Call it the Wolverine stack. That’s the name one of Peter Attia’s patients used to describe the peptide regimen a friend had been taking — a combination of injectable compounds promising accelerated healing, body recomposition, and something in the vicinity of biological restoration. The before-and-after photos were, by any honest measure, striking. More muscle. Less fat. A visible transformation over a matter of months.
Attia’s response, when the patient showed him the photos, was not to dispute what he was seeing. “This looks amazing,” he said. “What else was he doing?”
The friend had also started exercising. He was taking tirzepatide. His diet had changed as a result of the tirzepatide. He was on testosterone.
The Wolverine stack was one variable among six. It got all the credit.
This is, in miniature, the problem with how peptides are currently being sold, used, and understood by a large and growing number of people. Not that the compounds are fraudulent, necessarily, or that everyone selling them is acting in bad faith. The problem is more subtle and, in some ways, more difficult to address: a class of molecules that includes some of the most consequential drugs ever developed has been rhetorically collapsed into a single category — peptides — and that category has been assigned properties that belong only to its best members, if it belongs to any of them at all. The word has done an enormous amount of work that the science has not.
Insulin is a peptide. So are the GLP-1 agonists that have reshaped the treatment of obesity and type 2 diabetes over the past decade. So is BPC-157, a compound with thirty years of animal studies and no published human randomized controlled trials, sold by wellness clinics as a regenerative therapy for tendon injuries, gut disease, neurological conditions, and a list of other indications that has grown steadily without any of the individual items ever being rigorously established. The word “peptide” connects those molecules the way the word “drug” connects aspirin to fentanyl.
What follows is a framework for cutting through that conflation — five questions that, applied systematically to any peptide or drug, tend to reveal what the evidence actually supports and what the marketing has added on its own.
What the Word Actually Means
A peptide is a short chain of amino acids. That is the entirety of the chemical definition. It says nothing about what the molecule does, how it behaves in the human body, whether it is safe, or whether it has been tested in any meaningful way. It is a structural description, not a quality signal.
The naturalistic appeal of the word is almost entirely manufactured. Most peptides used in wellness contexts are synthetic molecules, deliberately modified versions of natural sequences engineered to bind receptors more tightly, resist degradation more effectively, or hit targets the original compound never reached. They are drugs by any functional definition. The fact that the body also produces amino acid chains does not change the pharmacology of injecting a synthetic, modified version at supraphysiological doses.
The peptide drug pipeline is legitimate and, in specific areas, genuinely exciting. Muttenthaler and colleagues, writing in Nature Reviews Drug Discovery in 2021, documented approximately 80 approved peptide therapeutics at the time of publication, with roughly 150 more in clinical development and several hundred in preclinical stages. Metabolic disease, oncology, infectious disease, and diagnostics are the areas where the field’s specificity advantage — peptides can be designed to hit a target with a precision that small molecules often cannot match — translates most reliably into clinical benefit. The blood-brain barrier remains a persistent obstacle for CNS applications. Broad claims about tissue repair and biological optimization are, for structural reasons, among the hardest things this class of molecules can credibly promise.
The pipeline supports real optimism. It does not support the conclusion that any given peptide in a wellness catalog inherits that optimism by association.
Five Questions, in Order
The framework worth applying to any peptide — or, for that matter, to any drug — runs through five questions. They are not original. They are the questions that formal drug development is designed to answer. The reason they’re worth stating explicitly is that the gray-market wellness context often bypasses all of them.
Is there a viable mechanism of action?
A mechanism is not a marketing claim. “Reduces inflammation” and “supports recovery” are not mechanisms; they are descriptions of desired outcomes dressed up as biology. A real mechanism of action specifies a molecular target, a downstream signaling effect, and a plausible chain from that effect to the clinical outcome being claimed.
Mechanism matters for two reasons. First, it makes a claim falsifiable. If a molecule is supposed to produce effect X by binding receptor Y and activating pathway Z, those claims can be tested independently. If the receptor is unknown, the pathway is unspecified, and the clinical effect has never been measured in humans, no experiment can definitively refute the claim — which means no experiment can confirm it either. Second, mechanism identifies failure modes. A molecule may reach the target but not in relevant tissues. It may produce the expected biomarker change without producing the clinical outcome. It may activate pathways with effects that counteract, or dangerously amplify, the intended result. None of these problems are visible without a defined mechanism.
Among FDA-approved drugs, the fraction with genuinely unclear mechanisms of action is estimated at roughly 3%. That number is worth holding on to when evaluating a compound whose mechanism is described as “multifactorial” or “pleiotropic” in ways that seem to expand rather than specify.
Is there evidence of meaningful benefit in humans?
Animal models are hypothesis generators, not outcome validators. Between 30 and 50 percent of compounds that clear preclinical testing and enter Phase 1 trials fail to advance to Phase 2 — frequently because human biology does not reproduce what the animal data predicted. The history of drug development is dense with molecules that produced compelling results in rodents and failed, sometimes dangerously, in people.
Biological activity in humans is not the same as clinical benefit. A molecule that raises a biomarker, activates a signaling pathway, or produces a measurable physiological change has demonstrated biological activity. The relevant question is whether those changes translate into outcomes that matter to patients: improved function, reduced pain, faster healing, longer life. Many compounds that satisfy the first criterion have been unable to satisfy the second, including some that were studied rigorously for years.
Do we understand safety, dosing, and pharmacokinetics?
How much of the compound reaches circulation? How long does it remain active? What dose was actually studied, in what population, via what route of administration? What are the short-term and long-term risks? What requires monitoring?
These questions define whether a drug can be used in a controlled, predictable way — whether there is a practical framework for deciding how much to give, when to stop, and what to watch for if something goes wrong. For gray-market compounds, these questions are frequently unanswerable because the studies that would answer them have not been conducted.
Does the likely benefit justify the risk for this specific person?
Risk is contextual. A serious adverse effect may be entirely acceptable when the alternative is a life-limiting disease. The same adverse effect is not acceptable when the expected benefit is modest or speculative and the therapeutic alternatives are well-characterized. A mitochondrial-targeting peptide that might make sense for a child with Barth syndrome — a severe, fatal mitochondrial disorder — does not automatically make sense for a healthy adult seeking better energy and performance. The risk-benefit calculation changes entirely when the population and the stakes change.
Is there a better-characterized way to achieve the same result?
The question is not whether a gray-market compound might work. It’s whether it works better than available alternatives, with better-understood risks and better-established dosing. If the answer is that a less-characterized compound is being chosen primarily because it’s more compelling, more novel, or more marketable, that is worth naming clearly.
BPC-157: A Case Study in Every Red Flag
Against those five questions, BPC-157 fails each one.
The compound is described as a fragment of a naturally occurring gastric body protection compound. Its origin is contested in unusual ways: the parent protein has never been fully characterized or published in peer-reviewed literature. BPC-157’s own sequence does not clearly correspond to any known human gastrointestinal peptide. Several mechanisms have been proposed — effects involving VEGF (vascular endothelial growth factor), angiogenesis, nitric oxide signaling, and monoamine neurotransmitter systems — but none have been established in humans, and the primary receptor through which the compound operates remains unidentified. Lacking a classic receptor is not inherently disqualifying; some drugs act through non-receptor mechanisms. But lacking a receptor is different from lacking a mechanism. A well-defined mechanism can describe non-receptor action. BPC-157’s does not.
The human evidence is, in a word, absent. A 2024 review in Frontiers in Pharmacology confirmed what a PubMed search makes immediately apparent: the published literature on BPC-157 is overwhelmingly preclinical, originates primarily from one academic group — the laboratory of Predrag Sikirić at the University of Zagreb, whose members hold patents on BPC-157 salts and production methods — and contains no peer-reviewed, published randomized controlled trials in humans for any indication. One Phase 2 trial in inflammatory bowel disease (NCT05386550) was registered in 2022; as of mid-2026, no results have been published and the trial remains listed as recruiting. Three decades of preclinical claims. Not one completed human RCT.
Dosing and pharmacokinetics in humans are unknown. The protocols circulating in wellness communities are not extrapolations from human pharmacokinetic data — that data does not exist. They are guesses, however confidently presented.
The risk picture deserves particular attention. VEGF signaling and nitric oxide pathway activation — the mechanisms BPC-157 proponents most frequently invoke — are not benign biological levers. They are implicated in angiogenesis, tumor vascularization, and tissue remodeling. This does not mean BPC-157 causes cancer. The evidence to make that claim does not exist either. But if proponents are genuinely asserting that the compound stimulates healing through proangiogenic pathways, intellectual consistency requires taking seriously the possibility that it stimulates biology one might not want overstimulated — particularly in someone with undetected early-stage malignancy.
The FDA reached its own conclusion: in 2023, BPC-157 was placed on the agency’s list of bulk drug substances that may not be used in compounding, citing insufficient evidence of clinical utility and unresolved safety concerns. This is not, by itself, proof that the compound is dangerous or ineffective. It is a data point about what the regulatory evaluation found — and what it did not find.
Then there is the pattern of the claims themselves. BPC-157 began with wound healing. The indications expanded to tendons, then ligaments, then muscle, then gut disease, then inflammation, then pain, then performance, then recovery, then neurological conditions including multiple sclerosis. Legitimate drug development moves in the opposite direction: one indication, proved rigorously, opens the investigation for a second, which is then proved or disproved on its own evidence. GLP-1 agonists did not arrive claiming to treat obesity, heart disease, kidney disease, fatty liver disease, and neurodegeneration simultaneously. They received one indication, established it with controlled trials, and earned additional ones as independent evidence accumulated. BPC-157 has expanded its claimed indications continuously while the first one was never established. The list of things it purportedly treats is inversely related to the rigor with which any of those claims has been tested.
CJC-1295 and the Gap Between Activity and Benefit
BPC-157 occupies what might reasonably be called the bottom of a three-tier evidence classification: no validated mechanism, no credible human evidence, claims that expand over time without scientific justification. CJC-1295 sits in the middle tier — and in some ways illustrates a subtler problem.
CJC-1295 is a GHRH (growth hormone-releasing hormone) receptor agonist. It binds its target. It raises growth hormone and IGF-1 in humans. These are not contested claims. The biology is plausible, and the biological activity is real.
The compound was developed by ConjuChem — the name is a direct abbreviation of that company — and reached Phase 2 clinical trials before development was discontinued. What it could not demonstrate, under controlled conditions, was that its biological effects translated into outcomes people actually care about: meaningful improvements in strength, physical performance, recovery capacity, or quality of life.
To understand why, it helps to look at what happens when the same pathway is stimulated more directly. Growth hormone replacement in adults who are genuinely growth hormone deficient can matter clinically. In the specific context of HIV-associated lipodystrophy, GHRH agonism via tesamorelin — a closely related molecule, developed around the same time as CJC-1295, with a defined patient population and clean Phase 3 data — received full FDA approval in 2010. The divergence between tesamorelin and CJC-1295 had nothing to do with patentability or pharmaceutical indifference. Tesamorelin succeeded because it had better data for a defined indication. CJC-1295 did not.
In growth hormone-replete adults — which is to say, nearly everyone using CJC-1295 for wellness purposes — the results of growth hormone administration have been consistently underwhelming. Modest changes in body composition appear in some trials. Lean body mass, measured by the metrics these studies typically use, can increase — but a meaningful fraction of that increase reflects water retention and non-contractile tissue rather than functional skeletal muscle. Strength, physical performance, recovery, and quality of life show little or no benefit. Given that outcome, the argument for an indirect GHRH stimulator producing dramatically better results in the same population carries an unusually high burden of proof — one that CJC-1295’s clinical development history did not come close to meeting.
The middle tier, then, is not a gray zone where cautious optimism is warranted. It is where biological plausibility has been tested and found insufficient.
Why Testimonials Cannot Answer the Question
The most honest response to testimonial evidence is not dismissal — it is precision about what testimonials can and cannot establish.
A personal account describes what happened after someone took a compound. It cannot describe what would have happened without it. That counterfactual is the entire question.
Musculoskeletal injuries are instructive here. They fluctuate. They tend to improve on their own, particularly from a nadir. People almost universally start a new intervention at their worst point — which is precisely when regression to the mean would predict improvement regardless of treatment. Layer on the other variables almost always present: rest, physical therapy, reduced training load, anti-inflammatory medication, improved sleep, dietary changes, and in many cases additional pharmacological agents including anabolic compounds. The peptide is one element in a dense, uncontrolled variable set, and receives disproportionate credit because it is the most novel, the most deliberate, and the most marketable element in that set.
The placebo effect in the specific context where peptides are most heavily promoted — pain and recovery — is not a modest or easily dismissed phenomenon. Wartolowska and colleagues, in a 2014 systematic review published in the BMJ, documented clinically meaningful pain improvements from sham surgical procedures. The Moseley trial, published in the New England Journal of Medicine in 2002, found that arthroscopic knee surgery for osteoarthritis produced no significant benefit over placebo surgery. Injections amplify the effect further: subcutaneous administration carries an implicit authority that oral supplements do not. The ritual has clinical weight. The story around the compound — regenerative, advanced, biologically targeted, succeeding where conventional medicine failed — has clinical weight. These are not reasons to discount someone’s experience. They are reasons why individual experience cannot substitute for controlled data.
The STEP 1 trial offers a useful calibration. Semaglutide 2.4mg weekly produced mean weight loss of 14.9% versus 2.4% in the placebo group over 68 weeks (Wilding et al., NEJM, 2021). The placebo group lost weight. Not because placebo is a drug, but because trial participation — the follow-up, the structured support, the expectation of treatment — moved outcomes in the control arm. The RCT’s purpose is to isolate how much of the observed effect belongs to the molecule above and beyond all of that.
For BPC-157 and most gray-market wellness peptides, that isolation has never been performed.
The Molecule Is Not the Drug
One misconception underlies much of the gray-market peptide trade, and it is worth addressing directly: the belief that an amino acid sequence is a drug.
It is not. A pharmaceutical is a specific product, manufactured under specific processes, formulated to specific standards, with a specific pharmacokinetic profile demonstrated in the human subjects who took the actual product in the actual trials. The clinical evidence applies to that product. It does not automatically apply to every preparation that shares the same amino acid sequence.
Consider what turning a peptide sequence into a reproducible pharmaceutical actually requires: consistent synthesis at scale, rigorous purification, batch-to-batch analytical verification of identity and concentration, demonstrated sterility, stability under defined storage conditions. These are not bureaucratic requirements layered onto the molecule as an afterthought. They are the conditions under which the compound can be manufactured, administered, and monitored in ways that make the evidence interpretable. Change the manufacturing process, and the properties of the drug may change with it.
A 2023 FDA analysis of compounded semaglutide products — a direct analog for the broader gray-market peptide situation — found instances of incorrect dosing and contamination in sampled products from compounding pharmacies. Third-party testing with HPLC or mass spectrometry can confirm identity and approximate purity in a sampled vial. It says nothing about sterility or lot-to-lot consistency. A doctor’s prescription does not create missing clinical evidence for the molecule. A compounding pharmacy’s involvement does not cause an uncharacterized compound to inherit the clinical characterization of a regulated product.
These are not arguments against any particular vendor or practitioner. They are statements about what those safeguards can and cannot provide, and about the gap that remains even when they are present.
Where the Gray Market Actually Came From
The narrative that positions gray-market peptides as compounds the pharmaceutical industry ignored — because natural molecules cannot be patented, because there is no profit motive — does not survive contact with the actual development histories.
Patent law leaves substantial room for monetization around natural molecules. Modified analogs, novel sequences, salts, conjugates, delivery systems, manufacturing processes, and specific dosing regimens are all patentable. Rapamycin, metformin, and the statins all began as natural or nature-derived molecules and were all developed into profitable pharmaceuticals through exactly these mechanisms. GLP-1 itself is a naturally occurring peptide; a generation of companies is currently racing to develop synthetic analogs of GLP-1 biology for new indications. The absence of commercial development is not, in most cases, an artifact of patent law.
CJC-1295 is named after the company that abandoned it. It did not end up in the gray market because ConjuChem passed on it for lack of a patent strategy. It ended up there because the Phase 2 data were insufficient to justify Phase 3 investment. The gray market, for many of these compounds, is not an alternative to pharmaceutical development. It is the destination for compounds that pharmaceutical development tested and discarded.
Among drugs that enter clinical trials, approximately 90 to 95 percent never reach approval — eliminated by inadequate efficacy, unacceptable safety profiles, poor pharmacokinetics, or an inability to outperform existing treatments. That failure rate is not a flaw in the system. It is the system functioning as intended, separating compounds that work at the required standard from those that do not. Many of the most popular gray-market wellness peptides are compounds that did not clear that bar.
The Falsifiability Standard
What would prove a given peptide claim wrong?
For compounds with clear mechanisms, defined clinical endpoints, and human trial data, the answer is specific: a well-powered RCT showing no significant effect on the primary endpoint, replicated independently, would substantially undermine the claim. That answer is available because the claim was stated in a form that evidence could address.
For BPC-157, the answer is harder to find. Every disappointing outcome gets attributed to dose, timing, supplier, stacking protocol, or individual response. Every absence of human data is explained by pharmaceutical suppression or the difficulty of funding natural compounds. The claim does not narrow in response to evidence. It expands.
A hypothesis that cannot be falsified is not a scientific hypothesis. A field whose claims expand rather than narrow over time, as the foundational questions remain unanswered, is not moving in the direction that real drug development moves. Legitimate development accumulates precision: better trials, narrower indications, clearer dosing, identified subpopulations. The list of what BPC-157 claims to treat is longer now than it was ten years ago. The list of what has been rigorously demonstrated in humans is unchanged.
Peptide science is not in question here. The pipeline is real. The approved drugs are real. The areas of genuine near-term promise — metabolism, oncology, diagnostics, infectious disease — are real, and the specificity advantages that peptides carry in those domains are advantages that may produce the next generation of consequential medicines. The registered NCT05386550 trial for BPC-157 in inflammatory bowel disease, if it completes and publishes results, will produce actual evidence for evaluation. That is the process working.
The gray-market wellness ecosystem is a separate matter. It is a system in which hope has been packaged as a product, and in which the compounds most aggressively promoted — for brain optimization, recovery, tissue repair, and longevity — are precisely the applications where peptide science faces its steepest evidentiary climb. The word “peptide” is doing work that the individual molecules have not earned. Running any specific compound through five honest questions tends to make that visible, faster than any marketing will.
Sources
- Muttenthaler M, King GF, Adams DJ, Alewood PF. Trends in peptide drug discovery. Nature Reviews Drug Discovery. 2021;20(4):309–325.
- Wilding JPH, Batterham RL, Calanna S, et al. Once-weekly semaglutide in adults with overweight or obesity. New England Journal of Medicine. 2021;384(11):989–1002.
- Wartolowska K, Judge A, Hopewell S, et al. Use of placebo controls in the evaluation of surgery: systematic review. BMJ. 2014;348:g3253.
- Moseley JB, O’Malley K, Petersen NJ, et al. A controlled trial of arthroscopic surgery for osteoarthritis of the knee. New England Journal of Medicine. 2002;347(2):81–88.
- U.S. Food and Drug Administration. Bulk Drug Substances Under Consideration for Use in Compounding: BPC-157 [Category 2 Listing]. 2023.
- Biotechnology Innovation Organization (BIO). Clinical Development Success Rates 2006–2015. 2016.
- Frontiers in Pharmacology. Review of BPC-157 preclinical evidence base and absence of human trial data. 2024. [Review article; full citation to be confirmed with publication details at editorial review.]
- Sikirić PC et al. Multiple peer-reviewed publications on BPC-157 in animal models, 1993–2024. University of Zagreb; patent filings on BPC-157 salts and production methods publicly available via USPTO and EPO databases.
- ClinicalTrials.gov. NCT05386550: BPC-157 for Inflammatory Bowel Disease. Phase 2; registered 2022; status: recruiting as of mid-2026.
- U.S. Food and Drug Administration. FDA analysis of compounded semaglutide products. 2023.
