Pete Davidson told Variety he put on 15 pounds of muscle for a movie role, and then he did something no actor does. He named what he used. BPC-157 and sermorelin, from what Futurism called the black market.
His review was two words. "They work." Then he went after every actor who credits chicken and rice for a vein running down their forehead, which, fair.
Good for Pete. But "they work" skips the step nobody in Hollywood talks about. Before anything in a vial can work, it has to actually be in the vial. And a lab just measured how often it isn't.
What a lab found in 6,441 vials
In April, researchers Christopher Mendias and Tariq Awan posted an analysis of 6,441 peptide samples that people had sent to an independent testing lab, Finnrick Analytics, over about 15 months. Fourteen compounds, including the two Pete named, plus retatrutide, tirzepatide, TB-500 and GHK-Cu. It's a preprint, so it hasn't been peer reviewed yet, but 6,441 samples beats every vendor anecdote you've ever heard.
They held every sample to the bar a compounding pharmacy has to clear. At least 98% pure, and between 90 and 110% of the amount printed on the label.
41.6% failed. On purity, on the amount, or on being the right compound at all.
156 vials didn't contain the peptide on the label. TB-500 was the worst, about 1 in 10. And 243 of the samples also got tested for endotoxin, which is leftover bacterial material that shows up when production isn't clean. 15% of those came back with a measurable amount.
The average vial was fine. That's the trap.
Here's the part that surprised me. The median sample came in at 101.8% of what the label said. So the typical vendor isn't scamming anybody.
The problem is the spread. You never buy the median. You buy one vial from one batch, and you can't tell which side of that 41.6% you landed on by staring at the powder.
That's the whole reason a COA exists. And most people read them wrong.
How to read a COA in two minutes
1. Look for mass spec, on top of HPLC. HPLC tells you how pure the powder is. It can't tell you what the powder is. Mass spectrometry weighs the molecule and confirms it's the peptide on the label. A 99% HPLC result with no mass spec can be 99% pure something else, and 156 vials in that dataset weren't the compound on the label.
2. Match the batch. The lot number on the COA should match the lot number on your vial. A clean report from a batch made last spring tells you about last spring.
3. Look up the lab. A real third-party lab has a name, a website, and usually a way to verify a report by its ID. If the only letterhead on it is the vendor's, that's a receipt they wrote for themselves.
4. Look for endotoxin. Plenty of COAs skip it. In that dataset purity and endotoxin didn't track each other at all, so a perfect HPLC number tells you nothing about it.
Before your next order, ask for the COA on the exact batch they're shipping you. The one on the product page could be from any batch. A vendor that has it sends it in five minutes. A vendor that needs to "check with the lab" just answered your question.
Got a COA you're not sure about?
Post it in the Discord. Lot number, lab name, the whole page. Somebody in there will tell you what you're looking at, and a second set of eyes costs you nothing.
P.S. If TB-500 is on your list, that's the COA I'd pull first. It had the worst identity failure rate in the whole dataset. Roughly 1 in 10 vials sold as TB-500 had no TB-500 in them.