This transcript has been edited for clarity.
Akshay B. Jain, MD: Today we are very excited to have Dr. Leggio join us all the way from the National Institutes of Health (NIH). He is an addiction physician scientist in the intramural research program at NIH. Welcome, Dr. Leggio. Thanks for joining us.
Lorenzo Leggio, MD, PhD: Thank you so much.
Dr. Jain: We’ll get right into this. Your session was, in my mind, extremely informative. The session looked at glucagon-like peptide 1 receptor agonist (GLP-1 RA) therapy and its potential effects on mitigating alcohol misuse syndrome, so, reduction of alcohol addiction potentially.
We’ve seen in some previous clinical trials, including many from your group, that alcohol use is known to be reduced — the overall risk of incidence, as well as recurrence of alcohol use — in individuals who are on GLP-1 RA therapy.
Can you share more insights about the data already out there?
Dr. Leggio: At the preclinical level, we have a very robust line of studies, experiments, and publications looking at the effect of GLP-1 RAs, starting from exenatide up to, more recently, semaglutide. They show that these GLP-1 RAs do reduce alcohol drinking. They used different animal models of excessive alcohol drinking, using different species — for example, mice, rats, nonhuman primates — models that reflect the excessive alcohol drinking behavior that we see in patients, like physical alcohol dependence or binge-like alcohol drinking, and other behaviors in animal models that reflect the human condition.
In addition to that, we recently have seen an increase in human evidence that GLP-1 RAs may reduce alcohol drinking. For example, there is some anecdotal evidence and some analyses using social media showing that people on GLP-1 RAs report drinking less alcohol.
There are also some pharmacoepidemiology studies which are very intriguing and quite promising. In this case, people have been looking at electronic medical records; they have used the pharmacoepidemiology approaches to match patients on GLP-1 RAs because of diabetes or obesity, and have compared and matched to patients on different drugs as the controls.
A study was recently published Nature Communications by a group in Cleveland in collaboration with Dr. Nora Volkow from the National Institute on Drug Abuse. This study shows the association between being on a GLP-1 RA and the lower incidence of alcohol use disorder and lower drinking.
There is also some promise from prospective randomized clinical trials. In particular, there was one clinical trial from Denmark, a well-known and -conducted clinical trial where they looked at exenatide, and they didn’t see an effect of exenatide compared with placebo in the main analysis. But in a subanalysis, they did see that exenatide reduced alcohol drinking, but only in patients with alcohol use disorder and obesity.
This suggests that these medications may work for some patients and not for other patients. That’s fine, because just like in any other field in medicine, including diabetes, obesity, hypertension, Parkinson’s, and depression, not all medications work for everybody. If these medications will work for alcohol addiction, we do not expect that they will work for everybody.
One ongoing question in the field is to try to identify the phenotypes or the subgroup of people who may be more responsive to these medications.
Dr. Jain: This is such a fascinating field, and all these studies are coming out. In your review of all the literature so far, do you think this is dose dependent? Also, we see that, for instance, with certain individuals, when they take GLP-1 RA therapy, they might have a lot of gastrointestinal (GI) side effects. Recent studies have shown that the rate of these GI side effects does not necessarily correlate with the amount of weight loss. In the alcohol addiction field, do you think that the GI side effects, things like nausea, could also have a potential role in mitigating the alcohol addiction?
Dr. Leggio: This is a great question. They may play a role; they may contribute, too, but we don’t think that they are the driving mechanism of why people drink less, for at least a couple of reasons.
One is that, similar to the obesity field, the data we have so far don’t necessarily show a relationship between the GI side effects and the reduction in drinking. Plus, the reduction in drinking is likely to happen later when many GI side effects are gone or attenuated.
The second reason is from the neuroscience field. We are starting to better understand the mechanism at the brain level as to how these medications work. We don’t see that the nausea or, more generally, not feeling well — malaise, etc. — are driving mechanisms for how these medications work.
Again, it’s not to discount completely that the GI side effects may play a role, but I would say that, if anything, they may be more contributing to. And if they do, that will not be unique to this class of medication. For example, we have three medications approved by the US Food and Drug Administration (FDA) for alcohol use disorder.
One challenge we have in the addiction field is that many people don’t know that these medications exist — many primary care providers don’t know — and they are completely underutilized. Everybody here who is listening to us knows that roughly 85% of people with diabetes receive a medication for diabetes. For alcohol use disorder, the number is 2%. These are medications approved by the FDA.
One of them is naltrexone, which does give GI symptoms — in particular, nausea and vomiting. The other medication is acamprosate, which does give diarrhea.
You have medications approved for alcohol disorder where you do have some GI symptoms, but they are not the mechanism either for how these medications help people to curb craving and reduce alcohol drinking.
Dr. Jain: What about the dose-dependent action? Do you think that GLP-1 RAs, at a lower dose, may not have an effect on alcohol use disorder vs at a higher dose, or is everyone a little different?
Dr. Leggio: That’s a wonderful question. The short answer is, we don’t know, to be honest. Now, in some of the animal studies — my team has been in collaboration with other scientists in the NIH intramural research program, and also with scientists in academia, for example, at Scripps, UCLA — we see a dose response where the higher the dose, the higher the effect of the drug. In this case, semaglutide reduced binge drinking in a rat model of a physical alcohol dependence.
That said, I would be very cautious about claiming, based on the rodent data, that humans will have a dose response. It’s an open question. We really don’t know. Some of the pharmacoepidemiology data suggested that even lower doses — for example, using semaglutide for diabetes without going up to the obesity dose — may be just as effective as a higher dose in reducing the incidence of alcohol use disorder.
It’s important also to keep in mind that the pharmacoepidemiology data are always an association. Reduction in alcohol disorder is associated with the prescription GLP-1 RA, but they don’t really replace the more gold-standard, double-blind, placebo-controlled randomized clinical trial. Nonetheless, with the pharmacoepidemiology data, I think there is an argument to at least hypothesize that people may respond well, even to lower doses.
This also may be important from a safety standpoint.
Basically, we need to wait for results in the next years to come from randomized clinical trials to better unfold the question about doses. For example, just anecdotally, I will tell you that in the clinical trial we are conducting right now at the NIH Intramural Research Program, for which I’m the principal investigator (PI), we are going up to 2.4 mg — the highest dose of semaglutide.
We are collaborating with Kyle Simmons, PhD, from Oklahoma State University. Our two studies are not like a two-site clinical trial, but they are harmonized. In Dr. Simmons’ clinical trial, they’re going up to 1.0 mg. We are excited about this team approach because the trials are slightly different, but they’re harmonized to the point that, once the studies are done, we’ll be able to combine and compare data to better answer the question about dosing, and many other questions.
Dr. Jain: From a clinical perspective, we see that many people who are battling alcohol use disorder may not have obesity. They might actually be on the leaner side, and hence, we may not want to use a high dose of GLP-1 RA therapy. It’ll be very exciting to see when these results come out.
This brings me to the next question. I think everyone would love to know why this happens. Why is GLP-1 RA having this effect on alcohol use disorder? I know that your group has done many animal studies, as you pointed out, and one of the postulated theories was the effect on the GABA neurotransmission pathway.
Can you tell us more about what you feel is the underlying mechanism of action here?
Dr. Leggio: I will start by saying that we don’t fully know. There are many open questions. If I can sidetrack for one second: We come up with the idea that, first of all, alcohol use disorder and substance use disorder are addictive behaviors, addictive disorders. We define addiction as a brain disease.
Granted that addiction is a brain disease, it doesn’t mean that addiction works just in the brain in isolation. As we all know, the brain works in concert with the rest of the body. One specific approach my team has been taking is working on the analogy and the similarities between obesity and addiction to try to understand how the body-brain connection, such as the gut-brain-neuroendocrine pathway, may play a role in patients with addiction.