21 peptides graded. Only two got an A

LONGEVITY LATEST ISSUE 25 · 26 AUGUST 2026

LONGEVITY LATEST

The Evidence-Based Edge on Living Longer and Better

Issue 25 · The Tier List · 26 August 2026

WELCOME

👋 Welcome

Seven peptides went in front of an FDA advisory committee last month. The agency's own scientists recommended against every one of them. The committee voted for six anyway.

The vote is advisory — it recommends adding them to a compounding list, and it approves nothing. But it turned on demand rather than data. Six of the eight recently appointed members run clinics that prescribe, produce or promote peptides, and reviewers had been through the human evidence for all seven and found, for four of them, no human safety studies at all.

Blood pressure has moved to next week; this couldn't wait. I've spent a fortnight in the briefing documents and the trial registries and graded twenty-one compounds from A to F. Two earned a straight A. Nine sit at D or F.

In this issue:

• 🔬 Top 3: the peptide with 13,165 patients behind it, the licensed one nobody talks about, and the famous one tested in fewer than thirty people

• ⭐ Spotlight: the vote the FDA's own scientists lost — and the fifteen-day problem

• 🚨 Hype Check: 6,487 vials tested independently, and what was in them

• 📖 Deep Dive: the full tier list — twenty-one peptides, graded

THIS WEEK'S ANALYSIS

🔬 Top 3 Interventions Under the Microscope

Three peptides, and the distance between them is the point. One has outcome data on 13,165 people. One has held a licence since 2010. The third is the most popular peptide on the internet and has been given to fewer than thirty people in a published study.

1. Tirzepatide — Evidence Grade: A

What it is. A 39-amino-acid chain that switches on two receptors at once, GLP-1 and GIP. Which makes it, unavoidably, a peptide — worth holding on to when someone tells you peptides are fringe medicine. So is insulin. The category contains the best-evidenced drugs of the decade and vials of unidentified powder from an industrial estate.

Human evidence. SURMOUNT-1 gave a 20.9% mean reduction in body weight at 72 weeks. But the trial that matters is SURPASS-CVOT, in the New England Journal in December: 13,165 people with type 2 diabetes and established cardiovascular disease, randomised against dulaglutide rather than placebo, median treatment just under four years. Three-point MACE: hazard ratio 0.92, non-inferior, as designed.

A post-hoc analysis this year applied a wider cardiorenal endpoint and found 23.7% against 27.4%, hazard ratio 0.84 (0.79–0.90). Beating an active comparator that had already beaten placebo is a harder trick than the coverage suggested. Post-hoc — but the direction isn't ambiguous.

Cautions. Gastrointestinal events hit 42.5% against dulaglutide's 35.9%; not a gentle drug to start. Lean mass comes off alongside the fat unless you train and eat for it. Contraindicated with a personal or family history of medullary thyroid carcinoma or MEN2.

Takeaway. Grade A, and the only peptide here with hard outcome data rather than surrogate endpoints. Everything else this week is measured against it.

2. Tesamorelin — Evidence Grade: B+

What it is. The strongest human file outside the GLP-1 class, and almost nobody in the scene mentions it. A growth-hormone-releasing hormone analogue: rather than flooding the system with growth hormone, it prods the pituitary into releasing its own, in the pulses it is meant to. Licensed by the FDA in 2010 for abdominal fat in HIV-associated lipodystrophy.

Human evidence. Falutz's 2007 registration trial cut visceral fat by roughly 18% over 26 weeks. Stanley's 2014 JAMA trial found visceral adipose tissue down 34 cm² against a placebo group that gained 8, and a twelve-month trial in HIV-associated fatty liver stopped fibrosis progressing. January's meta-analysis pooled five randomised trials: −27.71 cm², interval −38.37 to −17.06. Tight, consistent, replicated.

And it has been tested outside HIV. Baker's 2012 trial in Archives of Neurology gave GHRH to older adults with mild cognitive impairment, and to healthy older adults, for twenty weeks — and found cognitive benefit in both groups. Fourteen years, and still no decent follow-up.

Cautions. It raises IGF-1, which anyone who has read the ageing biology knows is not straightforwardly a thing to want. Glucose tolerance worsens in some. The fat comes back when you stop.

Takeaway. Grade B+, and the plus is doing specific work. Several consistent trials, nearly all in one clinical population, all on surrogate endpoints, and not one asking whether anybody lived longer or better for it.

3. BPC-157 — Evidence Grade: D

Start with the number, because the number is the review. Fewer than thirty people. That is the total published human exposure to the most popular peptide in the world.

What it is. A fifteen-amino-acid sequence derived from a protein found in gastric juice, credited with speeding repair in tendon, muscle, gut and nerve. The animal literature is enormous — over a hundred preclinical papers, most from one Croatian group, going back to the early nineties.

Human evidence. Three published pilot studies: an open-label knee-pain series in sixteen patients, an interstitial cystitis pilot in twelve, and a 2025 intravenous safety study in two adults. None placebo-controlled. A 2025 systematic review swept up 36 papers spanning 1993 to 2024 and found exactly one clinical study among them. A Phase I trial in 42 volunteers, registered in 2015, has never published a result.

Cautions. The proposed mechanism runs partly through VEGF and new blood-vessel formation — precisely the pathway nobody has checked in someone carrying an undiagnosed tumour. Prohibited at all times under the WADA code.

Takeaway. Grade D. I wanted to grade it higher — the mechanistic work is good and the safety signals so far are quiet. But “no harm found in thirty people” isn't a safety record. It's an absence of looking.

SPOTLIGHT

⭐ Spotlight: The Vote the FDA's Own Scientists Lost

On 23 and 24 July the Pharmacy Compounding Advisory Committee took seven peptides — BPC-157, KPV, TB-500, MOTS-c, emideltide, Semax and epitalon — and voted on whether pharmacies should be permitted to compound them. FDA's scientific staff had reviewed all seven and recommended against all seven. The committee backed six. Semax got through 8–5; MOTS-c 7–5 with two abstentions; emideltide lost 7–6.

Pros. The harm-reduction case is real and I won't be glib about it. Enormous numbers of people already inject these compounds, bought from websites, with no clinician near the decision. Route that through a pharmacy and you put a professional in the room and a supply chain under some form of inspection. Several members said exactly that, and they weren't wrong about the problem.

Cons. They may be wrong about the fix. Read the briefing documents and the file is stark. FDA identified no human safety studies at all for KPV, TB-500, MOTS-c or epitalon. For the three with any — BPC-157, emideltide and Semax — the studies were small, the safety reporting patchy, and treatment ran to no more than fifteen days.

Fifteen days is the longest anyone has been followed on these compounds — for conditions people intend to treat for years.

Most didn't even use the route of administration people actually use. Reviewers flagged aggregation, synthesis impurities and immunogenicity — the risk your immune system starts reacting to a molecule your own body also makes.

HYPE CHECK

🚨 Hype Check: “Research Use Only”

The Hype: Vials at $49 to $100 from outfits called Legendary Peptides, Texas Peptides, Lone Star Peptide — labelled for laboratory research, not for human consumption. Nobody involved believes that sentence. This month Eli Lilly sued six of them over retatrutide alone, having reported more than 14,000 listings across a hundred-odd countries.

The Evidence: In April two sports-medicine physicians pulled 6,487 independent test reports covering fourteen peptides from 203 vendors and ran them against pharmaceutical release standards. Against the looser compounded-medicine benchmark, 41.6% failed on purity, dose or identity. Against manufactured-drug standards, 71.1% failed. One hundred and fifty-six vials contained none of the peptide on the label. For TB-500 that was one vial in ten.

Endotoxin was measurable in 15% of the subset tested for it — and purity did not predict it. Not weakly. At all. A certificate reading 99.8% pure tells you nothing about whether the vial gives you a fever.

Why It's Misleading: “Research use only” is a legal shield, not a quality standard. A certificate of analysis covers two checks out of roughly ten a licensed peptide clears. No sterility. No stability. And because vendors submit samples voluntarily, the authors are explicit that these are floor figures.

Sit with that. The authors add one observation I haven't been able to shake: we understand the human pharmacology of several notorious street drugs considerably better than we understand BPC-157.

SUPERFOOD

🥣 Superfood Spotlight: Skyr, and the Peptide That Actually Builds Muscle

After four sections of what doesn't work, something that does. Every growth-hormone peptide in this issue is sold on a promise of muscle and recovery. The intervention with the best evidence for both sits in the chilled aisle.

Morton's meta-analysis in the British Journal of Sports Medicine pooled 49 randomised trials and 1,863 people: protein supplementation meaningfully increases the lean mass and strength you get from resistance training, plateauing around 1.6 g per kilogram a day. A larger, cleaner evidence base than anything in the C tier of this week's Deep Dive, and it isn't close.

Skyr and Greek yoghurt are the cheapest way there: 10 to 11 g of protein per 100 g, dense in leucine — the amino acid that flips the muscle-building switch, and the one collagen powder is short of. That matters most if you're on a GLP-1, where lean mass comes off with the fat unless protein and training hold it in place.

A 450 g tub is about £2. Two of those a week costs less in a year than one vial of BPC-157.

DEEP DIVE

📖 Deep Dive: Twenty-One Peptides, Graded

Twenty-one compounds, one A-to-F scale. Two earned a straight A, and both are drugs a GP can already prescribe.

But the thing that stopped me was further down the table. The six compounds sitting at D are, precisely, six of the seven the advisory committee voted on in July. Not roughly. Exactly. That took a fortnight to notice and about ten seconds to explain once I had, and the explanation is the reason this list sorts the way it does rather than the way the gym tier lists do.

The full article has all twenty-one, the reasoning behind each grade, and why growth hormone gets the only split verdict on the list.

👉 Read the Deep Dive: Twenty-One Peptides, Graded →

BIOHACKING CORNER

🌡 Biohacking Corner: The Growth-Hormone Protocol That Costs Nothing

Take the GLP-1s out and most of the peptide market aims at one place: the growth-hormone axis. Most of your daily growth hormone arrives in pulses tethered to slow-wave sleep — which is free. So: two things that reach the axis without a syringe, and three questions for anyone selling you the version with one.

1. Protect the first three hours of sleep above everything else. That's where the big pulse lives. Alcohol suppresses slow-wave sleep specifically, which is why a nightcap costs more than it looks like it costs.

2. Lift things. Resistance training is the only item here with unambiguous human evidence for what most peptide buyers actually want: muscle they keep.

3. Measure your waist, not your weight. Visceral fat is what tesamorelin moves, and a tape measure tracks it better than the scales do.

4. If a clinic offers you a peptide, ask three questions. Which randomised human trial supports this, for this indication, by this route? Who ran the batch test, and may I see it? What happens when I stop?

5. If you already use something, put it on your medical record. Most people I've spoken to who take these compounds haven't told their GP — which means that if anything does go wrong, nobody can attribute it. Including them.

I asked those three questions at a clinic last winter, out of curiosity rather than intent. I got a mechanism diagram, a certificate with no laboratory name on it, and a slightly wounded silence.

READER PULSE

📊 Reader Pulse

Last week I asked whether you knew your ten-year cardiovascular risk score. Twenty-two per cent had checked within the year, 19% had let it lapse, 44% worked it out on the spot while reading, and 15% would rather not. That 44% is the highest action rate any poll here has produced — on the cheapest thing I have ever recommended.

This week, given the subject: have you ever bought a peptide?

• “Yes — from a clinic, on prescription.”

• “Yes — from a website.”

• “No, but I've come close.”

• “No, and this issue settled it.”

Vote at longevitylatest.com/poll-25 — and if you know somebody about to spend £200 on a vial, forward this before they do.

CLOSING

🎯 Closing

Peptides aren't a category of treatment. They're a category of molecule — which is a bit like saying “pills”. Tirzepatide and epitalon are both peptides in the way that whisky and windscreen wash are both liquids, and the shared word is doing all the marketing.

Next week, blood pressure: whether home readings beat the clinic's, what the evidence actually says about treating to 120, and why the commonest measurement error happens in the thirty seconds before the cuff inflates.

Stay curious and stay healthy!

— Christian Thomsen, Editor

Longevity Latest is published weekly by FrontWave Media Ltd. The content is for educational purposes and does not constitute medical advice. Several compounds discussed in this issue are not approved for human use in any jurisdiction; nothing here should be read as encouragement to obtain or self-administer them. Peptide drugs are prescription medicines and decisions about them belong with a clinician who knows your history. Do not start, stop or alter any prescribed medication on the basis of this newsletter. If you are pregnant or breastfeeding, have a personal or family history of thyroid cancer or MEN2, or have diabetes, pancreatitis, liver or kidney disease, speak to your doctor before making any change. Consult your physician before starting any new regimen.

© 2026 FrontWave Media Ltd · Longevity Latest 1