Skip to content
PeptideHormone

Can a pharmacy compound it? How the peptide rules work

Every compounding headline leaves the same question open: can a pharmacy lawfully make this today? The answer never depends on the news event. It depends on four tests in federal law, and a molecule must pass all of them. BPC-157 fails at one, semaglutide at a different one, which is why they are different stories. What 503A and 503B are, the three ways an ingredient qualifies, why leaving Category 2 put a peptide nowhere, how a shortage switches the copy rule off, and four questions that sort any future headline.

9 min read · reviewed November 2026

The question under every headline

“FDA panel backs BPC-157.” “Peptides removed from the do-not-compound list.” “Compounded semaglutide is over.” Headlines like these arrive every few months, and each one leaves a reader to guess at the same thing: can a pharmacy lawfully make this for a patient today, yes or no?

The answer is never in the headline, because it does not depend on the news event. It depends on a short series of tests written into federal law, and a molecule has to pass all of them. A compounded drug is, by definition, one that has not been through FDA approval. It is legal only because the law carves out an exemption, and the exemption has conditions. Learn the conditions once and most of the headlines answer themselves. This piece is the standing explainer; the door swings both ways covers what moved in 2026.

Two kinds of compounder, two sections of law

The exemptions live in two sections of the Federal Food, Drug, and Cosmetic Act, and the first thing to ask of any claim is which one it is about.

  • Section 503A: the pharmacy. a licensed pharmacist or physician making a preparation for one identified patient, against a valid prescription. This is traditional compounding. It is regulated mainly by state pharmacy boards, and it is exempt from new-drug approval and from manufacturing-grade quality rules
  • Section 503B: the outsourcing facility. a registered facility that may make larger batches and need not hold a prescription for each patient. In exchange it registers with the FDA, is inspected by it, and must meet manufacturing-grade quality rules. The category was created in 2013

The two have separate ingredient lists and separate copy rules (21 U.S.C. 353a, 353b). A statement that is true of one is often false of the other, which is why “it can be compounded” with no section attached tells you very little. Most of the peptide argument is a 503A argument, so that is the path traced here.

The four gates

For a peptide to be lawfully compounded by a pharmacy, it has to get through four gates in turn. Failing any one is enough.

140 amino acids or fewer?longer chains are biologicsSTOPS HEREhCG, growth hormone2Does the ingredient qualify?monograph · approved drug · 503A listSTOPS HEREBPC-157, TB-5003Not a copy of a marketed drug?unless that drug is in shortageSTOPS HEREsemaglutide, tirzepatide4A patient and a prescription?licensed pharmacist or physicianSTOPS HEREresearch-use-only salesMay be compounded for that patient
The pharmacy (503A) path. A molecule must clear every gate; the names on the right are where each one stops today.

The gates are worth taking one at a time, because the two molecules most people ask about fail at different ones. BPC-157 stops at the ingredient gate. Semaglutide clears that gate easily and stops at the copy gate. Treating the two as the same story is the most common way to misread the news.

Gate one: is it small enough to be a drug at all?

Federal law draws a line between drugs and biological products, and since 2020 the FDA has drawn it for amino-acid chains by length. A chain with a defined sequence of more than 40 amino acids is a protein, and a protein is a biological product. Neither compounding exemption covers biological products.

That one rule removes a whole shelf from the discussion. Human chorionic gonadotropin, growth hormone and insulin are all on the far side of the line. Most of the molecules in the research-peptide conversation are short enough to stay on the drug side, which is why the argument about them is fought at the next gate instead.

Gate two: does the ingredient qualify?

A pharmacy compounds from a bulk substance, the raw active ingredient. Under 503A that substance has to qualify in one of three ways:

  • It has a monograph. an official quality standard in the United States Pharmacopeia or the National Formulary
  • It is a component of an approved drug. the same active ingredient that sits inside a product the FDA has approved
  • It is on the 503A bulks list. a list the FDA builds by regulation, after consulting its advisory committee and the Pharmacopeia

Semaglutide and tirzepatide get through on the second route. BPC-157, TB-500, MOTS-c, epitalon and the rest of the research shelf have no monograph and sit inside no approved drug, so their only way in is the third: the list.

The list is short and it moves slowly. The first final rule, in 2019, placed six substances on it (FDA). Many more had been nominated, and reviewing them one at a time through rulemaking takes years. So the agency set up an interim policy to say what it would and would not act against while the queue was worked through. That interim policy is where the categories come from.

What the categories were, and what they were not

Under the interim policy, a nominated substance was sorted into one of three bins while it waited:

  • Category 1. nominated with enough information to evaluate, and no safety concern identified so far. The FDA said it did not intend to act against pharmacies compounding with these while review continued
  • Category 2. nominated with enough information, but the FDA identified significant safety risks. No tolerance: the agency said it would consider action against compounding with these
  • Category 3. nominated without enough information to evaluate. No tolerance either

In September 2023 the FDA moved a group of popular peptides, BPC-157 among them, into Category 2. That is the event usually described as the peptides being “banned.”

The distinction that matters

A category was never a legal status. It described how the FDA intended to use its enforcement discretion toward a substance that was already not on the list. Only Category 1 offered any shelter. Being in Category 2, being in Category 3 and being in no category at all have the same practical meaning: the ingredient does not qualify, and a pharmacy that uses it has no protection.

This is why April 2026 changed less than it appeared to. The nominators of twelve Category 2 peptides withdrew their nominations, and the FDA took those peptides out of the category (FDA). Many reports called this removal from the “do-not-compound list.” But a substance that leaves Category 2 does not land in Category 1, and it does not land on the bulks list. It lands nowhere. It still fails gate two. The FDA has also said it will not sort new nominations into categories at all, so the interim bins are winding down, and the only question left for any substance is the real one: is it on the list?

How something gets onto the list

There is one route, and it has fixed steps. The FDA reviews the substance against criteria that include how well it is chemically characterised, what is known about its safety, what evidence exists that it works, and how long it has been used in compounding. It consults the Pharmacopeia and its Pharmacy Compounding Advisory Committee. It publishes a proposed rule. The public comments. It publishes a final rule. Only the last step changes what a pharmacy may do.

The July 2026 committee meeting was the consultation step. The committee recommended six peptides for the list, by narrow margins and against the advice of the FDA’s own reviewers, and declined a seventh (docket FDA-2025-N-6895). The agency is free to follow the recommendation, reject it, or take its time. For the molecule with the most riding on it, the evidence the reviewers were weighing is laid out in the trial that hasn’t reported.

Gate three: is it a copy of something already sold?

An ingredient that qualifies can still be stopped here. A pharmacy may not compound, regularly or in inordinate amounts, what is essentially a copy of a commercially available drug. The reasoning is simple: the exemption exists for patients whose needs an approved product cannot meet, not as a second supply of the approved product itself. Outsourcing facilities face a stricter version of the same rule.

The rule has a switch built into it. A drug on the FDA’s shortage list is not treated as commercially available, so a copy of it is not a prohibited copy. When demand for semaglutide and tirzepatide outran supply, both went on the shortage list, the copy gate swung open, and a large compounding industry formed in the gap. Tirzepatide came off the list in December 2024 and semaglutide in February 2025. After short grace periods the gate closed again.

Nothing about the molecules changed, and nothing about their safety was decided. The legal basis was the shortage, and the shortage ended. What survives is the narrow case the rule always allowed: a preparation that differs from the approved product in a way that matters for a particular patient, as their prescriber determines. In April 2026 the FDA also proposed keeping semaglutide, tirzepatide and liraglutide off the outsourcing-facility ingredient list, which would close the 503B route separately.

Gate four: is there a patient and a prescription?

The last gate is the oldest. Pharmacy compounding is for an identified individual, on a valid prescription, by a licensed pharmacist or physician. This is the gate that the phrase “research use only” is written to avoid. Material sold under that label is not claiming to be a compounded drug at all. It is sold as a laboratory chemical, outside the pharmacy system, with no prescriber, no pharmacist and none of the quality obligations either section imposes. That is a different legal position from a compounded preparation, and a weaker one for the buyer: see cataloged vs. reachable and how to read a certificate of analysis.

Four questions for the next headline

With the gates in hand, any compounding story can be sorted in under a minute.

  • Which section?. A change to the outsourcing-facility list says nothing about what a neighbourhood compounding pharmacy may do, and the reverse
  • Which gate?. An ingredient story (lists, categories, committee votes) and a copy story (shortages, grace periods) are different problems with different fixes
  • Advice, proposal or final rule?. A committee vote is advice. A proposed rule is an invitation to comment. Only a final rule, published and in effect, moves a gate
  • Did the substance arrive anywhere?. Leaving a restricted bin is not the same as joining the permitted list. Ask where it is now, not where it used to be

Run 2026 through those four and the year reads plainly. The semaglutide story was a copy-gate story, settled when the shortage ended. The research-peptide story is an ingredient-gate story, and it is not settled: the peptides left Category 2, drew a favourable advisory vote, and when this page was reviewed still sat on no list. Each molecule’s current standing is on its monograph, with the date it was last checked and a link to the record it was checked against.

The short version

  • A compounded drug is an unapproved drug with a legal exemption. The exemption has conditions, and every one of them has to be met.
  • Four gates. Small enough to be a drug rather than a biologic; an ingredient that qualifies; not a copy of a marketed product; a real patient and prescription.
  • Three ways for an ingredient to qualify. A pharmacopeia monograph, a place inside an approved drug, or the 503A bulks list.
  • Categories were enforcement posture, not law. Only Category 1 gave shelter. Leaving Category 2 put a substance nowhere.
  • The shortage list is a switch on the copy gate. It opened GLP-1 compounding and, when the shortages ended, closed it.
  • Only a final rule moves a gate. Votes and proposals are steps toward one.

Educational reference on how United States federal compounding law is structured, summarized and simplified from the statute and FDA’s public pages. It is not legal or medical advice, state law adds its own requirements, and the rules change; check the linked primary sources for the current position. Compounds are named to explain the rules, not to endorse obtaining or using any of them.