In this study, the investigators will stratify depressed subjects a priori based on CRP levels to test the hypothesis that eicosapentaenoic (EPA) would be more efficacious to treat depression in subjects with high CRP levels compared to subjects with low CRP levels.

Depressed subjects, with ongoing stabilized antidepressive treatment who remain clinically depressed, will be enrolled in an "Inflammation group" or in a "Non-inflammation group" depending on baseline levels of CRP. Subjects in both groups will receive EPA enriched omega-3 fatty acids for 8 weeks, added to their pre-stabilized antidepressant medication.

Background

Increased mean levels of peripheral inflammatory markers have repeatedly been reported in individuals with major depression compared to controls, although there is considerable overlap between groups. As further evidence for a role of inflammation in major depression, 20-60% of patients receiving treatment for viral hepatitis and certain forms of cancer with the cytokine interferon-alpha will develop depressive symptoms. Interestingly, some authors have suggested that the association between inflammation and depression is symptom specific, i.e. there might be a subtype of "inflammation-related depression" with a specific phenotype. Although it still remains unclear if the immune abnormalities often seen in depressed subjects derive from changes in the periphery or in the central nervous system (or both), animal studies have shown that by counteracting the effects of pro-inflammatory cytokines in either the blood or the brain, depressive-like behavior in animals can be mitigated. These animal studies, along with data from several clinical studies pertaining to this, suggest that interventions primarily targeting peripheral inflammation may be useful in treating psychiatric symptoms. Although the exact mechanisms by which increased inflammation may give rise to depressive symptoms remain unclear, there are several potential downstream pathways that may be involved including accelerated cellular aging, mitochondrial dysfunction, and oxidative stress.

In search for a potential antidepressant that could be tested in clinical trials in which subjects are selected a priori based on inflammatory markers, the investigators turn to eicosapentaenoic (EPA), an omega-3 (n-3) fatty acid with anti-inflammatory properties. The ability of dietary n-3 fatty acids to mitigate the inflammatory response has been shown in human and animal studies. EPA and Docosahexaenoic acid (DHA) are the two major n-3 fatty acids found in oily fish and fish oil supplements, and both have shown anti-inflammatory properties. Supplementation of EPA and DHA in individuals with cardiovascular disease results in decreased plasma levels of CRP. N-3 fatty acids also have several other anti-inflammatory properties including i) Decreased chemotaxis of neutrophils and monocytes, ii) Decreased expression of adhesion molecules (e.g. ICAM & VCAM) on the surface of immune cells and in the circulation, iii) Decreased production of prostaglandins, iv) Increased synthesis of anti-inflammatory molecules such as resolvins and protectins, and v) Inhibition of T-cell proliferation. The mechanisms underlying these effects are not fully understood but likely involves n-3 fatty acids acting via cell surface and intracellular receptors controlling inflammatory cell signaling and gene expression patterns. In addition to the well-established anti-inflammatory effects of n-3 fatty acids, they may also have beneficial effects on oxidative stress and cell aging parameters such as leukocyte telomere length and telomerase activity. However, more research is needed in order to confirm these relationships, and therefore the investigators will, in addition to assessing inflammatory markers, also study the effects of n-3 fatty acids on markers of cellular aging and oxidative stress.

Some, but not all, previous studies have shown that EPA is superior to placebo in treating unipolar or bipolar depression. Several caveats have, however, been issued when interpreting the results from these studies, including small and perhaps clinically irrelevant effect sizes, as well as potential publication bias. Generally, n-3 preparations with high doses of EPA relative to DHA have been shown to be more efficacious in treating depression. Interestingly, a double blind placebo-controlled RCT showed that that EPA (but not DHA) was effective in preventing interferon-alpha induced depression in hepatitis C subjects, consistent with the notion that EPA may have antidepressant effect in "inflammatory depression". In line with this notion, a recent a proof-of-concept study showed that high inflammation at baseline was associated with a better antidepressant effect of EPA, but not DHA, enriched n-3 compared to placebo. That study stratified subjects post-hoc based on inflammatory markers, whereas our approach will be to select study subjects a priori based on validated cut-offs for CRP. This is the next step in developing a personalized medicine paradigm for depression.

The main aims of the study are to test if i) EPA enriched n-3 (added to stabilized ongoing treatment) is efficacious in treating depressed patients, but only in subjects with prospectively ascertained elevations in baseline CRP, ii) Changes in inflammatory markers over the course of treatment mediate this effect, and iii) Clinical trial designs utilizing prospectively-ascertained biomarkers to predict response are feasible and thereby pave the way for personalized medicine in psychiatry.

Specific objective 1: To determine whether the antidepressant effect of n-3 EPA is greater in the Inflammation group than in the Non-inflammation group, controlling for baseline depression rating.

Hypothesis (H) 1: The antidepressant effect of n-3 EPA is greater in the Inflammation group than in the Non-inflammation group.

Specific objective 2: To determine whether changes in inflammatory, oxidative stress and cell aging markers from baseline to end of treatment correlate with antidepressant effect.

H 2: Change in inflammatory and cell aging markers with n-3 EPA treatment will be directly correlated with changes in depression ratings.

Inclusion Criteria:

1. Male/female, aged 18-80.
2. Fulfilling the DSM criteria for a current depressive episode, unipolar (symptom duration> 4 weeks) as determined by the study physician
3. HAM-D-17 score ≥ 15 (Rapaport, Nierenberg et al. 2016)
4. A Clinical Global Impression Severity Score ≥ 3 (Rapaport, Nierenberg et al. 2016)
5. All subjects should be stable on antidepressants or mood stabilizers ≥6 weeks.
6. Willing to not significantly modify their diet from the time they sign consent through the end of study participation.

Exclusion Criteria:

1. Serious or unstable medical illness that in the investigator's opinion could compromise response to treatment or interpretation of study results. Examples: Malignancy not in remission for at least 1 year, Active autoimmune disorder or inflammatory bowel disease, Insulin-dependent diabetes mellitus.

2. Known or suspected allergy to the study compounds.

3. Ongoing infection.

4. Ongoing pregnancy or breast-feeding

5. A diagnosis of psychotic disorder, bipolar disorder, mental retardation dementia, or individual whom, due to other causes, lack the ability to make an informed decision.

6. Ongoing ECT.

7. Concomitant use of anticoagulants or known bleeding disorder.

8. Patients who, in the investigator's judgment, pose a current, serious suicidal or homicidal risk.

9. A diagnosis for any Substance Use Disorder (except nicotine or caffeine) in the 3 months prior to the screening visit.

10. Any medications (within 1 week of baseline or during the trial) that might confound the biomarker findings, including: Regular ingestion of NSAIDs or COX-2 inhibitors, or any use of oral steroids, immunosuppressants, interferon, chemotherapy (Patients will be instructed not to take an NSAID, COX-2 inhibitor or Aspirin in the 24 hours prior to a biomarker assessment visit).

11. Patients who have taken supplements with omega-3 fatty acids for more than three consecutive days in the preceding month.

12. Within 4 weeks of starting psychotherapy or planning to start psychotherapy during the study

13. Active participation in other clinical studies with ongoing study visits

    -

Raters are blind to "Inflammation group" or "Non-inflammation group" allocation