Semaglutide Microdosing: What the Evidence Actually Shows

"Microdosing" is now one of the most heavily marketed ideas in compounded GLP-1 sales: a small, fixed weekly dose, held flat, promised to deliver most of the appetite effect with few of the side effects and a fraction of the cost. It is a genuinely appealing pitch. It is also a schedule that no randomized trial has ever tested, and the word itself has been borrowed from a part of pharmacology where it means something else entirely.

This article does three things, in order. It reports what the current Wegovy and Ozempic labels actually titrate to, and where the numbers being marketed as microdoses sit on those labels. It reports the one trial that genuinely did test low semaglutide doses, and what its results show about the trade-off being promised. And it is direct about the gap between the two — because the honest answer here is neither "microdosing works" nor "microdosing does nothing," and pretending otherwise in either direction would be a bad-faith reading of the evidence.

Nothing below is a dose recommendation. Every number is a number a label or a trial published, reported as such.

What "microdose" already meant, before the marketing

In pharmacology, a microdose study is a specific, defined thing: a sub-therapeutic tracer dose, given so that a drug's absorption and clearance can be measured in humans without producing a pharmacologic effect at all. The scale is the point. A published assay-development paper supporting a human microdose study of an oral GLP-1 receptor agonist describes a total dose of 0.005 mg — five micrograms — administered intravenously in a preclinical proof-of-concept run before the human study⁵. That is a measurement dose, not a treatment dose, and nobody expects it to do anything.

The marketing sense of "microdose" is unrelated to that. It means a fraction of the labeled maintenance dose, taken with the intention of producing a real therapeutic effect. Those are two different claims wearing the same word, which is worth knowing before reading anyone's copy about it.

What the labels actually titrate to

Both current semaglutide labels start low. Neither one stays there.

Wegovy injection. The recommended starting dosage is 0.25 mg subcutaneously once weekly, and the label's own Table 1 escalates it: 0.25 mg for weeks 1 through 4, 0.5 mg for weeks 5 through 8, 1 mg for weeks 9 through 12, 1.7 mg for weeks 13 through 16, and a maintenance dosage from week 17 onward. The label states the reason for the ramp in one clause — "Follow the dosage escalation in Table 1 to reduce the risk of gastrointestinal adverse reactions" — and the maintenance dosage for weight reduction in adults is either 1.7 mg or 2.4 mg once weekly, with 2.4 mg recommended and a 7.2 mg once-weekly maximum available for patients who tolerate 2.4 mg for at least four weeks¹.

Ozempic. Initiation is 0.25 mg once weekly for four weeks, and after those four weeks the label instructs increasing to 0.5 mg. The recommended maintenance dosages are 0.5 mg, 1 mg or 2 mg once weekly, with a maximum of 2 mg².

Read those two together and one fact does most of the work in this article: 0.25 mg appears on both labels only as a starting or initiation dosage, and on neither label's list of maintenance dosages. Its labeled job is to get a patient through the first four weeks with less nausea, not to be the dose. The lowest maintenance dosage anywhere on either label is 0.5 mg once weekly, on Ozempic, for glycemic control in type 2 diabetes. Doses marketed as microdoses generally sit at or below the starting rung — which is to say, below anything either label describes as a place to stay.

The one trial that actually tested low doses

There is a real dose-ranging trial, and it deserves to be read carefully rather than used as a slogan by either side.

O'Neil et al. published a randomized, double-blind, placebo- and active-controlled, multicenter, dose-ranging phase 2 trial in 957 adults with a BMI of 30 or more and without diabetes, across 71 sites in eight countries. Participants were assigned to semaglutide at 0.05, 0.1, 0.2, 0.3 or 0.4 mg, to liraglutide 3.0 mg, or to matching placebo, and the primary endpoint was percentage weight loss at week 52. The published findings, in the trial's own words: "Estimated mean weight loss was -2.3% for the placebo group versus -6.0% (0.05 mg), -8.6% (0.1 mg), -11.6% (0.2 mg), -11.2% (0.3 mg), and -13.8% (0.4 mg) for the semaglutide groups. All semaglutide groups versus placebo were significant³."

Two things follow from that, and both belong in an honest answer.

First: low doses of semaglutide are not inert. The lowest arm in the trial beat placebo by roughly 3.7 percentage points of body weight over 52 weeks, and the difference was statistically significant. Anyone claiming a small dose of semaglutide does literally nothing is contradicted by this trial.

Second: the dose-response gradient runs the wrong way for the marketing claim. The pitch for microdosing is that most of the benefit survives the dose reduction. The trial's own numbers say the benefit tracks the dose fairly closely across an eight-fold range, from -6.0% at the bottom to -13.8% at the top. The trial also reports that the most common adverse events were "dose-related gastrointestinal symptoms, primarily nausea³" — so the side effects fall with the dose too. That is a real trade-off, and it is a trade-off, not a free lunch. For scale on what the top of the labeled range does: the trial that established 2.4 mg once weekly for weight management randomized 1,961 adults 2:1 to 68 weeks of treatment or placebo and reported a mean body-weight change of -14.9% versus -2.4%⁴.

⚠ And now the part that most coverage of this trial leaves out. Every arm in O'Neil et al. was a once-daily injection, and every arm was escalated — "initiated at 0.05 mg per day and incrementally escalated every 4 weeks³." Not once weekly. Not held flat. The trial establishes that semaglutide has a dose-response relationship in obesity. It does not establish anything about a fixed, unescalated, sub-label weekly dose, because that is not what anyone in it received. Converting its daily milligram figures into weekly ones to argue for or against a marketed protocol would be inventing a comparison the trial never made, and this article will not do it.

What the "microdosing" literature actually consists of

If you search the published literature for microdosing GLP-1 receptor agonists, what comes back is not trials. It is correspondence.

Komé et al. published "One Size Does Not Fit All: Understanding Microdosing Semaglutide for Diabetes in Multidose Pens" in Diabetes Care in 2025⁶. It is indexed as a Letter — not a trial, not a cohort study — and its subject is patients with diabetes dialing intermediate doses out of a multidose pen, which is a different clinical situation from the weight-loss microdosing being sold. It drew a published Comment, also a Letter, the same year⁷. That exchange is the substance of the peer-reviewed record on the term.

The most direct clinical commentary is a 2026 brief report in the Journal of the American Association of Nurse Practitioners, which situates the practice precisely where it actually sits. Its own abstract describes affordability pressure "fueling a shift toward unsupervised strategies such as 'microdosing' and the purchase of illicit 'research-grade' peptides from online sources and unregulated providers including medical spas," and names the drivers of what it calls "subtherapeutic microdosing" as "gastrointestinal tolerability and 'vanity weight'." It flags "the specific side effects and dosing errors encountered with these practices" and "the safety risks associated with pen manipulation, compounded vials, and medication sharing⁸." That is a clinician's field description of a practice, published in a peer-reviewed journal — and it is not evidence that the practice works, which is a distinction the piece itself is careful about.

What does not exist, as of this writing, is a randomized trial in which people were assigned to a fixed sub-label weekly semaglutide dose, held there, and followed for weight or metabolic outcomes against a comparator. The marketed schedule has not been tested. That sentence is the whole article.

Why the vial makes this worse, not better

Microdosing and compounding travel together for an obvious reason: an approved pen dials a labeled dose, and a compounded vial does not. That means the microdose is not something the product delivers — it is something the patient measures.

The arithmetic itself is simple. Volume in milliliters equals the dose in milligrams divided by the vial's concentration in mg/mL, and on a U-100 insulin syringe that volume times 100 gives the units. What makes it hazardous at small volumes is that approved semaglutide injections alone span 0.68 mg/mL to 3.2 mg/mL¹², so the same syringe marking means a different milligram amount from one vial to the next — and the smaller the intended dose, the larger the proportional error from a misread marking. FDA has stated that it received "multiple reports of adverse events, some requiring hospitalization, that may be related to dosing errors associated with compounded injectable semaglutide products," arising from "patients measuring and self-administering incorrect doses" and from health care providers miscalculating them⁹.

There is also a purity question that no dose figure can answer. FDA states that semaglutide sodium and semaglutide acetate "are different active ingredients than are used in the approved drugs" and that it is "not aware of any lawful basis for their use in compounding⁹" — our explainer on what compounded semaglutide actually is covers that salt-form problem in full, and our dated regulatory timeline covers how the legal basis for selling it at scale narrowed. A dose you cannot verify, drawn from a vial whose active ingredient you cannot verify, is two unknowns stacked, not one.

If it is the label numbers you actually want, our full semaglutide dosing reference sets out all three current ladders — injection, tablets and Ozempic — with the missed-dose rules and the concentration table. Our reconstitution calculator runs the volume arithmetic on a concentration and a dose you supply; it does not suggest what either number should be, and that is deliberate.

And if the real driver is cost — which the published clinical commentary says it usually is⁸ — the more useful comparison is between sellers rather than between schedules. Our semaglutide provider comparison reads pharmacy, regulatory category and standing price off each company's own pages, and our cross-board pricing index puts every provider's real price for every compound in one sortable table. A cheaper standing price on a labeled schedule is a better answer to an affordability problem than a schedule nobody has tested.

The bottom line

Semaglutide has a dose-response relationship, and the trial that measured it found that lower doses produce less weight loss and fewer gastrointestinal side effects than higher ones — a trade-off, not a shortcut. Both current labels use 0.25 mg once weekly as a starting dose whose stated purpose is tolerability during the first four weeks, and neither label lists it as a maintenance dosage. No randomized trial has tested a fixed sub-label weekly dose held flat, which is the specific thing being marketed. The published record that uses the word "microdosing" consists of two letters and a clinical brief report, and the brief report's own framing is that the practice is driven by cost and tolerability rather than by evidence. What dose belongs to any individual patient is a prescriber's decision made on that patient's own clinical picture — this article's only claim is about what has and has not been studied.