How ketamine treats resistant depression

Depression is very common. Estimates are that as many as one in six people are affected, and up to a third of those might have depression termed “treatment resistant”, which fails to respond to the current gold-standard therapies. In recent years this has led researchers to explore the “drug space” more widely to look for compounds that might bring relief to those with these harder-to-manage conditions. Hallucinogens and ecstasy have shown some benefits, alongside the anaesthetic agent ketamine, which appears to be very effective short term, although neuroscientists aren’t sure why. To shed some light on what it might be doing, and speaking with Chris Smith, Alan Anticevic, at Yale, has been brain scanning healthy volunteers both on and off the drug…

Alan - Fundamentally, we still don't understand the abnormalities in the brain of people who suffer from those symptoms. And ultimately that limits how we're developing drugs for specific mechanisms that may be altered. And so this programme of research is designed to try to understand how drugs that are potential therapies affect the human brain.

Chris - Which drugs specifically?

Alan - We have studied a drug that's called ketamine. A substance that binds to a receptor called the NMDA receptor. And at a particular dose, this drug has been shown to be a potentially revolutionary medicine for people suffering from depression. Ironically, we still don't fully understand how it works. And so this study was designed to interrogate the effects of ketamine in healthy adults and to understand the patterns of brain activity that arise while people are given a dose of ketamine known to be therapeutic for depression.

Chris - Because it's an anesthetic agent in high doses, isn't it? We put people to sleep with ketamine or we use it when we do certain procedures, because It also makes them forget what's happening to them...

Alan - Indeed it is. And what's fascinating about ketamine, it has what's called a dose dependent effect on the brain. Which means that at different doses of the drug you can expect it to do different things. So in fact, at a higher dose it becomes an anesthetic. Yet at a far, far lower dose, the drug is able to change communication between brain cells in such a way that very rapidly affects a person's mood who has been suffering from depressive symptoms. And here we are studying that low dose pattern, not the anesthetic dose pattern.

Chris - And how long does the therapeutic effect last? And do we have any insights into who derives benefit? If you give this to somebody with what sort of certainty can you say they're going to feel better?

Alan - That's the billion dollar question. We as a field don't yet have a full understanding of who is most likely to benefit. And the effect of this drug is relatively short acting. So you will have a rapid antidepressant response in some people and that response will not last. And there are complicated reasons why that's the case. So it is our job and one of the reasons why we executed this study is to understand how ketamine affects the brains of different people and to identify whether there are really different patterns in the human brain of the effects of ketamine, which could give us a lens onto its potentially diverse therapeutic potential for different people.

Chris - So how did you actually go about testing how it works?

Alan - We asked 40 healthy volunteers to participate in a brain scan during which they were administered either an infusion of placebo - saline - or ketamine. We imaged their brain using a technique that's called functional magnetic resonance imaging, fMRI, for short to measure activity over time. This allows us then to analyse the patterns of the functional change that is induced by ketamine in different people.

Chris - So you've got a sort of before and after you can see what the connectivity map is, what bits of brain are talking to what other bits of brain, then you put the ketamine in and you can see how that shifts?

Alan - That's exactly right. So we, and we can do that for every single person in this study and we can understand if there is a different pattern of that shift across different people. And because the effect of ketamine is relatively strong at that dose, we could identify these different signatures in a sample of 40 people.

Chris - Do you see a consistent signature or do you see a mixture of signals that differ between individuals? Because, obviously, the pharmacologists - the people who make drugs - would love you to say, we see the same thing happening in these parts of the brain in everybody and then they know exactly where to target with the next wonder drug. But if you see a mixture of things happening, it's much harder to argue a case for exactly what's going on. So which is it?

Alan - Yeah, so that's a brilliant question. Exactly to your point, there is this assumption or, or in a lot of cases the hope that if you give a drug that is therapeutic, it's going to do the same thing in every person. Right? And therefore its biology is what we would call clean. But what we have found, we used a procedure, a mathematical procedure that allows us to test if there are more than one dimension or more than one way that people respond to ketamine. And we have found that in fact there are more than one way and two different patterns emerged as statistically significant, which we have then studied and interrogated further in this analysis. So the implication is in fact ketamine doesn't do the same thing to every person.

Chris - One study I read suggested that about two thirds of people with chronic depression that's been really hard to treat, get some benefit from being treated with ketamine. So if you take your findings, do you think that you can spot some kind of signature in the brain activity that would give you a prediction as to who is likely if they were to take ketamine to get benefit so we know who to focus on?

Alan - Yeah, exactly. So using these kinds of approaches, we can can hone in on a pattern of the effects of ketamine that may be therapeutically predictive and isolate it from patterns that are not therapeutically predictive. And so again, it goes back to your, one of your original questions, which is if the drug is inducing the same response or very similar responses in everybody, then you're working with the same features. In other words, the same effects in everybody. But if it is different, that individual variation is critical to map and that will improve the predictive models because we're focusing on the right kind of variation, the variation that is therapeutically predictive.