IRA FLATOW: This is Science Friday. I’m Ira Flatow. We’re continuing our conversation about what we do and don’t know about how antidepressants work. Sometimes finding the right medication is basically trial and error. And that’s because scientists are still trying to figure out why some antidepressants work for some people but not for others.

Researchers at the Veterans Administration wanted to know if genetic testing might help doctors with prescribing the antidepressant best suited for their patients. Specifically, they looked at genes that indicate whether or not someone is able to properly metabolize a medication. Joining me to explain this research and its broader implications is my guest, Dr. David Oslin, professor of psychiatry at the Crescenz Department of Veterans Affairs Medical Center, University of Pennsylvania Perelman School of Medicine, based in Philadelphia, PA. Welcome to Science Friday.

DAVID OSLIN: Thanks, Ira. It’s great to be here today with you.

IRA FLATOW: Thank you. Let’s start off by talking about, why did you decide to test for these specific genes linked to metabolism?

DAVID OSLIN: As you outlined in the beginning, a critical issue for us in the psychiatry field is not necessarily knowing which medicine is going to work with which patient. And that often leads to this trial and error kind of approach where we’ll try what we think is a good standard medicine in somebody, but if that doesn’t work, we will need to move on to another. Sometimes we’re lucky, and that first one works quite well, but often not.

So there’s a lot of factors that go into that decisional process. And a factor that we have known about for a number of years is variability in metabolism that is dictated by your genes. And so there are a number of companies now that offer genetic testing or pharmacogenetic testing that helps us better understand how you metabolize antidepressants or even other medicines. And the goal of our project was to see if that actually helped treat patients or lead to better outcomes for patients.

IRA FLATOW: And did it?

DAVID OSLIN: Well, the short answer is yes, but it’s a little bit more complicated than that. So, a big piece of genetics is that the genetic tests for a given medicine probably don’t tell us an answer for every patient. And the way to think about that in the study that we published was that in the control group, which did not have the pharmacogenetic testing, at the end of the day, about 20% of them ended up getting put on a medicine where genetic testing would have told us that we probably should have selected something different. The other 80% did not have a drug-gene interaction. So the test was really mostly helpful for that 20%. And we have to know that limitation sort of going into it.

IRA FLATOW: And so you then adjust for that limitation?

DAVID OSLIN: So, our design didn’t let us adjust for that. So we’re going to do some further analyses and explore that issue. Even with the fact that the genetic test was only likely relevant for about 20%, we saw a positive difference.

The way this study was designed was it’s a typical randomized clinical trial where patients were randomly assigned to, in this case, getting back a gene test, test results that helped them and their provider work on which medicine they wanted to use in the treatment of their depression, or they didn’t get a test result back and they were treated as usual. And then in the treatment with the genetic test group, over the course of six months, the patients did better. They did better in terms of the percentage of patients that ended up in remission as well as the patients that just responded to some extent.

IRA FLATOW: Yeah. I know that you were initially skeptical that this approach might be useful until you did this research. Why is that?

DAVID OSLIN: So, yes, Ira, I was skeptical going into this because my clinical experience is that these medications have a fairly broad therapeutic index, meaning I don’t have to get the dose exactly right for them to be effective. And we’ve never really had literature that says that serum drug levels are really good at predicting outcomes. So the idea that metabolism might make a difference just didn’t have a lot of scientific background in a lot of ways. So I’ve been pleasantly surprised that it’s made a difference in the patients that we enrolled in the study.

IRA FLATOW: Is it possible to figure out why that percentage of 20% of the patients found success and the others didn’t?

DAVID OSLIN: So, that’s a great question. It wasn’t necessarily that that 20% found success. Lots of patients found success. The study wasn’t designed to tease out who the 20% are.

So what the 20% reflects is, in the control condition, 20% of those patients ended up being started on a medicine that had a drug-gene interaction, but those providers and patients didn’t have that gene result to steer them in another direction. Overall, the difference in outcomes was about 5% to 10% different. So about 5% to 10% more patients remitted over the course of six months than in the usual care group.

IRA FLATOW: So what does this tell you? What do you learn from this, and how do the patients benefit from what you’ve learned?

DAVID OSLIN: What I think of this area of medicine as is being in sort of its infancy. These particular tests that are available now are really a first generation of genetic tests for us. We’d really like to have tests that really predict which medicine is going to work better for which person, and that’s not what these tests do.

So I think this is a really good step forward. I think in my practice, I’ll probably be using these tests more often. Probably not in every patient, but certainly in more patients than what we were using before. And I feel like we’re moving the field forward to a time where we don’t have to have quite as much trial and error with these medications.

IRA FLATOW: Will they help you choose which medicines to give your patients, even though we’re talking about this 20%?

DAVID OSLIN: In the world of antidepressants, we have a number of options, none of which really work any better than the other. So if I have one that I don’t know how to dose well, I can pick another that’s not genetically influenced and otherwise equal in terms of its chance of helping that person. And that should lead to better outcomes and did in the study.

IRA FLATOW: That’s interesting. What about uses to help those who are taking multiple psychotropic medications, right? There are a lot of people who do this. Might it help determine how well patients are actually absorbing the drugs and if they can reduce the amount of medications that are not working?

DAVID OSLIN: We don’t have the answer to that. It’s an area that I’m particularly interested in. So, in medicine these days, we often add more medicines when something’s not working, particularly with chronic illnesses like depression. So if I start you on treatment X, and you get some response to it but not a full response to that treatment, I’m more likely to add something else on to the regimen to see if I can boost that response.

So people end up on lots of medicines. And that’s the unfortunate part of that practice. So the idea of genetics could be that some of those medicines just shouldn’t have been good mixes or matches for that patient. And if I have a genetic test that suggests that a medicine is poorly metabolized or rapidly metabolized, maybe taking that one away and swapping it out or just taking it away could actually lead to a better outcome for that patient.

IRA FLATOW: Some people may be confusing this with genetic testing for, let’s say, cancer screening, something like that, where you’re actually determining whether the drug works. This is not what you’re doing, right?

DAVID OSLIN: That’s correct. So, the cancer field is much further ahead than we are in the mental health field with how to use genetics in practice. A good example might be the BRCA gene for breast cancer, where if you have a particular mutation in the cancer, it’s actually going to dictate which treatment you get.

Here, we’re really, again, sort of in our infancy of applying genetics to the treatment of mental health and really talking about how people metabolize the medication, which will have effects. So if, as an example, if I am an ultra-rapid metabolizer of a particular antidepressant, I might end up at a standard dose with a much lower serum level of that medicine than will be effective. On the other end of the spectrum, I could be a slow metabolizer, and I will actually end up with a much higher dose than what’s prescribed. And most of our medications, adverse events are dose response related, and so I’m more likely to have side effects or adverse events from that medicine from a slow metabolizer.

IRA FLATOW: There are very few diseases where one gene controls the disease, very few of them. And I think people are probably expecting to find, maybe, the genetic underpinnings of depression. But that may not turn out to be the way it really works, right?

DAVID OSLIN: Yeah, so most chronic illnesses are probably not one disorder. Even simple disorders like hypertension, high blood pressure, really aren’t one disease. They’re really likely many different diseases that have different pathways that manifest themselves in an elevated blood pressure.

Depression’s the same way. Some patients have depression that’s really driven by biology, and those patients might be much more responsive to an antidepressant or a particular antidepressant. Some people’s depression may be much more driven by their environment and their psychosocial circumstances. So we don’t have great ways of teasing out those different types of depression right now. And that is the hope for the future.

IRA FLATOW: Can the genetic screening tell us what deficiencies patients may have in certain key neurotransmitters?

DAVID OSLIN: That is sort of the holy grail of trying to tease out some of the depression. So there are some of the commercial products that are on the market that do say that they can do this. I would say the science is lukewarm on whether those tests are useful or not. And so I think we still have a little ways to go to be there.

IRA FLATOW: This is Science Friday from WNYC Studios. Well, let me expand this topic. Can these genes be used to help figure out how well someone can metabolize other types of medications, not just antidepressants?

DAVID OSLIN: Yeah, so that’s actually another very important point. The reality is that these metabolic genes really tell us how well the liver works in general. These are all genes that help the liver perform well. And they’re pretty agnostic to the medication. So any medication that is metabolized by the liver can be affected by these genes. And that’s a real varied number of medications– cardiovascular medicines, medications for infections, medications for GI problems, et cetera.

IRA FLATOW: You know, people are listening at home, and they might be thinking the question I was thinking– you can buy these home genetic testing services, right? Can those services tell us how well I’m metabolizing my antidepressants?

DAVID OSLIN: Some of the commercial firms that advertise to patients directly or to people directly do include some of these metabolic genes in their panel and do report back to you. I really think that that’s a conversation that you should be having with your health care provider, though, in terms of interpreting those genes and understanding which medicines those genes may be influencing. I would hate for somebody to stop a medicine that’s been working effectively just because they did a home test.

IRA FLATOW: Right. Right. So, are you actually using any of these tests on your patients at the VA yet?

DAVID OSLIN: Yes, we do. We actually have available a couple different options for providers to order the test. And we do get both patients that are interested but also providers that use the test in helping them select the medications, not just in psychiatry but, again, across a broad array of disorders.

IRA FLATOW: Yeah, because you still have to figure out who is best to test.

DAVID OSLIN: Yeah. That’s actually– so that was another learning lesson from this study. And we are trying to do some secondary analysis to see if we could actually better target who to test and who might benefit. One of the better stories here might be testing people with their first episode of depression and not necessarily their third or fourth medicine. But we have more work to do in that area.

I mean, there’s two approaches here. One is, you just test everybody. These genetic tests are actually getting really inexpensive to do. And if you think about the cost of a test over somebody’s lifetime, it’s really not that expensive. Or the other approach is to target the testing when you think it’s most valuable.

IRA FLATOW: What about other illnesses? What’s next on your research? How can you apply this to others?

DAVID OSLIN: So, I’m a psychiatrist, and I’m mostly interested in mental health disorders. So we are trying to think about how to combine genetic testing with other useful techniques to target treatment. So in our group, we’re looking at neurocognition as a way to think about who might benefit best from different forms of psychotherapy. We’re interested in PTSD and how to improve outcomes from PTSD. And there’s a lot of applications here about how to do a better job of having what we call precision mental health care.

IRA FLATOW: Yeah, that really is a challenge, isn’t it?

DAVID OSLIN: I think it’s a challenge, but it’s really exciting. I mean, I think a lot of people think of mental health as this black box of only sitting down and lying on the couch and talking things out. But we really do have a lot of different evidence-based approaches to treating different illnesses. And the challenge these days for me is really figuring out which one is going to work best for which patient.

IRA FLATOW: Thank you, Dr. Oslin, for taking time to be with us today.

DAVID OSLIN: Thank you, Ira.

IRA FLATOW: Dr. David Oslin, professor of psychiatry at the Crescenz Department of Veterans Affairs Medical Center. That’s at the famous University of Pennsylvania Perelman School of Medicine based in Philadelphia, PA.

Copyright © 2022 Science Friday Initiative. All rights reserved. Science Friday transcripts are produced on a tight deadline by 3Play Media. Fidelity to the original aired/published audio or video file might vary, and text might be updated or amended in the future. For the authoritative record of Science Friday’s programming, please visit the original aired/published recording. For terms of use and more information, visit our policies pages at http://www.sciencefriday.com/about/policies/.