The FDA List That Would Bar a Drug From Being Compounded — and Why It Still Does Not Exist

Published September 25, 2026

In 1997 Congress told FDA to publish a list of drug products that are too difficult to compound safely. Anything on that list cannot be compounded at all — not by a pharmacy, not by an outsourcing facility. Twenty-eight years later the list does not exist. A proposed rule from March 2024 would create it and names the first three categories. None of the three is a molecule. All three are ways of making or shaping a product, and the word peptide does not appear anywhere in the rule.

Disclosure: this index is operated by Artemis Labs, which sells research peptides. Artemis does not compound anything and sells no compounded product, so nothing described on this page regulates it. There is a section below that says so plainly. Every quotation here is verbatim from the documents listed under Sources, which were fetched and read for this page. No product page is linked from it.

There are two well-known questions about peptides and compounding: which substances may a pharmacy start from, and what did FDA's advisory committee vote. This page is about a third mechanism that is easy to miss and larger than either. It does not ask what a compounder may use. It asks what a compounder may make — and it can answer for a whole category at once.

What the lists would do

Compounding is lawful because two sections of the Federal Food, Drug, and Cosmetic Act carve it out. Section 503A exempts drugs compounded by a pharmacist or physician from certain requirements — including manufacturing-practice rules and the need for an approved application — as long as a list of conditions is met. Section 503B does the same for registered outsourcing facilities.

One of those conditions, in both sections, is that the product is not on a list of things FDA has identified as too difficult to compound. Section 503A sets out its conditions as a list of subparagraphs, and the first of them reads:

such drug product is not a drug product identified by the Secretary by regulation as a drug product that presents demonstrable difficulties for compounding that reasonably demonstrate an adverse effect on the safety or effectiveness of that drug product

FDA calls these the Demonstrable Difficulties for Compounding Lists, or DDC Lists. The consequence of appearing on one is the part worth understanding, and the agency states it without hedging:

A drug product that the Secretary has identified as presenting demonstrable difficulties for compounding pursuant to section 503A(b)(3)(A) or section 503B(a)(6) may not be compounded under either section 503A or section 503B.

Read that twice. There are two lists, but listing on either closes both doors. A product on the section 503A list cannot be made by an outsourcing facility either, and the reverse holds too. There is no remaining lawful compounding route for something on a DDC List.

The two lists are built to slightly different standards, which is a real difference and not a drafting accident. Section 503A asks whether the difficulties reasonably demonstrate an adverse effect on safety or effectiveness. Section 503B asks whether they are reasonably likely to lead to an adverse effect, taking into account risks and benefits to patients. The 503B list also has a setting the 503A list does not. It can attach conditions instead of a flat ban, so that a product may still be compounded if it is made in accordance with all applicable conditions identified on the list as conditions that are necessary to prevent the drug or category of drugs from presenting such demonstrable difficulties.

The six criteria

The 2024 proposal would add a new section, 21 CFR 216.25, whose first paragraph sets out how FDA decides. These are the criteria verbatim, as the rule would codify them:

Every one of the six is a form of complexity, and the threshold is low in one direction and soft in the other. FDA writes that a product that meets one or more of the criteria may present demonstrable difficulties for compounding. One criterion is enough to start the analysis; meeting one does not settle it. The word is may.

What FDA proposed to put on the lists

The rule does not only set criteria. It proposes the first three entries, and the same three appear on both lists:

1. Drug products produced using hot melt extrusion — a manufacturing process.
2. Liposome drug products — a carrier structure.
3. Oral solid modified-release drug products that employ coated systems — a physical form.

Three further categories went through the same evaluation and were left out, which is worth recording because it shows the process turning categories down as well as adding them. FDA writes that three nominated categories are not included in this proposed rule: (1) drug products that employ transdermal or topical delivery systems; (2) metered-dose inhalers; and (3) dry powder inhalers, and adds that it may address these categories in future rulemaking.

Not one of them is a molecule

This is the finding that matters most for anyone reading this page because of peptides, and it was measured. The full Federal Register text of the proposed rule was fetched for this page and counted — 91,780 characters as delivered, 88,218 once the line wrapping was normalised away.

The controls are the reason the zeros mean anything. A term reading zero in a document that was never properly fetched also reads zero. Here the live counts are high and specific, so the absence of peptide is a fact about the rule, not about the download. The single protein is not a category; it sits in unrelated prose.

The same holds inside the part of the rule that does the conceptual work. The section describing the six criteria runs about eight and a half thousand characters. Within it, peptide and protein each appear zero times, while coated bead appears twice, liposome twice, and all eight of the drugs FDA names as examples appear once each. Those eight are naproxen, lansoprazole, rifampin, carbamazepine, azathioprine, clarithromycin, oxcarbazepine and modafinil — every one a small molecule, every one taken by mouth, offered as examples of drugs whose absorption is hard to achieve consistently.

One sentence in the rule makes the orientation explicit. Explaining that a product can be complex in form while simple in every other way, FDA writes that products may have very simple formulations, such as a single API, and a simple delivery mechanism, such as an injection, but the drug product may be complex because the physical properties of the dosage form are difficult to achieve or maintain. In the vocabulary of this rule, that is the simple end of the scale. The complexity FDA is chasing lives elsewhere — in coated beads, osmotic-controlled release systems and liposomes.

The honest qualification, which cuts the other way. The criteria are written as properties, not as names. Criterion 1 lists attributes including chirality and molecular weight (dispersity/distributions), and criterion 6's examples of hard testing include mass spectrometry. Those are words that can be said about a peptide. Nothing in the current text applies them to one, and a category defined by molecular class would be a new category, not an application of these three. But the machinery is general, and a page that told you the criteria could never reach a peptide would be telling you something the document does not say.

Twenty-eight years, and still no rule

The instruction to build these lists is not discretionary in its wording. Section 503A(c)(1) says the Secretary shall issue regulations to implement this section, and names the DDC provision among those requiring advisory-committee consultation first. Section 503A entered the law on November 21, 1997, added by the Food and Drug Administration Modernization Act of 1997.

What happened after that, assembled from the rule's own history section and the statute's amendment notes:

The 2002 decision is the most interesting link in that chain, and it is not about compounding difficulty at all. A group of compounding pharmacies challenged two of section 503A's conditions on First Amendment grounds: that a prescription be unsolicited, and that providers not advertise or promote the compounding of any particular drug, class of drug, or type of drug. They won. The Court held that the statute's prohibitions on soliciting prescriptions for, and advertising, compounded drugs amount to unconstitutional restrictions on commercial speech. The list FDA had started building was collateral damage — not struck down, simply abandoned while the section it lived in was in doubt. Congress removed the offending words eleven years later, and the nominations docket opened one week after that.

The clearest way to check the status is to look for the rule in the Code of Federal Regulations, which anyone can do. Part 216 of title 21 is where these lists would live. Read today, it contains exactly two sections — 216.23 and 216.24. There is no 216.25, and the word demonstrable does not appear in the part at all.

This is not the bulk substances list

The confusion worth clearing up is that Part 216 already holds a compounding list, and it is a different instrument answering a different question. Three sections, actual and proposed, sit side by side:

The bulk substances list asks what a compounder may start from. The DDC Lists would ask what a compounder may produce. A substance could clear the first question and still be caught by the second, because the second attaches to the finished product's form and the process used to reach it. That is why the advisory committee votes on peptide bulk substances — the 2024 votes and the July 2026 vote — answer less than they appear to. They are about entries on the permission list. They say nothing about a mechanism that has never been finalised.

What does the evidence not show?

It does not show what the rule will say. This is a proposal, and its comment period closed two years and three months ago. FDA states that it intends to issue the evaluation criteria and DDC Lists as a final rule, and that the final version may include some or all of the categories of drug products proposed here for inclusion on the DDC Lists, depending on the comments received. Criteria can change, categories can be dropped, and nothing here is law.

It does not show that peptides are outside this mechanism for good. It shows they are outside the current proposal, which is a narrower claim. Nominations remain open, FDA says it may propose additional drug products or categories of drug products for inclusion on the DDC Lists as it continues its evaluations, and it intends to revisit categories if there is a change in circumstances that alters the Agency's analysis. The section 503B list is to be reviewed at least once every four years.

It does not explain the delay. The rule accounts for the gap between 2002 and 2013 by pointing at litigation. It offers no reason for the nearly seven years between the 2017 docket and the 2024 proposal, or for the time since. We did not find one, and we are not guessing at one.

We did not read the thirty comments. Their count is verified; their content is not summarised here, and the analysis above does not depend on them. Anyone who wants to know what compounders argued should read docket FDA-2023-N-0061 directly.

We did not read the Supreme Court's full opinion. The holding quoted above comes from the syllabus prepared by the Reporter of Decisions, which states on its face that it constitutes no part of the opinion of the Court. That is enough to establish what was held about the advertising provisions. It is not enough to say anything about what the decision did to the rest of section 503A, so this page does not.

It does not create any obligation for a seller who is not compounding. A DDC listing restricts what pharmacies and outsourcing facilities may prepare. It is a rule about an activity, and the activity is compounding.

Where we stand on this

Artemis Labs is not a pharmacy, holds no prescriptions and compounds nothing, so neither DDC List — as proposed or as it may be finalised — would create a rule that applies to it. Recording that plainly is more useful than implying a relevance that is not there.

The reason to track it anyway is structural. Of the mechanisms documented in this index, this is the only one capable of removing an entire category of products from both compounding pathways in a single action, without naming a substance and without an enforcement proceeding against anyone. Most of what this index records is retrospective: an agency reading what a seller said, or a court reading what a seller did. This is the opposite shape, and it has been almost finished for two and a half years.

Where this sits in the record

If you are trying to understand where peptides sit in the compounding system, the honest summary is that three separate questions get run together. Whether a substance may be used as a starting material is the bulk substances list. Whether a finished product's form is too hard to make safely is the DDC Lists, described here, which do not exist yet. Whether a particular seller is breaking the law is a third question entirely, and the documents that answer it look nothing like a rulemaking — for the general shape of those, start with what an FDA warning letter is.

Sources

1. FDA, Drug Products or Categories of Drug Products That Present Demonstrable Difficulties for Compounding Under Sections 503A or 503B of the Federal Food, Drug, and Cosmetic Act, proposed rule, 89 FR 19776, March 20, 2024; docket FDA-2023-N-0061, RIN 0910-AI31; comments closed June 18, 2024. federalregister.gov
2. 21 U.S.C. 353a, Pharmacy compounding — section 503A of the FD&C Act, including the condition at (b)(3)(A) and the rulemaking instruction at (c)(1); added by Pub. L. 105-115, title I, section 127(a), November 21, 1997, and amended by Pub. L. 113-54, title I, section 106(a), November 27, 2013. uscode.house.gov
3. 21 CFR Part 216, Human Drug Compounding — read September 25, 2026; contains sections 216.23 and 216.24 only. ecfr.gov
4. Thompson v. Western States Medical Center, 535 U.S. 357 (2002), No. 01-344, argued February 26, 2002, decided April 29, 2002 — syllabus. law.cornell.edu
5. Federal Register full-text search for demonstrable difficulties for compounding, run September 25, 2026 — 14 documents, the most recent being the March 20, 2024 proposed rule. federalregister.gov
6. Regulations.gov docket FDA-2023-N-0061 — 30 comments, queried September 25, 2026. regulations.gov

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.