The present trial consists of 2 sub-studies that investigate important novel aspects of treatment with erythropoietin (EPO) on cognitive dysfunction in bipolar disorder (BD) and recurrent unipolar depressive disorder (UD) (defined as minimum 2 treatment-requiring depressive episodes). The aims of the trial are three-fold. We aim to investigate the effects of 12 weekly recombinant human EPO infusions on cognition in (i) healthy people with cognitive impairment (substudy 1) and (ii) patients with remitted BD or recurrent UD (substudy 2), and (iii) explore early treatment-associated neural activity changes that may predict subsequent cognitive improvement.

It is hypothesized that:

i. 12 weekly EPO infusions improve cognition in healthy first-degree relatives and remitted BD patients in comparison with saline.

ii. EPO vs. saline-treated participants will display early cognition-related neural activity in the frontal lobes, which will correlate with cognitive improvement.

This trial will include healthy people (sub-study 1) and patients with bipolar disorder or recurrent unipolar depressive disorder in partial or full remission (defined as a score of ≤14 on the Hamilton Depression Rating Scale 17-items and the Young Mania Rating Scale, respectively (sub-study 2) with objectively-verified cognitive dysfunction. Participants will be recruited from Psychiatric Centres in The Mental Health Services in the Capital Region of Denmark, consultant psychiatrists in the Capital Region of Denmark, as well as through advertisements on relevant websites.

The study design comprises 4 major assessments (baseline, week 3, week 13, and a 6 month follow-up after treatment completion) and weekly safety monitoring and study medication infusions during a 12 week treatment period. The baseline assessment is divided into 2 days, 1-3 days apart for practical reasons and to avoid attrition. On the first day of the baseline assessment, participants will perform an fMRI scan. On the second baseline day, participants complete an assessment comprising neuropsychological testing, verbal IQ assessment, and filling in questionnaires concerning subjective cognitive complaints, quality of life, level of functioning, and functional capacity, as well as mood symptom severity ratings. Functional capacity is assessed using a clinician-rated interview and a performance-based task. After 2 weeks of treatment (i.e., 2 doses of EPO or saline) an fMRI scan, neuropsychological testing, mood ratings, and questionnaires on subjective cognitive difficulties are repeated. After treatment completion (week 13) and at the 6 month follow-up, the neuropsychological tests, questionnaires concerning subjective cognitive complaints, quality of life, and functional capacity (self-reported and performance-based) are repeated. Sleep quantity and quality in the past three days is assessed before each of the 4 major assessment time point. Blood samples are collected at baseline, week 3 and 13 for assessment of potential blood-based biomarkers of pro-cognitive effects.

Pharma Consulting Group AB (www.pharmaconsultinggroup.com) has conducted block randomization for each sub-study group, stratified for gender and age (sub-study 1: < or >=30 years; sub-study 2: < or >=35 years). Power calculation was also carried out by Pharma Consulting Group based on findings from a previous RCT in our group assessing the effect of 8 weeks of EPO treatment on the same cognitive composite score. In this trial, the clinically relevant differential change between EPO and saline groups following 12 weeks of treatment is assumed to be at least 0.4 SD (corresponding to a moderate effect size) on the primary outcome with SD of the change of 0.5. Assuming a 10% drop-out rate, we plan to recruit up to n=58 for each sub-study to achieve complete data sets for n=52 participants per sub-study.

Data from the primary, secondary, and tertiary outcomes will be analyzed using Mixed Models Design and Intention to Treat (ITT) analyses. Functional MRI data are pre-processed and analyzed with FMRIB Expert Analysis Tool (FEAT) and the 'randomize' algorithm implemented in FSL (FMRIB Software Library; www.fmrib.ox.ac.uk/fsl). Functional MRI data is analysed using region of interest (ROI) analyses to assess potential differences in neural activity within the dorsal prefrontal cortex and the hippocampi between EPO and placebo groups after 2 weeks of treatment. Exploratory whole-brain analyses are conducted to assess treatment-related activity change in other brain regions. Any differences in neural activity between treatment groups are correlated with potential changes in the primary cognitive composite measure after 2 weeks of treatment (week 3) and after treatment completion (week 13). If this correlation is significant, multiple regression analyses will be performed with adjustment for mood symptoms, age, and gender to assess the potential predictive value of early neural activity change for potential pro-cognitive efficacy after 12 weeks of EPO treatment.

Inclusion Criteria:

• Fluent Danish skills and objective cognitive impairment (a total score below cutoff, or scores below cutoff on a minimum of two out of the five subtests (Verbal Learning Test - Immediate, Working Memory Test, Verbal Fluency Test, Verbal Learning Test - Delayed and Processing Speed Test) on the Screen for Cognitive Impairment in Psychiatry - Danish version (SCIP-D).
• Patients must meet the ICD-10 diagnosis of BD (type I and II) or recurrent depressive disorder confirmed using the Schedules for Clinical Assessment in Neuropsychiatry (SCAN) interview.
• Healthy people are eligible even when diagnosed with a less severe mental disorder defined as ICD-10 codes ≥F40.

Exclusion Criteria:

• Schizophrenia or schizoaffective disorder
• Current alcohol or substance misuse disorder (3 months prior to inclusion)
• Daily use of benzodiazepines > 22.5 mg oxazepam
• Diabetes
• Kidney disease
• Renal failure
• Untreated/insufficiently treated arterial hypertension
• Heart diseases (previously diagnosed or abnormal ECG findings during screening)
• Previous serious head trauma
• Neurological illness (including dementia)
• Previous or current epilepsy in patient or first degree family
• Malignancies or thromboses
• Known allergy or antibodies against erythropoietin
• Initial hematocrit > 50% (males) or > 48% (females)
• Initial thrombocyte numbers over normal (>400 billions/L)
• Initial reticulocyte numbers <1‰
• Contraindications against prophylactic thrombosis treatment
• Myeloproliferative disorder, polycythemia
• Pregnancy or breast feeding
• Use of contraceptive medication or other hormonal contraceptives
• Sexually active women in the fertile age, who do not or do not want to use double barrier anticontraceptive methods
• Previous or current history of thromboembolic events or thromboses in patient or first degree family (increased risk of thromboembolic events)
• Overweight (BMI>30) or body weight <45 or >95 kg.
• Previous electroconvulsive therapy (ECT) treatment within last 3 months
• Dyslexia
• Claustrophobia
• Having a pacemaker or other metal implants inside the body
• Reluctance or inability to comply with the protocol requirements