(1) Identification of neurocognitive and neuroimaging biomarkers of affective disorder​s

Emotional bias in monozygotic and dizygotic twins at heritable risk of depression (UBFAL)

Negative cognitive bias and aberrant neural processing of emotional faces are trait-marks of depression. Yet it is unclear whether these changes constitute an endophenotype for depression and is also present in healthy individuals with hereditary risk for depression. In the first study of monozygotic twins, we investigated never-depressed twins at high risk of depression (indexed by depression in their co-twins) and healthy twins at low risk of depression (no personal or co-twin history of psychiatric disorder) with neurocognitive tests and fMRI. The results revealed different neural response and functional connectivity within fronto-limbic and occipito-parietal regions during emotional face processing and enhanced fear vigilance in healthy high-risk twins. This adds to the growing evidence for abnormalities in the processing of emotional faces as a key endophenotype for depression.

In a second study of dizygotic high risk or low risk twins from the same cohort, we observed an unexpected reduced fear vigilance and lowered recognition of fear and happiness in high risk twins. During face processing in the scanner, high risk dizygotic twins also displayed distinct negative fronto-limbic connectivity and reduced fronto-occipital response to all emotional faces. The findings point to putative neurocognitive and neuronal compensatory mechanisms in dizygotic twins who remain healthy despite their familial risk.


Neuromapping of Endophenotypes for Affective disorders: A twin study of neurocognitive, neuroimaging, cellular and epigenetic markers (NEAD)

The NEAD study aims to identify better diagnostic tools that can support earlier correct diagnoses. Specifically, the study investigates neurocognitive, neuroimaging and biochemical biomarkers (endophenotypes) for affective disorders in three groups of monozygotic twins recruited through the Danish registers: 1) twins who both have an affective disorder, 2) twins who are at genetic risk indexed by co-twin history of affective disorder and 3) twins who are both healthy and with no psychiatric history.

During face processing in the scanner, high-risk twins displayed greater medial and superior prefrontal response than affected twins, which correlated with stronger attentional avoidance of emotional faces in high-risk twins. Greater recruitment of medial and superior prefrontal cortex during implicit emotion processing may thus represent a compensatory or resilience mechanism in high-risk twins. High-risk twins also showed less prefrontal and anterior cingulate activity during reappraisal of unpleasant pictures than low-risk participants, possibly indicative of deficient prefrontal top-down regulation of emotional reactivity. During reward processing, affected twins showed reduced frontostriatal response to expected value and reward prediction error, while high risk co-twins showed a trend reduction in reward-related signals relative the he low risk group. At a structural level, high risk twins showed greater bilateral hippocampal volumes compared to affected twins, while affected twins had smaller dorsomedial prefrontal cortex volume compared with low risk twins. Whether the identified functional and structural changes in high risk and affected twins reflect compensatory mechanisms or vulnerability markers will be investigated in our planned follow-up of this twin cohort. This will help advance our understanding of key pathophysiological mechanisms in the etiology of affective disorders and, longer-term, aid earlier diagnosis and treatment.


The Bipolar Illness Onset study – work package 3 (BIO-3)

The BIO-3 study aims to identify a brain-based biomarker by comparing neurocognitive and neuroimaging measures of brain function in individuals with newly diagnosed/first-episode bipolar disorder, their unaffected first-degree relatives and healthy controls. The study is part of the largest longitudinal bipolar disorder cohort study to date (Bipolar Illness Onset; BIO). We found in the baseline part of the BIO-3 study that newly diagnosed patients with bipolar disorder show broad difficulties within non-emotional cognition compared to their unaffected first-degree relatives and healthy controls, suggesting that the impairments reflect illness-related processes that manifest around illness onset. Using a data-driven approach, we found evidence for three neurocognitive subgroups of patients with either global cognitive impairments (23%), selective impairments (31%) or no impairments (46%). Both cognitively impaired patients and their first-degree relatives also displayed aberrant emotional cognition, including difficulties with facial expression recognition and with down-regulating negative emotions.

At a neural level, patients showed hypo-activity in dorsolateral prefrontal cortex during attempts to dampen reactions to aversive images. In contrast, their relatives displayed enhanced negative functional connectivity between the amygdala and dorsomedial prefrontal cortex, suggestive of greater compensatory prefrontal top-down control of limbic reactivity. Finally, patients and their unaffected relatives both showed abnormal prefrontal cortex processing of reward prediction errors. This may indicate deficient engagement of prefrontal regions involved in inhibitory control during error prediction.


Prediction of risk and resilience in bipolar disorder: What can neuroimaging and neurocognitive biomarkers tell us? (PREDICT)

There is currently limited insight into biomarkers that predict the course of bipolar illness. It is also unclear whether the disorder is marked by neuroprogression or neurocognitive impairments are stable due to the paucity of longitudinal studies. In our ongoing PREDICT study, we will address these challenges through longitudinal investigations of newly diagnosed patients, their first-degree relatives and healthy controls using neuroimaging, neurocognitive and clinical psychiatric methods. The study builds on baseline assessments of 250 newly diagnosed patients, 95 first-degree relatives and 190 healthy controls from our Bipolar Illness Onset study, the largest longitudinal bipolar illness cohort study to date. PREDICT will investigate (A) neurocognitive and neuroimaging predictors of new mood episodes and resilience and (B) whether new mood episodes are accompanied by a decline in neurocognitive functions and brain changes. The results will provide a platform for development of targeted prophylactic interventions and will elucidate the trajectory of neurocognitive functions in bipolar disorder.

(2) Development of neurocognitive screening tools and methods to reduce the risk of affective disorder

Cognitive processing of infant stimuli in pregnant women with and without affective disorders and the association to postpartum depression (HEAPAD)

Pregnancy and childbirth are among the strongest risk factors for depression, affecting 10-15% of women. Notwithstanding the importance of mothers' mental health for the early mother-child interactions, there are no available methods to accurately predict who will develop postpartum depression. In a recently completed study, we investigated interpretation of infants' signals of emotion in pregnant women with and without affective disorders in remission.

Healthy pregnant women rated infant cries less negatively than non-pregnant women, which may indicate an increased ability to tolerate infant distress perhaps as a preparation for motherhood. In contrast, pregnant women with unipolar disorder exhibited a negative bias in ratings of infant cries, which may be a marker of more sensitivity to infant signals of distress. Pregnant women with bipolar disorder showed a positive bias in ratings of infant faces and greater recognition of positive than negative facial expressions. Across all pregnant women, bias in the ratings of infant cry was associated with more risk of postpartum depression. The findings point to putative neurocognitive mechanisms involved in risk of postpartum depression.


Mother-infant Interaction, Psychophysiological and neurocognitive responses to infant emotion in mothers with Affective Disorders (MIPAD)

For parents with bipolar disorder, there is a 50-60 % risk that the child will develop a psychiatric disorder. Genetic factors play a key role in the intergenerational transmission of psychiatric disorder, but it is unclear how early environmental risk factors interact with genetic vulnerability in this process. Using a multidisciplinary approach, we demonstrated that symptomatically stable mothers with affective disorders display aberrant psychophysiological and neurocognitive responses to infants' emotional signals. Healthy mothers showed increased vigilance toward infant stimuli compared with non-mothers, but this 'maternal vigilance' was attenuated in mothers with affective disorders. Mothers with unipolar disorder expressed more negative facial emotion when listening to infant cry and their attention was more directed towards sad and away from happy infant faces. In contrast, mothers with bipolar disorder expressed less attuned facial expressions while watching infant distress signals and – during fMRI - displayed reduced dorsal prefrontal response while watching own infants' faces. Importantly, the decreased maternal vigilance and emotional biases were associated with some difficulties in mother-infant interaction and delays in infant development. The findings have implications for understanding neurocognitive changes in mothers that may confer early subtle environmental risk for the infant and, ultimately, contribute to the intergenerational transmission of risk.


Prediction of depressive symptoms after birth: An online pregnancy risk profile screening tool (PreScreen)

Finally, in an ongoing pilot study we are testing the feasibility of a new online neurocognitive screening tool and its ability to predict postpartum depression. Pregnant women in third trimester of pregnancy are recruited via the national Danish registers. Participants receive an invitation to participate in the study via an online digital email box with a link to the online screening tool. The use of the Danish registers for recruitment allows us to invite a large sample representative of pregnant women and conduct large-scale simultaneous data collection. We aim to get full datasets for approximately 400 women.


(3) Development of ​​novel biological and psychological treatments targeting cognitive dysfunction

Effects of erythropoietin on depressive symptoms and neurocognitive deficits in depression and bipolar disorder (EPO trials)

There is a need for more effective treatments that aid cognitive functions in depression and bipolar disorder. Erythropoietin (EPO) is involved in neuroplasticity and is a candidate for treatment of cognitive dysfunction and mood symptoms affective disorders. We therefore conducted two parallel randomized controlled clinical phase II trials in treatment-resistant (unipolar) depression and bipolar disorder in remission to examine the effects of 8 weeks of weekly EPO vs. saline infusions on mood and cognition in these groups. Beneficial (mood-independent) effects of EPO were found on cognitive function across both patient groups. In contrast, there was no clear beneficial effect of EPO on the primary measure of mood in treatment-resistant depression (although EPO significantly improved self-rated depression and quality of life; tertiary outcome measures in the trial). The findings highlight EPO as a promising candidate treatment to target deficits in neuroplasticity and cognition in affective disorders.


Prefrontal Target Engagement as a biomarker model for Cognitive improvement - Erythropoietin (PRETEC-EPO)

The lack of treatments with solid and enduring pro-cognitive efficacy is related to major methodological challenges in the field, including the absence of a brain-based biomarker model to select among candidate treatments. In our PRETEC-EPO trial, we therefore investigate whether weekly EPO vs. placebo (saline) treatment for 12 weeks can improve cognitive functions in cognitively impaired patients with affective disorders as well as the early neuronal changes associated with such effects. The study is ongoing with expected completion in the autum 2021.


Effects of Short-Term Cognitive Remediation on Cognitive Dysfunction in Partially or Fully Remitted Individuals with Bipolar Disorder (REMEDI)

Cognitive remediation (CR) is a psychological intervention that aims to improve cognitive function, compensational skills, coping skills and, consequently, psychosocial function. Based on substantial evidence for beneficial effects of CR in patients with schizophrenia, we fonducted the first RCT of CR in bipolar disorder. The trial examined whether 12 weeks group-based CR improves cognition in remitted patients with bipolar disorder who experienced moderate to severe cognitive difficulties. No significant beneficial effects were observed on objectively measured cognitive function, suggesting that more intensive, individualised interventions may be necessary.


Effect of Action-Based Cognitive Remediation on cognitive impairment in patients with remitted bipolar disorder (PRETEC-ABC)

In our most recent RCT, we investigated the effects of a new intensive 10-week cognitive remediation intervention, Action-Based Cognitive Remediation (ABCR), in bipolar disorder. The intervention involved two weekly group-based training sessions and daily computer training between sessions. The parallel control group involved weekly unstructured group sessions supervised by a therapist. ABCR produced no significant improvement in a global cognition (primary outcome) but a strong improvement in executive functions (secondary outcome) and subjective cognitive functions (tertiary outcome). Importantly, ABCR treatment benefits on executive function was predicted by greater baseline executive dysfunction. 

(4) Exploring neurocognitive and neuroimaging mechanisms of established and novel trea​​tments

The neural correlates of EPO-associated improvement of cognition

As part of the EPO studies, we conducted prospective neuroimaging assessments at baseline and after treatment completion to explore the neuronal correlates of any beneficial effects of EPO on cognition and mood. This revealed that the memory improvement in EPO-treated patients was accompanied by structural increase in the left hippocampus and functional changes in neuronal activity in the dorsal prefrontal cortex and functionally related regions. The EPO-associated improvement in executive func tion was also accompanied by increase in task-relevant activity in the dorsal prefrontal cortex. Together, these findings point to modulation of task-related prefrontal cortex activity as a key neurobiological target for pro-cognitive treatments. Based on these findings, we are in the process of investigating whether modulation of prefrontal cortex activity provides a human-based biomarker model for pro-cognitive effects in additional studies using other treatments to target cognition, including ABCR.


The neural correlates of improved executive function after action-based cognitive remediation

In the ABCR study, participants underwent fMRI at baseline and after two weeks of ABCR/control treatment. This enabled investigation of the early change in the dorsal prefrontal cortex response in remitted patients with bipolar disorder, and whether the observed neural change predicted improved executive function following 10 weeks of treatment. Importantly, we observed a working memory-related activity increase in dorsolateral prefrontal cortex similar to the activity increase observed after EPO treatment. Further, this neural activity increase correlated with and predicted the observed improvement in executive function after treatment completion. Together, the fMRI findings from the EPO and ABCR studies provide key proof of principle for the use of neuroimaging biomarker models to detect pro-cognitive effects of novel pro-cognitive interventions. Future cognition trials should include fMRI to confirm the validity of this putative biomarker model across neuropsychiatric disorders with cognitive impairments.


Effects of electroconvulsive therapy on neurocognitive and neuronal measures of brain function

Heightened neuronal activity in the amygdala and ventromedial prefrontal cortex to negative information is a consistent observation in depression and seems to resolve with antidepressant drug treatment. Using fMRI, we investigated whether another evidence-based treatment for depression, electroconvulsive therapy (ECT) has similar effects on neurocircuitry response to emotional information in patients with depression. The study showed that a single active vs. sham ECT modulated the neuronal response in areas of the prefrontal cortex to emotional images, faces and self-referent words. The neurocircuitry activity changes were suggestive of less neurocognitive reactivity to negative material and facilitated recollection of positive self-referent information. The observed effects are in line with the putative neurocognitive mechanisms of action of antidepressant drugs, suggesting that modulation of neurocognitive response to emotional information may be a common mechanism of different treatments for depression.