Is compounded semaglutide or tirzepatide still legal in the US, and what changed?

The short answer

Compounded semaglutide and tirzepatide were never a permanent category. They existed because both drugs were on the FDA shortage list, and US law permits compounding a copy of an approved drug only while a shortage lasts. Both shortages are resolved — tirzepatide since 19 December 2024, semaglutide since 21 February 2025 — and the grace periods ran out in spring 2025.

Compounders sued, twice, and lost both times. In April 2026 the FDA went further and proposed removing all three GLP-1s from the list of substances outsourcing facilities may compound from bulk at all.

What still circulates is therefore either a permitted personalisation, a prohibited copy, or something sold outside the compounding system entirely.

Evidence tier: Tier 1 — FDA statements, Federal Register actions and court records. US law only; European rules are covered separately. Educational content, not legal or medical advice.

The key points:

• Shortages resolved: tirzepatide 19 Dec 2024, semaglutide 21 Feb 2025.
• Grace periods ended between February and May 2025.
• Compounders lost preliminary injunctions in March and April 2025.
• FDA proposed excluding all three from the 503B bulks list on 30 April 2026.
• Retatrutide cannot be compounded at all — there is no approved product to copy.

503A and 503B are two different things

Evidence tier: 1 — statutory framework.

Nearly every confused discussion of this topic comes from treating "compounding pharmacy" as one category. US law creates two, under sections 503A and 503B of the Federal Food, Drug, and Cosmetic Act.

The practical consequence is that these two routes fail in different ways. A 503A pharmacy compounding a copy of an approved drug is constrained by the shortage rules. A 503B outsourcing facility is constrained by both the shortage rules and whether the active ingredient appears on the bulks list — which is the lever the FDA reached for in 2026.

The shortage was the whole basis, and it ended

Evidence tier: 1 — FDA CDER statements.

Compounding what is "essentially a copy" of a commercially available approved drug is generally prohibited. The shortage list is the exception that made compounded GLP-1s possible, and the timeline is more tangled than most summaries admit.

The October-then-December sequence for tirzepatide is where most accounts go wrong. FDA resolved it, was challenged, reconsidered, and resolved it again. The operative date is 19 December 2024, not October.

Compounders sued twice and lost twice

Evidence tier: 1–2 — court records; appellate outcome unverified.

The Outsourcing Facilities Association brought parallel challenges in the Northern District of Texas — one over tirzepatide, one over semaglutide — arguing in substance that the FDA had disregarded continuing evidence of shortage and had not given compounders adequate notice or participation.

Both preliminary injunctions were denied: the tirzepatide motion on 5 March 2025, the semaglutide motion on 24 April 2025. Search results indicate the Fifth Circuit affirmed for the FDA on 27 August 2026, but we did not open the appellate record and are marking that unverified — if the outcome of the appeal matters to you, check it directly.

What the FDA said compounding may still be

Evidence tier: 1 — FDA statement of 1 April 2026.

On 1 April 2026 the FDA published a clarification of what remains permitted now that supply has stabilised. It is a restatement rather than new policy, and three points in it are worth knowing.

The "essentially a copy" test has a numeric edge. A compounded product that differs from the approved drug only trivially is a copy. The FDA's own worked example is semaglutide combined with vitamin B12: if the semaglutide strength is within 10% of the approved product's, the combination still counts as essentially a copy. Adding an inert companion ingredient does not create a different drug.

There is a volume threshold for 503A. The statement sets out an enforcement position around compounding no more than four or fewer prescriptions per calendar month of a drug that is essentially a copy, in the circumstances described. That is a narrow clinical-need allowance, not a business model.

And the closing line settles the status question: tirzepatide and semaglutide appear on neither the 503B bulks list nor the shortage list.

A genuine clinical need — a documented allergy to an excipient in the commercial product, or a dose the approved presentations cannot deliver — remains a lawful basis for compounding for an identified patient. "The branded version is expensive" is not one.

April 2026: the FDA moved to close the bulks route

Evidence tier: 1 — FDA press announcement.

On 30 April 2026 the FDA proposed to exclude semaglutide, tirzepatide and liraglutide from the 503B bulks list, on the basis that there is no clinical need for outsourcing facilities to compound them from bulk substances. The proposal was published in the Federal Register as document 2026-08552, with comments due 29 June 2026.

This matters because it is belt-and-braces. The shortage route had already closed. Exclusion from the bulks list would remove the remaining bulk-compounding pathway for outsourcing facilities independently of any future shortage determination — so a renewed shortage would not automatically reopen it.

As of September 2026 we could not confirm that a final determination has been issued. The Federal Register site repeatedly timed out for us. Treat the proposal as proposed until you have checked the docket.

Retatrutide is a different question entirely

Evidence tier: 1 — regulatory status.

Retatrutide cannot be compounded under either section, and the reason is more fundamental than shortage rules.

Compounding from a bulk substance requires, broadly, that the substance be a component of an FDA-approved drug, or appear on the 503B bulks list, or be the subject of an applicable USP monograph. Retatrutide satisfies none of these. There is no approved retatrutide product anywhere, and Lilly has indicated a first-quarter 2027 filing — so there is no approved drug for it to be a component of, and no shortage of an approved product to invoke.

Anything sold as compounded retatrutide is therefore outside the compounding framework, not a borderline case within it. We cover the availability question in full in can you legally buy retatrutide.

Enforcement has reached compounders directly

Evidence tier: 1 — court filings and FDA warning letters.

Two strands are worth separating.

Private litigation. Eli Lilly's six lawsuits of 12 August 2026 were not confined to research-chemical websites. The defendants included a compounding pharmacy and a medical spa alongside four research-use-only sellers — so practitioner-facing businesses are inside the enforcement perimeter, not adjacent to it. We cover those suits in our analysis of the crackdown.

FDA warning letters. The letters issued on 24 August 2026 and published on 1 September show the agency reading entire websites rather than labels. Selling bacteriostatic water alongside peptides was cited as making the water itself an unapproved drug, and in one letter a peptide dosing calculator was cited as evidence that products were intended for human use. We covered that in the tesamorelin analysis.

The safety questions that remain

Evidence tier: 2–3 — partially verified.

Two recurring problems are worth naming, with the caveat that we did not complete verification on either.

Concentration confusion. Compounded products are supplied at concentrations that differ between pharmacies and differ from the approved pens. Dose errors arise when a patient reasons in millilitres or "units" rather than milligrams, or switches suppliers mid-course. This is the same failure mode behind the accidental overdose we covered in the September signal report — where a patient on 7.5 mg estimated having injected around 21 mg after misreading pen markings.

Salt forms. Products described as semaglutide sodium or semaglutide acetate are not the same active ingredient as the semaglutide base in approved products. We flag this as an area the FDA has addressed but did not complete our verification of the agency's exact position, so confirm before relying on it.

What we could not verify

• Whether a final determination on the 503B bulks exclusion has been issued. The Federal Register site timed out repeatedly.
• The 1 May 2026 publication date of document 2026-08552.
• The Fifth Circuit outcome reported as 27 August 2026 — we did not open the record.
• FDA's precise position on semaglutide salt forms.
• Documented adverse event counts specific to compounded GLP-1s.

Limitations

This is educational content. It is not legal advice, and it is not medical advice.

• This is US law only. European personal-import and compounding rules are entirely different and are covered in our regional guides.
• The regulatory position is moving. A proposal is not a final rule, and we have said where that distinction applies.
• We describe the framework, not what to buy. Nothing here is guidance on obtaining unapproved or compounded medicines, and we name no pharmacies or vendors.
• A lawful compounded product still differs from the approved drug in ways that are not always documented.
• Marko Maal, MSc Pharmacy reviewed this article. Reviewer attribution does not constitute a doctor-patient relationship.

The bottom line

The compounded GLP-1 era rested entirely on a shortage, and the shortage is over. Tirzepatide's resolution was finalised in December 2024 and semaglutide's in February 2025; the grace periods expired by May 2025; and the trade body's two attempts to overturn those determinations both failed at the preliminary stage.

The April 2026 proposal to strip all three from the 503B bulks list is the part worth watching, because it changes the architecture rather than the facts. Removing them from the bulks list would close the outsourcing-facility route even if a shortage were ever declared again — converting a temporary closure into a structural one.

What remains lawful is narrow and clinical: a documented patient-specific need, for an identified patient, not a copy of the commercial product. The FDA has told compounders what a copy is, down to a 10% strength margin and a four-prescriptions-a-month threshold. Anything operating at scale on price alone is outside that, and the enforcement record of the last month shows both the FDA and Lilly reading the whole operation rather than the disclaimer on it.

Related on this site

• Can you legally buy retatrutide?
• Research-grade vs pharmaceutical-grade vs compounded peptides
• Lilly sues six retatrutide sellers
• Zero-peptide vials and the 2026 counterfeit wave
• How to verify a peptide vendor
• Are peptides legal? US law, importing, and by-compound status
• GLP-1 cost-saving strategies, done safely
• Peptides in Europe: import rules by region

References

• FDA, CDER. FDA clarifies policies for compounders as national GLP-1 supply begins to stabilize — shortage resolution dates, 503A and 503B grace periods, the "essentially a copy" test, the semaglutide plus B12 worked example, the 10% strength margin, and the four-prescriptions-per-month enforcement position. Statement dated 1 April 2026. FDA
• FDA. FDA proposes to exclude semaglutide, tirzepatide and liraglutide from the 503B bulks list, 30 April 2026 — Federal Register document 2026-08552, comments due 29 June 2026. Press announcement
• Outsourcing Facilities Association v. FDA — tirzepatide, No. 4:24-cv-00953 (N.D. Tex.), preliminary injunction denied 5 March 2025; semaglutide, No. 4:25-cv-00174 (N.D. Tex.), preliminary injunction denied 24 April 2025. Appellate outcome reported as 27 August 2026 (No. 25-10600) — unverified.
• Eli Lilly and Company, 12 August 2026 — six lawsuits including a compounding pharmacy and a medical spa. Investor release
• FDA warning letters, dated 24 August 2026, published 1 September 2026 — bacteriostatic water treated as a drug; peptide calculator cited as evidence of intended human use. Peptide Partners LLC · Royal Peptides LLC