Weight & Metabolic Health

GLP-1 "microdosing": what the evidence shows, and why the word is doing legal work

By the US Health Digest editorial team · August 6, 2026

"Microdosing" means taking a GLP-1 below the approved label's maintenance doses — usually compounded semaglutide or tirzepatide bought through telehealth. On PubMed searches run August 6, 2026 we could locate no randomized trial of these regimens for weight management in adults without diabetes; every efficacy figure below comes from trials of the approved schedule, where lower doses produced smaller average weight reduction. "Personalized dosing" is less a clinical claim than a legal one: the argument that a compounded product at an unapproved strength is not "essentially a copy" of an approved drug.

What the word means, and what it is being sold as

There is no regulatory or clinical definition of "microdosing" a GLP-1. In marketing it means a weekly dose below the label's maintenance doses, sold as personalized dosing or a low-dose protocol, and it bundles three unequally supported claims: fewer gastrointestinal side effects, lower cost, and weight maintenance without the full dose. Two practices share the name — stretching an approved multi-dose pen, and, the subject here, buying compounded product at a strength no approved label contains. See our explainers on compounded versus branded GLP-1s, which covers the marketing, and how GLP-1 telehealth works.

What the approved labels actually specify

Both drugs already start low, and both labels say so for a reason: Wegovy escalates "to reduce the risk of gastrointestinal adverse reactions," and Zepbound's 2.5 mg step "is for treatment initiation and is not approved as a maintenance dosage." The table records the labeling for the two injections in the adult weight-reduction indication — not advice; both labels also cover other indications at other maintenance dosages, and the Wegovy label now covers an oral tablet too.

Wegovy injection's labeled range now runs from 0.25 mg to 7.2 mg once weekly, so "low dose" is relative — see our piece on the 7.2 mg maximum semaglutide dose. Two things follow. The approved maintenance range for adult weight reduction starts at 1.7 mg, not 2.4 mg, so a 1.7 mg weekly dose is a labeled maintenance dosage rather than a microdose. And the label already contains lower strengths: a 0.25 mg or 1 mg weekly dose is not one only a compounder can make.

What the evidence at lower doses does — and does not — show

The in-label dose-response

SURMOUNT-1 randomized 2,539 adults with obesity, excluding diabetes, to once-weekly tirzepatide 5 mg, 10 mg or 15 mg or placebo for 72 weeks. On its treatment-regimen estimand — effects regardless of discontinuation, intention-to-treat — mean weight change at week 72 was −15.0% at 5 mg, −19.5% at 10 mg, −20.9% at 15 mg and −3.1% with placebo. One trial, one estimand: the gradient is a real dose-response signal. STEP 1 tested a single dose — once-weekly semaglutide 2.4 mg in 1,961 adults over 68 weeks — and on its primary estimand, effects regardless of discontinuation or rescue interventions, mean weight change was −14.9% versus −2.4% with placebo. Having tested no lower dose, it says nothing about one.

The one dose-ranging trial we located

One trial did test semaglutide across a dose range in adults with obesity and without diabetes: a phase 2, double-blind trial of 957 participants, published in The Lancet in 2018 by O'Neil and colleagues. On its intention-to-treat ANCOVA estimand with missing data derived from the placebo pool, estimated mean weight loss at week 52 was −2.3% for placebo versus −6.0%, −8.6%, −11.6%, −11.2% and −13.8% across the five semaglutide groups in ascending dose order.

That is the result most likely to be misused. Those doses were given once daily, not weekly, on a schedule matching no approved product; the trial ran 52 weeks; and the estimand differs from either pivotal trial's. These figures must not be set beside the STEP 1 or SURMOUNT-1 numbers above.

What we searched for and did not find

On August 6, 2026 we searched PubMed via the NCBI E-utilities API. A title search for microdosing, microdose or micro dose with semaglutide, tirzepatide or GLP-1 returned five records: two 2025 Diabetes Care letters on microdosing semaglutide from multidose pens in diabetes, a 2026 Obesity commentary on tirzepatide pens, a 2026 brief report in the Journal of the American Association of Nurse Practitioners, and an assay paper. None is a trial. A second search, for randomized trials combining subtherapeutic, submaintenance or maintenance-dose terms with semaglutide or tirzepatide and weight loss or obesity, returned zero records.

The nearest published comparison is not a microdosing trial either. STEP 2 randomized 1,210 adults with overweight or obesity and type 2 diabetes to once-weekly semaglutide 2.4 mg, 1.0 mg or placebo for 68 weeks, weight change coprimary on an intention-to-treat estimand — a lower weekly dose tested head to head, but in a population with diabetes and at a strength the manufacturer already sells.

Stated as a search result, not an absolute: as of August 6, 2026 we could locate no published randomized trial of the sub-maintenance weekly regimens now being sold, for weight management in adults without diabetes, and neither search surfaced a trial of compounded product at such doses. The 2018 dose-ranging trial above is not that trial: its schedule was daily. That is not evidence the practice fails — the question is unanswered, and anyone selling an answer does not have one.

Why "personalized dosing" is a regulatory argument

Compounded drugs escape FDA approval only under narrow conditions, one being that the product must not be "essentially a copy" of an approved drug. FDA's January 2018 final guidance for 503A pharmacies sets the test: a product is essentially a copy if it has the same active ingredient, "the same, similar, or an easily substitutable dosage strength," and the same usable route — with "similar" meaning within 10% of the available strength. The exception is narrow: a prescriber must determine "there is a change, made for an identified individual patient, which produces, for that patient, a significant difference," documented on the prescription. "Other factors, such as a lower price, are not sufficient …"

The 503B outsourcing-facility guidance is where strength matters. A drug "identical or nearly identical" to an approved drug — same active ingredient, route, dosage form, dosage strength and excipients — is essentially a copy outright, with no prescriber escape hatch, unless the drug is in shortage. But a product that "differs on one or more of these characteristics" is generally not identical or nearly identical. An unapproved strength clears that hurdle — and lands in 503B(d)(2)(B), under which a compounded drug built on a bulk substance also found in an approved drug is still essentially a copy "unless there is a change that produces, for an individual patient, a clinical difference, as determined by the prescribing practitioner."

A different strength thus moves a product from a rule with no exception to a rule with one, and "personalized dosing" gestures at the determination that exception requires. One caution: FDA has published no "essentially a copy" determination about these products, so applying the 2018 guidances to low-dose GLP-1s is our reading, not an agency determination.

The agency has addressed microdosing by name elsewhere, though. Its May 1, 2026 bulks-list notice records a nomination referring to "[i]njectable dosing, or microdosing," for controlling side effects, and an Outsourcing Facilities Association comment that "[m]icrodosing trends have been widely reported and allow patients otherwise intolerant to standard doses to continue treatment." FDA's answer — reached on clinical need, not on the copies rules — was that "[v]ague statements that a healthcare provider may determine that a different dose is needed for side effects and improved patient adherence, without identifying the dose, patients, or any attribute that makes the approved drugs unsuitable for such patients, do not establish clinical need." Announcing warning letters to 30 telehealth companies on March 3, 2026, Commissioner Marty Makary said compounders "should not try to compound drugs in a way that circumvents FDA's approval process."

The practical risks

First, dosing errors. FDA states it "received multiple reports of adverse events, some requiring hospitalization, that may be related to dosing errors associated with compounded injectable semaglutide products," from "patients measuring and self-administering incorrect doses." A low-dose regimen requires drawing a small volume from a multi-dose vial; confusion between milligrams, milliliters and syringe units is the recurring failure mode.

Second, what is in the vial: FDA notes compounded semaglutide products "may be the salt forms," which "are different active ingredients than are used in the approved drugs." Third, volume: as of May 31, 2026 FDA reported 990 adverse-event reports for compounded semaglutide and more than 730 for compounded tirzepatide (page current June 15, 2026). Such reports do not establish causation, but they are the signal that exists. See our guides to side effects in women and semaglutide results for women.

If the FDA finalises the 503B exclusion

Announcing that proposal on April 30, 2026, FDA said it "did not identify a clinical need for outsourcing facilities to compound" the three drugs from bulk substances. The comment period, extended by a notice published June 26, 2026, closed July 30, 2026, and a Federal Register search of docket FDA-2018-N-3240 run on August 6, 2026 returned nothing later: no final decision has been published as of today. We track the rulemaking in our explainer on the proposed exclusion, and the cost pressure behind it in GLP-1 costs without insurance.

Questions worth asking a prescriber who offers this

• What published trial supports the dose proposed? "Clinical experience" is a different kind of evidence.
• Is it compounded by a 503A pharmacy or a 503B outsourcing facility, and what determination of significant or clinical difference is documented on my prescription? That is what the exemption turns on.
• Is it a salt form? A multi-dose vial — and what am I asked to measure: milligrams, milliliters or syringe units?

More in our GLP-1 FAQ for women. We neither endorse nor discourage any dosing approach.

This article is general information, not medical or legal advice. It describes a practice we are not endorsing. Discuss any medication decision with a licensed clinician who knows your health history.