Using Inflammatory Biomarkers to Guide Depression Treatment Choices

(Part 4 of a 4-part series)

In this video, Psych Congress 2020 cochair Charles Raison, MD, discusses the role of inflammatory biomarkers in guiding depression treatment choices. Dr. Raison, a leading researcher in the field of immune-brain interactions, spoke at Psych Congress on the role of inflammation in mental health.

Dr. Raison is the Mary Sue and Mike Shannon Chair for Healthy Minds, Children & Families; Professor, Human Development and Family Studies, School of Human Ecology; and Professor, Department of Psychiatry, School of Medicine and Public Health, University of Wisconsin-Madison. He is also Director of Clinical and Translational Research for the Usona Institute, which conducts research on the therapeutic effects of psilocybin.

Part 1: Health Effects of Chronic Increased Inflammation

Part 2: The Link Between Inflammation and Response to Psychotherapy

Part 3: The Relationship Between Inflammation and Mental Illnesses

Read the transcript:

I think we are coming closer and closer to being able to recommend or suggest the use of inflammatory biomarkers to guide treatment choices in mood disorders in general, and perhaps, especially in major depressive disorder.

I say “closer to” because the studies we have to date are suggestive, but not conclusive, because they weren't designed to directly test the role of inflammatory biomarkers as predictive variables and because their sample sizes generally were small, not always.

The studies we have to date paint a pretty consistent pattern, and it's this: anti-inflammatory agents do not work, in general, in people with depression despite a couple of meta-analyses of non-steroidal anti-inflammatory agents, which I think are flawed.

The best data really does not support the idea that if you just block inflammation in a group of random depressed people that you're going to see anything better than a placebo response. What you see though is that as inflammation increases, people are more likely to get an antidepressant response from an anti-inflammatory agent.

This makes a lot of sense because if inflammation is causing your depression, or contributing to your depression, and you block it, it makes sense that that might make people feel better.

That's one way in which inflammatory biomarkers might be useful down the road as predictors of what to do in folks that are depressed. Certainly, I think the data suggests that you do not want to use anti-inflammatory agents in depressed people who do not have elevated levels of inflammation.

Taken as a whole, the studies we have suggest, in fact, that in those people, blocking inflammation may actually have far less benefit than just giving them a placebo. It really seems to be counter-productive. This has been shown with cytokine antagonists. It's been shown with omega-3 fatty acids.

Other than telling people to take omega-3 fatty acids, most of us who are psychopharmacologists are not routinely prescribing anti-inflammatory agents to treat depression or other mood disorders, and I think that's really smart. What we do use are regular psychotropic agents. In major depression, antidepressants. In bipolar disorder, often atypical antipsychotics.

Here, a fairly consistent story is beginning to emerge, and this is the essence of that story. People who are depressed and have lower levels of inflammation—say, a C-reactive protein (CRP) level less than one milligram per liter—preferentially respond to selective serotonin agents, or agents that have significant serotonin reuptake capacity.

On the other hand, as inflammation increases—as CRP rises above 1—it appears that people have a preferential response to agents that interact with catecholamines, either norepinephrine or dopamine.

There's a large study from Europe suggesting that at CRP levels above 1, people with depression do better with nortriptyline, for example—an old adrenergic TCA [tricyclic antidepressant] than they do with a selective serotonin reuptake inhibitor.

Madhukar Trivedi has replicated those data in smaller samples, again, showing that if you take people and you put them on an SSRI plus placebo versus SSRI plus bupropion, which is an adrenergic-dopaminergic agent, when people have low levels of CRP, so they're not inflamed, they actually do better with just the serotonin agent and not the addition of the bupropion.

Conversely, when CRP levels rise above 1, you begin to see a falling off of efficacy for the SSRI and increasing efficacy for the addition of that adrenergic-dopaminergic agent. We know that inflammatory activation targets dopaminergic circuits, especially in the ventral striatum, so that makes biological sense.

Now, I have done several large studies with the folks at Sunovion looking at lurasidone—which is an atypical antipsychotic—as monotherapy for bipolar depression in both adults and children and adolescents.

We see a very striking, similar pattern—it’s especially apparent in the adults—which is that as CRP rises, as inflammation rises in these people with bipolar depression, you see a larger and larger benefit for lurasidone compared to placebo.

This is because people get a larger response to lurasidone with increasing inflammation and a smaller response to placebo. A slightly more complex pattern is seen in children, but it's basically the same idea.

It's interesting. In that large study in adults with lurasidone, if your CRP was low, you have no more likelihood of responding to lurasidone than you do to just a placebo pill. If your inflammation is high—say, greater than five milligrams per liter—the effect size for lurasidone is stupendous compared to placebo.

Taken all in, I now tell people that I think that there's not much harm in looking at C-reactive protein in people before you start an antidepressant if you have the capacity to do so. It is a standardized laboratory test. It's not always covered by insurance, but it's reliable between labs.

Generally, when we're treating somebody, especially with an antidepressant for depression, you're sort of throwing a dart at the dart board. We don't have much empirical guidance.

Given that we don't have much empirical guidance, something like this is not likely to harm us, and the data would suggest that we were going to get some increased success out the door by looking at inflammatory biomarkers.

Again, I say this with some caution because what has not been done up to the current time is a large, prospective study where people look at inflammation and see whether it predicts response, so we're about halfway there.

Given that it's pretty benign to get a CRP, I think it's justifiable now to do that as we try to think about what antidepressant to put a particular depressed person on.