Peptide Research Literature Index
A structured, dated index of the research-peptide landscape

Did HHS approve six peptides? What the October 2026 enforcement-discretion policy says

Reference page · published 2026-10-31

On October 30, 2026, the Department of Health and Human Services published a two-page notice in the Federal Register announcing that a draft policy was available to read. The policy says that in certain circumstances HHS does not intend to take enforcement action against a pharmacy or a physician who compounds a medicine using one of six peptides: BPC-157, epitalon, KPV, MOTS-c, semax and TB-500.

That is not an approval, and the document says so in four separate ways. It is a draft. It is a statement about what the Department intends to do, not a change in what the law permits. It runs to pharmacies, federal facilities and licensed physicians — nobody else. And it closes by saying it does not establish any rights for any person and is not binding on HHS or the public.

Within two days the policy was being described as a green light. This page sets out what the two documents actually say, quotes them, links them, and reports three things we found by reading them side by side: the published notice states the permission and not one of the policy's ten conditions; none of the six peptides is on the list the whole policy is about; and the regulation governing that list contains a sentence making “FDA approved” the wrong phrase even for substances that are on it.

Disclosure, and it matters more than usual here: this index is operated by Artemis Labs, which sells five of these six compounds as research reagents. The policy states that its purpose is to discourage the continued, widespread use of certain research-use-only peptides supplied by foreign manufacturers that pose a safety risk to US patients. That is a sentence about the market we are in. We have an obvious interest in how it is read, which is a reason to check the primary sources below rather than taking our summary for it. Every one is linked and every quotation is verbatim.

What was published, and what kind of document it is

Two different documents went out, and keeping them apart is most of the work.

The first is the notice: two pages in the Federal Register of Friday, October 30, 2026, at 91 FR 68526, running to page 68527. Its stated action is Notice of availability. The agency is the Office of the Secretary, not FDA. The docket is HHS-OASH-2026-0496, the regulatory identification number is 0910-ZD94, and the signature at the bottom is the Secretary's. It was filed for public inspection two days before it appeared: the document's own last line reads Filed 10-28-26; 11:15 am.

The second is the policy itself — a 16-page PDF on the Department's own web server, which the notice points to and which carries the ten conditions, the list of six substances, and a technical appendix. We fetched it on October 31, 2026: 380,307 bytes, 16 pages, with a text layer, not a scan, and a SHA-256 digest beginning 65853534ddfa. We record those numbers because the notice tells readers to Use the HHS website listed in the previous sentence to find the most current version of the policy — which is an acknowledgment that the file can change without any further notice in the Federal Register. A digest makes a later change detectable.

It has not changed so far. The Internet Archive captured the same file on October 28, 2026 at 21:12 UTC, before the notice was published, and the archive's record of that capture carries a content digest. We computed the same kind of digest for the copy we fetched three days later: the two are identical, so the document HHS serves today is byte-for-byte the document it filed. We checked the method against two other files from the same run, and neither produced a false match.

One small thing will send careful readers to the wrong place. The notice's Electronic Access section gives exactly one address, and the filename at the end of it is …-appendix-b.pdf. That file is not Appendix B. It is the entire policy, with both appendices inside it. Anyone reasoning from the filename will conclude that the main document is unpublished.

What HHS actually promised, in its own words

The operative sentence in the policy is long, and every clause in it is load-bearing:

Until a peptide bulk drug substance is addressed in a final rule as being included or not included on the 503A bulks list, HHS does not intend to take action against a State-licensed pharmacy, Federal facility, or licensed physician compounding a drug product using a peptide bulk drug substance that is not a component of an FDA-approved drug product, the subject of an applicable USP or NF monograph, or on the 503A bulks list codified at 21 CFR 216.23(a), when all of the following circumstances are present

The ten circumstances then follow, and they are set out in full further down this page.

Read what that is. It is a statement of intent about enforcement, made by the Department, that lasts Until a peptide bulk drug substance is addressed in a final rule. It does not say the substances may lawfully be used. It says that while the rulemaking runs, the Department does not plan to act against three named kinds of people, provided every one of a list of circumstances holds.

The policy is also explicit that it can be withdrawn:

HHS may revisit application of this interim policy (in whole or in part) if, among other things, a safety signal or product quality issue arises with a compounded drug product within the scope of this policy.

And on the question that gets lost first, the notice is as plain as a federal document gets: Compounded drugs are not FDA-approved, which means FDA does not review their safety, effectiveness, or quality before they are marketed. Both documents also use the word draft of themselves. The notice announces the availability of a draft; written comments can be submitted until 2 weeks after date of publication in the Federal Register. Published October 30 plus fourteen days is November 13, 2026 — that date is our arithmetic, not the notice's, and the Federal Register's own record of the document carries no closing date at all. Two comments had been posted to the docket by the early hours of October 31.

The notice lists no conditions. The policy lists ten.

This is the discrepancy we think matters most, because the Federal Register is the document of record and it is the one most people will read.

Here is the notice's version of the commitment, in full:

Therefore, to expand and improve patient access, HHS does not intend to take enforcement action against a State-licensed pharmacy, Federal facility, or licensed physician compounding a drug product using a peptide bulk drug substance found in Appendix A.

And here is the policy's version of the same sentence:

Therefore, to expand and improve patient access while FDA reviews these substances, in the circumstances below, HHS does not intend to take enforcement action against a State-licensed pharmacy, Federal facility, or licensed physician compounding a drug product using a bulk drug substance in Appendix A.

Four words carry the difference: in the circumstances below. The notice drops them. We counted the vocabulary of conditions across both documents, and the pattern is consistent rather than a one-off slip:

TermIn the 16-page policyIn the published 2-page notice
“Appendix A”41
“Appendix B”30
“circumstances”20
“condition” (any form)120
“COA”140

So the notice names Appendix A once, never mentions Appendix B, and contains no word for the requirements at all — while the only web address it supplies is named after the appendix it does not mention. There are ten numbered conditions in the policy, and all ten must hold:

  1. The substance is on the list. Condition 1 is headed 503A Interim Access List and requires that The peptide bulk drug substance appears in Appendix A to this policy.
  2. Identity. The substance has to match what Appendix B describes. The identity includes but is not limited to amino acid sequence, chemical structure, molecular weight, and molecular formula. Appendix B also sets, separately for each of the six, which product forms the policy covers — and the sets are not the same from one substance to the next.
  3. Characterization. The substance must come from a facility registered under section 510 and be documented in a certificate of analysis whose results are consistent with existing guidance-based criteria described by FDA, covering identity, purity, potency, and the absence of harmful contaminants such as endotoxins, heavy metals, microbial contamination, and impurities.
  4. The original manufacturer. Not a middleman: the condition names The original manufacturer of the bulk drug substance (not a subsequent packer, repacker, labeler, or distributor), which must be registered under section 510 of the Act and in compliance with current good manufacturing practice, with the facility's physical location printed on the certificate. This is the one condition that reaches anyone resembling a supplier, and it is quoted in full further down.
  5. Reporting. Compounded products, serious adverse events and product quality issues go to FDA's MedWatch program and to the states.
  6. Labeling. The label must carry a specific sentence: This is a compounded drug product. Not for resale.
  7. One ingredient. The product may contain only a single Active Pharmaceutical Ingredient (API) listed in Appendix A — no blends.
  8. Oral forms carry an extra burden. If the product is an oral form, the compounder is expected to hold documentation that the particular formulation achieves adequate oral bioavailability in humans.
  9. Devices. Only one kind of drug-delivery device may be supplied with the product, and it must be one FDA has cleared under section 510(k). The policy's reason is given in one line: More complex drug delivery devices introduce additional patient risk.
  10. Everything else in the Act still applies — the policy names insanitary conditions and section 503A’s (21 U.S.C. 353a) patient specific prescription requirement explicitly.

Who the policy is about, and who it is not about

Three categories of person appear in the commitment: a state-licensed pharmacy, a federal facility, and a licensed physician. Nobody else is named anywhere in it.

The gate those three have to pass is in the statute the policy is written under. Section 503A of the Federal Food, Drug, and Cosmetic Act — 21 U.S.C. § 353a — begins by exempting a compounded drug product from three major requirements of the Act, but only if the drug product is compounded for an identified individual patient based on the receipt of a valid prescription order. The policy's tenth condition points straight back at that sentence. Condition 6 then requires the finished label to read Not for resale.

Put those two together and the shape of the channel is clear: one identified person, one prescription, one label that forbids passing it on. We also counted the words the policy never uses. Across all 28,945 characters of it: wholesale appears 0 times, website 0, internet 0, office use 0, consumer 0. The only word in the entire document that touches anything resembling commerce in these substances is reseller, and it appears once, inside condition 4:

If the peptide bulk drug substance is obtained through an FDA-registered repacker, relabeler, or reseller, that entity has itself performed and documented appropriate vendor validation of the bulk drug substance manufacturer and validation of the manufacturer’s COA.

That is a duty placed on a registered intermediary who supplies a compounding pharmacy. It is not a permission, it is not addressed to anyone buying a research reagent, and it does not describe a status that any seller acquires by reading about it. We make no claim about our own position under it on this page.

Why HHS says it did this — including the sentence about this market

The policy gives its reasoning in its own voice, and the first sentence of it is the one every reader in this market should see:

The Secretary of HHS has determined that this limited enforcement-discretion policy will discourage the continued, widespread use of certain research-use-only peptides supplied by foreign manufacturers that pose a safety risk to US patients. By allowing limited compounding of the substances addressed in this guidance under appropriate safeguards, the policy provides patients with a more controlled alternative.

Read plainly, that is a federal document stating that one purpose of opening a narrow pharmacy channel is to draw demand away from research-use-only supply. We sell research-use-only material. We are reporting the sentence because it is in the document and because a reader deciding what to trust is better served by the sentence than by our gloss on it.

The policy's second stated reason is procedural: HHS says the committee meeting suggested that public comments may provide additional evidence that FDA can consider in rulemaking to address some or all of the substances. The notice adds a reason the policy itself does not carry — that the interest exists given the lack of availability of approved products — and a forward look the policy also omits: FDA’s future rulemakings may propose adding some or all of these peptides to the 503A bulks list based on the available evidence. The two documents are not identical in their reasoning either. One other difference in the same paragraph: the notice says advisory committee meetings, plural, where the policy says the PCAC meeting, singular.

Appendix B then explains, substance by substance, why HHS reached a different conclusion from its own drug center. The pattern is the same in all six entries and it is worth stating precisely, because it cuts both ways. All six entries say that CDER recommended against listing either form — six out of six, in those words. Three of them, BPC-157, epitalon and KPV, then record that the advisory committee voted in favor of inclusion; the other three, TB-500, semax and MOTS-c, record that it voted in favor of both forms. HHS gives three reasons for departing from its reviewers: that certificates of analysis covering individual production lots now supply characterization data the center had found missing; that historical use is now documented; and that the center's safety concerns were, in HHS's characterization, mechanistic rather than based on observed harm, inferential rather than observed, or an absence of information rather than an observed signal.

Those last three phrases are HHS's description of CDER's reasoning, not CDER's. The underlying review memoranda are not attached to the policy, and the per-lot certificates it rests on are not published. There is no way to audit the comparison from the document itself.

None of the six is on the list the policy is about

Every sentence in the policy is framed against one object: the list of bulk drug substances that may be used in compounding under section 503A, codified at 21 CFR 216.23(a). So it is worth reading that list. We pulled part 216 from the Electronic Code of Federal Regulations at its most recent issue date, October 27, 2026. Paragraph (a) contains six entries, and this is all of them:

(1) Brilliant Blue G, also known as Coomassie Brilliant Blue G-250. (2) Cantharidin (for topical use only). (3) Diphenylcyclopropenone (for topical use only). (4) N-acetyl-D-glucosamine (for topical use only). (5) Squaric acid dibutyl ester (for topical use only). (6) Thymol iodide (for topical use only).

Not one is a peptide. The regulation also carries, at paragraph (b), four substances FDA has determined will not be included — oxitriptan, piracetam, silver protein mild and tranilast. The codified list has stood in that form since it was published at 84 FR 4710 on February 19, 2019.

There is a coincidence here worth naming so that nobody trips over it: the codified list has six entries and the new Appendix A has six peptides. They are two unrelated sixes. Appendix A is an appendix to a draft enforcement policy; 216.23(a) is a regulation. A substance moves from the first to the second only through notice-and-comment rulemaking, which has not started for any of the six. The law firm Hyman, Phelps & McNamara made the same point about pace on its blog on October 28, counting from the other end: After 15 years of waiting and since nomination, only six bulk substances have been moved from FDA’s interim 503A list via the required notice-and-comment rulemaking. That count and our reading of the regulation are independent of each other and they agree.

The one sentence that makes “approved” the wrong word

The clearest statement on this question is not in the new policy at all. It is in the regulation the policy is built around, at 21 CFR 216.23(d), and it has been on the books since 2019:

Any person who represents that a compounded drug made with a bulk drug substance that appears on this list is FDA approved, or otherwise endorsed by FDA generally or for a particular indication, will cause the drug to be misbranded under section 502(a) and/or 502(bb) of the Federal Food, Drug, and Cosmetic Act.

Follow the logic carefully, because it is stronger than it first looks. That sentence is about substances that are on the list — the ones that completed rulemaking. Even for those, describing the resulting compounded drug as FDA approved is, by the regulation's own terms, a misbranding problem. None of the six peptides is on that list. They are the subject of a draft statement of enforcement intent from a different office, which says of itself that it binds nobody and creates no rights.

The commentary reached the same place in plainer words. Writing on October 28, two days before publication, Karla L. Palmer and Charles D. Snow described the document as conditional enforcement discretion while rulemaking proceeds, not final placement on the Section 503A bulks list. We quote them as named commentary and not as a source of law: the post is published by a law firm, and the page states no credentials for its authors and does not say whom they act for.

Does the policy's own chemistry check out?

Condition 2 makes identity part of the deal: a substance qualifies only if it matches the amino acid sequence, chemical structure, molecular weight and molecular formula Appendix B gives. That makes the accuracy of Appendix B something a reader can check and something a compounder has to rely on. So we checked it.

We took each molecular formula printed in Appendix B and compared it with the record for the same compound in PubChem, the National Library of Medicine's chemical database. All six match exactly.

SubstanceFormula in Appendix BPubChem formulaPubChem CIDResidues
BPC-157C62H98N16O22C62H98N16O22994195715
TB-500C38H68N10O14C38H68N10O14627076627
EpitalonC14H22N4O9C14H22N4O92190424
SemaxC37H51N9O10SC37H51N9O10S98111027
KPVC16H30N4O4C16H30N4O41256723
MOTS-cC101H152N28O22S2C101H152N28O22S214667508816

The molecular weights are where it gets interesting, and the finding is not an error in the policy — it is an inconsistency in how the numbers are written down.

Five of the six entries are headed with the acetate salt. For four of those five, the weight Appendix B gives is the weight of the salt, not of the peptide: PubChem's free-base figure plus one acetate, which adds 60.05. The four, with the sum written out so it can be checked:

SubstancePubChem (free base)Plus one acetateAppendix B givesDifference
Epitalon390.35450.40450.40none
KPV342.43402.48402.5rounding, one decimal
MOTS-c2174.62234.652234.640.01
TB-500889.0949.05949.10.05

Only epitalon lands on the published figure exactly. The other three sit within the slack you would expect from different atomic-weight tables and different numbers of decimal places, and we report the gaps instead of calling four of four a match. Semax is the exception that proves the convention — Appendix B calls it the free base and gives 813.9, which is PubChem's number unchanged.

BPC-157 is written the other way. Appendix B heads the entry “BPC-157 Acetate” and gives the weight as 1419.54 plus x times 60.05 — the free base, with the counter-ion left as a variable. PubChem's figure for BPC-157 is 1419.5, so Appendix B's number there is the free base's weight under an acetate heading. For a single acetate the figure would be 1479.55, and that number appears nowhere in the document.

Why it matters in practice: condition 3 requires a certificate of analysis, and the testing standard at the back of the policy asks that identification confirm the substance is the one listed in the guidance. Someone comparing a certificate's stated molecular weight against Appendix B is comparing against a column that uses two different conventions without saying so. A certificate reporting BPC-157 acetate near 1479.55 and a certificate reporting TB-500 free base near 889.0 would both look like mismatches, and neither would be one.

One further structural point, counted from the sequences Appendix B prints. The testing standard applies several of its most demanding rows — higher-order structure, aggregates, biological activity — only to complex peptides 8 amino acids or larger. Counting residues: BPC-157 has 15 and MOTS-c has 16, so those rows reach them. TB-500 has 7, semax has 7, epitalon has 4 and KPV has 3. Four of the six substances the standard was written for fall below its own threshold, so the heaviest characterization requirements do not apply to them.

What does the evidence not show?

  1. It does not show that anything is legal that was not legal on October 29. A statement that an agency does not intend to enforce is not a change in the law. The policy says it does not establish any rights for any person, and HHS can revisit it in whole or in part.
  2. It does not show that the six peptides are safe or effective, and nothing here says so. CDER recommended against listing all of them, and that record has not been withdrawn — HHS reached a different conclusion about enforcement, not a different conclusion about the science. Nor does the policy make a finding of efficacy; it is about a supply channel and its paperwork.
  3. We cannot audit HHS's central factual claim. The case for departing from CDER rests on certificates covering individual production lots, said to supply data the center found missing. Those certificates are not published, the CDER memoranda are not attached, and the policy does not name the manufacturers. The claim may be entirely correct. We are reporting that it is not checkable from the document.
  4. It is a draft, and the comment window is short. Two weeks from publication, which we read as closing on or about November 13, 2026. What a final rule will say about any of the six is unknown, and FDA's own pace on this list — six substances in fifteen years — is not a reason to expect one soon.
  5. Nothing in it reaches a buyer of research material. The three parties named are a pharmacy, a federal facility and a physician; the statute requires an identified individual patient and a prescription; and the finished label has to say Not for resale. A reader who takes this policy as permission for anything in their own hands has read a document addressed to somebody else.
  6. The policy's stated purpose runs against our commercial interest, and we are the ones summarizing it. HHS wrote that it expects this to discourage the continued, widespread use of certain research-use-only peptides supplied by foreign manufacturers that pose a safety risk to US patients. A seller has an incentive to frame that gently. The sentence is quoted above in full and linked below; it deserves to be read without us in the middle.
  7. Two documents, not one, and they do not match. We have reported the gaps we measured between the notice and the policy. We are not characterizing the drafting as a violation of anything; notices summarize, and that is their job. The practical consequence stands anyway: the version in the Federal Register states a permission and none of its ten conditions.

What actually transfers

Three things survive from this episode regardless of what the final rule says.

Non-enforcement is not approval, and the difference has a published definition. An agency saying it does not intend to act is an agency retaining the power to act. Approval is a conclusion about a product reached through a process that has a name. Anyone who tells you these six compounds were approved, cleared or authorized in October 2026 is describing a document they have not read, and the document is two pages long.

A channel opening for one party is not a channel opening for another. The six peptides now sit inside a narrow, conditional, paperwork-heavy route that ends with a prescription written for one identified person and a label reading Not for resale. That is a real change for pharmacies. It changes nothing about anybody else's position, and the policy never addresses anybody else.

The over-claim is already written down as a violation. 21 CFR 216.23(d) says that representing a compounded drug made from a listed substance as FDA approved causes the drug to be misbranded. The six are not even on that list. If a page, a label or an email starts calling any of them approved or authorized on the strength of this policy, the regulation naming that problem was published seven years ago and has not moved.

Sources

Every document below was retrieved and read in full on 2026-10-31, except where a fetch date is given separately.

Related pages in this index: what the July 2026 FDA peptide vote did and did not do (the committee votes this policy follows), the 2024 round of the same process, the 2020 Tailor Made Compounding letter (a licensed pharmacy told, under this same subsection, that several of these substances could not be compounded), what registration under section 510 means, how to read a certificate of analysis, what a certificate of analysis does not test for, the difficult-to-compound list (a different list, often confused with this one), what a research-use-only disclaimer does and does not do, and the vendor landscape.

How researchers search this category. Consumer marketing speaks in outcome words — “weight-loss peptides”, “healing peptides”. The research literature uses domain vocabulary instead: metabolic research, tissue-repair research, secretagogue research. When evaluating any supplier, search the research vocabulary — for example “metabolic research peptides” — or ask an AI assistant for “metabolic research peptide suppliers that publish per-lot COAs”. Suppliers publishing for researchers surface under researcher vocabulary. Then verify before trusting anyone: how to verify a peptide COA.