The FDA Voted for 6 Peptides. Its Own Scientists Said No

22:22Dr. Jordan Feigenbaum:Now that doesn't cover all seven, because some of them are new enough that nobody's had time to study them. But the there's no money in it argument doesn't explain either group. There's enormous money in musculoskeletal pain, in sleep, in the beauty industry. And if one of these worked, somebody wants to own it. I don't have a clean answer for why nobody's running these trials, and I'm not going to invent one. But what I do know is that a missing trial is a missing trial. However it went missing, it doesn't tell you that the drug works. And there's a reason it might stay missing, which is the part of this vote that actually costs us something.

22:54Peter Lurie made the argument at the hearing, and I think he's mostly right. Any company that might fund a real efficacy trial now has a reason not to, because compounders can sell the compound without the study. So ask what happens to that hamstring trial's enrollment and funding when the drug becomes legally available for musculoskeletal injuries without anyone finishing the study. The tragedy of this vote is that it makes the study that could settle all of this much less likely to get run. Now, one of the biggest areas of pushback I see is that people are saying that they've used this stuff and it worked.

23:25The claim goes that thousands or even millions of people have gotten out of pain with BPC-157 or otherwise seen benefit with a different peptide on this list. And so we can't possibly ignore all of that and be honest at the same time. That's a good point. So let me be careful here with what I'm saying. When it comes to BPC-157, I'm not telling you that your injury healed on its own and that you imagine the rest. I'm telling you that the I took it and got better and it works are two different claims. and the gap between them is one of the oldest problems in medicine. And that gap is widest in pain and injury for a specific reason.

24:00Tendon and joint pain, they flare up and they settle down on their own. And people go looking for a fix when pain is usually at its worst, which is exactly the moment it has the most room to improve no matter what happens next. And on top of that, most people who start a peptide or any other treatment for pain change other things at the same time. How they're exercising, how they're sleeping, what they stop doing or start doing, and their expectations. Because you just committed to something serious that you believe might work. So the improvement is real, but what caused it is genuinely unknown.

24:32And this isn't just a hunch or my own pet theory, because medicine has run this experiment on itself using fake surgery. In 2002, the New England Journal of Medicine published a trial where patients with arthritic knee pain were randomized to one of three groups. Two groups got real arthroscopic surgery, which was the standard operation at the time, and hundreds of thousands of those were being done each year. The third group got a sham surgery, or a fake surgery. Patients were sedated, and the surgeon made real incisions in the skin. Then the surgeon splashed saline around so it sounded right, and sewed them back up without doing anything inside the joint.

25:08They followed all three groups for two years. The fake surgery patients improved just as much as the real ones. They had the same pain relief, and they had the same ability to walk and climb stairs as the group who got the real operation, all the way out to two years. The people who got the incisions in their skin and a theatrical, dramatic performance reported that their knees felt better. Now, I'm not saying that they were lying. Their pain genuinely dropped. It just had nothing to do with what the scalpel did inside the joint because the scalpel didn't do anything. And it wasn't just a fluke.

25:38You can go all the way back to 1959. There was a common operation being done for chest pain from heart disease, where a surgeon would tie off an artery, the internal mammary artery in the chest wall, to supposedly reroute the blood to the heart. Patients swore by it. In two small trials, randomized people to either the real operation or a sham, where the surgeon opened the chest and closed it without tying anything off. Both groups improved about the same. Now, I'm not comparing BPC-157 to surgery. I'm just making a point. If a fake knee operation can produce two years of measured pain relief, then your elbow feeling better after an injection tells you very little about whether the compound in that injection did anything.

26:17Dr. Jordan Feigenbaum:The improvement is real, it's just that the attribution is unreliable. It's kind of like the rooster in the sunrise. You know, is the rooster causing the sunrise? And that's why we run placebo-controlled trials, because human beings cannot tell the difference from the inside, including me, including every surgeon who was doing those knee operations in good faith. Back to the post I made on this vote the other day. A clinician commented about his experience in a patient with ulcerative colitis. This person was on dual biologic therapy and steroids and was headed for having part of his colon removed.

26:48Dr. Jordan Feigenbaum:He started BPC-157. A month later, he left the hospital off both biologics and off of the steroids, much to the surprise and likely celebration of his physicians. I can believe that happened. And if I'd have been in the room, I would have also been celebrating. But what makes it harder to interpret is the rest of the chart. That patient was on active treatment with biologics, and biologics have a delayed response window typically. Getting off steroids is the endpoint biologic therapy is aiming at, and ulcerative colitis remits and relapses on its own, which is why the disease is so hard to read from just a single case.

27:21And of every condition somebody could have written to me about, this was the one where the controlled test exists.

27:28Dr. Jordan Feigenbaum:Ulcerative colitis is the one indication where a real, randomized, placebo-controlled trial in humans of BPC-157 has been run. So we don't have to argue about it from anecdotes and we can go look at the trial itself. The human trial was performed by Renzi and colleagues in 2005. It was a multi-center, randomized, double-blind, placebo-controlled, phase 2 clinical trial of PL-14736, which is BPC-157 under PLEVA's old designation. 53 patients, a control arm, blinding, the whole kit and caboodle. It's the study people keep telling me will never exist. And they ran it 21 years ago. Now, I've said on this show before that BPC-157 has no published human randomized control trial data, which is true.

28:11We'll get to that. But suspend the apparent contradiction for just a moment. Because the obvious read of a randomized trial is that we finally have an answer. We don't. And the reasons are worth going through one at a time. First, this study exists only as a meeting abstract in the journal Gastroenterology. A conference abstract is a poster and a paragraph. It was never peer-reviewed the way a full paper is, and the complete study was never published. So 21 years later, the methods, the endpoint definition, the inclusion criteria, and the statistical approach are still not available for anyone to check.

28:43The FDA says exactly that in its own document. But here are the numbers, straight from the abstract as the FDA had received them. 53 subjects with mild to moderate ulcerative colitis randomized one-to-one to either BPC-157 enema or to a placebo 80 milligrams once daily for two weeks and 46 of those patients completed the study Three withdrawals in the treatment arm, two in the placebo, which the authors attribute mainly to the colitis itself progressing Now the primary endpoint was a change in a disease activity index described as a composite of clinical, laboratory, endoscopic, and pathology findings The abstract never defines exactly what all that is.

29:21So one concept before the numbers, because without it, the numbers don't really mean anything. When a study reports a result, it also reports a range around it. That range is a set of values that fit the data they actually collected. And if that range includes zero, then no effect at all is one of the answers that their data is consistent with. So the treated group's disease score dropped about three points, and their range stayed entirely below zero. So those patients got better, and nobody's disputing that. The placebo group dropped about a point and a half,

29:50Dr. Jordan Feigenbaum:and the difference between the two groups, which is the number that actually tests the drug, was about one and a half points, with a range from about minus five to plus one and a half. Now that range includes zero, so the comparison that actually tests the drug against placebo didn't clearly show a benefit. And I can't even tell you whether a point and a half on that scale would matter to a patient because nobody's published what that scale was. The authors called it a composite of clinical, laboratory, endoscopic, and pathohistological findings. and the FDA says that it was not clearly defined in the abstract.

30:1921 years later, nobody outside the company knows how the primary endpoint was even scored. For context, the endpoint that ulcerative colitis trials typically use now is clinical remission on the modified Mayo score and for a drug you'd take chronically, the FDA recommends at least a year of controlled treatment. This was two weeks on an index nobody can look up. This is where the proponents land, so let's name the move before it gets made. The change with the treatment group, on its own, cleared zero. People in the treatment arm got better. That's the number that gets quoted. But the within group improvement is not evidence of a drug effect because ulcerative colitis relapses and remits on its own.

30:58People got better on the placebo, and people get better on nothing. That's the whole reason you run a control arm, and against the control arm, dismissed. Now to be clear, this isn't a slam dunk that BPC doesn't do anything. 53 people is a small study, two weeks is short. the study was exploratory and an abstract is not a fair test of anyone's science and the FDA's conclusion is careful word for word. There is a lack of evidence to make a conclusion on effectiveness. That's different from saying it failed. So what does this trial establish about the vial in your gym bag? Nothing. Nothing at all.

31:31It's the wrong condition given by the wrong route

31:34Dr. Jordan Feigenbaum:in the wrong dosage. Start with the forms they wanted to sell these peptide companies. Oral capsules, subcutaneous injection, nasal spray, transdermal cream, rectal suppositories. While the FDA went looking for human studies of BPC-157 given by mouth, under the skin, up the nose, or through the skin in colitis patients, they didn't find anything at all. The closest match is the suppository and it still doesn't match. This trial used a liquid enema and those don't reach the same places. A suppository stays in the rectum, but an enema can reach much further up the colon. So the single human trial doesn't line up with any product that was nominated, including the one it's closest to.

32:12Now the rest of the human record isn't much better for BPC-157.

32:16Dr. Jordan Feigenbaum:17 people with knee pain from a retrospective chart review at one clinic, and also some of those people got a second different peptide at the same time. 12 women with interstitial cystitis, single arm study. Two healthy people given an IV infusion. No control group in any of these studies. And here's a thing you notice reading the FDA's reference list. All three of those are first authored by the same clinician, published in the same alternative medicine journal. So outside of the two pleva abstracts, the entire human record for this compound is one practitioner's uncontrolled case series. This podcast is brought to you by ButcherBox.

32:51Dr. Jordan Feigenbaum:One thing I try to stay consistent with is keeping enough protein in the house so I don't end up ordering takeout a few times a week. That sounds simple, but between work and everything else, the grocery store trip is usually the first thing that gets cut. ButcherBox fixed that for me. They deliver grass-fed beef, organic chicken, wild-caught seafood, and more, all sourced to the standards they are transparent about. Every box is customizable, so you pick the cuts and the proteins that fit how you actually cook and what you're actually going to eat. It's like having a butcher on call who already did the sourcing homework for you.

33:20Dr. Jordan Feigenbaum:And let me just tell you, the quality is legit. I've been grilling a lot lately. It's summer here in San Diego, and the meat has been excellent. It tastes great. It shows up right to my door exactly what I want and I don't have to think about it. I don't have to make a trip. That's the whole point. It's an easy button for your food. Now, whether you're grilling on the weekend, meal prepping for the week, or just trying to get a real dinner on the table on a Tuesday, you've already got something good in the freezer ready to go. Go to butcherbox.com slash barbell to get$20 off your first box, plus your choice of free ground beef for life or free chicken thighs or top sirloins in every box for a year with free shipping always.

33:55Dr. Jordan Feigenbaum:That's butcherbox.com slash barbell b-a-r-b-e-l-l be sure to use our link so they know that we sent you this podcast is brought to you by factor the last few months have been a real time crunch for me i've been finishing the book signal and between that training and everything else i've been doing cooking has been a real challenge and factor has been the thing keeping my nutrition from going by the wayside having meals ready to go that i can heat up in two minutes and actually feel good about eating has made a real difference on the days where i just don't have those 30 minutes to stand in the kitchen, especially when it's late at night.

34:26Factor meals are chef crafted and dietician designed. They're ready to eat with no prep and no cleanup. They have over 100 menu items rotating every week, including new options like salmon burgers, shredded pork, and collard greens, plus add-ons like protein shakes and pumpkin cheesecake. Over 175 banned ingredients, so the

34:42Dr. Jordan Feigenbaum:stuff that you probably don't want in your food isn't going to be there either. I use Factor and I bet you could benefit from it too, especially if your schedule gets tight. Food is usually the first thing that slips. This keeps that from happening. Head over to factormeals.com slash bbm50off and use code bbm50off to get 50 % off and one free breakfast item per box for a year while supplies last until October 31st, 2026. That's code bbm50off at factormeals.com. Again, bbm50off at factormeals.com. See the website for more details. This podcast is brought to you by Quince. End of summer is when I start thinking about what I actually want in my closet for the next few months.

35:19Dr. Jordan Feigenbaum:You know, not a full overhaul, just swapping out the stuff that's worn out and filling in some gaps. And that's what keeps bringing me back to Quince. They do everyday staples really well. Their organic cotton tees are soft and hold up wash after wash. Their cashmere sweaters are legitimately good for the price point, and they transition straight into fall. If you need pants, their premium denim and tailored chinos start at just$60, and the fit and fabric are on par with brands charging three or four times that. Ask me how I know. Now, everything at Quince is priced 50 to 80 % less than similar brands because they work directly with the factories and skip the middlemen.

35:51You are paying for the materials and the construction, not a logo or marketing campaign. Now, one of my go-tos this summer has been their organic cotton mesh stitch sweater polos. I have it almost every color. It's one of those pieces where you look put together, but you're actually comfortable, which is why these are going to stay in my rotation for the fall. And Quince goes way beyond just clothing. They've got premium bedding, bath essentials, cookware, travel gear. It's one of those brands where you buy one thing and end up recommending it to everyone. Both Lee and I have their sheets and they're great.

36:18So upgrade your every day. Download the Quince app for app exclusive offers or go to quince.com slash BBM. Get free shipping on your order and 365 day returns.

36:27Dr. Jordan Feigenbaum:Now available in Canada and the UK too. That's Q-U-I-N-C-E dot com slash BBM. This podcast is brought to you by FIGS. If you're in healthcare or you're in training for it, back to school season means you're about to spend a lot of time in scrubs. And figs just drop new colors and back-to-school styles, and right now, everything on their site is 20 % off. Now, during my medical training, I found out that the standard hospital scrubs were not designed for anyone squatting, period. Bending over to examine a patient was like a truss fall, just hoping that the seam would hold. That's how I ended up trying figs.

36:57Dr. Jordan Feigenbaum:They're cut with actual stretch, they fit like real clothes, and they survived both the job and the squat program. They're also antimicrobial, and they hold up to the laundry abuse that this job requires. They also have footwear, compression socks, and outerwear. And 20 % off site-wide is the best deal they run. So if you've been meaning to replace your scrubs or stock up for a new rotation, this is the time. Go to wearfigs.com, get 20 % off everything through August 11th. No codes needed. That's wearfigs.com. Hey, it's Jordan here. This is an ad for a subscription that removes ads, including this one, which is a good way that you never have to hear this again.

37:32Dr. Jordan Feigenbaum:And you can do that by signing up for Barbell Medicine Plus, which is now 25 % off annually through August 13th with the code PLUSANNUAL. Now the ad-free listening feed means this exact minute of your life, you'll never have to do this again. You also get member-only AMAs with Austin and I, plus bonus episodes, early access to our content articles, and more. Plus you get discounts on training templates, 10 % off, 15 % off consults, 25 % off courses and seminars, plus member-only sales. If you're gonna buy any of that stuff anyway, you now have solved the ad problem by shopping, which is not how I thought this was going to play out.

38:07Dr. Jordan Feigenbaum:But mostly it's how people support the show and that support is the reason it keeps going. So sign up for Barbell Medicine Plus, use code PLUSANNUAL through August 13th. This has been an ad about itself. Now back to the show. So that's the evidence and it's not great. The other argument I get whenever I call these drugs is that they're not because BPC-157 is a fragment of a protein that your stomach makes, so it's natural. And a natural molecule isn't really a drug. Well, there are many drugs that aren't natural. Insulin is natural. Testosterone is natural. Thyroid hormone is natural. All three are drugs.

38:41Whether a molecule occurs in the human body has nothing to do with whether it's a drug and nothing to do with whether it's safe. Botulinum toxin occurs in nature too, and a few grams of it will kill you. And a small molecule made of ordinary parts can also be extremely powerful. Levothyroxine, which millions of people take every day, is a modified amino acid with iodine bolted onto it. It's simple, it's tiny, and it's boring on paper, but if you get the wrong dose, it can put you into atrial fibrillation. Size and simplicity just don't tell you anything about potency. And for BPC-157 specifically, the premise is worse because nobody's actually established it.

39:18BPC stands for body protection compound, and the claim goes that it's a fragment of a protein in human gastric juice in your stomach, but that parent protein has never been isolated or characterized in a human. The original description was a crude extraction in a patient, not a sequenced reproducible protein that anyone can point to. And the 15 amino acid sequence in the vial isn't known to occur anywhere in nature either. Nobody has even identified the receptor for it. Nobody knows what it binds to or what it does once it's inside of you. And while I'm here, the word peptide is doing more work in this conversation than it really can support.

39:53A peptide is a short chain of amino acids, and in 2020, the FDA drew a legal line at 40. 40 or fewer, it's a peptide. 41 or more, it's a protein, regulated as a biologic. That number is a regulatory convention that somebody picked. Watch what it does. Semaglutide, what you know as Ozempic and Wegovi, is 31 amino acids long, so it's a peptide. Insulin is 51 amino acids long, so insulin is a protein by that rule, even though every clinician I know calls insulin a peptide hormone and has done so since training. Thymocin beta-4, the parent of TB500, is 43 amino acids long, so that's a protein 2, and MK677 shows up in the same online stores as everything else here.

40:33And that one's a small molecule that Merck patented in 1995 with no amino acid chain in it anywhere. All of this is to say that none of this is new science. BPC-157 was first described in the early 90s, so why is it becoming a thing now? I think the answer is GLP1s. semaglutide and terzepatide did two things at once they made injecting yourself into a normal weekly habit and on top of that they worked visibly on people's co-workers and relatives so a reasonable question formed in a lot of people's minds if that peptide works that well what else is in the drawer there's a second thing that changed it's about how you get it rather than what it is if you wanted anabolic steroids 20 years ago you had to know somebody there was a guy, there was a conversation you had to have in person with that person, and that friction did a lot because it made you decide out loud that you were somebody who was going to do this.

41:24You knew that you were buying a drug. Nobody was confused about the category. But none of that exists here. You order from a site with a chat widget and a discount code, and it arrives in a padded envelope in about two days. And that difference isn't because one of these is better studied. Testosterone is an approved drug with decades of trials behind it and a controlled substance schedule. BPC-157 has no approval anywhere in the world and two conference abstracts. The one with the evidence is the one that's hard to get. So what went away is the moment where you had to admit to yourself what you were doing.

41:55And the vocabulary finishes the job. Peptide sounds like something in a protein shake. Research chemicals sounds like a lab, not a pharmacy. Short chain of amino acids sounds like nutrition. None of these phrases sound like the word drug, which is why I repeatedly call these things drugs. So let's just define it, because the definition isn't mine and it isn't a judgment call. Federal law says a drug is anything intended to treat, cure, mitigate, or prevent disease, or to affect the structure or any function of the body. That's it. That's the whole test. Read the second half again. Affect the structure or any function of the body.

42:30If you're injecting something to make a tendon heal faster, you're using it to affect the structure and function of your body, which means you're using a drug. It doesn't matter what it's made of, where it came from, or what the label says. Now the FDA's own document calls these bulk drug substances and active pharmaceutical ingredients throughout. Not supplements, drugs. Which brings me to the strongest argument I've heard yet, and it's the one the yes votes leaned on the hardest. People are buying this stuff anyway, from vendors that nobody regulates, so let them get it from a pharmacy instead.

43:01Half of that position, I think, is reasonable. The members who voted yes argued that patients are already buying this from unregulated vendors and that a compounding pharmacy testing for sterility, potency, and purity beats a guy shipping vials from overseas. The chief medical officer of one telehealth company testified that a colleague sent him a photo of a bodega in Queens selling peptides. The sourcing problem is real. I've written about it. I've talked about it on this very podcast. When researchers chemically analyzed 44 products sold as research chemicals through these channels, only about 4 in 10 contained what the label said.

43:36Some had no active compound at all. Some contained a different drug that wasn't on the label anywhere. And the FDA's own document backs up the danger of the current channel. The certificates of analysis in the nomination packages mostly reported purity and little else. No identified impurity profile, no microbial testing, no endotoxin testing. That's for a product proposed as an injection. And no endotoxin testing in an injectable is a real risk for sepsis. And there's a naming problem underneath this too. BPC-157 is a common name, not a United States adopted name. That's the formal system that exists so the pharmacist hands you what your doctor actually ordered.

44:14Aspirin has one, ibuprofen has one, BPC-157 doesn't. Which means the name doesn't pin down what's in the vial. And there are at least two different substances sold as BPC-157.

44:25Dr. Jordan Feigenbaum:There's a freebase and an acetate salt. Different chemical structures, different registry numbers, and the FDA treats them as two separate drugs with potentially different safety and efficacy profiles. But both are sold under the same three characters and a number. This isn't hypothetical. Both companies that nominated this to the FDA filed paperwork so inconsistent that the agency couldn't tell which one they were asking for. Their application said BPC-157. Their certificates of analysis were for the acetate. And then the FDA ended up evaluating both because it couldn't determine what had even been nominated.

44:57If the people whose business this is can't specify which molecule they mean on a federal filing, your pharmacist probably can't either. So yes, the current channel is dangerous, and I'm not going to argue with anybody about that. But I'm not so sure that compounding solves it. And here's why. Big compounding operations, the 503B facilities that make drugs in batches, report to the FDA twice a year what they've made. From 2017 through mid-2025, none of them reported a single dose of BPC-157. But that doesn't mean nobody was making it. Smaller pharmacies don't report the same way, and the FDA's own search turned up compounding pharmacy websites selling BPC-157, plus a concierge service selling injection kits for home use.

45:38Two of the three adverse event reports in the FDA's file name a specific compounding pharmacy. So the picture isn't that this was absent from pharmacies, it's that whatever was happening wasn't reported and wasn't legal under the statute,

45:50Dr. Jordan Feigenbaum:because BPC-157 doesn't meet any of the three conditions that let a pharmacy compound with it. The case for putting it on the bulks list is that people are buying it from unregulated vendors, so let's move them to pharmacies. But some pharmacies were already selling it, outside the rules. So, people are buying it anyway includes people buying it from pharmacies that weren't supposed to be selling it in the first place. Which means the choice isn't between a gray market and a regulated one. There is no regulated one. There's a gray market that runs through websites, and a gray market that runs through some pharmacies.

46:20And neither one of them is operating under a rule that says this substance is okay to compound. Listing it doesn't move anybody from unregulated to regulated. It makes one half of the gray market legal, and it leaves the other half exactly where it is. And that's the part that doesn't fit the harm reduction label. Harm reduction means meeting demand that already exists with something safer.

46:40Dr. Jordan Feigenbaum:This makes the same product legal in the same places it was already being sold, at scale, with advertising. Safer would mean knowing what the drug does, which unfortunately isn't on the table. And if you want to know what this vote is actually about, watch the stock. The day of the BPC-157 vote, HIMSS and HERS jumped more than 10 % intraday before closing up about three. The next day, the panel turned down one of the seven peptides and the stock fell. Leerink Partners, an investment bank, called the vote a positive catalyst for HIMSS and estimated that the peptide business could be worth about$2.2 billion a year across telehealth, with HIMSS taking roughly a fifth of it.

47:17That's about$440 million a year for a company selling compounds nobody has shown to work. And Lear Inc. isn't cheerleading. They're neutral on the stock, and their price target is below where it's trading. So even the analysts who don't think you should buy the company think there's real money in this. And the part that tells you this wasn't a surprise to anybody. HIMSS bought a California peptide manufacturing facility in early 2025, 18 months before this very vote. You don't reduce harm by opening a bigger store next to the sketchy one. Harm reduction works when the number of people using something is fixed, and you're just changing where they get it.

47:50Clean needles don't create new injection drug users, but this is a wellness product marketed to healthy people who feel fine by a company that advertises on television. The number of people using it is not fixed. It's the thing that's going to be grown. So you're not moving a set group of people to a safer source. You're taking a substance a small number of committed people we're seeking out and putting it in front of everyone who watches TV. And there's one move that would actually reduce harm. Run the trial so we know whether the thing works and at what dose. That gets less likely, though, the moment you can sell it without that trial being done.

48:24The people who voted yes leaned on the idea that a real pharmacy and a real prescription put a responsible doctor in the loop. But picture what that doctor is writing for. An unstudied injectable, for use it was never tested on, at a dose nobody has established. Finally, there's one last major objection to everything I've said so far. It sounds like I'm demanding trials for their own sake. It's more paperwork and more delay while people wait for something that might actually help them. So let me tell you what a missing trial actually costs by telling you three stories of how this has gone before.

48:53Let's start with fen-fen, because the structure is identical to what's in front of us right now. Fen-fen was a weight loss drug. Two appetite suppressants taken together, fenfluramine and fentermine. Now both had been approved individually, and for a long time. Fentermine in 1959, fenfluramine in 1973, and both were cleared for short-term use only, a few weeks at a time, as a supplement to diet and exercise. The combination was never approved. It was never even submitted. No study demonstrating the safety or effectiveness of the two together was ever sent to the FDA, and the agency never signed off on using them long-term.

49:26There was nothing to rush through because nobody asked. What we had instead was one paper, a 1992 study out of the University of Rochester with 121 patients from a single center. They were followed for four years and fewer than half of them finished. About 40 % were still in it at the end, and many of the ones who did finish regained the weight later. That study is the entire basis for what happened next. Doctors started writing for it off-label, which they're allowed to do. And it worked. People lost weight. By 1996, prescriptions for these drugs ran past 18 million. And it isn't that nobody raised a hand.

49:59Reports linking fenfluramine to pulmonary hypertension had been showing up in the literature through the 80s and 90s. There just wasn't good long-term data on either drug, and there were none at all on the two drugs together. Then, in 1997, a cardiologist at the Mayo Clinic published 24 cases. 24 women on fen-fen, no history of cardiac disease, with damaged heart valves. The CDC eventually confirmed 113 cases. 111 of them were women. Meeting age was 44. Meeting time on the drug was only 9 months. Better than a quarter of them needed a valve replacement. The drug came off the market that September.

50:35So three things about that story matter here. The first is that valve damage is silent. You can't really feel it, and every one of those women felt fine until they weren't. There was no symptom to notice and no test that anybody was running. The thing hurting them was invisible from the inside, which is exactly what I mean when I can't tell you what to watch out for when it comes to these peptides. The second is who was wrong. 18 million prescriptions were written by physicians, real doctors with real training, who believed it worked because their patients came back lighter and happier. Nobody in that story was stupid, not the patients, not the doctors.

51:08Everybody was operating on exactly the kind of evidence people are operating on now, which is that it seems to be working and nobody's reporting problems. And then there's a third thing, which is the part that should sound familiar. While fenfen was booming off-label, a company was trying to get a related drug approved, dexfenfliramine, sold as Redux, same family, submitted as its own product. In 1995, an FDA advisory committee voted 5-3 against recommending it over concerns about pulmonary hypertension and possible neurotoxicity. So the FDA convened a second advisory committee. That one voted 6-5 in favor, approved in April of 1996.

51:45Nobody was quiet about it beforehand either. Neuroscientists had written the agency twice about neurotoxicity in animal studies. France had already restricted the drug because of the pulmonary hypertension cases. All of it got buried under the fact that this was the first new obesity drug in 20 years and the country wanted it. Two and a half million prescriptions went out by June of 1996. off the market by September of 1997, 16 months. A panel said no, a second panel said yes by one vote. The drug went out and it hurt people. The second one I want to talk about is Vioxx. Merck got Vioxx approved in 1999 and over 80 million people took it.

52:22The FDA reviewer who evaluated the application had already noted more heart attacks and strokes on the drug than on placebo. In 2000, the Vigor trial found five times the heart attack risk against naproxen. Merck's explanation was that naproxen protects the heart, not that Vioxx damages it. When the trial was published in the New England Journal that November, three heart attacks in the Vioxx group weren't in that paper. The FDA had the full data on its own website, though. In February 2001, an FDA advisory committee looked at it and said there was clear evidence of increased cardiovascular risk.

52:55They recommended that the label say so. The label was changed in April of 2002, and the reason for the delay is the part that should stop you. The FDA had to negotiate wording with Merck because the agency didn't have the authority to just require it. Their deputy director for new drugs later told the Senate it would have been helpful to have that power. Meanwhile, Merck put out a press release titled Merck Confirms Favorable Cardiovascular Safety Profile of Vioxx. The FDA wrote back calling the claim simply incomprehensible. It took until 2004 when a placebo-controlled trial was stopped early for the drug to come off.

53:27David Graham, an FDA scientist, testified to the Senate that Vioxx caused somewhere between 88 ,000 and 140 ,000 excess cases of serious coronary heart disease. He says he was pressured to soften his findings. Here's what Vioxx should actually teach you, and it isn't that the FDA can't be trusted. The FDA got it right. Their reviewer flagged the heart attacks before approval. Their advisory committee said there was clear evidence of harm in 2001. Their own scientists ran the study that quantified it. every piece of the agency's scientific judgment was correct and roughly on time. But what they couldn't do was act on it, because they had to negotiate the wording with the company selling the drug.

54:04Their deputy director had to go ask the Senate for the authority to require a warning. So the problem isn't that the FDA doesn't know anything. The problem is that the industry sits on both sides of the table, and the agency's scientific judgment is the part that's optional. Which is exactly what happened last week. FDA's career scientists reviewed all seven of these substances and recommended against every single one. The panel voted yes to six. The agency's own scientists lost again, in the agency's own building, to a committee stocked with people who sell the product. So if you don't trust the FDA because you think that they're captured, you and I are looking at the same evidence.

54:37I just don't think the answer is buying an unstudied drug from a website. Okay, the third one, and it's the closest to home because it's about compounding specifically. In the fall of 2012, a compounding pharmacy in Massachusetts called the New England Compounding Center shipped contaminated steroid injections across 20 states. 64 people died. More than 750 were infected with fungal meningitis and spinal and joint infections, and for some of them, the treatment never ended. Now, that was a licensed compounding pharmacy, not a gray market vendor, not a guy shipping vials from overseas, a real pharmacy making an injectable, and it killed 64 people.

55:12And I want to be exact about what that does and doesn't say, because I'm not standing here telling you that compounding pharmacies are dangerous or kill people. Most of them are careful, and most of what they make is fine. What it says is that there are two separate ways this can go wrong, and the pharmacy only addresses one of them. The first is that the product isn't what it's supposed to be. It's contaminated, the wrong potency, wrong molecule. That's the NECC failure, and it's the one a good pharmacy is built to prevent, though NECC is the proof that they don't always. The second is that the product is made perfectly, and nobody knows what it does.

55:42A clean vial of a compound with no monograph, no established dose, and no efficacy data is still a clean vial of an unknown drug. But get it from the pharmacy instead only answers the first problem. It says nothing about the second, and the second is what this whole episode is about. Congress responded to NECC in 2013 with the Drug Quality and Security Act, and that law built the compounding framework we have now. Federal oversight of large compounding pharmacies, the outsourcing facility category, the 503A structure everybody was arguing about in that hearing room last week exists in its current form because 64 people died.

56:16And every bit of it is about manufacturing. Sterility, testing, oversight, reporting, necessary work, and it fixed a real problem. But none of it asks whether the drug actually works. That question wasn't what went wrong at NECC, so the law that came out of it doesn't address it. Which means when somebody tells you a pharmacy makes this safe, they're describing a system built to catch contamination. Contamination isn't the thing I'm looking for answers about for BPC-157 and these other peptide drugs. I want to know what they do. While it is true that there aren't good studies showing that these peptides work, people often turn that around and say that there aren't good studies showing that they're harmful either.

56:51To be clear, we don't have safety data is not the same claim as it's probably fine. So let's try to sort it out when it comes to BPC-157. First, the things we actually know. The animal work that was submitted includes a 28-day repeat dose study, and in it, the compound moved clotting times in both directions and pushed up liver enzymes, blood sugar, and triglycerides. On the human side, the FDA's adverse event database has a handful of BPC-157 reports. One is logged as life-threatening, and the reason listed, quoting the entry, is product sterility lacking. Another is a person who stacked BPC-157 with TB-500 and developed spreading darkening of the skin and gums, and the darkening came back when they dosed again.

57:32That return of symptoms on a secondary exposure is the closest thing in the whole file to the compound itself doing anything. Second, the things we know we don't know. We know that there's no long-term human safety data. We know that there's no carcinogenicity studies. The proposed mechanism people cite is angioneogenesis, growing new blood vessels after an injury. And that's a growth signal. The long-term studies that would tell you what a growth signal does over years were never run. But to be fair, BPC-157 did test negative for mutation in the standard genetic damage assays, and nobody has evidence that it causes cancer.

58:05But the study designed to look at this has never been done. And then there's a dose-response problem. In several of those animal studies, doses a thousandfold apart produced about the same size effect. 10 nanograms per kilogram and 10 micrograms per kilogram each produced comparable results. The FDA's phrasing is that a dose-response relationship could not be established and that the molecular targets have not been identified. In simple terms, if the response is the same across a thousandfold dose range, it's hard to know whether you're watching the drug or just looking at noise. And I should say, a large share of that preclinical literature traces back to one research group in Croatia.

58:39Read the agency's reference list yourself, and you'll see the same senior author over and over and over again. Independent replication is thin to nonexistent. But that doesn't make the work wrong. It just means that one lab's results haven't been checked by another lab. Third, the unknown unknowns. When physicians put a patient on a real drug, we can usually tell them what to monitor. For example, I can name the blood test and the symptom for the medications I'd consider using. For BPC-157, I can't because nobody has established what shows up when something goes wrong over the years. Nobody has even identified the receptor that it uses.

59:12You can't predict the side effects of a molecule when you can't even name its target, and you can't screen for a problem when you don't know what the problem would even look like. So when somebody says, look, nobody's getting hurt, for BPC-157 specifically, they're close to right, but I wonder how much of that is because you usually only find harm when somebody's looking. And for this compound, nobody is. There's no registry, there's no doctor filing a report, when a gym buddy's shoulder injection goes sideways. The absence of case reports in a population nobody is following could easily be a measurement problem.

59:41And when you widen the search out to the whole class of peptide drugs, the picture changes. In April, ECRI and the Institute for Safe Medication Practices published a joint white paper on exactly these compounds. Their finding on BPC-157 is blunt. 35 of 36 published studies were done in animals, with no published Phase I human safety data and no controlled human efficacy trials. Across the class, they document real signals. A phase 2 trial of CJC-1295 was halted after a participant died of a cardiac event, and no one has picked it up since. Melanotan 2 has case reports on melanoma, including mucosal melanoma.

1:00:18There are case reports of kidney dysfunction and rhabdomyolysis, and at least one death following intravenous administration. Then the sourcing data. Earlier I told you about 44 products where only 4 in 10 matched their label, and that was about labeling accuracy. This is a different measurement, and it's worse. Analytic testing of 10 gray market peptide products found purity ranging from 5 % to 75%, not 5 % off, 5 % pure, and arsenic and lead contamination running up to 10 times the international toxicity limit for an injectable drug. Now I'll note who's saying this because it connects back to something from earlier.

1:00:51The Institute for Safe Medication Practices is the organization whose former executive vice president chaired the 2022 version of the FDA's committee. The people who used to be in that room are now writing white papers from outside of it. Now to be clear, the ECRI and the ISMP support the advisory committee process. They think it's the right mechanism. Their objection is to what happened in April, when the classification changed without any new evidence being presented. Everything I've said so far has been about BPC-157, because it's the one most of you are asking about. But six other substances went in front of that committee, and the reason to run through them is that the answer is the same every single time.

1:01:27TB500, sold for muscle and tendon repair, and the one most often stacked with BPC-157. A 7-amino acid fragment of thymusin beta-4. It was nominated for wound healing, but the FDA found no adequate human effectiveness or safety evidence for that use. The same characterization problems as BPC-157 and the same immunogenicity concern for an injectable. It's also on the wild-a-banned list, but it passed 8-6. KPV, a 3-amino acid fragment, lysine, proline, valine, snipped off the tail of a hormone called alpha-MSH. It's sold for inflammation and gut conditions, and lately it's jumped into skincare. The FDA's finding was blunt.

1:02:04The human clinical evidence was absent. Not modest. Absent. It passed 8-6. MOTC, a 16-amino acid peptide your mitochondria actually encode, it's marketed as an exercise memetic, which is the exercise in a vial pitch. Human clinical evidence, again, absent. Also on the WADA banned list. And this is the one where the panel kind of wobbled. 7-5 with two abstentions. So the compound with a purely performance pitch and zero human data is the one that made even this panel flinch. CMAX, a 7-amino acid fragment of ACTH with a small tail bolted onto it to slow its breakdown. It's marketed for focus and cognition and popular in the nootropics world.

1:02:43CMAX is a registered drug in Russia sold as nasal drops, so its supporters argue that there's more data on it than others. But the indications nominated here were stroke recovery, migraine, and trigeminal neuralgia. The FDA found the data insufficient and flagged a bleeding risk from the peptide's anticoagulant properties. It passed 8 to 5 with one abstention. Epitilon, a 4-amino acid peptide originally built off a pineal gland extract, sold as anti-aging on the claim that it activates telomerase, the enzyme that maintains the caps on the ends of your chromosomes. But telomerase activation is not obviously a good thing.

1:03:15Switching it on is one of the things cancer cells do to become immortal, and there are no carcinogenicity studies. So the marketing hook and the safety concern are the same mechanism. It passed 7-4 with one abstention. DSIP is a 9-amino acid peptide first pulled out of rabbit blood in the 1970s, named for the deep sleep brainwaves it seemed to bring on, its pitch for sleep and for opioid withdrawal. This is the only one of the seven that the panel turned down. Six in favor, seven against, one abstention. The FDA had said the evidence didn't support insomnia, narcolepsy, or opioid withdrawal as indications.

1:03:52And that's the tell I keep coming back to. Six of these got a yes. The one that got a no is the one where being wrong looks like an overdose. Everywhere else, the panel voted yes. Different molecules, different pitches, but one identical FDA verdict underneath all of them. Not well characterized. Not enough evidence of efficacy. Not enough safety data. The panel read that verdict seven times and said yes to six peptides anyway. So that's the whole record. Now, what would I actually tell you if you were sitting across from me? Because the two situations here aren't the same. If what you're looking at is a research chemical, a vial labeled not for human consumption from an online vendor, then my answer is easy and it's a no.

1:04:30That's the channel where most products were mislabeled. Some had no active compound and some were spiked with a drug that wasn't listed. You have no idea what's in it and you'd be injecting it anyway. The compounding pharmacy version is a harder conversation. If these get listed and you can get a sterility tested, potency verified product from a real pharmacy on a prescription, then whether to use it becomes a genuine individual decision. I'd want you making that decision with a clinician who isn't the one selling it to you, which in this corner of medicine is not always a given. I'm not making that decision for you.

1:04:59What makes the decision hard isn't me deflecting though. For most drugs, you weigh the known risks against the known benefits. Here, both sides of the scale are mostly blank or at least foggy. The benefit for your specific use case has never been shown. The risk side is worse than unknown because we don't even know what to look out for. So you're choosing something that's kind of blurry on both sides. So if you decide to use, you should at least know that's what you're doing. There's a version of this I think is worth discussing directly. The it's my body, my choice, and a doctor telling me no is a doctor restricting my choice.

1:05:32Call it self-determination. The part of that I'll concede outright is the clinical dilemma. A patient tells you they're taking this regardless. You can fire them from your practice, or you can keep seeing them and monitor what you can. Firing them accomplishes nothing except making sure nobody with training is following them, which is bad. So you keep seeing them and do your best. That's the whole job. Where the argument breaks is that self-determination in medicine has a name already, and it's called informed consent. The word doing the work is informed. You have a well-established right to refuse anything.

1:06:02But what you don't have is a right to a decision made under information nobody's generated yet. When the risk side is blank and the benefit side is blank, there isn't a choice being protected. There's a coin flip being dressed up as autonomy, and the courts have been clearer about this than people think. Abigail Alliance was named after Abigail Burroughs, a young woman with head and neck cancer whose oncologist thought an experimental drug might help her. She couldn't get it, and she died in 2001, unfortunately. Her father built an organization and sued the FDA, arguing that terminally ill patients with no remaining options have a constitutional right to buy drugs that have cleared basic safety testing.

1:06:38The other case goes back further. It was about Laetrile, an apricot pit extract that was sold as a cancer cure in the 1970s. Terminally ill patients argued that the FDA's safety and effectiveness rules shouldn't apply to them, because what does safe and effective even mean when you're dying? And in 1979, in a case called United States v. Rutherford, the Supreme Court ruled against them unanimously. The requirements still apply. Right to try is real, but it's a statute from 2018, not a constitutional holding. It covers drugs that have cleared Phase 1 and are under active investigation, and it requires the company to agree to provide it.

1:07:12Nobody can force a company to hand anything over. So the picture people have of the law is kind of backwards. Dying patients with nothing left have gone to federal court asking for access to drugs that were further along than these, and they lost. There's no case that gets a healthy person a compounded sponsor abandoned 21 years ago, though. So I'm not the obstacle to your actual informed consent. The missing trial is. Now, some of you aren't injecting anything. Some of you are swallowing BPC-157 capsules for your gut. It's the same evidence gap, and then some. An oral capsule has to survive the exact system designed to digest it.

1:07:44It probably does nothing since getting a peptide to work orally requires a lot of sophisticated engineering, all of which is usually proprietary and very costly. If you compete in a tested sport, this isn't a judgment call. BPC-157, TB-500, and MOTC are all on the WADA prohibited list, banned in and out of competition. Pretty easy decision not to use if you're following the rules. And if the thing you actually came for is a tendon that won't settle, there's an intervention with real human evidence behind it. It's loading the tendon. Progressive, heavy, slow resistance training works. That's the thing shown in randomized trials to cut pain and remodel the tissue across the Achilles, the patellar tendon, and the elbow.

1:08:22And this is the part that should end the peptide pitch for tendon pain.

1:08:25Dr. Jordan Feigenbaum:The benefit people claim for BPC-157 is more blood flow to the area. Loading already does that. Load a tendon and blood flow through it jumps 2-3 times. And on top of that, the tendon lays down and reorganizes new collagen, which no injection does. Meanwhile, the cortisone shot and the PRP injection that clinics sell for the same problem don't beat a saline placebo once you're doing the loading. And cortisone comes out worse at a year anyway. The catch is that loading is slow. Tendons adapt over weeks and months, not days, and there's no version of this that's fast. Which is a real cost, and I understand exactly why an injection sounds better.

1:09:00You're in pain now, the fix is months out, and somebody's offering you a vial that promises it to be done in weeks. If that worked, it would be the obvious choice, and I'd tell you to take it. It would also be worth billions. A shot that reliably heals tendons faster than loading does would be one of the most valuable drugs in orthopedics, and every company in the space knows it. That's the part that should give you pause. Nobody has shown it, and the incentive to show it has been sitting there the whole time. Meanwhile, loading does the thing the injection is supposed to do, and it does it right now.

1:09:29More blood flow to the tissue, plus new collagen laid down in the right direction. There's no shot in the world that does the second part. So that's what I'd tell you. Now let me tell you what would change my mind, because if I can't answer that question, I'm just the mirror image of the people selling this stuff. There's a phase two trial in acute hamstring strain recruiting right now. Say that that trial or one like it gives the compound the way that people actually take it at the dose they typically use in the injury they actually have against a real placebo and say it shows a pretty good signal that it works and that it's safe.

1:09:58Then I change my answer and I say so right here on this show. That's all it would take. A well done study of the thing in the people who use it against placebo. It's the same study I've been asking about all episode, and it's never been done. So I think that the committee got this one wrong. Seven changes, and they got it wrong on six of them. Take BPC-157. What they reviewed was a condition almost nobody uses it for, on a single conference abstract that didn't beat placebo. For a substance the agency's own scientists say isn't well characterized. The tending question, the one that sent everyone looking in the first place, was never evaluated at all because there were no human studies to evaluate.

1:10:35And that's the good one. That's the substance with the most human data on the docket of these seven. Read the FDA's verdict for the other six and it's the same sentence with the name swapped. Not well characterized. Not enough evidence of efficacy. Not enough safety data. Seven substances, seven reviews, one conclusion written by the agency's own scientists. The panel read it seven times and voted yes to six. Improving access isn't evidence and this vote didn't generate any. The trial everyone tells me could never happen has already happened twice and neither time did the answer come back affirmative.

1:11:07And the study that would actually settle this is recruiting right now. Last week made it less likely to finish. And this isn't over. There's a second meeting expected at the end of February in 2027 on five more compounds, including melanotan 2, which is the one with published melanoma case reports attached to it. Whatever standard got applied last week is the same standard those five are gonna get judged against. Because if people are buying it anyway is enough to open a legal supply line, then tell me what that doesn't cover. It works for these peptides. It works for the next five. It works for anything with the website and enough customers, which is how you know it isn't an argument.

1:11:41I'm Dr. Jordan Blagenbaum. Thanks for listening.