Gray-Market GLP-1s vs Compounded: What Testing Misses

Why This Market Exists

The arithmetic that opens this market is not subtle. The branded semaglutide and tirzepatide pens list near or above a thousand dollars a month before discounts, and insurance frequently declines to cover them for weight management. Compounded versions from telehealth clinics run a few hundred dollars a month. A vial of "research use only" tirzepatide from an internet supplier can cost around $44. Anyone with a spreadsheet can see why the third option gets typed into a search bar, and pretending otherwise is where most articles on this subject lose their reader in the first paragraph.

Two other forces built the habit. Both drugs spent long stretches of the past few years on the FDA drug shortage list, which is the condition under which compounding pharmacies were permitted to prepare copies of them, and those years trained a very large number of people to think of GLP-1s as something you buy outside the brand-name channel. When the shortages were declared resolved and the room for compounded copies narrowed, some of that population kept moving down the price ladder. Add ordinary prescription friction, an intake form, a provider review, a real medical record, and the one-click vial site starts to look efficient rather than suspicious.

The coverage of this market splits into camps that talk past each other. News stories moralize about dangerous internet drugs without engaging the price gap that drives the behavior. Clinic blogs lean on fear. Forum threads do careful per-milligram math and skip the questions that math cannot capture. This page tries to do the whole job: define the three tiers precisely, take the tested-vial argument seriously enough to answer it technically, and be equally honest about the limits of the tier PeRx itself operates in.

The Three Tiers, Precisely

Most confusion in this conversation comes from blurring three legally distinct things into two. "Gray market" and "compounded" are not synonyms, and neither one is a discount version of the brand. Here is the actual map.

Two boundary lines deserve emphasis because marketing on both sides blurs them. First, compounded semaglutide is not a generic and it is not "the same thing as Ozempic but cheaper." The published trial results belong to the approved products; a compounded preparation has never been through the review that generated them. Second, a gray-market vial is not "compounding without the paperwork." A 503A pharmacy is a licensed facility preparing medicine for a named patient under a prescriber's order. A research-chemical seller is outside that system entirely, which changes everything about accountability even before you ask what is in the vial.

U.S. Food & Drug Administration, "FDA's Concerns with Unapproved GLP-1 Drugs Used for Weight Loss." Content current as of June 15, 2026. View study

One more data point for calibration: even the fully regulated tier has been attacked. In 2024 the World Health Organization issued a formal alert after falsified Ozempic was found in the regulated supply chains of Brazil, the United Kingdom, and the United States. If counterfeiters reached the channel with serial numbers and batch verification, the channel with neither has no floor at all. That is the context the rest of this page sits in.

World Health Organization, "Medical Product Alert N°2/2024: Falsified OZEMPIC (semaglutide)." June 19, 2024. View study

What "Research Use Only" Legally Means

Under the Food, Drug, and Cosmetic Act, a new drug cannot be introduced into interstate commerce for use in humans without FDA approval. Chemicals sold "for research use only" ride an exemption meant for laboratory work: reagents supplied to institutions running experiments, not medicine supplied to people. A seller who markets semaglutide for human use without approval is selling an unapproved new drug, and the FDA has issued warning letters to sellers of "research" GLP-1 products on exactly that basis.

So the label is doing legal work, not describing the product. "Not for human use" is how a seller ships an injectable drug to the public while formally disclaiming that anyone will inject it. It shifts liability: if the vial is contaminated, under-filled, over-filled, or something else entirely, the seller's position is that you used the product against its stated purpose. The label is not a purity grade, not a quality tier, and not a wink. It is the seller's exit from responsibility, printed on the box.

None of this makes the buyer a criminal target in practice; enforcement lands on sellers, not individuals. But it does mean the transaction happens in a space where nobody owes you anything. The general version of this argument, covering every peptide sold as a research chemical rather than just GLP-1s, is worked through in research peptides vs prescription peptides Read nextResearch Peptides vs Prescription Peptides (2026)An honest comparison for the patient who already uses peptides — what the 'Not for Human Use' label actually means, the real quality differences, and the practical steps to switch from research-chemical sourcing to a prescription supply.Read article . This page stays on the GLP-1-specific question, because GLP-1 buyers have an answer to the legal point that deserves its own section: testing.

The Tested-Vial Argument, Taken Seriously

The strongest case for the gray market is not naive. It goes like this: "I do not trust the label either. I send my vials for third-party testing, the certificate of analysis comes back over 99 percent pure, and the identity matches. At that point I know more about my vial than most patients know about theirs, so what exactly is the pharmacy premium buying?" This argument deserves a technical answer rather than an eye roll, because parts of it are correct. A real COA on a real sample is genuine information. The problem is the size of the question it answers.

What a COA actually proves

A certificate of analysis for a submitted vial typically confirms two things: that the molecule is what it claims to be, and what quantity and purity the tested sample contained on the day it was tested. That is the entire claim. It is a statement about one physical object at one moment. Used that way, it is honest data, and it is better than the nothing that most gray-market buyers operate on.

What it cannot prove

Start with the batch problem. Your tested vial came from a fill run of hundreds or thousands of vials, filled by a seller with no obligation to fill them consistently. A regulated pharmacy operation exists precisely to make vial ten and vial nine hundred the same; an unregulated filler faces no consequence when they are not. Testing one vial from a batch tells you about that vial. Extending the result to the vial you inject next week is an assumption, not a measurement.

Then the next-order problem, which is worse. The gray market has no continuity of supply. The seller who shipped the batch your clean COA came from may source different powder next quarter, or be a different operation wearing the same website. Published COAs are routinely recycled across batches, and there is no authority to complain to when the paper does not match the product. Each order resets your knowledge to zero, which means the honest version of the testing strategy is testing every order, forever, and most buyers quietly stop after the first good result.

Then the biology the standard tests do not see. Purity by chromatography is a chemistry claim. It does not tell you whether the vial is sterile, whether it stayed sterile through an uncontrolled shipping chain, or what the endotoxin load is, and endotoxin matters because it survives conditions that kill bacteria and it goes wherever the injection goes. A vial can be chemically clean and biologically unfit to inject, and the mixing step the buyer performs at the kitchen counter adds one more uncontrolled variable after the testing is done.

Finally, the testing layer itself is unregulated. No agency accredits consumer peptide-testing services, audits their methods, or penalizes a wrong result. Some are competent. You have no institutional way to know which, so the strategy of distrusting labels ends by trusting a different unregulated party. And one thing a clean COA cannot do under any circumstances is change the legal status of the transaction: selling an unapproved injectable drug for human use does not become legal because the drug was pure.

What buyers actually received when someone checked

In 2024, researchers at the University of Pecs and UC San Diego did the experiment the spreadsheet assumes has a happy ending. They ordered semaglutide from online sellers that required no prescription. Of six orders, three never arrived at all; those sellers came back demanding another $650 to $1,200 in fake customs fees. The three vials that did arrive contained substantially more semaglutide than their labels claimed, 29 to 39 percent over, while measured purity collapsed to between 7 and 14 percent against an advertised 99 percent. One sample also showed elevated endotoxin. Every delivered product failed the standard checklist used to distinguish genuine pharmaceuticals from counterfeits.

Ashraf AR, Mackey TK, Schmidt J, et al. "Safety and Risk Assessment of No-Prescription Online Semaglutide Purchases." JAMA Netw Open, 2024. PMID 39093567. View study

What 503A Compounding Actually Involves

A 503A compounding pharmacy is a state-licensed facility that prepares medications for individually named patients against prescriptions from licensed providers. It operates under its state board of pharmacy, follows recognized sterile-preparation standards for injectable products, assigns each preparation a beyond-use date rather than an open-ended shelf life, and keeps dispensing records tied to your name. If something is wrong with what it dispenses, there is a prescriber who ordered it, a pharmacist who prepared it, a board that licenses both, and a paper trail connecting all of it to you. That recourse chain, more than any single quality step, is what the tier is.

U.S. Food & Drug Administration, "Human Drug Compounding." Compounded drugs are prepared by licensed pharmacies under FDA and state board of pharmacy oversight. View study

This is the tier PeRx operates in. Compounded semaglutide is $249 per month and compounded tirzepatide is $399 per month, each prescribed by a licensed provider after an intake review and compounded by an FDA-registered 503A pharmacy against your individual prescription, with vitamin B12 co-formulated in the same vial. One vial covers 28 days, the next cycle only ships after a provider reviews how the current one went, and your PeRx provider will prescribe an optimal protocol rather than leaving titration math to you. Every vial ships fully reconstituted and ready to use, which removes the buy-your-own-supplies mixing step entirely; what PeRx publishes about its own testing process is on the quality testing page.

The honest limits of this tier

Now the part a clinic selling compounded GLP-1s is usually quieter about. Compounded semaglutide and tirzepatide have not been through FDA review. No agency evaluated the specific compounded formulations for safety or effectiveness before they reached patients, and the FDA does not endorse compounded versions of these drugs. The trial results that made this drug class famous were generated by the approved products, and borrowing them for a compounded preparation is exactly the blurring this page criticizes in the other direction.

The adverse event record is public and worth stating plainly. As of May 31, 2026, the FDA had received 990 adverse event reports associated with compounded semaglutide and more than 730 associated with compounded tirzepatide, with likely underreporting since 503A pharmacies are not required to submit them. Many reports trace to dosing errors: patients drawing the wrong volume from multi-dose vials, and in some cases prescribers miscalculating. The agency has also warned that some compounders used salt forms of semaglutide, which are different active ingredients from the approved drug. Poison-center data tells the same story at larger scale: over ten thousand GLP-1 exposure calls to US poison centers from 2012 through 2023, climbing steeply after the 2021 weight-loss approval, and mostly unintentional therapeutic errors rather than product contamination.

Miller J, Miller R, Varney SM, Han D. "National Poison Center Trends in GLP-1 Receptor Agonist Exposures Following FDA Approval for Weight Loss." J Med Toxicol, 2026. PMID 41634285. View study

The Decision Framework

Strip the moralizing out and the decision reduces to three questions. What are you actually buying at each tier? What does each tier really cost per month? And who answers when something goes wrong?

What you are buying. At the brand tier, you are buying the exact product the evidence describes, in a sealed pen, at the highest price. At the compounded tier, you are buying a legal prescription product made for you by a licensed pharmacy, with verified concentration and a recourse chain, minus FDA review of the preparation itself. At the gray tier, you are buying powder of unknown provenance plus a probability distribution: maybe accurate, maybe 39 percent over label, maybe never shipped. A COA narrows that distribution for one vial and leaves it untouched for the next.

What it costs. The $44 vial is not a $44 month. Price in the syringes and mixing supplies bought separately, the testing if you actually run it on every order rather than once, the orders that arrive wrong or never arrive, and the time spent re-vetting sellers each cycle; the study above lost half its orders to outright scams. The number climbs toward the legal tier faster than the spreadsheet suggests, and every dollar of the gap that remains is buying the absence of oversight. The legal compounded route at PeRx is $249 or $399 per month, flat, with the prescription, the pharmacy, the ready-to-use vial, and the 28-day provider review inside the price. The brand tier sits near or above a thousand a month without coverage, which is the gap that created this whole market.

Who answers. This is the question the spreadsheet cannot hold. If a compounded vial is wrong, you have a pharmacy accountable to its state board, a prescriber accountable to a medical board, an FDA reporting channel, and a medical record any emergency clinician can see. If a gray-market vial is wrong, you have a seller whose entire legal posture is that the product was never for you. People who have never needed recourse price it at zero. The 990 adverse event reports in the compounded column are, among other things, proof that a reporting channel exists there; the gray market's equivalent number is unknowable because there is nowhere for the reports to go.

Two adjacent notes for the price-motivated reader. If the monthly cost of staying on therapy is the real pressure, the lever with actual evidence behind it is the dose conversation, covered in GLP-1 maintenance dosing and microdosing Read nextGLP-1 Maintenance and Microdosing After Goal WeightReach your goal weight on semaglutide or tirzepatide and the advice splits in two. The label and most medical sites say stay on your full dose indefinitely. Half of Reddit says stretch it to every ten days or shrink it to a fraction, and nobody with a prescription pad seems willing to discuss either. This page discusses both: what people actually do in maintenance, what the pharmacology and the real-world data say about each path, and why every one of these changes belongs on a prescription rather than in a spreadsheet.Read article , not a riskier supply channel for the same dose. And if the spreadsheet exists because you are budgeting across several compounds, read the evidence ranking in peptides to take with a GLP-1 Read nextPeptides to Take With a GLP-1: The EvidenceSearch this question and you get a stack. Every result recommends four or five peptides, none of them ranks the options, and none of them tells you which ones have human data behind them. This page sorts the same list into three tiers and says out loud which tier each one lands in.Read article before adding anything; most of the add-ons marketed alongside GLP-1s have far weaker cases than the GLP-1 itself.

Common Questions