Heart risks, serotonin, and fen-phen: 5 Questions for 5-HT receptor researcher Bryan Roth

Roth discusses what we know about drugs that stimulate 5-HT receptors and their effect on the heart.

Bryan Roth got interested in psychedelics by accident. After finishing his biochemistry PhD dissertation on opiate receptors, he started a postdoctoral research fellowship at the National Institutes of Health. There, his advisor suggested he study what was then a little-known group of receptors called 5-hydroxytryptamine (5-HT), or serotonin receptors. Roth’s research led him to two particularly interesting classes of drugs that interact with those receptors: psychedelic drugs, which serve as “agonists” and activate the receptor, and atypical antipsychotic drugs, which do the opposite. His first paper on 5-HT receptors was published in 1984, and he has published dozens of papers on the subject since.

One drug that acts on serotonin is fen-phen, a weight loss drug prescribed to over 18 million people in the 1990s. Physicians noticed that patients taking fen-phen were developing valvular heart disease, a condition that occurs when heart valves are damaged. Some fen-phen users developed symptoms including chest pain and heart murmurs within just a month of taking the medication. The drug also increased the risk of pulmonary hypertension and in 1997 the FDA removed fen-phen from the market. The drugmaker eventually paid $3.75 billion in legal settlements to thousands of people who took the drug. After the fen-phen’s withdrawal, Roth’s research lab began studying the link between valvular heart disease and 5-HT serotonin receptors.

Most psychedelics, including psilocybin, MDMA, LSD, DMT, and mescaline, also act on serotonin receptors. The Microdose spoke with Roth about what we know about drugs that stimulate 5-HT receptors and their effect on the heart.

Your lab published a paper back in 2000 analyzing the binding behavior of fenfluramine, the “fen” in fen-phen, to understand why it’s associated with valvular heart disease. You also posited that other serotonergic medications, which are used to treat psychiatric disorders and migraines, may also put people at higher risk of developing valvular heart disease. Why did you decide to work on that issue? 

After fen-phen was withdrawn from the market, that’s when my lab got interested in this area. Why was it that this drug causes valvular heart disease? We found that a metabolite of fenfluramine called norfenfluramine is a very potent serotonin receptor agonist, and that these 5-HT2 serotonin receptors are found at high levels in the heart and other peripheral tissues. We showed that a number of other drugs previously known to induce valvular heart disease in humans also activated the 5-HT2B receptor, and we published that data in 2000. At around the same time, there was a group from the pharmaceutical companies DuPont and Merck that published a similar data set. 

We began screening other medications and found that a number of drugs used for treating Parkinson’s disease and migraine headaches also activate the 5-HT2B receptor. We notified the FDA and drug companies. To make a long story short, they ignored our recommendations. A few years later, there were large-scale retrospective clinical studies that showed that the drugs we identified had also caused valvular heart disease in up to 30 percent of patients taking them, and several of those drugs have subsequently been withdrawn from the market as well.

Several psychedelics, including psilocybin, LSD, and MDMA, also bind to serotonin receptors. Should people interested in taking these substances and researchers studying potential treatments using these drugs be concerned about possible heart risks? 

As far as I can tell from the FDA’s draft guidance, the agency agrees with us that many psychedelic drugs are serotonergic agonists that have the potential to induce valvular heart disease. The science is solid, and anyone involved in drug discovery has already known this for twenty years. It’s really only news to psychedelic consumers. But if you work for a pharma company, you know this already. Any drug in development is tested against this receptor, and if the drug activates it, they throw it away. It’s well-established as a target to avoid.

It’s a risk, and going forward, clinical trials will need to be designed to determine what is the safe dosing regimen. Psychedelics have never been formally assessed for cardiovascular safety, and they need to be.

What would a formal assessment look like, and do you think the FDA will mandate that?

Typically, people might start with animal studies. They’re not particularly predictive of what might happen in humans. In humans, typically what’s done is a clinical trial where they do baseline echocardiograms in all patients, then again midway through therapy and after. 

At some point, the FDA is probably going to mandate those sorts of trials be done, but it’s not clear from the current draft guidance if they’re going to mandate that for planned phase 3 clinical trials or not. All those studies have one or two doses of psilocybin. If I were a betting man, I would say there’s little chance that that type of dosing would be an issue, but we still need to do clinical trials. The big concern is with repeated dosing, and lots of people are doing that now with microdosing. That’s extraordinarily risky until the appropriate studies have been done; you’re basically recapitulating the situation that led to fen-phen. 

You’ve been studying these receptors for three decades now. Have you noticed any change in how interested people are in this area of work?

In 2018, I gave a Presidential Special Lecture at the Society for Neuroscience conference. There were 10,000 people in the audience.  At the talk, I highlighted my lab’s work on psychedelics, and mentioned that the next day, a postdoc from my lab would present work on the structure of LSD bound with serotonin receptor. That work was subsequently published in Cell and it was a huge finding. Only five people came to his poster. It’s really only been recently that people are interested in this work. 

Many studies compare psychedelics’ efficacy in treating mental health issues to the efficacy of selective serotonin reuptake inhibitors, or SSRIs. When SSRIs were first brought onto the market, they were marketed as a drug that could “fix” depression by increasing brain serotonin levels. Do SSRIs have the same risk for valvular heart disease?

They do not. SSRIs don’t interact with serotonin receptors; they interact with a certain serotonin transporter. With a single dose, there is some elevation of brain serotonin, but it’s really peripheral. There have been studies that measure serotonin in people’s blood, and they’ve found that chronic SSRI treatment actually lowers serotonin levels. 

No neuropsychiatric disease is due to a single neurotransmitter. That said, SSRIs are effective; they’re not for everybody but as a practicing psychiatrist, I’ve seen some truly remarkable recoveries with SSRIs. When I tell people depression is not a serotonin disease, they might say, ‘Well, then, I shouldn't take an SSRI?’ I would not come to that conclusion. They work — we don’t know why they work, but there’s no question they can be effective. What we know from animal studies and some human data is that SSRIs ultimately enhance plasticity in hippocampal and cortical neurons. That’s been known for 20 years, so they share that with psychedelics. But it’s important to note that every drug causes plasticity: morphine, cocaine, amphetamines, PCP. That’s what drugs do — they change how neurons function and that occurs through plasticity. It’s clear that psychedelics also cause plasticity, but the assumption is that this is in some way related to therapeutic actions. It probably is, but the details are still unclear.

This interview has been edited and condensed for clarity and length.