Recombinant human erythropoietin for treating treatment-resistant depression: a double-blind, randomized, placebo-controlled phase 2 trial

Miskowiak KW, Vinberg M, Christensen EM, Bukh JD, Harmer CJ, Ehrenreich H, Kessing LV

Neuropsychopharmacology. 2014 May;39(6):1399-408.

Abstract

Pharmacological treatments for depression have insufficient efficacy in 30-40% of patients and fail to reverse cognitive deficits. Erythropoietin (EPO) has neurotrophic actions and aids neurocognitive function. The aim of this exploratory study was to determine whether recombinant human EPO improves mood and memory in treatment-resistant depression. Forty treatment-resistant depressed unipolar patients with Hamilton Depression Rating Scale-17 (HDRS-17) score ≥ 17 were randomized to eight weekly EPO (Eprex; 40,000 IU) or saline infusions in a double-blind, placebo-controlled, parallel-group design. Patients were assessed at baseline and at weeks 5, 9, and 14. Primary outcome was reduction in HDRS-17 score. Global assessment of function (GAF) was reported in addition. Secondary outcome was remission rate, and tertiary outcomes were changes in Rey Auditory Verbal Learning Test (RAVLT), Beck Depression Inventory-21 (BDI-21), and World Health Organization Quality of life-BREF (WHOQOL-BREF). Exploratory outcomes were depression and cognition composite scores. HDRS-17, GAF, and remission rates showed no effects of EPO over saline at week 9 (P-value ≥ 0.09). However, EPO improved BDI (P=0.02) and WHOQOL-BREF (P=0.01), and this was maintained at follow-up week 14 (P-values ≤ 0.04). EPO enhanced verbal recall (P=0.02) and recognition (P=0.03), which was sustained at follow-up (P-values ≤ 0.04). Exploratory analysis in patients fulfilling depression severity criteria at trial start revealed ameliorated HDRS-17 in EPO (N=14) vs saline groups (N=17), which was sustained at week 14 (P-values ≤ 0.05). Exploratory analysis in the complete cohort showed that EPO reduced depression composite at weeks 9 and 14 (P-values=0.02). The findings of this exploratory study highlight EPO as an interesting compound for treatment-resistant depression, which deserves further investigation.

Different neural and cognitive response to emotional faces in healthy monozygotic twins at risk of depression

Miskowiak KW, Glerup L, Vestbo C, Harmer CJ, Reinecke A, Macoveanu J, Siebner HR, Kessing LV, Vinberg M

Psychol Med. 2015 May;45(7):1447-58.

Abstract

BACKGROUND: 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 are also present in healthy individuals with hereditary risk for depression.

METHOD: Thirty healthy, never-depressed monozygotic (MZ) twins with a co-twin history of depression (high risk group: n = 13) or without co-twin history of depression (low-risk group: n = 17) were enrolled in a functional magnetic resonance imaging (fMRI) study. During fMRI, participants viewed fearful and happy faces while performing a gender discrimination task. After the scan, they were given a faces dot-probe task, a facial expression recognition task and questionnaires assessing mood, personality traits and coping strategies.

RESULTS: High-risk twins showed increased neural response to happy and fearful faces in dorsal anterior cingulate cortex (ACC), dorsomedial prefrontal cortex (dmPFC), pre-supplementary motor area and occipito-parietal regions compared to low-risk twins. They also displayed stronger negative coupling between amygdala and pregenual ACC, dmPFC and temporo-parietal regions during emotional face processing. These task-related changes in neural responses in high-risk twins were accompanied by impaired gender discrimination performance during face processing. They also displayed increased attention vigilance for fearful faces and were slower at recognizing facial expressions relative to low-risk controls. These effects occurred in the absence of differences between groups in mood, subjective state or coping.

CONCLUSIONS: Different neural response and functional connectivity within fronto-limbic and occipito-parietal regions during emotional face processing and enhanced fear vigilance may be key endophenotypes for depression

Risk markers for affective disorder, a seven-years follow up study of a twin cohort at low and high risk for affective disorder

Vinberg M, Miskowiak K, Kessing LV

J Psychiatr Res. 2013 May;47(5):565-71

Abstract

This study aims to investigate whether: familial history of affective disorder, subclinical depressive symptoms and life events (LEs) are predictive of a later development of mood disorder (onset). In a high-risk study, 234 healthy monozygotic and dizygotic twins with and without a co-twin history of affective disorder (high and low risk twins, respectively) were identified through nationwide registers and assessed from 2002 to 2005. Participants were followed longitudinally at 6-months intervals for up to nine years and finally reassessed with a personal interview to obtain information on whether they had an onset. During the follow-up period (mean time 7.0 years), 36 participants (15.4%) developed onset. Onset was significantly associated with risk status (Hazard ratio (HR) = 1.38, 95% CI 1.08-1.76), female sex, HR = 2.70, 95% CI 1.19-6.97, age HR = 0.97, 95% CI 0.93-0.99), and also with baseline Hamilton 17 score (HR = 1.30, 95% CI 1.13-1.48), Becks Depression Inventory 21 (HR = 1.14, 95% CI, 1.05-1.24) and neuroticism (HR = 1.08, 95% 1.02-1.12). Finally, the experience of LEs lifetime before baseline predicted onset (HR = 1.20, 95% CI 1.01-1.46) and the experience of LEs during follow-up also predicted onset (HR = 1.06, 95% CI 1.01-1.11). These findings suggest that young individuals at familial risk of affective disorders are at enhanced risk of onset and at further risk when having female sex and more subclinical depressive symptoms at baseline. Further, they seem to experience more LEs and to be more vulnerable to these.

Brain Derived Neurotrophic Factor (BDNF) levels as a possible predictor of psychopathology in healthy twins at high and low risk for affective disorder

Vinberg M, Miskowiak K, Kessing LV

Psychoneuroendocrinology. 2014 Jan;39:179-83

Abstract

Brain Derived Neurotrophic Factor (BDNF) is a potential biomarker of affective disorder. However, longitudinal studies evaluating a potential predictive role of BDNF on subsequent psychopathology are lacking. The aim of this study was to investigate whether BDNF alone or in interaction with the BDNF Val66Met polymorphism predict onset of affective disorder in healthy individuals at heritable risk for affective disorder. In a high-risk study, we assessed whole blood levels of BDNF in 234 healthy monozygotic and dizygotic twins with or without a co-twin history of affective disorder (high and low risk twins, respectively). Participants were followed up longitudinally with questionnaires at 6-month intervals for mean seven years and then reassessed with a personal interview to obtain information about whether they had developed psychiatric illness. At follow-up 36 participants (15.4%) had developed psychiatric disorder. Cox regression analysis revealed that BDNF levels at baseline were not associated with onset of illness in this explorative study. Further, two-way interactions between BDNF levels and the Val66Met polymorphism or between familial risk and the Val66Met polymorphism did not predict illness onset.

Serotonin transporter genotype, salivary cortisol, neuroticism and life events: impact on subsequent psychopathology in healthy twins at high and low risk for affective disorder

Vinberg M, Miskowiak K, Kessing LV

Prog Neuropsychopharmacol Biol Psychiatry. 2014 Jan 3;48:193-8.

Abstract

OBJECTIVE: To investigate if cortisol alone or in interaction with other risk factors (familial risk, the serotonin transporter genotype, neuroticism and life events (LEs)) predicts onset of psychiatric disorder in healthy individuals at heritable risk.

MATRIAL AND METHODS: In a high-risk study, 234 healthy monozygotic and dizygotic twins with or without a co-twin history of affective disorder (high and low risk twins) were baseline assessed. Participants were followed up for seven years and then reassessed with a personal interview revealing whether they had developed psychiatric illness.

RESULTS: 36 participants (15.4%) developed psychiatric disorder. Using Cox proportional hazards ratio (HR) estimates neither morning nor evening salivary cortisol at baseline did predict illness onset. In multivariate Cox models, the two-way interaction between morning cortisol and LEs lifetime before baseline was significantly associated with onset. Further, the HR of onset was higher concerning individuals carrying the short allele of the 5-HTTPLR and having experienced more LEs lifetime. Familial risk for affective disorder predicted illness and the risk of onset was further increased in individuals at familial risk carrying the short allele of the 5-HTTPLR.

CONCLUSIONS: Cortisol levels alone do not increase the risk of onset of psychiatric illness but the interaction of a lower cortisol level and the experience of more LEs do. The 5-HTTLPR genotype seems to interact and contribute to increased stress vulnerability in combination with other stress indicators of illness thereby adding to the risk of subsequent psychopathology.

Impairment of executive function and attention predicts onset of affective disorder in healthy high-risk twins

Vinberg M, Miskowiak KW, Kessing LV

J Clin Psychiatry. 2013 Aug;74(8):e747-53.

Abstract

OBJECTIVE: To investigate whether measures of cognitive function can predict onset of affective disorder in individuals at heritable risk.

METHOD: In a high-risk study, 234 healthy monozygotic and dizygotic twins with and without a co-twin history of affective disorder (high- and low-risk twins, respectively) were identified through nationwide registers and assessed at baseline using the Schedules for Clinical Assessment in Neuropsychiatry, the 17-item Hamilton Depression Rating Scale (HDRS), and the cognitive tests Trail Making Test Parts A and B, the Stroop test, and the Cambridge Cognitive Examination-Revised (CAMCOR). Participants were followed longitudinally at 6-month intervals for up to 9 years and finally reassessed with a personal interview to obtain information on whether they had developed psychiatric illness. The study was conducted between 2003 and 2012.

RESULTS: 36 participants (15.4%) developed psychiatric disorder, mainly affective and anxiety disorders (31 diagnoses) (ICD-10). Onset was predicted by decreased executive function as reflected by performance on the Trail Making Test A – B (hazard ratio [HR] = 1.02; 95% CI, 1.00-1.03) when adjusted for sex, age, years of education and HDRS score at baseline. Reduced global cognitive function as indicated by a lower CAMCOR score at baseline showed a trend toward an association with subsequent illness onset (P = .08). With regard to the 5 CAMCOR subscales, lower scores on attention (HR = 0.71; 95%, CI, 0.54-0.94) and language (HR = 0.76; 95% CI, 0.58-0.99) were significantly associated with subsequent illness onset.

CONCLUSIONS: Among healthy individuals at heritable risk for affective disorder, discrete cognitive deficits, especially within executive function and attention, seem to predict subsequent onset of affective illness.

Optimising screening for cognitive dysfunction in bipolar disorder: validation and evaluation of objective and subjective tools

Jensen JH, Støttrup MM, Nayberg E, Knorr U, Ullum U, Purdon SE, Kessing LV, Miskowiak LV    

Journal of Affective Disorders (in press)

Abstract

Introduction: Cognitive impairment is common in bipolar disorder and contributes to socio-occupational difficulties. The objective was to validate and evaluate instruments to screen for and monitor cognitive impairments, and improve the understanding of the association between cognitive measures and socio-occupational capacity.

Methods: Patients with bipolar disorder in partial or full remission (n=84) and healthy controls (n=68) were assessed with the Screen for Cognitive Impairment in Psychiatry (SCIP), Cognitive Complaints in Bipolar Disorder Rating Scale (COBRA), and established neuropsychological tests and subjective rating scales. Socio-occupational function and affective symptoms were evaluated with the Functional Assessment Short Test, and the Hamilton Depression Rating Scale 17-items and Young Mania Rating Scale, respectively. Concurrent validity of the SCIP and COBRA were assessed by correlation with established objective and subjective cognitive measures, and decision validity was determined with Receiver-Operating-Characteristic analyses. Correlations and linear regression analyses were conducted to determine the associations between objective and subjective cognitive impairment, and socio-occupational difficulties.

Results: The SCIP and COBRA correlated strongly with established objective and subjective cognitive measures, respectively. The SCIP yielded higher sensitivity and specificity for detection of cognitive dysfunction than the COBRA or a combined SCIP-COBRA measure. Correlations between objective and subjective cognitive impairment were weak but both were associated with socio-occupational difficulties.

Limitations: Influence of ageing was not investigated.

Conclusions: The SCIP and COBRA are valid for detection of objective and subjective cognitive impairment in bipolar disorder. Screening for cognitive dysfunction should be conducted with an objective measure like the SCIP.

Assessment of subjective and objective cognitive function in bipolar disorder: Correlations, predictors and the relation to psychosocial function

Demant KM, Vinberg M, Kessing LV, Miskowiak KW

Psychiatry Res. 2015 Jun 3. pii: S0165-1781(15)00291-7.

Abstract

Cognitive dysfunction is prevalent in bipolar disorder (BD). However, the evidence regarding the association between subjective cognitive complaints, objective cognitive performance and psychosocial function is sparse and inconsistent. Seventy seven patients with bipolar disorder who presented cognitive complaints underwent assessment of objective and subjective cognitive function and psychosocial functioning as part of their participation in two clinical trials. We investigated the association between global and domain-specific objective and subjective cognitive function and between global cognitive function and psychosocial function. We also identified clinical variables that predicted objective and subjective cognitive function and psychosocial functioning. There was a correlation between global subjective and objective measures of cognitive dysfunction but not within the individual cognitive domains. However, the correlation was weak, suggesting that cognitive complaints are not an assay of cognition per se. Self-rated psychosocial difficulties were associated with subjective (but not objective) cognitive impairment and both subjective cognitive and psychosocial difficulties were predicted by depressive symptoms. Our findings indicate that adequate assessment of cognition in the clinical treatment of BD and in drug trials targeting cognition requires implementation of not only subjective measures but also of objective neuropsychological tests.

Is there an association between subjective and objective measures of cognitive function in patients with affective disorders?

Svendsen AM, Kessing LV, Munkholm K, Vinberg M, Miskowiak KW

Nord J Psychiatry. 2012 Sep;66(4):248-53.

Abstract

BACKGROUND: Patients with affective disorders experience cognitive dysfunction in addition to their affective symptoms. The relationship between subjectively experienced and objectively measured cognitive function is controversial with several studies reporting no correlation between subjective and objective deficits.

AIMS: To investigate whether there is a correlation between subjectively reported and objectively measured cognitive function in patients with affective disorders, and whether subjective complaints predict objectively measured dysfunction.

METHODS: The study included 45 participants; 15 with bipolar disorder (BD), 15 with unipolar disorder (UD) and 15 healthy individuals. Participants’ subjectively experienced cognitive function and objective cognitive function were assessed with the Massachusetts General Hospital Cognitive and Physical Functioning Questionnaire (CPFQ) and the Screen for Cognitive Impairment in Psychiatry (SCIP), respectively. Patients were rated for affective symptoms with Hamilton Depression Rating Scale (HDRS) and Young Mania Rating Scale (YMRS).

RESULTS: Patients demonstrated subjective and objective cognitive dysfunction relative to controls (P-values ≤ 0.01) but there were no differences between patient groups (P > 0.1). We found no correlation between subjectively experienced and objectively measured cognitive dysfunction in BD (P = 0.7), and a non-significant trend towards a correlation in UD (P = 0.06), which disappeared when controlling for gender (P = 0.1).

CONCLUSION: Our results suggest that it is not necessarily patients who have cognitive complaints that are most impaired. If confirmed in a larger sample, our findings suggest that neuropsychological assessment is warranted to elucidate the potential role of cognitive dysfunction in patients’ everyday lives and to inform treatment strategies targeting these difficulties.

Is there a difference in subjective experience of cognitive function in patients with unipolar disorder versus bipolar disorder?

Miskowiak K, Vinberg M, Christensen EM, Kessing LV

Nord J Psychiatry. 2012 Dec;66(6):389-95.

Abstract

BACKGROUND: Cognitive dysfunction in unipolar disorder (UD) and bipolar disorder (BD) may persist into remission and affect psychosocial function. Executive and memory deficits during remission may be more pronounced in BD than UD. However, patients’ subjective experience of cognitive difficulties is poorly understood, and it is unclear whether BD and UD patients experience different cognitive difficulties.

AIMS: To investigate whether there are differences in the quality and magnitude of subjective cognitive difficulties between UD and BD, and which factors influence the subjective cognitive difficulties in these patients.

METHODS: Patients with BD (n = 54) or UD (n = 45) were referred to the outpatient mood disorder clinic at Department of Psychiatry, Copenhagen University Hospital, following hospital discharge. Affective symptoms and patients’ experience of cognitive symptoms were assessed at their initial consultation at the clinic.

RESULTS: Patients experienced mild to moderate cognitive impairment despite being in partial or full remission, but there were no differences in subjective difficulties between BD and UD. Subjective cognitive dysfunction was predicted by depression severity, anxiety and mania symptoms rather than by diagnosis, age, gender or alcohol misuse.

CONCLUSION: The absence of difference in subjective cognitive difficulties between UD and BD contrasts with evidence of greater objective dysfunction in BD. This highlights a potential discord between subjective and objective measures of cognitive function. Subjective cognitive function was predicted by affective symptoms, perhaps suggesting that this reflects mood symptoms rather than objective deficits. This points to a clinical need for objective assessment of cognitive function in these patient groups.