GLP-1 Microdosing: What the Term Means and What the Evidence Shows

Microdosing a GLP-1 means taking less of the drug than the label schedule calls for, usually at or below the approved starting dose, and staying there instead of titrating up. No regulator uses the term. Neither does any pivotal trial. There is no randomised controlled trial of microdosing as a weight-loss strategy, so every claim about how well it works is an extrapolation from data collected at higher doses. This page sets out what the word covers, what the approved schedules say, and what the published numbers support.

What people mean by “GLP-1 microdosing”

Three separate practices hide under one word, and they carry different risks.

The first is staying at the label initiation dose indefinitely rather than stepping up. For injectable semaglutide that is 0.25 mg weekly. For tirzepatide it is 2.5 mg weekly. Both of those doses exist in the label, so a prescriber can write them, but neither was studied as a dose someone stays on.

The second is dosing below the initiation step. Clinic and industry sources describe roughly 0.05 to 0.25 mg weekly for semaglutide and 0.5 to 2.5 mg weekly for tirzepatide. Those ranges describe what the term is used to mean in practice. They are not label doses, they were never tested, and quoting them here is a description of what is happening, not a suggestion.

The third is a cost behaviour: making a prescribed pen cover more weeks than it was labelled for. That is not a dosing strategy, it is a money problem, and it raises dose-accuracy and sterility questions. This site does not explain how any of it is done.

What a standard label dose looks like

Four products matter for this conversation, and each has a published schedule.

Injectable semaglutide (Wegovy, Ozempic) titrates 0.25 to 0.5 to 1.0 to 1.7 to 2.4 mg weekly, four weeks per step. Tirzepatide (Zepbound, Mounjaro) starts at 2.5 mg weekly for four weeks, then uses 5, 10 or 15 mg weekly as maintenance. Oral semaglutide is a 25 mg once-daily tablet for weight management. Orforglipron (Foundayo) is a once-daily oral small molecule at 6, 12 or 36 mg with no food or water restriction.

One point about that list is worth holding onto. No product has an approved dose below its initiation step. There is no approved microdose of anything in this class, which is why every microdose protocol is off-label by definition.

Why the approved starting dose is not a therapeutic dose

Starting doses were chosen so people could keep the drug down, not because they work. Gastrointestinal side effects are the main reason people stop taking GLP-1s, and stepping the dose up over months is the label's answer to that.

STEP 1 escalated participants from 0.25 mg to 2.4 mg semaglutide over 16 weeks. Not one arm stopped at 0.25 mg, so the trial produced no estimate of what sustained 0.25 mg dosing does to body weight. SURMOUNT-1 is the same shape at the bottom end: 2.5 mg tirzepatide was a four-week ramp, and the lowest dose anyone was randomised to stay on was 5 mg.

So “the starting dose” and “a low dose that was tested” are not the same object. A person sitting at 0.25 mg semaglutide for a year is in territory no trial has visited.

What the trials show about dose and weight loss

Every dose-response curve published in this field points the same way: more drug, more weight loss.

SURMOUNT-1 is the clearest dataset because it randomised three maintenance doses. Across 2,539 participants at 72 weeks, mean body-weight loss was 16.0% on tirzepatide 5 mg and 22.5% on 15 mg, against 2.4% on placebo. The gap widens sharply at the top end of response. Reaching 25% or more body-weight loss happened for 16.5% of the 5 mg group, 35% of the 10 mg group and 39.7% of the 15 mg group, versus 0.3% on placebo.

STEP 1 measured semaglutide 2.4 mg only, and found 14.9% mean body-weight loss at 68 weeks against 2.4% on placebo, with 86.4% of the treated group losing at least 5% versus 31.5% on placebo. ATTAIN-1 tested orforglipron at 6, 12 and 36 mg daily and reported 5.1%, 7.0% and 9.6% mean loss at week 72 against 2.5% on placebo under the treatment-regimen estimand, with 12.4% at 36 mg under the efficacy estimand.

Those numbers come from different trials with different durations, populations and placebo responses, so they are not directly comparable to each other. Read each one against its own placebo arm.

What evidence for microdosing exists, and what does not

The evidence is thin to the point of absence. There is no randomised controlled trial of GLP-1 microdosing as a weight-loss strategy, at any dose, for any product in the class.

What exists instead is three weaker things. Titration-phase observations show what happens during the four weeks someone spends at a starting dose on the way somewhere else. The SUSTAIN diabetes programme studied semaglutide at 0.5 and 1.0 mg, but with glycaemic primary endpoints in a diabetes population, so those weight figures do not transfer cleanly to people taking the drug for obesity. And there is self-report, which is real for the person reporting it and tells you nothing about the average.

That means any specific weight-loss percentage attached to a microdose is unsourced by this site's standard. If a page gives you a number for 0.1 mg semaglutide weekly, ask which trial arm produced it. There isn't one.

Why the trend grew

Price is most of the answer. Branded GLP-1s are expensive, coverage is patchy, and a lower dose means fewer milligrams per month.

Supply is the rest of it. During the shortages, US compounding pharmacies could legally make semaglutide and tirzepatide, and that cheap route disappeared. FDA declared the tirzepatide shortage resolved in December 2024 and the semaglutide shortage in February 2025, with 503A and 503B enforcement grace periods closing between February and May 2025. On 1 May 2026 FDA published a notice proposing not to include semaglutide, tirzepatide and liraglutide on the 503B outsourcing bulks list, tentatively finding no clinical need for them. That is a proposal with a comment period, later extended, not a final determination.

Side-effect avoidance is the third driver and the only one with a physiological basis. Nausea and vomiting do track with dose. Whether staying low to avoid them is a good trade is a different question, covered on the trade-off page.

Risks of dosing outside the label

Some risks scale down with dose. Several do not, and those are the ones people get wrong.

Semaglutide carries a boxed warning for thyroid C-cell tumours, and a personal or family history of medullary thyroid carcinoma or MEN 2 is a contraindication. Tirzepatide carries the same. Neither is a dose band. A smaller dose does not make a contraindicated person eligible. Pancreatitis and acute kidney injury are label risks for these products. On 6 June 2025 EMA's PRAC concluded that NAION, a sudden form of optic-nerve vision loss, is a very rare side effect of semaglutide at up to 1 in 10,000 users, and recommended it be added to EU product information.

Then there is the benefit that does not carry down. SELECT enrolled 17,604 patients and found a 20% reduction in major adverse cardiovascular events over a median 39.8 months, at semaglutide 2.4 mg weekly. No trial has tested whether any of that survives at a lower dose. Assuming it does is the single most consequential mistake in microdosing content.

Supply route carries its own risk, separate from dose. A 2024 FDA investigation found some compounded semaglutide products contained salt forms such as semaglutide sodium, which are not the approved active ingredient.

What the microdose calculator can and cannot tell you

This site has a microdose calculator. Here is what it is.

It reads published dose-response curves backwards. You give it a goal and a timeframe, and it returns the dose band whose trial-observed average rate of loss is closest to what you described. It shows the trial name, the dose and the week that produced each figure, and it flags loudly when your input falls outside the range any trial measured, which for microdose-level inputs it almost always will.

What it cannot do: it cannot tell you what to take. It has no information about your kidneys, your pancreas, your thyroid history, your other medicines, or how you tolerated the last step. It reports population averages from trials whose participants were selected, monitored and supported in ways you are not. Below the label initiation dose it is extrapolating past the end of the data, and it says so on screen. Treat the output as a reference figure to take to an appointment, not as a plan.

Questions to bring to a prescriber

The useful move is not deciding a dose at home. It is arriving with the right questions.

• What is the lowest dose you would consider for me, and what would you expect from it?
• Are any of my conditions or medicines a reason to avoid this class entirely?
• If side effects are why I want to stay low, what else can we do about them first?
• How would we know within three months whether a lower dose is working for me?
• If I reach my goal weight, what happens to the dose after that?
• What is the plan if I cannot afford the maintenance dose?

The evidence, one row per claim

Sources

1. FormBlends GLP-1 hub (secondary; verify against FDA label absence)
2. STEP 1 (NEJM 2021) summary
3. Eli Lilly / NEJM SURMOUNT-1
4. American College of Cardiology, SELECT
5. FDA, 'FDA clarifies policies for compounders as national GLP-1 supply begins to stabilize' (Internet Archive snapshot)
6. FDA compounding pages (verify exact citation before publishing)
7. ClinicalTrials.gov posted results for ATTAIN-1 (NCT05869903), first posted 14 August 2026

Keep reading

• Semaglutide Microdosing: The Ladder, the Data and the GapsThe label ladder laid next to the STEP programme, so you can see exactly how far a microdose sits from anythin
• Tirzepatide Microdosing: What SURMOUNT-1 Says About Low DosesTirzepatide has the only published maintenance dose-response curve in the field, and a microdose sits below th
• GLP-1 Dose Chart: Label Schedules and Trial Dose-ResponseThe reference table: what each label says, next to what each trial measured, with no microdose column because
• Microdosing vs Standard Dosing: What the Trade-Off Actually IsThe four axes people are really weighing, and which of them have evidence at a low dose. Two do, one does not.
• Low-Dose GLP-1 for Maintenance: What Is Known After Goal WeightThe strongest legitimate case for low-dose GLP-1, and the specific trial that has never been run.
• Do GLP-1 Side Effects Improve at Lower Doses?Which adverse effects actually track with dose, and which categories a lower dose changes nothing about.