Microdosing a GLP-1 means using the drug at a fraction of the doses it was approved and studied at. The term appears in no FDA label and no clinical guideline. It is a market practice, sold as a named product by at least one telehealth company, and the honest summary of its evidence base on 2026-09-17 is short: three registered studies, no posted results, and labels that describe the lowest doses as starting doses rather than treatment.
This page reports what the practice is, who sells it, what the studies registered to test it measure, and what the approved labels say about the low doses it borrows from. Every figure carries its source and date. Nothing here is a dose, a schedule or advice for anyone; decisions about any medication belong with the prescriber, as our editorial standards set out. What each individual molecule's label and trial doses are is reported per compound on Peptifact's semaglutide dosage and tirzepatide dosage pages; this page is about the practice.
What the word means, depending on who uses it
There is no single definition, and the lack of one is the first fact about microdosing.
| Source | What "microdosing" refers to |
|---|---|
| UTHealth Houston trial NCT07325500 | 0.5 mg semaglutide weekly, after 12 weeks of titration up to 2 mg, to test weight regain |
| AgelessRx trial NCT07092605 | Semaglutide by sublingual or subcutaneous route in a compounding base, in healthy adults seeking "health and longevity" |
| Letters in Diabetes Care (2025) and Obesity (2026) | Dialling partial doses out of multi-dose semaglutide and tirzepatide pens |
| Seller marketing (AgelessRx, read 2026-09-17) | "Microdosing Tirzepatide" and "Longevity Microdosing GLP-1" sold as monthly treatments; the page read did not state a milligram figure |
The UTHealth definition is worth noticing: 0.5 mg of semaglutide weekly is a dose the Wegovy label lists among its escalation steps. In that trial, "microdosing" means a dose a quarter the size of the 2 mg participants had just reached, not something far below the label.
Who sells the practice
The practice is sold openly. On 2026-09-17, the AgelessRx website carried a banner reading "Lowest price ever on Microdosing Tirzepatide", priced at $159 a month against a struck-through $239, and listed "Longevity Microdosing GLP-1" among its treatments. The page carrying it, a blog post on combining microdosed GLP-1 with low-dose naltrexone for inflammation, was marked as last updated 16 September 2026.
AgelessRx also sponsors the oldest registered microdosing study, described below. A study sponsored by a seller of the product it tests is not invalid for that reason. It is a fact a reader needs when the study's results arrive, and it is recorded here so it is not lost.
The registered evidence: three studies, no results
A ClinicalTrials.gov search on 2026-09-17 for microdosing together with semaglutide, tirzepatide or GLP-1 returned three studies.
| Study | Sponsor | Design | Primary outcome | Status | Results |
|---|---|---|---|---|---|
| NCT07092605 | AgelessRx (industry) | Early phase 1, randomised, single-masked; placebo vs sublingual vs subcutaneous semaglutide | Pain, blood counts, metabolites, mood, heart rate variability, sleep, physical activity, at 6 months | Enrolling by invitation; started 2024-11-01; primary completion estimated 2025-11-01 | None posted |
| NCT07325500 (REINFORCE) | UTHealth Houston | Phase 2, randomised; titration to 2 mg semaglutide, then 0.5 mg weekly or stopping | Percent weight regain, weeks 12 to 60 | Recruiting; started 2026-02-12; completion estimated 2028-05-31 | Not yet due |
| NCT07588984 (FLOURISH / THRIVE) | Noom Inc. (industry) | Prospective cohort of GLP-1 medication with behavioural programmes | — | Not yet recruiting | Not yet due |
Two details in the AgelessRx record deserve attention:
- Body weight is not a primary outcome. The study is framed around health and longevity measures in healthy adults aged 18 to 65.
- Its estimated primary completion date has passed. The record was last updated on 30 July 2025 and shows no results. That can mean enrolment is still under way and the record has not been refreshed; it cannot be read as either a positive or a negative finding.
The UTHealth trial is the one designed to answer the question most people are asking — whether a much lower weekly dose holds the weight off after a full course — and it is not due to finish until 2028.
What the literature holds
A PubMed search for microdosing with semaglutide, tirzepatide or GLP-1 in the title or abstract returned 9 records on 2026-09-17. Read one by one:
- Four letters and commentaries on multi-dose pens, from or responding to a University of North Carolina group (Diabetes Care 2025, two items; Obesity 2026, two items). Their subject is the mechanics of dialling a partial dose from a pen built to deliver fixed ones, and the evidence gaps that come with it.
- One narrative review (Cureus, July 2026), which summarises possible benefits. A narrative review selects its sources; it is not a trial.
- One practice report for nurse practitioners (J Am Assoc Nurse Pract, May 2026), which describes the drivers as gastrointestinal tolerability and "vanity weight", and the risks as pen manipulation, compounded vials and medication sharing.
- Three records unrelated to the practice: an assay for a pharmacokinetic "microdose" study of an oral agonist, a microneedle gene-therapy paper, and a psilocybin study in mice, which the search term swept in.
None is a randomised trial of GLP-1 microdosing.
What the labels say about the low doses microdosing borrows from
The approved labels do not discuss microdosing. They do describe the low doses precisely, and they are explicit that those doses exist to start treatment.
| Label, read 2026-09-17 | What it says about its lowest doses |
|---|---|
| Wegovy (NDA215256) | "0.25 mg, 0.5 mg, and 1 mg once-weekly dosages are initiation and escalation dosages and are not approved as maintenance dosages." Adult maintenance: 2.4 mg or 1.7 mg. |
| Zepbound (NDA217806) | "The 2.5 mg dosage is for treatment initiation and is not approved as a maintenance dosage." Maintenance: 5, 10 or 15 mg. |
| Mounjaro (NDA215866) | "The 2.5 mg dosage is for treatment initiation and is not intended for glycemic control." |
| Rybelsus and Ozempic tablets (NDA213051) | Starting dose "is not effective for glycemic control." |
The stated reason for starting low is tolerability — the oral semaglutide label writes its starting and escalation schedule "to reduce the risk of gastrointestinal (GI) adverse reactions". That is also one of the reasons the commentary gives for why people microdose. The labels and the practice agree that low doses are easier to tolerate; they part on whether low doses are a treatment.
Why a lower dose is easier on the stomach, and why the gastric effect fades over time, is on how GLP-1 works.
A lower dose is not the same thing as microdosing
A reader weighing this practice may not realise how much evidence exists for the lowest approved maintenance doses, because they are approved and unremarkable.
SURMOUNT-1 (NCT04184622), the 72-week tirzepatide trial behind Zepbound, randomised 2,539 adults to 5, 10 or 15 mg or placebo. The posted results:
| Arm | Mean body-weight change at week 72 | Reached ≥5% reduction |
|---|---|---|
| Placebo | −2.4% | 27.9% |
| Tirzepatide 5 mg | −16.0% | 89.4% |
| Tirzepatide 10 mg | −21.4% | 96.2% |
| Tirzepatide 15 mg | −22.5% | 96.3% |
So a dose a third of the maximum was tested at scale, over 72 weeks, and is on the label. Wegovy's label likewise allows 1.7 mg as a maintenance dose. Those are lower doses with trial records. Microdosing generally refers to amounts below them, which is where the record stops.
The practical risk the record does document
The one hazard with a documented record is not the dose itself but how a non-standard dose gets measured.
- Multi-dose pens deliver fixed doses. The Diabetes Care and Obesity letters exist because people count clicks to extract partial doses, which the pens were not designed or labelled for.
- Compounded vials are measured in units and concentrations the branded pens do not use. The FDA's page on unapproved GLP-1 drugs, current as of 1 September 2026, reports adverse events, some requiring hospitalisation, linked to dosing errors with compounded semaglutide, from patients measuring their own doses and in some cases clinicians miscalculating them.
- Dosing error is already the dominant reported problem in this class. Tirzepatide's single most-reported adverse-event term in the FDA's FAERS database is a dosing error, not a symptom, as covered on our GLP-1 overdose page.
The compounded market these vials come from, and what changed when the shortages ended, is on compounded semaglutide. What the labels and trials say happens to weight after stopping altogether is on stopping GLP-1.
What is not established
- Whether microdoses produce weight loss, or hold weight off, has no published trial result. The trial designed to test maintenance is due in 2028.
- Whether microdoses produce the "longevity", inflammation or mood benefits sellers describe has no published trial result. The study measuring those outcomes has posted none.
- There is no agreed microdose. Figures in circulation differ by seller, by drug and by device.
- Tolerability at low doses is established; effectiveness at doses below maintenance is not, and for oral semaglutide the label states the starting dose is not effective for glycaemic control.
How this page was built
Registry records were pulled from the ClinicalTrials.gov v2 API on 2026-09-17, including sponsor, arms, primary outcomes, dates and results status; SURMOUNT-1's posted outcome measures were read from the same API. Label dosing sections were read from openFDA. The PubMed census used microdos*[tiab] AND (semaglutide[tiab] OR tirzepatide[tiab] OR GLP-1[tiab]), and each of the nine records was read and classified. The seller page was requested directly and quoted verbatim.