false
Catalog
Long-Term Antidepressant Treatment: Let's Look at ...
Lecture Presentation
Lecture Presentation
Back to course
[Please upgrade your browser to play this video content]
Video Transcription
Hello and welcome. I'm Tristan Grindode, Deputy Medical Director and Director of Education at the American Psychiatric Association. Thank you for joining us for today's SMI Advisor webinar, Long-Term Antidepressant Treatment. Let's look at the evidence. SMI Advisor, also known as the Clinical Support System for Serious Mental Illness, is an APA and SAMHSA initiative devoted to helping clinicians implement evidence-based care for those living with serious mental illness. Working with experts from across the SMI clinician community, our interdisciplinary effort has been designed to help you get the information that you need. Now I'd like to introduce you to the faculty for today's webinar, Andrew Nirenberg. Dr. Nirenberg holds the Thomas P. Hackett Endowed Chair in Psychiatry at Massachusetts General Hospital, and he is the Director of the Daughton Family Center for Bipolar Treatment Innovation. He also serves as the Associate Director of the Depression Clinical and Research Program at MGH, and he is a Professor of Psychiatry at Harvard Medical School. Andy, welcome to SMI Advisor, and thank you so much for leading today's webinar. Oh, Tristan, thank you so much. It's my real privilege to be able to present for this presentation. I don't have any conflicts of interest, although I do work with a lot of the pharmaceutical companies, but mostly in the area of bipolar disorder these days. The learning objectives today are to assess the risk of recurrent major depressive episodes, to really revisit that, because I know a lot of us know that, to summarize the strengths and weaknesses of long-term antidepressant treatment, we make a lot of assumptions about it, and I want to be able to question those assumptions, and I want to be able to describe the strengths and weaknesses of psychotherapy alone to prevent recurrent depressive episodes. So the big question is, what is the risk of depressive relapse and recurrence? Do antidepressant treatments prevent relapse or recurrence, and if so, how much? And here's the most difficult question I think that we all face as clinicians, and that is, is treatment forever for some people? So this is a well-known graph that was originally published in 1991 by David Kupfer, and what it essentially does is looks at the long-term course of depression. On the left, it says increased severity as you go down. On the bottom, time is a cross, of course, and someone has theoretical euthymia. They start to develop symptoms, it progresses to the syndrome, and then there are the treatment phases of acute continuation and maintenance, and during that time, if somebody is among those people who can respond to treatment, they can either go on over time and continue into euthymia, or they can have a relapse during acute treatment, those first 6-12 weeks, or they could have a relapse during continuation treatments, 4-9 months, or they can have a recurrence thereafter. In some ways, there's a theoretical aspect to this, that the relapses are thought to be a continuation of the original episode, and a recurrence is thought to be a separate another episode, and those are really just assumptions. Depression waxes and wanes. Depression comes and goes. Someone can respond to treatment, and then you can lose that response over time, and I dare say that we don't understand exactly why that happens. If you look at a classic study done by Lavori and colleagues, if you look over a period of 15 years, right, that's 15 years, there's a cumulative probability of having a recurrence. Now, this is in a group of patients who were followed mostly naturalistically, and they were not treated systematically. This is purely observational. And what will strike you is that once you get out to about, oh, let's say 7 or 8 years later, you start to have more than two-thirds of people having a recurrence, and if you wait even longer, to 15 years, it starts to approach 90%. So the challenge is if you get somebody well, how do you keep them well, and how long do you really have to treat them for, and is it the best thing to do just to keep them on the medication that worked acutely, or can you wait until symptoms come back? These are all actually open questions. If you look at the STAR-D study, the Systematic Treatment to Relieve Depression, Systematic Treatment Alternatives to Relieve Depression, this was a massive study that was done that many of you may know about, and I want to point out a couple of things about it. Now, the first thing is that there were four steps. The first step was escitalopram alone. The second step was complicated because somebody can get all sorts of things during the second step. The third step, there were some other choices, and in the fourth step, there were other choices. And what is really remarkable from this study that was published in 2006 by John Rush is that if you look at what happens in naturalistic follow-up 12 months later, what is quite striking, first of all, after step one, you have about 55% of people who ended up having a relapse. If you look at the blue, and that's step two, then you have even more people who had a relapse and fewer people who are remaining well. If it was perfect, if people got well and remained well, you would just see a horizontal line right across the very top, right? That's where you would see it. And going down is the proportion of patients who are free of relapse goes down. So if you look at step two after step one didn't work, right, that's escitalopram, you've got about maybe 30% who are remaining well, meaning 70% did not. And then you have even steeper curves, but people weren't followed as long for steps three and steps four. And what this really indicates is how difficult it is to treat people with major depressive disorder and get them well and then keep them well, right? It is really challenging. If you look at those patients who didn't quite get all the way better, right, they didn't get to remission, they responded, and they chose to stay on those drugs, the long-term outcome is even worse. And if you look all the way on the right, that's about maybe, if you're lucky, 25% and even less are remaining well. What this suggests is that if you don't get somebody all the way to remission, they are at high risk of getting ill again in the next 12 months. Now, we have this full armamentarium of antidepressants, and you're all familiar with these. The question is, is any one of them or any class any better than any other class in keeping people well? And it turns out there's very little data that compares them, but I'll try to look at the extant data, and we could try to make sense of that together. Back in 2015, in the International Journal of Neuropsychopharmacology, Sim and colleagues were trying to do a full meta-analysis that would pull together all of the data of comparing continuing an antidepressant versus switching over to a placebo. Now, some of these data have been criticized in that the people who get switched over to placebo may experience some discontinuation reactions, which could be misinterpreted as having a relapse. In some of these studies, it was done perhaps too quickly. In other studies, it was done gradually to try to reduce that. So with that caveat, what this shows is that the relative risk of staying well favors the continuing of antidepressants, and the degree of that effect size, it's about twice as effective as going on to placebo, switching off of it. So the pooled risk ratio is 2.03, and again, the way to interpret that is antidepressants are twice as effective as placebo in preventing relapse or recurrence. That's pretty convincing, right, so that there may be some subpopulations who are more prone to have relapse and recurrence, and you'll see a bigger effect size in those subpopulations. Now what about psychotherapy? Now this is also really interesting because what Sim and colleagues did was they looked at all of the studies of acute psychotherapy, followed by continuing that psychotherapy, or it's not quite switching over to placebo, but it's switching over to something that's less than the full psychotherapy, and this is mostly cognitive behavioral therapy and things like that. What was striking to me when I read this and looked at it, looked over it, right, you've got mindfulness-based therapy, IPT, CBT, family psycho-ed, what was striking to me is that this was 1.4 times as effective as placebo, not quite the effect size of the antidepressants, a little bit less. Now these are very difficult studies to do, and even though there were 1,969 participant subjects in this, you'll see that each individual study was fairly small, and which the largest one with Caton in 2001 was 386. So I thought that was interesting. Now if you compare the drug continuation, the drug maintenance, and the psychosocial maintenance, and particularly if you look at the RR, that's one, two, three, four from the bottom, what you'll see is that the risk ratio hovers around two, which I mentioned earlier for the drug maintenance, for drug versus placebo, twice as effective as placebo in preventing it, and there's the 1.4. So this is actually that direct comparison, again, which is interesting. Now whether or not you can actually compare these is also a question. That is, we really don't know if the severity of the people who entered these studies was similar across the drug continuation maintenance and the psychosocial maintenance. So again, I think you have to be very cautious about comparing these sort of numbers. Let's dig into some of the newer meds and see what the evidence is for them also working long term, because you assume that if something's been approved acutely by the FDA that it would, in fact, work in the long haul. So here's a study of lazadone at 20 milligrams and 40 milligrams, with placebo being in the very light gray of lazadone 20 and the darker gray and the darker one, lazadone 40. And this is the estimated cumulative rate of relapse. That means if you're going up on the graph, you have a higher rate. There are several things that are really quite striking about this. That is that, first of all, the overall rates were really rather low over the 196 days. And although it looked like over the course the lazadone did somewhat better, if I'm reading this correctly, than placebo, the effect size wasn't that large. If you look at levomel nasopran, and this is, again, taking people who did well and then it's time to relapse, what you'll notice after 225 days is that the time to relapse and the proportion who relapsed is substantially lower with the levomel nasopran extended release versus placebo. Nevertheless, this is about a little more than six months, the relapse rate was not particularly high, and this may not have been a group that was particularly prone to have relapse over that period of time. Nevertheless, right, you've got a clear advantage of the levomel nasopran versus placebo. And what about vortioxetine? With vortioxetine, you have on the left is the survival probability. So again, I'm just trying to orient you to each of these graphs because they're all slightly different. In this one, it goes back to what I said earlier. If somebody survived the entire time and did great, you would have a horizontal line right at the top, right? And what you see here is that in the red, the vortioxetine, 5 to 10 milligrams was clearly better than the placebo here. And then they tried to figure out, is there any subpopulation where there's either a bigger effect or a higher probability of having a relapse? And one of the things they looked at was childhood trauma. And here again, you do see a separation between the vortioxetine and the placebo. But if you look at the placebo in the blue, right, you've got about 75% who remain well, meaning that 25% had a problem. If you go back earlier, you'll see that the overall group did somewhat better. So this is not surprising, right? I think we're learning more and more about the insidious and persistent and pervasive effects of having childhood trauma on top of major depressive disorder. And this is one of those pieces of evidence that shows that people can have a worse course even when you compare the placebo group to the entire group there. And what about ketamine? Now, we have all read a lot about ketamine. Believe it or not, it's now been 20 years since the first report, and it was only last year that S-ketamine got approved by the FDA. And many of our very treatment refractory patients are getting ketamine. But here's the question. If somebody responds to ketamine, how long do they sustain it for? And if somebody responds to acute ketamine, how long do you actually either give it for or wait? Or if you want to do something afterwards, what is it that you would possibly do? Do you just keep people on ketamine forever or for how long? You know, I don't think we quite know that yet. So here again, you have the proportion of patients without a relapse. If it worked perfectly, you would have a horizontal line at the top. And what you see here, after 27 days or so, you've got about 20% who are remaining well, meaning that 80% will have a relapse during that time. Now, if you look at just the 14 days, right? I'm sorry, in 10 days, you've got about 75% well, so you'll have a 25% relapse in that timeframe. And many of you may know that for some people with, who get ketamine, that they can stay well after one infusion, after even just one of these. But clearly, it is not something that's gonna keep somebody well in the long run. One of the things that Sanjay Mathew and colleagues tried to do, based on the mechanism of action, they thought that the ALS drug, the amyotrophic lateral sclerosis drug, riluzole, which has a similar mechanism of action to ketamine, they wanted to test the hypothesis that maybe after people got ketamine, maybe riluzole would then keep them well. And they were quite disappointed that they did not find that in this particular study. Right, there was no difference in the relapse rate between placebo and riluzole. So that was disappointing. And many of you may also know there's some controversy about what the actual mechanism of action of ketamine is, and it may be much more complicated than we originally thought. And here is S-ketamine, right, but that's approved by the FDA, ketamine is not. This is by Daly and colleagues back in 2018. And for these patients who got it, right, they did well and then mostly sustained it over about a week, mostly, right? The change in the Montgomery-Asperger depression rating scale was mostly flat once they got better. There was one patient who sort of didn't do as well, and that's the person at the top. They were also trying to look at different doses and trying to understand what the real doses needed to be, and the higher doses may have been somewhat better. This also looked at the same thing up to eight days. And here's another way of looking at this phenomenon by Popova and colleagues back in June of 2019, just last year. And here it was a little bit different. These are people who were taking an antidepressant and people got S-ketamine plus an antidepressant or they got their antidepressant plus placebo. So it's trying to understand if S-ketamine actually improved people over time. And what this showed is that it not only worked, but it worked up until day 28. What about repetitive transcranial magnetic stimulation? Right, one of the things that has also spread into the community is more and more people are doing the repetitive transcranial magnetic stimulation. And the original reports from George and colleagues, was Mark George and colleagues, it seemed like it had a modest effect, but many of you may know that in the community, people are finding that it really can be quite useful for many patients. So here is one of the few studies in terms of a follow-up, right? Because what happens if you have treatment versus no treatment? And that essentially is taking the people who responded acutely and either not giving them treatment or continuing the RTMS. And here again, you'll see that the continued RTMS clearly is better than no treatment. And after about six months, you've got about 75% who are staying well with the continued treatment and 25% did not do well, where it's almost the opposite with those who didn't get treatment. And they start to fall off even after two or three months. So by the end of six months, less than 20% are staying well, which means that 80% had a relapse. And here's the other really interesting question, but by Flint and colleagues. If you have somebody with psychotic depression and you have them respond to one of the approved antipsychotics to add to an antidepressant, how long do you continue the antipsychotic? And what's clearly shown here is that for those people who responded to sertraline and olanzapine, that they should continue taking it because it keeps them well, because those people who were switched over from sertraline and olanzapine to sertraline and placebo did a lot worse, right? So about 80% are still well with the combination at 36 weeks versus less than 50% are staying well with sertraline and placebo. Very important study, right? It shows that if somebody has psychotic depression, if it makes sense clinically and the side effects aren't that bad, probably should continue what worked acutely. And what about rexanilone? And many of you have heard about this for postpartum depression, that one of the remarkable things about it is that there is a 60-hour infusion and this is sustained up to day 30. This is one infusion over 60 hours. So the Madras score, as you can see, goes down into the normal range and essentially is still in that normal range at about day 30, which is really extraordinary. So it's one of the things about this, the thing with postpartum depression, but still the question is, if someone has a relapse or recurrence, what do you do? Do you redo the infusion or do you not do the infusion? Still a question. And here's the other thing that is an ongoing challenge. What do you do after ECT? And there is a remarkable dearth of studies that try to compare potential interventions. Now, many years ago, Harold Sackheim and colleagues did the seminal studies that showed that nortriptyline plus lithium sustained the effect of ECT better than nortriptyline or lithium and clearly better than doing nothing. All right, so the other thing that's happened over time and just more empirically than any other method is that if somebody has ECT, then sometimes you just continue it. So Kellner and colleagues, Charlie Kellner and colleagues, asked the simple question, what's the comparative effectiveness of continuation ECT versus the combination of meds of nortriptyline plus lithium? And one of the things that they did, and this will make sense to all of you, after they responded to ECT, they tapered off how often the ECT was done. So first once a week, then biweekly, then once per month, and then close to the 20th week, there was no more ECT until the observation was ended at the 24th week. And what is clear from this study is that the continuation ECT was about equally effective as continuation nortriptyline plus lithium with about, oh, let's say 60% or so of people remaining well, 40% not remaining well. And for the people who get ECT, that is actually pretty good. So you can have a choice like that, not too bad. Now, the other thing, DeRubis and colleagues published this rather interesting study that tried to understand if somebody had an antidepressant with or without cognitive therapy, would it keep them well? And there are four groups here. There's the people who got the antidepressant medication plus cognitive therapy where the antidepressant was maintained. That is in the dark, is that greenish, bluish? That's the dark graph at the top, the line at the top. Then there's the group who had the antidepressant withdrawn. Then you have an antidepressant only where the antidepressant was either maintained, that's the solid orange line, or the antidepressant was withdrawn. And what is striking about this is that the people who did well, whether or not they got cognitive therapy or they just got an antidepressant, those are the people who had the antidepressant maintained. The people who didn't do as well were those people who had the antidepressants withdrawn. If you look at the proportion who have a sustained recovery, what you'll have, and this is, wow, 1,100 days, right? That's a long time. You have about 40% who've been maintained and doing well while they took the antidepressant versus about half of that, about 20% who stopped the antidepressant. What this shows is that you really do need to continue the antidepressant even if someone had responded to an antidepressant plus cognitive therapy. And then there is this remarkable study from 2004. And what's remarkable about it is Giovanni Fava did a study where he gave patients who stopped their antidepressants a certain type of cognitive behavioral therapy versus clinical management. And what is buried in the details here is that the cognitive behavioral therapy is really well-being therapy. And what that is is it's cognitive therapy is sort of turned on its head so that you increase someone's hedonic experience. What's remarkable is that what Giovanni Fava did was he gave 10 one-half hour sessions over 20 weeks and then stopped it. And what happened was that that's when he started the clock for these different groups, the effect of the well-being therapy lasted for up to six years, again, extraordinary. Now this has not been replicated but it is currently under study. It's taken this long for other people to adopt the well-being therapy and in other groups they're trying to study it to see if they can replicate this rather remarkable effect. There was also Zindel Siegel and colleagues did mindfulness-based cognitive therapy either, and basically added to usual care and compared it to usual care. And what he found, his group found, was that the mindfulness-based cognitive therapy was able to sustain people in feeling better. In the original study that he had when he first published results about this was that, of course, this was in the group where they were at high risk and this was a high-risk group. So it's another form of therapy that helps people manage stress and cope. So Kennedy and colleagues in the Canadian Journal of Psychiatry also published some guidelines for maintenance, right? So the question that we all have as clinical people, clinicians, is who would you wanna do this for and who is at particular risk? And none of this will surprise you in terms of frequency, severity, chronicity, comorbidity, residual symptoms, or a history of difficult-to-treat episodes, right? These are the people who are really at risk and that are candidates to have ongoing treatment that needs to be reevaluated, right? You just don't know. So the fundamental questions that I wanted to be able to address and we'll have time to discuss with all of you, right? What's the risk of depressive relapse and recurrence? For many patients, it is fairly high, especially those who have these particular aspects of their history. The second thing is do antidepressant treatments prevent relapse or recurrence? And overall, it can, but you all know that it is not perfect and that many patients, a high proportion of patients, especially if they've had a difficult time, are extremely difficult to get well and keep well. And we currently don't really know what to do for the people who are on continuation or maintenance therapy. And if they're really taking the therapy, what do you do when they have a relapse or recurrence anyway? And the risk is anywhere between 30% and 70% depending on how difficult it has been to treat their episodes. And that presents us with a tremendous and difficult clinical problem to solve. And certainly that's an important area of research that has not resolved it, right? So we have that as a problem. It's not perfect. But then again, nothing is perfect in medicine. And if you compare the long-term treatment of depression to things like hypertension, it's really not a lot worse. The last question that I wanted to be able to address with all of you is treatment really forever. Does it mean that for some patients, they have to stay on an antidepressant? Because if they don't, they are at high risk of getting ill again. And you don't know when that might happen. And is it possible that you can treat somebody, let's say for four to six months, and then if they're doing well, you can taper them off of the antidepressant and then do watchful waiting and that you see if the symptoms recur, you can have a low threshold to be able to restart it and see how people do over time. That tactic has never been studied compared to just keeping people on antidepressants. And of course, when people are on antidepressants for a long time, the longer they're doing well, the less tolerant they are of the side effects, which is another paradox. So it becomes increasingly difficult for some people to stay on. For other people, definitely yes. And then the final issue is, is there a subpopulation of people who essentially get sensitized to the antidepressants and that's why they have to stay on it? And we don't know the answer to that either. And certainly psychotherapy can help, but the DeRubis study suggests that for some people, they probably do okay with psychotherapy alone, but many other people may need the antidepressant plus the continued psychotherapy. So with that, really thank you for spending your time with this in these very peculiar times that we're all in. And I'm happy to take any sort of questions that you have. Thank you.
Video Summary
In this video, Dr. Andrew Nierenberg discusses the long-term treatment of depression and the risk of relapse and recurrence. He highlights the need for ongoing treatment for patients with major depressive disorder and emphasizes the importance of medication maintenance. Dr. Nierenberg presents various studies and data on different treatment approaches, including antidepressants, psychotherapy, and other interventions like ketamine and transcranial magnetic stimulation. Overall, he concludes that while no treatment is perfect, antidepressant medications have been shown to be twice as effective as placebo in preventing relapse or recurrence of depression. He also discusses the challenges of treatment duration and determining when to stop or continue treatment for individual patients. Dr. Nierenberg suggests that certain subpopulations may require longer-term treatment due to the risk of relapse and inconsistencies in response to treatment. Psychotherapy, especially cognitive-behavioral therapy, can also be beneficial in preventing depressive episodes. He concludes by acknowledging the ongoing research and uncertainties in long-term treatment for depression and opens the floor for questions. This video was part of the SMI Advisor webinar series, organized by the American Psychiatric Association and the Substance Abuse and Mental Health Services Administration.
Keywords
depression
relapse
medication
psychotherapy
treatment duration
subpopulations
cognitive-behavioral therapy
Funding for SMI Adviser was made possible by Grant No. SM080818 from SAMHSA of the U.S. Department of Health and Human Services (HHS). The contents are those of the author(s) and do not necessarily represent the official views of, nor an endorsement by, SAMHSA/HHS or the U.S. Government.
×
Please select your language
1
English