Compounded semaglutide: what's actually left after the shortage ended

GLP-1 is a hormone your gut releases about fifteen minutes after you eat. It nudges your pancreas to release insulin, slows your stomach, and tells your brain you've had enough. The native peptide is gone in minutes. Semaglutide, Novo Nordisk's engineered analog, was built to outlast it.

The chemistry goes like this. Three tweaks to the peptide backbone - a substituted amino acid at position 8 that shrugs off the DPP-4 enzyme, an arginine swap at position 34, and an acylated C18 fatty-acid tail at position 26 that tethers the molecule to albumin in your blood - stretch its half-life from minutes to about a week.

That's the whole trick.

Once-weekly dosing is what that half-life buys you, and once-weekly dosing is what let the drug escape endocrinology clinics and land in the consumer category it now occupies. On paper it's a diabetes peptide. In practice it's something closer to a lifestyle product with a prescription pad attached.

What it does in the body is broader than most patients realize. In the pancreas it amplifies glucose-dependent insulin release while suppressing glucagon - the original Type 2 diabetes rationale. In the stomach it slows gastric emptying, which blunts the post-meal glucose spike and, more importantly for the weight-loss use case, makes you feel full sooner and for longer. Then there's the hypothalamus, where it suppresses appetite directly. That's the channel patients are describing when they say the "food noise" stops. GLP-1 receptors also sit on heart, kidney, muscle, and fat tissue, which is why the metabolic effects reach well past what you'd predict from a plain appetite suppressant.

The single point worth pausing on: this isn't a stimulant, it's not a fat blocker, and nothing about your metabolism is being sped up. The weight loss happens because you eat less and the food parks in your stomach longer. That mechanism shapes the side effects, what happens when you quit, and who actually tolerates the drug - all of which we'll get to.

Compounded semaglutide is, in the cleanest cases, the same active ingredient as Ozempic and Wegovy: semaglutide sodium, prepared by a licensed pharmacy into an injectable solution. In the messier cases, and there are plenty, what arrives in the vial is something else entirely.

Consider what the FDA actually found. A 2024 review of compounded GLP-1 products clocked potency anywhere from 42% to 170% of labeled strength in sublingual formulations.

That's not a rounding error.

A FAERS analysis running from 2018 to 2024, pulling from more than 81,000 reports, flagged higher rates of nausea, abdominal pain, diarrhea, gallbladder inflammation, and hospitalizations tied to compounded GLP-1s when you stack them against the brand-name versions.

Some of that signal is reporting bias. Not all of it.

The differences that matter, in order of importance:

The molecule itself. Brand-name Ozempic and Wegovy use semaglutide manufactured by Novo Nordisk under cGMP at scale, with the supply chain you'd expect from a top-five pharma company. Compounded semaglutide uses semaglutide-base or semaglutide-sodium sourced from a smaller set of API suppliers, some FDA-registered, some not. A legitimate 503A compounder will document their source. A bad actor won't, and the patient can't easily tell which they're buying from.

The formulation. Brand products show up as aqueous injection solutions in fixed-dose pens or single-use vials, carrying the excipient mix that cleared FDA review. Compounded products are a different animal. Most are vial-and-syringe injectables at whatever concentration the pharmacy picked. Some toss in B12 or other "research adjuncts" with no clinical evidence behind them, and a meaningful share are sublingual or oral tablets with basically no published bioequivalence data - which, in practice, account for a lopsided slice of the documented potency failures.

The legal status. Here's where most consumer pages garble the timeline. Next section.

The price. This is real and large. Brand-name Wegovy lists at roughly $1,350 per month without coverage. Compounded semaglutide from a vetted DTC provider currently runs $99 to $349 per month, depending on tier and titration phase.

The question is whether that price gap reflects what an honest market would price, or whether it reflects regulatory arbitrage that no longer legally exists.

Semaglutide entered the shortage list during the demand surge of 2022, alongside tirzepatide. Under FDA enforcement discretion, 503A compounding pharmacies (patient-specific compounders) and 503B outsourcing facilities (larger-scale, FDA-registered manufacturers) were permitted to compound semaglutide formulations to address the shortage. Telehealth providers built entire businesses on this window. So did the compounders. The economics worked because the law worked, and the law worked because Novo Nordisk genuinely could not make enough drug.

The shortage was declared resolved by FDA in February 2025. The compounding industry pushed back, litigation followed, and FDA set enforcement deadlines: April 22, 2025 for 503A pharmacies, May 22, 2025 for 503B outsourcing facilities. Past those dates, compounding what FDA describes as an "essentially-a-copy" of an approved semaglutide product is no longer permissible under the shortage exception.

In April 2026, the FDA went further - proposing to permanently knock semaglutide, tirzepatide, and liraglutide off the 503B bulk drug substances list. That'd close the door even on a future shortage-based pathway for outsourcing facilities. Public comment closed June 29, 2026.

A final rule is expected later this year.

So what's left? Two narrow legal theories, and neither is roomy. The first is 503A patient-specific compounding under a clinical-necessity rationale, meaning a prescriber documents that the patient can't use the commercially available product, usually because of an allergy to a specific excipient or a need for a dose strength the brand doesn't make. It's real. But it's meant to be exceptional, not routine.

The second theory is what some providers call "personalized" formulations: semaglutide combined with B12, glycine, or other add-ins, on the argument that the resulting product isn't "essentially a copy" of Ozempic or Wegovy. FDA has signaled it views many of these combinations skeptically, and at least one enforcement action in late 2025 turned on exactly this question. In practice, the agency seems unconvinced that tossing a vitamin into the vial gets you out of the copy rule.

The honest read: most of the volume of compounded semaglutide being prescribed in mid-2026 is operating in a gray zone that the 2022-2024 volume did not occupy. Some of it's clearly legitimate patient-specific compounding. Some of it's regulatory positioning that may or may not survive the next round of enforcement. A consumer entering the market today should know which side of that line their provider claims to be on, and should be skeptical of providers who can't answer the question.

A 503A pharmacy compounds for an individual identified patient against an individual prescription. Think of your local compounding pharmacy filling a custom hormone preparation. They're state-licensed, not FDA-registered as manufacturers, and they're not permitted to produce in bulk for office-stock dispensing or for prescriber inventory.

A 503B outsourcing facility is FDA-registered, inspected against cGMP standards comparable (though not identical) to brand manufacturers, and can produce in larger batches for office-use distribution. They're the entities that scaled compounded GLP-1s to the volumes seen in 2023-2024.

For a consumer evaluating a telehealth provider in 2026, the relevant questions are: which pathway are they using, can they name the specific pharmacy, and what is that pharmacy's regulatory standing? A good provider answers all three without friction. A provider that deflects, or that names a pharmacy you can't find on the state board of pharmacy's website, is not one to use.

The semaglutide evidence base is unusually strong for a drug in this category. The STEP program (semaglutide 2.4 mg subcutaneous, weight management) and SUSTAIN program (lower doses, T2D) together cover tens of thousands of patients across multiple geographies and trial designs. The Type 2 diabetes program, anchored by SUSTAIN 6 (NCT01720446, n=3,297), established that semaglutide at 0.5 to 1 mg weekly produces clinically meaningful HbA1c reduction with a cardiovascular safety signal that turned out to be a benefit signal, not a wash.

The weight-loss trials are the ones consumers actually care about, so here is what they showed.

STEP 1 enrolled about 1,961 adults with obesity but without diabetes and randomized them to semaglutide 2.4 mg or placebo for 68 weeks. The semaglutide group lost a mean of 14.9% of body weight versus 2.4% on placebo, a between-group difference of 12.44 percentage points. Most of the responders cleared the threshold of clinical significance: 86.4% lost at least 5% of body weight. Translate that into a 220-pound adult and the average loss is roughly 33 pounds in 16 months. That's large for a pharmaceutical weight-loss intervention. It's not, however, bariatric-surgery-large.

STEP 2 ran the same design in patients who also had Type 2 diabetes. The treatment effect shrank - 6.21 percentage points greater weight loss versus placebo. Diabetes blunts the response.

STEP 3 stacked intensive behavioral therapy on top of the drug and produced a 10.27-point gap. STEP 4 is where it gets interesting, because that's the trial that tells you what happens when you stop. Patients finished a 20-week run-in on semaglutide and were then randomized to continue or switch to placebo. Over the next 48 weeks the two arms diverged by 14.75 points, with the placebo-switch group clawing back most of what they'd lost.

Short version: the drug works while you take it.

That one finding has done more to reframe these drugs as chronic therapy, not a course you finish, than anything else in the literature.

STEP 5 took 304 patients out to two years and found mean weight loss of roughly 17.4% maintained at 104 weeks, with about 36% of patients clearing the 20%-loss threshold. STEP UP, an investigational arm at 7.2 mg, pushed mean loss to 20.7% with about a third of patients losing more than 25% of body weight - that dose is not yet FDA-approved as of early 2026, but it tells you where the dose-response curve is still going.

SELECT (NCT03574597) randomized roughly 17,604 adults with established cardiovascular disease, plus obesity or overweight, but without Type 2 diabetes, to Wegovy 2.4 mg or placebo. The primary endpoint was MACE - heart attack, stroke, or cardiovascular death - and the trial hit. The cardiovascular benefit separated from placebo within the first three months, with a hazard ratio of 0.63 (95% CI 0.41-0.95) at that early timepoint, before the weight-loss curves had meaningfully diverged. That timing is the part clinicians find genuinely interesting. Whatever is driving the cardiovascular protection, it's not waiting for weight loss to do the work - it's showing up early, which points toward direct effects on inflammation, endothelial function, or both.

SELECT is the basis for the March 2024 FDA expansion of Wegovy's label to include MACE reduction in adults with obesity or overweight and established cardiovascular disease. It's also the basis for an emerging clinical argument that some patients should be on a GLP-1 receptor agonist for the cardiovascular indication independent of their weight-loss goals. That argument applies to brand-name Wegovy, with brand-name clinical trial data behind it. Whether and how it extends to compounded preparations of the same molecule is a question the trial can't answer.

The brand-name landscape is worth knowing because it determines what the legitimate prescriber alternatives are. Ozempic, approved December 2017, is indicated for glycemic control in Type 2 diabetes at doses of 0.5, 1, and 2 mg weekly. Wegovy, approved June 2021 and expanded in March 2024, is indicated for chronic weight management at 2.4 mg weekly in adults with BMI ≥30, or ≥27 with at least one weight-related comorbidity, and for MACE reduction in the SELECT population. Rybelsus is oral semaglutide for Type 2 diabetes, approved September 2019, at 3, 7, and 14 mg daily, with the well-documented absorption quirks - empty stomach, 30 minutes before anything else, four ounces of plain water - that make adherence harder than the injection.

Ozempic prescribed off-label for weight loss is not the same regulatory situation as compounded semaglutide. Off-label prescribing of an FDA-approved drug is broadly permitted, well-trodden, and supported by the same brand-name supply chain. The 2022-2024 shortage was driven in real part by off-label Ozempic prescribing for weight loss, which is part of why Wegovy was approved with a different dose escalation and a different commercial program.

Skip a step and you'll likely quit. That's the practical reason the titration schedule exists - the gut side effects scale with both the dose and how fast you climb to it, and patients who rush the ladder tend to wash out of treatment entirely.

Wegovy's approved schedule is the reference point for weight-loss compounding: 0.25 mg weekly for four weeks, then 0.5 mg for four weeks, then 1.0 mg for four weeks, then 1.7 mg for four weeks, then 2.4 mg as the maintenance dose. You don't hit the therapeutic dose until month five. The 0.25 mg starter isn't a treatment dose. It's a runway, there so your gut can acclimate, and if a provider is telling you to expect real weight loss in week one, they're selling you something.

Month one is for tolerability. Full stop.

Compounded providers handle titration differently. Some follow the Wegovy ladder rung for rung. Others run a faster ramp, especially for patients who've already titrated on brand product and are just switching the source. And then there's the "micro-dosing" crowd, pushing 0.125 mg or lower, pitched either as a gentler entry or as some off-label longevity hack (a use case nobody has properly studied). The evidence under those protocols is thin to nonexistent, so treat them as experimental rather than a kinder version of the approved schedule.

The Ozempic schedule for Type 2 diabetes runs lower - 0.25 mg start, 0.5 mg therapeutic, 1 mg if needed, 2 mg ceiling - and reflects that glycemic control responds to lower doses than weight loss does. A compounding pharmacy or telehealth provider prescribing 1 mg weekly for weight loss is undershooting the trial-supported dose. Whether that matters for an individual patient depends on response, but it's worth noticing if your prescriber settles you at 1 mg and calls it the maintenance dose.

The trial averages are the right starting point. But the spread around that average is wider than most patients hear about. STEP 1 came in at 14.9% on average. People in the top quartile shed more than 20% of body weight; the bottom quartile lost under 5%. So the drug works very well for most and poorly for a meaningful minority. In practice, we don't yet have a reliable way to tell, at the individual level, which group a given patient will land in.

What seems to matter: dose reached, duration of treatment, baseline BMI (higher BMIs tend to drop more pounds in absolute terms, though percentage-wise it lands similar), and what the patient is doing behaviorally alongside the shot. STEP 3 layered on intensive behavioral therapy and didn't produce dramatically better numbers than STEP 1, which hints the drug is doing most of the heavy lifting on its own.

Maintenance is a different animal. That's where established habits appear to earn their keep, once the dose plateaus and the appetite suppression stops feeling novel and starts feeling like the new baseline.

Most of the variance is still unexplained.

The other thing the trials show, and that consumer pages tend to underplay, is the composition of the weight lost. A 2024 systematic review of six trials (n=1,541, PMID 38629387) found that lean mass loss ranged from essentially zero to 40% of total weight lost, depending on the trial and the population. The mean weight is primarily fat, but the variance is real, and the patients who lose the most weight the fastest are also the ones at highest risk of meaningful lean mass loss. For most younger patients this is a minor concern. For patients over 60, for patients with sarcopenia risk, and for patients losing very rapidly, resistance training and protein intake are not optional adjuncts - they're part of the protocol.

The side effect profile is the part of the conversation that has changed most as real-world data has accumulated.

The usual GI stuff is well-characterized, and for most patients, it fades. Nausea hits 28% to 44% across the GLP-1 class, with semaglutide posting the highest reporting ratio in FAERS disproportionality analysis (ROR 7.41 for nausea, 6.67 for vomiting, 3.55 for diarrhea, 6.17 for constipation). Most of it clusters during dose escalation. Then it settles at steady-state, which is what you'd expect from a drug that slows gastric emptying.

16% of patients in SELECT quit for adverse events versus 8% on placebo, and GI symptoms drove most of those exits. So the honest discontinuation gap to plan around is eight points. Meaningful, sure. Hardly catastrophic.

The safety questions worth taking seriously are pancreatitis, gallbladder disease, gastric emptying around anesthesia, and thyroid cancer (plus the boxed warning that comes with it). Here's what the evidence actually shows:

Pancreatitis. The FAERS disproportionality signal looks alarming on paper: ROR 19.1. But the 2024 meta-analysis of randomized placebo-controlled trials, pooling across dosing regimens, found no statistically real bump versus placebo. The honest read goes like this. Pancreatitis happens in this patient population at a base rate anyway, GLP-1 prescribers are watching closely for it, and the FAERS signal probably tracks monitoring intensity more than excess events. Patients with a personal history of pancreatitis should still steer clear.

Gallbladder disease. Real, documented in trials, mechanism plausible (rapid weight loss is itself a gallstone risk factor). Patients who develop upper-right-quadrant pain should be evaluated rather than reassured.

Gastric emptying and anesthesia. This one is more recent and more concerning to clinicians. A 2024 Lancet Gastroenterol Hepatol review (PMID 39096914) documented clinically real gastric emptying delay persisting longer than the original guidance assumed, with aspiration risk during endoscopy and surgery. Current anesthesia guidelines recommend holding GLP-1s for a week or longer before procedures requiring sedation. Any patient on semaglutide planning surgery needs to flag it to their surgical team well in advance.

Thyroid C-cell tumors. This is the basis for the boxed warning. Rodent studies showed C-cell tumors at a mechanism that's not clearly translatable to humans (rodents express GLP-1 receptors on thyroid C-cells at much higher density). A systematic review covering 46,719 patients found no signal for thyroid or other cancers in humans. The boxed warning remains because the rodent data exists and the FDA is appropriately conservative. The contraindication is real for patients with personal or family history of medullary thyroid carcinoma or MEN2; for everyone else, this is risk to disclose, not risk to fear.

Hypoglycemia. Low as monotherapy because the insulin-stimulating mechanism is glucose-dependent. Real when stacked with sulfonylureas or insulin, which is why combination regimens require dose adjustment.

What we tell patients who ask goes like this: the common side effects are real but manageable, the serious ones rare but worth monitoring, and the surgical-aspiration risk is the one most likely to actually bite them if they aren't paying attention. Most tolerate the drug fine. A meaningful minority can't tolerate it at all, and the only way to learn which camp you're in is to start the titration and watch.

The FDA-approved Wegovy criteria - BMI ≥30, or ≥27 with at least one weight-related comorbidity (hypertension, dyslipidemia, sleep apnea, cardiovascular disease, Type 2 diabetes) - are the right starting frame, and most legitimate prescribers lean on them or some close adaptation. On paper, fine. At the level of the individual patient, they're imperfect. Take a patient at BMI 26 with metabolic syndrome and a father who had a heart attack at 52, and compare her to someone at BMI 28 with no comorbidities at all - the cardiovascular case for the first is arguably stronger, and the criteria don't capture that.

Some patients shouldn't pursue this drug, or should only after a careful conversation. Personal or family history of medullary thyroid carcinoma or MEN2 is a boxed contraindication, full stop. Past pancreatitis is another. Severe gastroparesis. A planned pregnancy inside the next twelve months. Severe psychiatric history involving depression or an eating disorder, because the appetite-suppression effect can sideswipe established eating-disorder treatment in ways that are hard to predict. And patients facing surgery within the next two months who can't reliably pause the medication.

Now the underserved candidates - the ones who often should be on it and aren't. Patients with established cardiovascular disease and BMI ≥27, which is the SELECT population and where the evidence is strongest. Patients with Type 2 diabetes whose metformin response has plateaued. And patients who've lost weight through behavioral work and regained it, sometimes more than once. That last group is, in reality, who the trials were built around.

Then there's the candidate we'd push back on, gently: the person asking for semaglutide to drop ten or fifteen cosmetic pounds. The drug is approved for, and trial-validated in, patients with obesity or genuine overweight plus comorbidities. Below that, the risk-benefit calculus simply hasn't been worked out. Off-label cosmetic use is a clinical choice, and we've seen it handled well and handled poorly.

The poorly version goes like this: weight regained, muscle lost on the way down, drug discontinued, patient worse off than when they started.

Is compounded semaglutide FDA-approved? No. Compounded preparations are not FDA-approved drugs. Brand-name Ozempic, Wegovy, and Rybelsus are FDA-approved. Compounded semaglutide is a prepared-to-prescription product, which is a different regulatory category, and as of mid-2025 the shortage-based legal pathway that supported widespread compounding has closed. Some patient-specific compounding remains legal under narrower theories.

Is compounded semaglutide the same as Ozempic or Wegovy? It contains the same active ingredient when sourced and prepared correctly. It's not the same product, not subject to the same manufacturing standards, and not supported by the brand clinical trial data in any formal sense. Whether the clinical effect is equivalent depends on the quality of the specific preparation, which the patient can't independently verify.

Will the shortage come back, and would that make compounding legal again? FDA's April 2026 proposed rule would exclude semaglutide from the 503B bulk drug substances list permanently, which would prevent shortage-based compounding even if a future shortage occurred. The final rule is expected later in 2026. The honest read is that the regulatory door is closing, not reopening.

What happens if I stop the drug? STEP 4 is the relevant trial. Patients who discontinued semaglutide after a 20-week run-in regained most of their weight loss over the following 48 weeks. The clinical implication is that semaglutide functions as a chronic therapy, comparable to antihypertensives or statins in how it should be planned, not as an episodic intervention.

What about muscle loss? Lean mass accounts for a variable share of weight lost, from roughly zero to 40% depending on the trial and the patient. Resistance training and adequate protein intake reduce lean-mass loss meaningfully and should be considered part of the protocol, particularly for patients over 60 and patients losing weight rapidly.

What about the cardiovascular benefit - does it apply to compounded semaglutide? The cardiovascular benefit was showed in SELECT using Wegovy 2.4 mg. Whether the same benefit extends to compounded preparations of the same molecule is a scientifically reasonable inference but not a tested one. Patients for whom cardiovascular risk reduction is the primary indication have a stronger case for brand-name therapy.

Can I switch from brand to compounded, or compounded to brand? Clinically yes, with attention to dose equivalence and titration history. Most providers handle this routinely. The cost calculation is the more common reason patients consider switching, and it's worth running with current pricing rather than what you remember from 2024.

The honest planning frame goes like this: expect some nausea during titration, and understand that the appetite suppression isn't a side effect at all - it's the thing you're paying for. What's most likely to actually blindside you is the surgical-aspiration risk, which only bites if you don't flag the medication to your surgical team ahead of time.

That last one matters more than people realize.

Editorial note: This page is for informational purposes and doesn't constitute medical advice. Semaglutide therapy decisions should be made with a licensed healthcare provider familiar with your medical history. Patients with personal or family history of medullary thyroid carcinoma or multiple endocrine neoplasia syndrome type 2 shouldn't use semaglutide. Last reviewed by Dr. Sarah Henderson, MD, May 2026.