Most patients arrive having already formed a strong view, shaped largely by Instagram and by friends or fellow gym-goers who are convinced the results speak for themselves.

What they want from me is something specific: a legitimate source and a clinician who has actually read the literature telling them whether the science supports what they’ve heard. I have read the literature — all of it I can find — and the answer is more complicated and more interesting than either “it works” or “stay away.”

What follows is my genuine assessment. Not to discourage anyone from exploring this space, but because the people I look after deserve better information than they’re currently getting.

‍

The cultural moment

You can’t understand the volume of these conversations without understanding what happened to public trust in medicine over the past five years.

During COVID, Australians were mandated — or heavily pressured — to take vaccines developed on an accelerated timeline. Emergency-use authorisations, truncated Phase III schedules, rolling submissions. Some of those people didn’t feel well afterwards, and whether or not “compelled to participate in a medical experiment” is a fair characterisation of what happened, it is how a significant proportion of the population experienced it. Those same people are now being told they can’t easily have compounds they actually want. Compelled to take something experimental, denied something else experimental — I believe that irony is the emotional engine behind peptide demand, and it has very little to do with the science of any individual compound.

The distribution mechanism is social media, and the scale matters. This isn’t fringe. It isn’t bodybuilding forums. Fitness influencers, biohackers, and wellness entrepreneurs have moved these compounds into the mainstream with remarkable speed. BPC-157 alone has over 50 million tagged video views across Instagram and YouTube (Vasireddi et al. 2025, PMID 40756949). The content follows a reliable formula: personal testimonial, before-and-after imagery, a sprinkling of scientific terminology, and a link to a clinic or vendor. The algorithm does the rest.

‍

The hype machine

“I’ve been doing my research” is the sentence I hear most often, and what it usually means is: the algorithm served peptide content for three weeks, every video confirmed what the person wanted to hear, and they now hold a conviction about a compound that is inversely proportional to the evidence supporting it.

These patients are doing what we told them to do — taking an interest in their health, questioning conventional medicine, seeking alternatives when the system hasn’t delivered. But the information ecosystem they’re navigating was built to sell, not to inform. A fitness personality with 200,000 followers posts about their “recovery protocol” and mentions BPC-157 or CJC-1295/Ipamorelin. The content becomes more frequent, more confident, more specific. The audience grows. Sales grow and doctors don’t have much say in it.

At no point in that pipeline does anyone mention that BPC-157 has been tested in a total of 30 human subjects, none in a controlled trial. Nobody mentions that CJC-1295 and ipamorelin have never been studied as a combination. Nobody mentions that MK-677 was abandoned by Merck because of safety signals. The patient arrives at my consultation having already decided, and what they’re usually asking for is a prescription, not an opinion.

‍

The regulatory picture

In Australia, the TGA scheduled BPC-157 as S4 from June 2024 with explicit guidance advising against compounding. In October 2024, the TGA banned compounding of all GLP-1 receptor agonists following safety concerns around compounded semaglutide. The RACGP has warned GPs specifically about prescribing unapproved peptides without appropriate clinical justification, and penalties for unapproved peptide supply exceed AUD $200,000 with potential criminal prosecution. Compounding pharmacies can still produce certain peptides, but the regulatory tolerance for experimental peptide prescribing is clearly shrinking.

In the US, the story has become political. In February 2026, RFK Jr. announced that roughly 14 of the 19 FDA Category 2 peptides — including BPC-157 — would be reclassified to Category 1, making them eligible for compounding again. FDA advisory committee meetings are scheduled for July 2026. No new safety or efficacy data was submitted. The reclassification was driven by political advocacy through the Make America Healthy Again movement, not by scientific review.

The policy argument isn’t entirely without merit: people are already obtaining these compounds from grey-market sources with unknown purity, so allowing licensed compounding pharmacies to produce them at least improves quality control.

That’s harm reduction, and harm reduction is a legitimate public health framework. But a reclassification for compounding eligibility is not FDA approval, and the evidence gap remains exactly as wide as it was before the announcement.

Here is the part that concerns me most, and it should concern every Australian doctor working in this space: to my knowledge, no Australian professional indemnity insurer will cover a doctor prescribing experimental peptides for off-label use. This is not a hypothetical consideration — it is a career-ending exposure, and most doctors prescribing these compounds haven’t thought it through. The influencers posting their peptide protocols on Instagram are certainly not mentioning this part.

‍

The funding gap

This is the question underneath every other question in this space, and understanding it reframes everything else.

BPC-157 is not patentable. CJC-1295 is not patentable. TB-500, GHK-Cu — none of them are patentable. Without patent protection, no commercial entity will spend $200–500 million on a Phase III programme because the return on investment simply doesn’t exist. Government research bodies like the NIH and NHMRC rarely fund commercial-stage clinical trials for compounds already being sold, and the compounding pharmacy industry — which profits handsomely from selling these compounds in their current regulatory limbo — has no incentive to fund trials that might demonstrate they don’t work or aren’t safe.

To put concrete numbers on it: a properly powered Phase III trial for BPC-157 in a single indication — say, tendon healing after ACL reconstruction — would require around 500 patients per arm, twelve months of follow-up, standardised MRI endpoints, independent safety monitoring, and multi-site GCP-compliant recruitment. Estimated cost: $50–100 million minimum. Professor Sikirić’s entire thirty-year research programme at the University of Zagreb cost a fraction of that, and Croatian government research budgets are not going to fund international Phase III trials.

So we’re stuck in a structural trap. Compounds that can’t be patented can’t attract the investment required to generate the evidence required to know whether they work. They aren’t “alternative” because the science is bad or the ideology is fringe. They’re alternative because of economics. Universities are probably the only realistic path to rigorous human data, and even that requires philanthropic or government funding at a scale that hasn’t materialised.

When patients ask me “why hasn’t this been studied properly?” the answer I give them is: nobody will pay for the science. That’s not satisfying, but it’s true.

‍

The compounds that made it — and what they cost

Before I walk through the experimental compounds, it’s worth understanding what it actually takes to get a peptide from laboratory curiosity to something a doctor can prescribe with confidence. Four compounds have done it. Their stories are instructive.

Insulin is the benchmark. Discovered in 1921 by Banting and Best at the University of Toronto, first injected into a human — Leonard Thompson, a 14-year-old dying of diabetes — on 11 January 1922. Nobel Prize the following year. Banting sold the patent for one dollar because he believed insulin belonged to the world. One hundred years of global human data followed. That is the standard every therapeutic peptide should be measured against, not in terms of importance, but in terms of what it took to know.

Semaglutide is what serious investment looks like. Novo Nordisk took the body’s own GLP-1 — a gut hormone with a natural half-life of two minutes — and attached a C18 fatty diacid chain, a molecular anchor that binds to albumin in the blood and drags the half-life out to one week. Decades of development and billions of dollars produced the SUSTAIN programme in type 2 diabetes, the STEP trials showing 14.9% mean weight loss at 68 weeks, and the SELECT trial demonstrating a 20% reduction in major cardiovascular events in overweight patients without diabetes. A single Phase III trial in that programme cost more than BPC-157’s entire thirty-year research history. That is the scale of the evidence gap between approved peptide drugs and the compounds patients are asking me about.

Tesamorelin found a narrow path through the regulatory system via orphan drug designation. It’s a synthetic analogue of growth hormone-releasing hormone — the same mechanism that CJC-1295 targets — and was FDA-approved in 2010 for HIV-associated lipodystrophy based on two pivotal Phase III trials in 816 patients showing 15–17% visceral fat reduction by CT imaging. It made it through because it had three things CJC-1295 did not: a patentable molecular modification, a specific clinical indication with measurable endpoints, and a sponsor willing to fund Phase III. Same mechanism, different commercial trajectory, and that trajectory is the reason we have robust safety and efficacy data for one and essentially nothing for the other.

Bremelanotide — PT-141 — is the story I keep coming back to. In the 1980s, researchers investigating melanocortin peptides for sunless tanning discovered that Melanotan I caused unexpected spontaneous erections in a researcher who administered himself a dose. That accidental observation redirected two decades of work. PT-141 was developed as a nasal spray for sexual dysfunction and failed in 2008 because blood pressure elevations were too high. The same molecule was then reformulated as a subcutaneous injection at a lower dose, pivoted to hypoactive sexual desire disorder in premenopausal women — a completely different indication from its original target — and went through two identically designed Phase III trials enrolling roughly 1,200 women. The FDA approved it in June 2019 as Vyleesi. Forty-three completed studies. Approximately 3,500 subjects. A tanning experiment that caused an erection, spent fifteen years failing as a nasal spray, then succeeded as an injection for women’s sexual desire. That is what it takes.

Four compounds. Four stories. Every one of them required patentable intellectual property, a willing sponsor, and years — often decades — of controlled human trials with independent safety monitoring. The process is expensive, slow, sometimes absurd, and it produces something irreplaceable: actual knowledge. We know insulin saves lives. We know semaglutide’s cardiovascular benefit. We know tesamorelin’s dose and safety profile. We know bremelanotide works for premenopausal HSDD. After thirty years of BPC-157 research, we know none of those things with any certainty.

That’s the threshold. Now here’s why the experimental compounds haven’t crossed it.

‍

The science we do have

What follows is my read of the published evidence for the compounds I’m asked about most. The range is wider than most people expect — from genuinely interesting to genuinely alarming — and the gap between what the data actually shows and what Instagram claims is vast.

BPC-157

BPC-157 is a 15-amino acid synthetic peptide derived from a fragment of human gastric juice protein. It does not exist in the body in this isolated form. For thirty years, Professor Predrag Sikirić and his group at the University of Zagreb have published over 200 papers demonstrating healing effects across dozens of animal models — gut, tendon, ligament, bone, nerve, vascular tissue. The mechanistic work is detailed and, on its own terms, impressive: eNOS/NO pathway modulation, VEGFR2 upregulation for angiogenesis — the growth of new blood vessels — and FAK-paxillin-mediated fibroblast migration, which is essentially the cellular machinery that drives tissue repair (Chang et al. 2011, PMID 21030672; Seiwerth et al. 2021, PMID 34267654).

The animal data is genuinely interesting. I want to be clear about that because I’m about to be less generous about the human evidence, and I don’t want the takeaway to be that this compound is nonsense. It might not be. We don’t know yet, and the “don’t know” is the problem.

In humans, we have three small uncontrolled case series totalling 30 patients, all from the same research group, all published in the same low-impact journal. No control group. No blinding. That’s Level IV evidence approaching Level V on the Oxford Centre for Evidence-Based Medicine hierarchy — roughly one step above expert opinion (McGuire et al. 2025, PMID 40789979; Vasireddi et al. 2025, PMID 40756949).

Then there’s the silence, which I think is the most important part of the BPC-157 story and the part that gets discussed least. In December 2015, a quadruple-blind, placebo-controlled Phase I trial (NCT02637284) was registered on ClinicalTrials.gov. Forty-two healthy volunteers. Trial site in Tijuana, Mexico. Sponsor: PharmaCotherapia d.o.o., Croatia — a company with no other registered trials. Estimated completion: March 2016. As of April 2026 — a full decade later — no results have been published. No conference abstract. No data uploaded to the registry. Nothing has appeared.

A decade of silence after a Phase I safety trial in healthy volunteers is not normal. Phase I trials in healthy volunteers are specifically designed to be low-risk and publishable regardless of outcome. The most common reason completed Phase I trials go unpublished is that the results were unfavourable, and I can’t know that for certain in this case, but the silence is conspicuous. Anyone considering injecting BPC-157 deserves to know about it.

There’s also the VEGFR2 question, which I find genuinely troubling. BPC-157 upregulates VEGFR2 and promotes angiogenesis. That is precisely the pathway that cancer therapies are designed to block — bevacizumab (Avastin) exists specifically to inhibit VEGFR. The theoretical concern that BPC-157 might promote tumour angiogenesis is unquantified but biologically real (Józwiak et al. 2025, PMID 40005999; Yuan et al. 2026, PMID 41898733). We have no human data to assess this risk. We have a mechanism that should make any oncologically literate clinician pause, and we have no long-term safety data to tell us whether the concern is warranted or not.

CJC-1295 and Ipamorelin

CJC-1295 is a growth hormone-releasing hormone (GHRH) analogue — essentially a modified version of the signal your brain sends to stimulate growth hormone release — with a Drug Affinity Complex that extends its half-life from minutes to 5.8–8.1 days. Ipamorelin is a ghrelin receptor agonist with a two-hour half-life. Together they are the most widely prescribed peptide combination in what gets called “anti-ageing medicine,” and the combination has never been studied in any registered clinical trial.

For CJC-1295 alone, there is one published human study: a Phase I pharmacokinetics trial from 2006 in 65 healthy volunteers (Teichman et al. 2006, PMID 16352683). It showed reliable growth hormone elevation of 2–10 times baseline for six or more days after a single injection, with IGF-1 elevated 1.5–3 times for 9–11 days. No serious adverse events at the doses tested. That 49-day study is the entirety of published human data for CJC-1295. For ipamorelin, we have pharmacokinetics data and the observation that it selectively stimulates growth hormone without raising cortisol or ACTH. No efficacy data. No clinical outcomes (Sigalos and Pastuszak 2018, PMID 28400207).

The Teichman data contains a warning embedded in its selling point that I think deserves more attention than it gets. Sustained IGF-1 elevation is a potent mitogenic signal — it tells cells to grow and divide, including potentially cells you’d rather didn’t. Epidemiological evidence links chronically elevated IGF-1 to increased risk of colorectal, breast, and prostate cancer. CJC-1295’s multi-day half-life means IGF-1 is effectively continuously elevated on any weekly dosing protocol, with no washout period. Once you’ve injected it, you cannot turn it off for days. I’m not saying this means CJC-1295 causes cancer. I’m saying we have no data to assess the risk, and the biology gives us reason to want that data.

I mentioned tesamorelin earlier for a reason. It targets the same growth hormone mechanism as CJC-1295 and it made it through to FDA approval with 816 patients across two Phase III trials. The difference wasn’t biology. It was economics: a patentable modification, a specific indication, a willing sponsor. CJC-1295 had none of those advantages, which is why we have robust safety and efficacy data for tesamorelin and essentially nothing for the compound clinics are actually prescribing.

MK-677

MK-677 — ibutamoren — is the only oral growth hormone secretagogue, which means you can take it as a pill instead of injecting it, which partly explains its popularity. Developed by Merck in the 1990s, it is the one compound in this landscape with genuine Phase II randomised controlled trial data, including a two-year study. And the two-year data is instructive in ways that should give pause to anyone taking it.

The Nass et al. 2008 trial enrolled 65 healthy elderly adults and randomised them to MK-677 or placebo for two years. Fat-free mass increased by 1.1 kg (P<0.001). Appendicular skeletal muscle mass increased by 0.5 kg (P<0.001). Growth hormone and IGF-1 normalised to young-adult levels. Those are the numbers that get cited on forums and in clinic marketing materials, and they’re real.

But the same trial also showed that visceral fat was not reduced (P=0.68), which matters because visceral fat reduction is one of the primary reasons people take growth hormone secretagogues. Thirty-seven percent of MK-677 subjects developed impaired fasting glucose — a measure of how well your cells handle sugar — and fourteen percent progressed to the pre-diabetic range. Femoral neck bone mineral density paradoxically decreased. And functional outcomes — the things that actually matter in daily life, like gait speed, stair climbing, and balance — did not improve. The biochemical surrogates moved in the right direction. The clinical outcomes that matter to patients didn’t move at all.

Merck also ran a separate programme called AIMM in elderly patients and halted it for increased congestive heart failure events. That programme was never published, which is itself significant — the FDA references it specifically in warning letters. More recently, a 2025 case report documented the first published instance of drug-induced liver injury from MK-677 (Cobani et al. 2025, PMID 40675653). When a pharmaceutical company’s own development programme raises safety alarms serious enough to halt the programme, that tells you something. Merck walked away from whatever they had invested.

MK-677 is still widely sold online and prescribed by peptide clinics.

PT-141

I covered bremelanotide’s forty-year journey earlier because it’s the proof of concept for the entire approval process. What matters here is what that journey produced and what it didn’t.

For premenopausal women with hypoactive sexual desire disorder, we have real answers. Forty-three completed studies, approximately 3,500 subjects, two pivotal Phase III trials. We know the dose. We know the side effects. We know the risks. That’s rare in this landscape and it’s the direct result of a company spending the money and doing the work.

PT-141 is also widely prescribed off-label to men for sexual function, and for that indication it has essentially no controlled trial data. The approval confers credibility that does not extend beyond its specific indication, which is a pattern I see across this field — one result gets generalised into a dozen claims.

To put the efficacy question in broader context: semaglutide, a GLP-1 receptor agonist that went through the full regulatory process, demonstrated 14.9% body weight loss in the STEP trials and a 20% reduction in major cardiovascular events in the SELECT trial. That’s the standard of evidence the regulatory process is designed to produce when it works properly, and it’s the standard that most peptides in current use haven’t come close to meeting.

‍

The rest of the field

A few other compounds come up often enough to warrant brief comment.

TB-500 is the grey-market name for a synthetic fragment of thymosin beta-4, a naturally occurring peptide found in virtually every nucleated human cell. It migrated from veterinary use in racehorses to human use via the bodybuilding community. The best human data comes from a 2025 Phase IIb trial in 96 STEMI patients — people having heart attacks — across 18 Chinese hospitals (Zhang et al. 2025, PMID 41229390). The overall primary endpoint was negative, with a hypothesis-generating subgroup result. For the musculoskeletal indications that actually drive most patient interest, there are no human studies at all.

Semax, a synthetic ACTH fragment developed in Moscow in the 1980s, sits on Russia’s List of Vital and Essential Drugs, approved for stroke and cognitive disorders. It reportedly upregulates BDNF — brain-derived neurotrophic factor, a protein that supports nerve cell growth and survival. A study of 110 ischaemic stroke patients showed improved motor recovery (PMID 29798983). Credentialed scientists, published data, but no independent Western replication and methodological standards that diverge from the ICH-GCP framework that governs trial quality in most of the world.

Epitalon, from Professor Vladimir Khavinson’s laboratory in St. Petersburg, claims telomere elongation — essentially a slowing of cellular ageing — and mortality reduction in human studies spanning 40 years (PMIDs 12374906, 12195242, 14647006). The claims are extraordinary. Independent verification by Western standards doesn’t exist.

GHK-Cu has decades of topical cosmetic use for skin and wound healing but zero published human studies for injectable administration (Deng et al. 2023, PMID 36905132). Copper accumulates in the liver. The pharmacokinetics of injected copper peptide are fundamentally different from rubbing a cream on your face, and those pharmacokinetics have not been studied.

‍

Where I stand

I want to be direct about where this has brought me, because I think people need a doctor who has actually worked through the evidence rather than one who either dismisses these compounds reflexively or endorses them because the market wants it.

I’m definitely not a prohibitionist. My patients are already using these compounds — they’re arriving at consultations months into a protocol they sourced online. The conversation isn’t “should I try this?” It’s “what should I know?” Refusing to engage with that question serves nobody.

I believe in patient autonomy. If you understand the evidence limitations and the risks and you choose to proceed, that’s your right. My job is to make sure you’re making that decision with complete information, not Instagram’s version of it.

The biological mechanisms behind many of these compounds are credible, and the animal data for BPC-157 is genuinely interesting. But credible mechanisms and animal data are the beginning of the evidence process, not the end. Most compounds with promising preclinical data fail in human trials. That is the base rate of drug development, and there’s no reason to assume peptides are exempt from it.

The short-term safety track record is, based on available reporting, pretty good. These compounds have been in widespread off-label use for years and relatively few severe adverse events have been reported, which partly explains their popularity — people use them, they don’t drop dead, the word spreads. But absence of reported harm during unmonitored use is not the same as evidence of safety. We have no long-term data. We have no systematic surveillance. I genuinely don’t know what five or ten years of intermittent BPC-157 use does to cancer risk via the VEGFR2 pathway, and neither does anyone else.

If you choose to use these compounds, you are running an uncontrolled experiment on yourself (which I have no philosophical objection to - you’re an adult).

The purity of what you’re injecting may be unverified. The dose is extrapolated from animal models or online forums, not from human pharmacokinetic studies. And the person who sold it to you has a financial interest in you believing it works. Those are things worth sitting with before you inject anything.

If you are an athlete subject to anti-doping testing: every compound discussed in this article is WADA-prohibited.

I want to be transparent about something else, because it matters. I have my own observations from clinical practice, and they inform why I take this field seriously rather than dismissing it. I’ve seen friends and patients with significant injury repair — shoulders, elbows — that correlated temporally with BPC-157 use. I have a patient with autoimmune alopecia whose condition reversed with Thymosin Alpha-1. I have patients who’ve achieved meaningful body fat reduction on tesamorelin. These are anecdotes. They are the lowest level of evidence and I acknowledge that explicitly — they’re not published, not controlled, not blinded. But they’re real, and they’re part of why I keep reading the literature instead of filing this under “fad” and moving on.

I think we’re unlikely to see adequate research to bring most of these compounds to full regulatory approval without some kind of structural change — a new funding mechanism for non-patentable therapeutics, or a willingness from universities and governments to run the trials that industry won’t. I’ll keep watching the research and the regulatory developments because I want to give good advice and avoid harm, and right now that means engaging honestly with the evidence as it exists rather than as I might wish it were.

If you’re a clinician prescribing in this space: document your clinical justification meticulously. Use only reputable compounding pharmacies with GMP-equivalent standards. Inform your patients explicitly about the evidence limitations. Monitor them. And check your indemnity coverage, because the regulatory tolerance for experimental peptide prescribing is shrinking and the consequences of getting this wrong are not abstract.

The patient who arrives with a compounding pharmacy order already confirmed deserves better than what they’re getting from either side of this debate — better than the influencer who told them it’s a miracle, and better than the doctor who told them it’s rubbish without ever having looked at the data. They deserve someone who has actually read the evidence and will tell them what we know, what we don’t know, and what it would take to find out. That’s the best I can offer, and for now it has to be enough.

‍