On this page

• What “compounded semaglutide” actually means
• Does compounded semaglutide work?
• Is compounded semaglutide the same as Ozempic?
• What the impurity testing actually found
• Is compounded semaglutide safe?
• Why the shortage ending changed everything
• Can you still get compounded semaglutide?
• What the FDA has actually written
• The arguments, set out side by side
• Where the position is genuinely unsettled
• Is compounded semaglutide going away, and will it be banned?
• What happened to the biggest company that bet on this
• How long does compounded semaglutide last, and does it expire after 28 days?
• What it costs now, compared with the branded route
• What this means if you run or are planning a weight loss program
• What to ask your platform and your pharmacy this week
• Where we sit

Does compounded semaglutide work, and is it going away?

On the evidence available, compounded semaglutide can produce weight loss, and the reason is simple: where the product genuinely contains semaglutide at the stated dose, it is the same molecule that was tested in the branded trials. What has not been tested is the product itself. The regulatory half is less tidy. The shortage route closed in 2025, a narrower route remains, and what qualifies under it is argued rather than settled. Here is where both halves stand on 8 September 2026.

What “compounded semaglutide” actually means

Compounding is a pharmacy preparing a medication for a patient rather than a manufacturer producing a finished drug for the market. A compounded semaglutide product is made by a licensed pharmacy, usually as a vial of solution the patient draws from, sometimes combined with something else such as vitamin B12.

Three things follow from that, and every question further down this page comes back to one of them.

It is not an FDA approved drug. The FDA does not review a compounded preparation for safety, effectiveness or quality before it is sold. Approval attaches to a finished product from a manufacturer, so Wegovy and Ozempic are approved and a compounded vial is not, even when the pharmacy is licensed and the prescription is genuine.

It is not a generic. A generic goes through an abbreviated application and has to demonstrate bioequivalence to the branded product. Semaglutide has no approved generic in the United States. Anything described as “generic Ozempic” is being described wrongly, and that description is one of the specific framings the FDA has warned about.

Its quality depends entirely on the pharmacy. Two vials with the same words on the label can come from operations with completely different sourcing, testing and sterility practice. This is the part that matters commercially, because a business putting its brand on the outside of that vial is taking on the difference.

Does compounded semaglutide work?

The honest answer has two layers and most pages only give you the first.

The pharmacology layer. Semaglutide is semaglutide. If a vial contains the semaglutide base at the dose on the label, the molecule reaching the patient is the one studied in the STEP and SUSTAIN programmes, and there is no mechanism by which it would behave differently because a pharmacy rather than a factory put it in the vial.

The product layer. No compounded semaglutide product has been through a clinical trial. The trials that produce the familiar numbers tested Novo Nordisk’s finished product, made to a fixed specification under manufacturing controls, and their results transfer to a compounded vial only to the extent that the vial matches on identity, dose, purity and stability. That is an assumption, not a finding.

There is now some direct evidence rather than only inference. A retrospective real world study published in Health Science Reports in 2026 looked at weight outcomes in people using a non-comparable biotherapeutic subcutaneous semaglutide, which is the clinical literature’s term for a follow-on version rather than the originator product. It reported meaningful weight reduction, which is what you would expect if the molecule is present. Retrospective real world studies carry the usual caveats: no control group, patients who stay in the data are the ones who stayed on treatment, and the product studied is not necessarily the product any given pharmacy is making.

So the defensible version of the answer, and the one worth saying to a client, is this. Compounded semaglutide is not inert and it is not a placebo. Whether a specific product performs like the branded one depends on whether that specific product contains what it says it contains, and that is a question about the pharmacy rather than about semaglutide.

Is compounded semaglutide the same as Ozempic?

Not in the sense the question usually means, and the difference is worth being precise about because getting it wrong in your marketing is a named enforcement risk.

The active molecule can be the same. Ozempic and Wegovy are semaglutide, and a compounded product made from the semaglutide base contains the same molecule. What differs is everything around it: the formulation, the excipients, the concentration, the container, the manufacturing controls, the testing, and the regulatory status.

Then there is the case where the molecule is not the same. Some compounded products have been made from semaglutide salts, usually semaglutide sodium or semaglutide acetate, rather than the base. The FDA’s position is direct: it “does not have information on whether these salts have the same chemical and pharmacologic properties as the active ingredient in the approved drug” (FDA, Concerns with Unapproved GLP-1 Drugs Used for Weight Loss). A salt form is a different active ingredient, not a different presentation of the same one.

For a business the practical rule is short. Ask your pharmacy in writing whether the product is made from semaglutide base or a salt. If the answer is a salt, or if the answer is vague, you have your answer about the pharmacy as well as the product. And never let “same as Ozempic”, “generic Wegovy” or “identical active ingredient” appear in your own copy, whatever a supplier’s marketing pack says.

What the impurity testing actually found

This is the part of the evidence that gets least attention and deserves most, because it is the mechanism by which a compounded product could genuinely differ from the branded one.

A study published in Pharmaceutical Research on 30 July 2026 analysed products from 36 commercial suppliers: nine follow-on semaglutide drug substances, thirteen compounded injectables, eleven oral products and four liraglutide products, against the originator as comparator. Its findings, in the order that matters to somebody dispensing:

• The compounded injectables carried 44 impurities, of which 34 were not detected in the originator product. The follow-on drug substances carried 57 not found in the originator.
• The novel impurities included dimers, trimers, oxidised forms and formaldehyde adducts.
• Photostability was the sharpest difference. After light exposure, compounded product strength fell to between 86% and 95%, and impurities rose from between 1.3% and 2.5% up to between 5.1% and 14.2%. The originator product changed negligibly.
• An immunogenicity assay found that some of the impurities, the ones involving added or deleted amino acids, were presented on antigen presenting cells from healthy donors. That is a signal that they could provoke an immune response, and it is a question the branded trials never had to ask because those impurities were not in the branded product.

One disclosure that most write ups of this study omit and that you should know before you use it: the corresponding author is affiliated with Novo Nordisk, which manufactures the originator. That is a real interest in the result. It does not make the analytical chemistry wrong, and the impurity findings are the kind of thing another laboratory can check, but it belongs in the same sentence as the finding.

The practically useful part is the photostability result, because it is about handling rather than chemistry, and handling is something a clinic controls. Light exposure is a variable in your fridge, your counter and your courier, and it appears to matter more for a compounded product than for the branded one.

Is compounded semaglutide safe?

Safe compared with what is the question that has to come first, and the answer differs depending on whether you mean the molecule or the supply chain.

On the molecule, semaglutide’s side effect profile is well characterised and dominated by gastrointestinal effects. That is the same whoever prepared it.

On the supply chain, the FDA has logged a meaningful volume of reports. As of 31 May 2026 it had received more than 990 adverse event reports involving compounded semaglutide and more than 730 involving compounded tirzepatide, over 1,720 in total. Two caveats belong with that number in both directions. Adverse event reports do not establish causation, so the figure is not a count of harms caused. And the totals are likely undercounts, because state licensed pharmacies are not generally required to report adverse events at all.

The agency’s specific concerns, in its own words, are worth reading as a list of failure modes rather than as a general warning:

Dosing errors. The FDA received “multiple reports of adverse events, some requiring hospitalization, that may be related to dosing errors associated with compounded injectable semaglutide products”. The mechanism is mundane and therefore common. Branded semaglutide arrives in a pen that delivers a set dose. A compounded vial arrives with a syringe and a set of instructions, and the arithmetic moves between milligrams, millilitres and units. Patients have drawn up ten times the intended dose. So have some clinicians.

Doses beyond the approved label. Prescribing above the labelled maximum has produced the predictable pattern of nausea, vomiting, diarrhoea, abdominal pain and constipation.

Salt forms, covered above.

Products sold as research use only. The FDA notes products “sold directly to consumers for human use with dosing instructions” while carrying a research use disclaimer. That is a different market from a licensed compounding pharmacy and no business should be anywhere near it. We have written about how that labelling works, and does not work, in our article on whether peptides are legal.

The honest summary: the risks that show up in the data are mostly not exotic pharmacology. They are dosing arithmetic, handling and sourcing, and every one of them is a process a business can either run properly or not.

Why the shortage ending changed everything

Almost every business currently selling compounded semaglutide started during a specific and temporary legal window, and a lot of them never knew that was what it was.

When a drug is on the FDA’s shortage list, the ordinary bar on compounding a copy of a commercially available drug lifts. Semaglutide went on that list in 2022 and stayed there. That is the whole reason a compounded GLP-1 industry exists: not a judgment that compounded semaglutide was a good idea, but a supply gap that made compounding it lawful.

Then the gap closed.

Dates and deadlines from the FDA’s own statement to compounders.

After those dates the ordinary rules came back. And the ordinary rules are restrictive by design, because the whole point of them is that compounding is a clinical exception rather than a route to market.

Can you still get compounded semaglutide?

Yes, and the two routes work differently enough that they are worth separating before anything else.

503B outsourcing facilities. A 503B facility may compound from a bulk drug substance only where that substance is on the 503B bulks list or the drug is on the shortage list. Semaglutide is currently on neither. This is the route that supported large batch, non patient specific supply.

503A pharmacies. A state licensed pharmacy may compound for an identified individual patient on a valid prescription. The constraint it works under is section 503A(b)(1)(D), which addresses compounding, regularly or in inordinate amounts, drug products that are “essentially copies” of a commercially available drug product.

What counts as essentially a copy, and what takes a preparation outside that description, is where the whole argument in this category sits. The rest of this section sets out what the FDA has written, and what practitioners and compounders argue, without pretending the answer is settled where it is not.

What the FDA has actually written

These are quotations rather than characterisations, because in this area the paraphrases have drifted a long way from the text.

On the standard itself, a compounded drug is not essentially a copy where there is a change “made for an identified individual patient, which produces for that patient a significant difference”, as determined by the prescriber. The FDA’s own examples of adequate documentation are short and specific: “No Dye X, patient allergy”. “Liquid form, patient can’t swallow tablet”. “6 mg, patient needs higher dose”.

On strength, the agency “generally intends to consider two drugs to have a similar dosage strength if the dosage strength of the compounded drug is within 10% of the dosage strength of the commercially available drug product”. Separately it describes an easily substitutable strength by example: a compounded 50 mg tablet where the approved product is 25 mg, “because the patient could take two Drug X 25 mg tablets to achieve the required dose”.

On combinations, the agency “generally intends to consider a compounded oral drug product that combines drug X and drug Y in strengths that are within 10% of the strengths of the respective commercially available products to be essentially a copy of the commercially available drug product, unless a prescriber determination of a significant difference has been documented”.

On price, “a lower price” is “not sufficient to establish that the compounded drug product is not essentially a copy”.

On volume, “as a matter of policy, at this time FDA does not intend to take action against a compounder for compounding a drug product that is essentially a copy of a commercially available drug product regularly or in inordinate amounts if the compounder fills four or fewer prescriptions for the relevant compounded drug product in a calendar month”. And immediately after: “a prescription that documents such a prescriber determination would not be counted towards the four prescriptions”.

On what follows from not being a copy at all: “the regularly or in inordinate amounts provision is not implicated if the compounded drug is not essentially a copy”.

All quotations from FDA, Compounded Drug Products That Are Essentially Copies of a Commercially Available Drug Product Under Section 503A, final guidance, January 2018.

The arguments, set out side by side

Every rationale in circulation is below with the case for it and the case against it. Neither column is our view, and which of them prevails for any given patient is a question for the prescriber and for counsel, not for a page on the internet.

Two observations that hold whichever way an individual case falls.

The first is that the additive, the strength and the schedule are not alternatives to the prescriber’s determination. On the text above they sit inside the same analysis, with the determination as the thing that resolves it. A programme that treats a formulation choice as the reason it is compliant has skipped the step the guidance is actually about.

The second is that none of this turns on how many patients a practice has. The statutory limit is on compounding copies regularly or in inordinate amounts, and the guidance says a documented determination is not counted. Telehealth extends a clinician’s reach to patients who could not otherwise be evaluated, and an evaluation is not made less individual by being one of many. Where the evaluations are real, scale is a staffing question. Where they are not, the problem exists at four patients as much as at four thousand.

Where the position is genuinely unsettled

Being straightforward about this is more useful than a confident answer would be.

The guidance was written in January 2018, before a drug class sold as a titration ladder became the most compounded category in the country. Its worked examples are tablets. How the substitutability prong applies to an injectable titrated across five strengths is not something the agency has spelled out, and reasonable practitioners read it differently.

The FDA’s April 2026 analysis, in the separate 503B context, declined higher strengths, microdosing, alternative routes, excipient avoidance and multi-ingredient combinations as grounds for a general clinical need. That analysis was about whether a substance belongs on a list for outsourcing facilities, which is a different question from whether an individual prescription is a copy. Whether it signals anything about the agency’s reading of the 503A question is exactly the kind of inference that ought to be tested with counsel rather than assumed either way.

State law adds another layer. California’s regulations effective 17 June 2025 require a documented, patient specific difference before a pharmacy compounds a product otherwise identical to an approved one, and other boards have moved at their own pace.

None of which means a lawful prescription cannot be written. It means the basis for it is a clinical and legal judgement made per patient, by people who can see the patient and the file, and that anyone telling a business owner the question is simple is selling something.

Is compounded semaglutide going away, and will it be banned?

“Banned” is the wrong word for what is happening, and the right word matters if you are making a decision about a business.

Nobody has banned semaglutide. What is happening is that the specific legal permissions that allowed it to be compounded at scale are being closed one at a time, and the closing is not finished.

Here is the state of play as at 8 September 2026.

On 30 April 2026 the FDA announced that it proposes to exclude semaglutide, tirzepatide and liraglutide from the 503B bulks list, having found no sufficient clinical need for outsourcing facilities to compound them from bulk substances. Commissioner Marty Makary put the reasoning plainly: “When FDA-approved drugs are available, outsourcing facilities cannot lawfully compound using bulk drug substances unless there is a clear clinical need” (FDA announcement).

The notice was published in the Federal Register on 1 May 2026 under docket FDA-2018-N-3240. The comment period was extended by thirty days and closed on 30 July 2026. As at the date on this page the FDA has not published a final determination.

Two things follow, and they point in the same direction. First, this particular door is already closed in practice: a proposal to exclude a substance from a list it is not on does not change today’s position, it forecloses tomorrow’s. Second, the FDA’s stated reasoning rejected the arguments the industry actually makes. Higher strength injections, microdosing, buccal and sublingual routes, excipient avoidance, multi-ingredient combinations and device preference were all put forward as reasons for clinical need, and none of them persuaded the agency.

If you are building a plan around a rule change, that is the sentence to plan against.

What happened to the biggest company that bet on this

If you want to know what a compounded GLP-1 program looks like from the other end, the most useful case is the largest one, because it happened in public and in filings rather than in rumour.

Hims and Hers built a very large weight loss business on compounded semaglutide during the shortage, and had a commercial collaboration with Novo Nordisk. In June 2025 Novo terminated it, saying publicly that it was doing so over what it called illegal mass compounding and deceptive marketing. Hims continued.

In February 2026 two things happened within days. The FDA said it would restrict GLP-1 active ingredients used in unapproved compounded products, and named the company. Novo Nordisk sued, alleging patent infringement by the compounded semaglutide products. Hims backtracked on a planned compounded version of the newly approved oral product.

By 9 March 2026 the two companies had a new agreement. Hims now sells injectable and oral semaglutide at the same self-pay prices as Novo’s other telehealth partners, has discontinued selling compounded semaglutide except where a physician determines it is clinically necessary, and Novo dropped the lawsuit.

Nine months, start to finish, at a company with a market capitalisation and a legal department. The lesson is not that compounding is doomed. It is that the compounded programme was the transitional position and the branded one was the destination, and the companies that got there by choice arrived in better shape than the ones that got there by litigation.

How long does compounded semaglutide last, and does it expire after 28 days?

This is the question your clients will ask you, and the honest answer is more useful than the confident one you will find on vendor pages.

The date that governs is the beyond use date on the label your pharmacy applied. Not a number from the internet, and not the branded product’s expiry. A beyond use date is not a manufacturer expiration date. An expiration date comes from stability testing of a finished product in its final container. A beyond use date is the pharmacy’s assignment for a preparation it made, and under USP General Chapter <797> it depends on how the preparation was made, whether it passed sterility testing, and how it is stored.

The ranges are wide, which is exactly why a single number circulating online is unreliable. Under USP <797> a Category 1 sterile preparation may carry a beyond use date measured in hours. A Category 2 preparation made aseptically without sterility testing runs from about a day to 45 days frozen, and passing sterility testing extends that. Category 3, which requires stability data and additional controls, allows the longest dating. Two pharmacies can lawfully put very different dates on visually identical vials.

A note on vocabulary, because it trips people up: these USP categories have nothing to do with the FDA’s Category 1 and 2 lists for bulk substances that come up in peptide compounding. Same words, unrelated systems.

Where the 28 days comes from. It is the multiple dose container rule. A preserved multiple dose container, once punctured, must not be used beyond its assigned beyond use date or 28 days, whichever is shorter. So 28 days is a ceiling that applies after first puncture to a preserved multi dose vial. It is not a property of semaglutide, and it does not extend anything: if the label says 21 days, the answer is 21 days.

On refrigeration, follow the storage condition on the label, because the beyond use date was assigned for that condition and does not survive being stored somewhere else. And given the photostability findings above, protecting the vial from light is not fussiness.

The operational point for a business: whoever answers your phone should be reading the label, not remembering a number. Storage and dating questions are the most common client contact in a GLP-1 program and the easiest place to give a wrong answer with confidence.

What it costs now, compared with the branded route

The economics that made compounded GLP-1s obvious in 2023 are not the economics of 2026, and this is the change most operators have not repriced against.

Manufacturers have moved hard into self-pay. These are their published prices, not ours:

Sources: LillyDirect pricing as reported and NovoCare’s own savings page. Check both before quoting them, because introductory pricing has an end date on it and these have moved repeatedly.

Compounded product still undercuts those figures in most cases. But a price comparison is the wrong frame if one of the two options carries regulatory exposure, an unreviewed supply chain and a product your marketing has to be careful about describing. The gap that remains is narrower than the gap that built the category, and it is narrowing against a backdrop where the cheaper option is the one under active rulemaking.

What this means if you run or are planning a weight loss program

The strategic position is straightforward once the regulatory detail is set aside.

Plan for the rules to keep moving. The shortage route closed in 2025, the 503B proposal is pending, and the 503A questions above are contested rather than resolved. Whatever position a business takes today, it is worth building so that it survives the position changing.

The client relationship is the durable asset, not the molecule. Clients come to a med spa or a clinic for the relationship, the monitoring, the convenience and the accountability. Those survive a change of product. A program that can move a client from compounded to branded without losing them has a business. A program whose only advantage was price does not.

The record is what gets read. Whatever basis a practice prescribes on, the thing an inspector, a board or an opposing party sees is the file. Reading a random sample of your own prescriptions the way somebody unsympathetic would read them is cheap, and it is the only way to find out what they actually say.

Your pharmacy’s position becomes yours. This is the same lesson as the peptides question and it does not get less true. Ask which specific products they will supply, on what legal basis, from base or salt, and what their beyond use dating and sterility testing practice is. Get it in writing. If you buy through a platform rather than contracting the pharmacy directly, the platform has made that choice on your behalf, which is one of the things worth knowing before you sign: see our breakdown of what a white label telehealth platform actually includes.

Most enforcement so far has been about marketing. The FDA issued more than 55 warning letters to online sellers of compounded GLP-1s on 16 September 2025, and the theory in them was misleading advertising rather than the compounding itself. In February 2026 Commissioner Makary framed the agency’s focus as companies “mass-marketing illegal copycat drugs” and misleading direct-to-consumer advertising. Whatever a business concludes about its clinical basis, how it describes the product is a separate question with its own exposure.

Have the switch planned before you need it. If a final rule lands, or your pharmacy changes its position, or a manufacturer moves against your supplier, the question is not whether you can source a replacement. It is whether your clients hear it from you first, with a route already in place. That plan costs nothing to write now and is worth a great deal the week it is needed.

What to ask your platform and your pharmacy this week

Five questions. The answers are more revealing than any certificate.

1. Is the product made from semaglutide base or a salt? If the answer is a salt, or the answer is unclear, that settles it.
2. On what basis are you supplying it today? You are looking for a specific answer that engages with the essentially-a-copy analysis, not “our pharmacy is FDA registered”. Facility registration is not product approval, and a supplier who cannot articulate the basis has not thought about it.
3. Where is any clinically significant difference documented, and who made it? The guidance contemplates a prescriber determination about an individual patient, recorded on the prescription. Ask to see the shape of it.
4. What beyond use dating do you assign, on what basis, and do you sterility test? A pharmacy that cannot explain how it arrives at its dates is telling you something.
5. What is the plan if a final rule closes this? There is a right answer, which is a transition to an approved product with a named timeline, and a wrong one.

Where we sit

Local Healthcare’s weight management programs use GLP-1 therapy prescribed by an independent clinician where that clinician determines it is appropriate, and not everyone qualifies. Prescriptions are written by Beluga Health, an independent physician-led medical group licensed in all fifty states. Local Healthcare does not practise medicine, does not make prescribing decisions, and does not decide who is a candidate.

Where a program involves a compounded preparation, the decision to prescribe one, and the clinical basis for it, belong to the prescribing clinician and the dispensing pharmacy. Local Healthcare does not select formulations, does not direct prescribing and does not supply the clinical rationale for any prescription. We have set out the arguments above because partners ask about them, not because we are resolving them.

Compounded medications are prepared by state licensed pharmacies, are not FDA approved, and we make no claim that any compounded preparation is as safe or as effective as any branded alternative. We do not describe any compounded product as generic, as the same as a branded drug, or as FDA approved.

On pricing we publish what we can publish plainly. The Local Healthcare platform fee is $50 per treatment per month. Medication and fulfilment are billed at cost and quoted directly, because pharmacy product pricing varies by product, strength and pharmacy, and any figure we printed would be wrong for most partners. Every competitor in this category hides both numbers; we would rather show one and explain the other.

The partner sells the program to their client; Local Healthcare provides the platform and collects payment as the partner’s billing agent; the clinician is independent.

Any figures shown are illustrative, not a promise of income. What a partner earns depends on the retail price it sets, its patient volume and retention. The partnership agreement is the controlling document.

Availability varies by state.

How current this is

Every regulatory fact on this page was checked against a primary source on 8 September 2026: the FDA for the shortage deadlines, the adverse event figures and the salt form position, the Federal Register for the 503B proposal and the comment period extension, Pharmaceutical Research and Health Science Reports for the product evidence, USP for beyond use dating, and the manufacturers’ own pages for self-pay prices. Where a fact came from a trade publication rather than a primary source, the sentence says so.

This article is general information about regulatory status and published evidence. It is not legal advice, it is not medical advice, and it has not been reviewed by outside counsel or by a clinician. The FDA’s proposal on the 503B bulks list is pending and a final determination could change parts of this page. Confirm anything you intend to rely on with your own counsel and your own pharmacy, and check the date above before treating any of it as current.