care. Also, the mechanistic link that we discovered between irritability and anxiety is also consistent with our prior data showing that irritable youth are at increased risk to develop anxiety disorders as adults. Potentially, these associations between anxiety and irritability may also be relevant to treatment. Last year, we began a double-blind trial designed to ascertain whether citalopram (a serotonergic reuptake inhibitor (SRI) antidepressant that is effective in the treatment of pediatric anxiety) plus stimulant is more effective than placebo plus stimulant in the treatment of SMD. Importantly, stimulant and SRI treatment tend to be less toxic than the atypical antipsychotic treatment that is considered first-line treatment for bipolar disorder, yet stimulants and SRI's are relatively contraindicated in patients with bipolar disorder because of concern about possibly inducing a manic episode. Therefore, this treatment trial has considerable public health importance. Thus far, we have randomized approximately 35 children into the trial, and youth are tolerating the experimental treatment well. In addition, we are assisting collaborators at UCLA on a similar trial that is funded extramurally. We are also now extending our treatment work to include an adjunctive trial of computer-based training designed to shift subject's perception of faces from angry toward happy. This work is based on our prior work demonstrating face emotion labeling deficits in youth with SMD, and work by our collaborators (Dr. Marcus Munafo at the University of Bristol and Dr. Yair Bar-Haim at Tel Aviv University) that such face labeling training can be associated with decreases in trait anger and irritability.

In response to the concern about the appropriate diagnosis for children with chronic, severe irritability, we defined a group of children whom we described as having severe mood dysregulation (SMD). These children have extremely severe irritability and symptoms of hyperarousal (the latter being similar to those seen in attention deficit hyperactivity disorder (ADHD)); although they frequently receive the diagnosis of bipolar disorder (BD), they do not indeed meet diagnostic criteria for BD. Our SMD phenotype formed the basis for the new pediatric diagnosis of mood dysregulation disorder with dysphoria (DMDD) in DSM-5. (DMDD differs from SMD primarily in that DMDD does not require the hyperarousal symptoms, related to attention deficit hyperactivity disorder (ADHD)). Since the inception of this project, approximately 270 youth with SMD/DMDD have been recruited into the project. Approximately 25 new patients were recruited this year. It is very important to note that these youth with DMDD suffer very severe psychiatric impairment, and indeed are as ill as youth with BD in terms of number of medications prescribed, number of psychiatric hospitalizations, and standardized measures of function. Also as noted above, in previous years we demonstrated differences between youth with SMD and those with BD in terms of clinical course, family history, and the brain mechanisms associated with behavioral problems. Thus, while we continue to compare youth with BD and those with severe irritability in terms of mediating brain mechanisms, we have become increasingly interested in the pathophysiology of irritability in its own right. Indeed, irritability is one of the most common presenting symptoms in pediatric psychiatry clinics. As noted above, the construct of irritability is particularly well-suited for a transdiagnostic, translational approach consistent with the Research Domain Criteria (RDoC) approach. Therefore, much of our work this year involved examining irritability, not only within the severely irritable DMDD phenotype, but also in other groups that exhibit considerable irritability, including youth with anxiety disorders, ADHD, or bipolar disorder (BD). Across all groups, we characterize irritability as a continuous variable. A major focus of our neuroimaging work includes the use of frustrating tasks, since a hallmark of irritability is difficulty tolerating frustration. In prior work, we demonstrated behavioral and neural differences between irritable and non-irritable youth while completing a frustrating attentional task. We have improved that paradigm and are now applying it in a large sample of youth with DMDD, BD, ADHD, and/or anxiety disorders, as well as healthy youth. This approach allows us to examine the neural correlates of irritability, operationalized dimensionally, while also comparing DSM-5 diagnostic groups in terms of the brain circuitry engaged during a frustrating task. We have also continued our work comparing the neural circuitry mediating face emotion processing among youth with DMDD, BD, and healthy youth. As with the frustration work, we now incorporate a dimensional aspect to the research, examining neural associations with continuous measures of irritability across diagnoses. In work being prepared for publication, we found differences between BD and DMDD in associations between irritability and amygdala and superior temporal gyrus activation, suggesting that the circuitry mediating irritability may not be the same across diagnoses. This emphasizes the potential importance of integrating both categorical (i.e., diagnostic group) and dimensional (e.g., irritability) approaches into research on psychiatric illness. A major focus of our work continues to be treatment. We have continued our double-blind trial designed to ascertain whether citalopram (a serotonergic reuptake inhibitor (SRI) antidepressant that is effective in the treatment of pediatric anxiety) plus stimulant is more effective than placebo plus stimulant in the treatment of severe irritability in youth. Importantly, stimulant and SRI treatment tend to have a lower side-effect burden than the atypical antipsychotic treatment that is considered first-line treatment for bipolar disorder, yet stimulants and SRI's are relatively contraindicated in patients with bipolar disorder because of concern about possibly inducing a manic episode. Therefore, this treatment trial has considerable public health importance. Thus far, we have randomized approximately 50 children into the trial, and youth are tolerating the experimental treatment well. We are also now extending our treatment work to include the development and testing of two novel psychosocial approaches to the treatment of irritability in youth. The first intervention involves computer-based training designed to shift a subject's perception of ambiguous faces from angry toward happy. This work is based on our prior work demonstrating face emotion labeling deficits in youth with SMD, and work by our collaborators showing that such face labeling training can be associated with decreases in trait anger and irritability. We first demonstrated that youth with DMDD are more likely than healthy youth to judge ambiguous faces as angry rather than happy. We then demonstrated in both healthy youth and those with DMDD that we can change this judgment through computer-based training and that, in patients with DMDD, this shift in judgment is associated with decreased irritability. However, this work in DMDD was an open trial, without a control condition. We recently received permission to conduct a controlled trial of this new Interpretation Bias Treatment, along with brain imaging before and after training. In addition, we received permission to develop a cognitive behavioral treatment designed specifically to target irritability. Unlike other available treatments, this would have a strong emphasis on behavioral techniques i.e., exposing the child to tolerable, but frustrating situations, to improve the childs ability to tolerate such situations. The design of this treatment will be based on a neuroscience-based model of irritability that we devised, based in part on hypothesized deficits that irritable children have in the process and content of instrumental learning. We are developing neuroimaging paradigms that can be used pre- and post- treatment to document changes in the circuitry mediating these processes during successful treatment.

Given concerns about the appropriate diagnosis for children with chronic, severe irritability, we defined severe mood dysregulation(SMD)to capture youth with extremely severe irritability and symptoms of hyperarousal. Our SMD phenotype formed the basis for the new pediatric diagnosis of mood dysregulation disorder with dysphoria(DMDD)in DSM-5. Since the inception of this project (ZIA MH002786-17), approximately 375 highly irritable (i.e., those with SMD, DMDD, or sub-threshold DMDD) have been recruited into the project, along with more than 100 youth with ADHD. (Many DMDD patients also have ADHD, and youth with ADHD tend to have less irritability than those with DMDD but more than healthy youth, making them an appropriate comparison group for your work.) Approximately 40 new patients were recruited this year. Youth with DMDD suffer severe psychiatric impairment, in terms of medications received, psychiatric hospitalizations, and standardized measures of function. As noted above, irritability is one of the most common psychiatric symptoms in children, but there has been little brain-based research devoted to it, and there are few evidence-based treatments. This year, we published a translational mechanistic model of irritability suggesting that core deficits in pediatric irritability include aberrant responses to frustration, aberrant approach responses to threat, and deficits in instrumental learning that prevent adaptation to changes in environmental contingencies. Irritability is particularly well-suited for the transdiagnostic, translational approach of the Research Domain Criteria (RDoC). Our work involves examining irritability characterized as a continuous variable, in DMDD and other groups with irritability, including youth with anxiety disorders or ADHD. Our neuroimaging work includes the use of frustrating tasks, since a hallmark of irritability is difficulty tolerating frustration. Previously, we demonstrated behavioral and neural differences between irritable and non-irritable youth during a frustrating attentional task. We improved that paradigm and obtained fMRI data from approximately 200 youth with DMDD, ADHD, and/or anxiety disorders, as well as healthy youth. In a manuscript recently accepted for publication, we found that irritability is associated with increased prefrontal and striatal engagement when youth attempt an attention orienting task after receiving frustrating feedback. Since prefrontal engagement is important in emotion regulation, one interpretation of this finding is that prefrontal emotion regulation mechanisms are less efficient in irritable vs. non-irritable youth. Alternatively, in irritable youth, increased prefrontal engagement may be required after frustration to down-regulate exaggerated subcortical limbic responses e.g., in the striatum. These effects of irritability are present even when ADHD and anxiety are taken into account. We have piloted two additional frustration tasks. The second frustration task differs from the first in the timing of frustration (short blocks of frustration, interspersed randomly with non-frustrating blocks, vs. a long block of non-frustration followed by a long bout of frustration) and in the cognitive task (attention orienting vs. cognitive flexibility). This second task allows us to test whether, across different task timing and cognitive tasks, increased irritability is associated with increased prefrontal engagement after frustration. We are now analyzing data from 80 healthy youth or those with DMDD, anxiety, or ADHD. Our third frustration task allows us to determine if irritability is associated with instrumental learning deficits at baseline and after frustration. Instrumental learning is the process by which people learn which of their behaviors will be rewarded. Deficits in such learning could lead to increased frustration in irritable youth. We successfully piloted this task outside the scanner and are now acquiring imaging data. Children typically present with multiple symptoms and diagnoses. It is important to identify pathophysiological mechanisms specific to different symptoms, and how the neural mechanisms mediating different symptoms interact. Such research furthers efforts toward personalized medicine. We are interested in neural mechanisms that differentiate anxiety and irritability, given cross-sectional and longitudinal associations between these traits and possible shared circuitry mediating them. Both irritable and anxious youth have an attention bias toward threatening faces. Therefore, we scanned approximately 200 children with a range of common symptoms, including anxiety, irritability, and ADHD, while they completed an attention bias task. In a manuscript published this year, we used cutting-edge statistical techniques to dissociate neural activation associated with parent- or child-reported irritability or anxiety. We found that irritability is associated with increased amygdala, striatal, and prefrontal engagement when subjects were asked to attend away from threat. In contrast, anxiety was associated with normal neural activity but abnormal connectivity between prefrontal and limbic regions. A major focus of our work is treatment. We completed our double-blind trial designed to ascertain whether citalopram (a serotonergic reuptake inhibitor (SRI) antidepressant that is effective in the treatment of pediatric anxiety) plus stimulant is more effective than placebo plus stimulant in the treatment of severe irritability. In a sample of 49 youth, we demonstrated that indeed, stimulant plus citalopram is more effective in reducing irritability than is stimulant plus placebo. These data are currently being prepared for publication. Stimulant and SRI treatment tend to have fewer side-effects than atypical antipsychotic treatment, which is used frequently in youth with severe irritability. Therefore, this work has considerable public health importance. We recently developed and are testing two novel psychosocial approaches to treating irritability. The first involves computer-based training designed to shift a subject's perception of ambiguous faces from angry toward happy. In work published last year, we found that youth with DMDD tend to judge ambiguous faces as angry rather than happy; that we can change this judgment with computer-based training; and that this is associated with decreased irritability. This was an open trial, without a control condition. In 2016, we began a controlled trial of this new Interpretation Bias Treatment, along with brain imaging before and after training. Our target enrollment is 40 youth, and 35 have completed the protocol. Therefore, we anticipate finishing this trial soon and preparing the results for publication. This year, we completed and published an open pilot study of a manual-based cognitive behavioral treatment targeting irritability. This treatment has a strong emphasis on behavioral techniques i.e., exposing the child to tolerable, but frustrating situations, to improve the child's ability to tolerate such situations. The treatment is modelled on exposure therapy for anxiety disorders, and draws heavily on our neuroscience-based model of irritability, including irritable children's hypothesized deficits in the process and content of instrumental learning, and their aberrant responses to frustration. We designed and are now enrolling subjects for a multiple baseline design study of this new treatment. Such a design allows for more structured piloting whose results can be subjected to statistical analysis. Imaging on the first frustration paradigm and the threat paradigms, described above, will be conducted pre and post treatment. Outcome variables will include not only clinical ratings, but also real-time parent and child reports of temper outbursts using ecological momentary assessment.