Microdosing GLP-1 (Tirzepatide and Semaglutide): Who It's For, Doses, Evidence, Risks

Key takeaways

• "Microdosing" means taking less than the label starting dose (for example 1 mg of tirzepatide or 0.1 mg of semaglutide a week), stretching the interval to 10 or 14 days, staying on the starting step for months, or counting clicks on a multidose pen. It is done with compounded vials or manipulated pens; the brand single-dose pens cannot do it.
• No randomized trial has tested a dose below the label starting dose of either drug. The only sub-2.5 mg tirzepatide data is the 1 mg arm of a 26-week phase 2 trial in type 2 diabetes: HbA1c fell about one point, weight by under a kilogram, and 23 percent still had gastrointestinal side effects.
• The dose-response points the other way. In SURMOUNT-1, 5, 10 and 15 mg of tirzepatide produced 15.0, 19.5 and 20.9 percent weight loss; in STEP 2, semaglutide 1 mg produced 7.0 percent against 9.6 percent at 2.4 mg. Less drug does less, and nobody has measured where the effect disappears.
• The labels already contain the legitimate low-dose option: delay any step by four weeks if it is not tolerated, and hold the lowest approved maintenance dose, 1.7 mg of semaglutide or 5 mg of tirzepatide. Those are studied doses. A microdose is not.
• The risks are specific: microdose volumes are 2 to 10 units on an insulin syringe, the range in which FDA has documented 5- to 20-fold errors; weight returns when the drug stops; and a compounded "microdose program" is still a compounded copy under the rules that applied once the shortages ended.

"Microdosing tirzepatide" and "GLP-1 microdose" are among the most searched phrases in weight-loss medicine in 2026, and almost none of the pages that rank for them cite a trial, because there is none to cite. This page is the honest version: what the term means, what the dose-ranging studies of the brand drugs found at their lowest doses, what the Wegovy and Zepbound labels permit, and where the practice goes wrong. It sits alongside our guides to GLP-1 injection sites and to the GLP-1 patches, drops and gummies sold as needle-free alternatives, and it assumes the background in the compounded semaglutide versus tirzepatide hub. The label schedules are in the tirzepatide dosage chart and the semaglutide dosage chart; the syringe arithmetic lives at how many units are in 1 mL.

What people mean by "microdosing"

The word covers four practices that share a goal, less drug, and a problem, no data.

1. Starting below the label. The Zepbound label starts everyone at 2.5 mg once weekly and the Wegovy label at 0.25 mg; a microdose program starts at a fraction of that, often 0.5 to 1.25 mg of tirzepatide or 0.1 mg of semaglutide. A 2026 commentary in Expert Opinion on Pharmacotherapy defines microdosing as "25-50% of manufacturer-recommended starting doses or extended dosing intervals" (Beck 2026).
2. Staying on the starting step. Holding 2.5 mg of tirzepatide or 0.25 mg of semaglutide for months, even though the Zepbound label states that 2.5 mg "is for treatment initiation and is not approved as a maintenance dosage" (Zepbound label).
3. Stretching the interval. Injecting every 10 or 14 days instead of weekly, usually after reaching a goal weight, on the theory that a smaller average dose will hold the loss.
4. Fractional doses from a pen or vial. With a compounded vial, this means drawing a few units into an insulin syringe. With an Ozempic multidose pen, it means "click counting": dialing by the roughly 72 unnumbered clicks in the pen instead of the labeled dose. The pharmacists who described that method in Diabetes Care call it off-label and not endorsed by the manufacturer, and note that it depends on health literacy, eyesight, dexterity and the ability to follow complex instructions (Komé 2025). Wegovy, Zepbound and Mounjaro single-dose pens cannot be split at all.

The motives are consistent across the literature. A 2026 brief report for nurse practitioners names gastrointestinal tolerability and "vanity weight" as the drivers, with affordability pushing patients toward compounded vials, "research-grade" peptides and medical spas once the shortage-era supply dried up (Trainer 2026). The pharmacotherapy commentary adds cost and access during shortages, and says the practice spread through patient communities, telemedicine weight-loss programs and clinician anecdote rather than through any study (Beck 2026). Cost is a real pressure: compounded tirzepatide runs about $180 to $300 a month and brand Zepbound vials $299 to $449 through the manufacturer's program, as our cost guide records for October 2026, and a program that sells a tenth of a dose can price the vial accordingly.

What the evidence actually shows

As of October 2026, a PubMed search for microdosing of semaglutide or tirzepatide returns commentaries, a narrative review and one assay paper, and no randomized trial of any sub-label regimen. What exists is the dose-ranging data from the trials that led to approval, which tested lower doses than the labels use, and that data is worth reading closely because it is the only data there is.

Tirzepatide below 2.5 mg: one arm of one trial

The phase 2 trial that established tirzepatide's doses randomized 316 adults with type 2 diabetes (mean BMI 32.6) to 1, 5, 10 or 15 mg once weekly, dulaglutide or placebo for 26 weeks. The 1 mg arm is the only dose below 2.5 mg ever studied in humans. It lowered HbA1c by 1.06 percentage points (placebo 0.06), and body-weight change across the tirzepatide arms ran from 0.9 kg at 1 mg to 11.3 kg at 15 mg, against 0.4 kg on placebo. Gastrointestinal events were dose-related: 23.1 percent at 1 mg, 32.7 percent at 5 mg, 51.0 percent at 10 mg, 66.0 percent at 15 mg and 9.8 percent on placebo. Decreased appetite, reported as an adverse event, was 3.8 percent at 1 mg against 20.0 percent at 5 mg and 2.0 percent on placebo (Frias 2018). Read plainly: the lowest dose did something to glucose, little to weight, and still caused side effects in a quarter of people.

Semaglutide at low doses: the dose-ranging trial and STEP 2

Semaglutide's obesity program began with a 52-week phase 2 trial in 957 adults with obesity and no diabetes, using once-daily injections of 0.05, 0.1, 0.2, 0.3 or 0.4 mg. The lowest dose, 0.05 mg a day (about 0.35 mg a week, in the range of the label's 0.25 and 0.5 mg starting steps), produced 6.0 percent weight loss against 2.3 percent on placebo; 0.1 mg a day produced 8.6 percent and 0.4 mg a day 13.8 percent, with dose-related nausea (O'Neil 2018). That is the floor of the human data: no semaglutide dose below it has ever been tested for weight loss.

STEP 2 is the nearest thing to a weekly dose comparison. In 1,210 adults with overweight or obesity and type 2 diabetes, semaglutide 2.4 mg produced 9.6 percent weight loss at 68 weeks, the 1.0 mg diabetes dose 7.0 percent, and placebo 3.4 percent; gastrointestinal adverse events occurred in 63.5, 57.5 and 34.3 percent respectively (Davies 2021). Cutting the dose by more than half removed about a quarter of the weight loss and almost none of the nausea.

The dose-response at the top

SURMOUNT-1 fixed 2,539 adults with obesity at 5, 10 or 15 mg of tirzepatide for 72 weeks: 15.0, 19.5 and 20.9 percent weight loss against 3.1 percent on placebo (Jastreboff 2022). The curve is steep at the bottom and flat at the top, which is why 5 mg is a maintenance dose and 1 mg is not: 5 mg delivers roughly three-quarters of the maximum effect, and the one data point below it delivers almost none.

Microdosing claims, graded by the evidence behind them

Claim	Grade	What the evidence shows	Sources
Tirzepatide below 2.5 mg produces meaningful weight loss	C	One arm of one trial: 1 mg weekly for 26 weeks in type 2 diabetes lowered HbA1c by 1.06 points and weight by about 0.9 kg (11.3 kg at 15 mg); 23.1% had gastrointestinal events. No study in people without diabetes, no study below 1 mg, no phase 3 data below 2.5 mg.	Frias 2018
Low doses of semaglutide, around the label's starting steps, cause weight loss	B	Phase 2, 957 adults with obesity, 52 weeks of daily injections: 0.05 mg a day (about 0.35 mg a week) produced 6.0% loss versus 2.3% on placebo; 0.1 mg a day 8.6%; 0.4 mg a day 13.8%. Nothing below 0.05 mg a day has been tested.	O'Neil 2018
Lower doses mean less weight loss	A	SURMOUNT-1: 15.0% at 5 mg, 19.5% at 10 mg, 20.9% at 15 mg. STEP 2: 7.0% at semaglutide 1 mg versus 9.6% at 2.4 mg (placebo 3.4%). Phase 2 tirzepatide: 0.9 kg at 1 mg to 11.3 kg at 15 mg.	Jastreboff 2022; Davies 2021; Frias 2018
Microdosing gives the benefit with fewer side effects	C	Side effects are dose-related, so less drug produces fewer of them: gastrointestinal events 23.1% at tirzepatide 1 mg versus 66.0% at 15 mg; 57.5% at semaglutide 1 mg versus 63.5% at 2.4 mg. The benefit falls with the dose, and a microdose still causes events above placebo (9.8% and 34.3%).	Frias 2018; Davies 2021
A stretched interval (every 10 to 14 days) holds weight after goal	D	Untested. The withdrawal trials show what happens when the drug stops entirely: 14.0% regain within a year off tirzepatide, and two-thirds of the loss regained within a year off semaglutide. No trial has studied anything between full dose and none.	Aronne 2024; Wilding 2022
Microdosing is an established clinical protocol	D	The published literature is commentary. A 2025 Diabetes Care piece describes click counting as off-label and unvalidated by any trial; a 2026 commentary calls high-quality evidence 'essentially absent'; a 2026 brief report frames the practice as a response to cost and tolerability, not a studied regimen.	Komé 2025; Beck 2026; Trainer 2026
A compounded microdose vial is safer because the dose is small	D	The opposite risk is documented: small doses are small volumes, and FDA's 2024 alert describes patients drawing 50 units instead of 5 and clinicians miscalculating by 5 to 10 times. A 2025 secret-shopper study found 4 of 21 facilities supplying clinics were not licensed for sterile compounding.	FDA dosing-error alert 2024; DiStefano 2026
Slower titration and the lowest approved maintenance dose reduce side effects within the label	B	Both labels say to delay escalation if a dose is not tolerated; Wegovy allows a 1.7 mg maintenance dose for weight reduction, Zepbound a 5 mg maintenance dose and says to consider a lower maintenance dose if a higher one is not tolerated. These doses were trial arms; 5% of adolescents in the Wegovy pediatric trial finished on 1.7 mg as their maximum tolerated dose.	Wegovy label; Zepbound label

What the labels say about maintenance and the lowest approved doses

Both labels were written around the trial doses, and both contain more flexibility than the microdosing pages admit.

Sources: (Wegovy label) (Zepbound label). For type 2 diabetes, the Ozempic schedule makes 0.5 mg a maintenance dose, which is why "0.5 mg of semaglutide" is a label dose for diabetes and a sub-label dose for obesity; the semaglutide dosage chart lays both out. The practical reading is that a clinician has three on-label tools before reaching for a fraction: slow the climb, stop the climb at the lowest maintenance dose, or step back down to it. The Wegovy label records that 5 percent of adolescents in its pediatric trial finished on 1.7 mg as their maximum tolerated dose, which is what tolerability-driven dosing looks like inside a trial (Wegovy label).

The search phrase "Zepbound maintenance dose after goal weight" has no label answer because the question assumes a phase the trials never had. SURMOUNT-4 is the closest: after 36 weeks on tirzepatide (20.9 percent loss), 670 adults were randomized to continue at 10 or 15 mg or switch to placebo for a year; the placebo group regained 14.0 percent and the continuing group lost a further 5.5 percent (Aronne 2024). The semaglutide version, the STEP 1 extension, found two-thirds of the lost weight back within a year of stopping (Wilding 2022). Nobody has run the trial of 2.5 mg every other week after goal, so nobody knows whether it lands nearer the continuing group or the placebo group.

What is plausible, and what is marketed

Appetite suppression at low doses is real for some people, and unquantified. The phase 2 trials show that low doses do something: 0.05 mg of semaglutide a day produced 6 percent weight loss, and 1 mg of tirzepatide produced a measurable glucose effect and a small appetite signal (decreased appetite in 3.8 percent, against 2.0 percent on placebo and 20.0 percent at 5 mg) (O'Neil 2018) (Frias 2018). Someone who says a quarter dose blunts their hunger may be describing a genuine pharmacological effect. What no one can tell them is how much weight it will cost them in a year, because the effect at that dose was never measured against placebo in the population they belong to.

"Fewer side effects" is true in the way that less of anything does less. Gastrointestinal events fell from 66 percent at 15 mg to 23 percent at 1 mg of tirzepatide, but 23 percent is still more than double the placebo rate (Frias 2018). In STEP 2, halving the semaglutide dose cut nausea-type events by six percentage points and weight loss by 2.6 points (Davies 2021). A microdose trades effect for tolerability at a rate nobody has published.

A stretched interval is a lower dose by another name. Semaglutide's elimination half-life is about a week and tirzepatide's about five days (Wegovy label) (Zepbound label). Weekly injections let the level build to a steady plateau; a 14-day gap lets it fall by roughly half before the next dose, so "2.5 mg every two weeks" is not a steady 2.5 mg, it is a sawtooth that averages well below it. Whatever is being tested by that schedule, it is not the dose the label describes.

Claims with no human data at all. "Metabolic reset," "anti-inflammatory microdose," "longevity dosing," and microdosing for "vanity weight" in people whose BMI would have excluded them from every trial (STEP 1 and SURMOUNT-1 enrolled adults with a BMI of 30, or 27 with a weight-related condition). The early reports of benefit summarized in the commentary literature are anecdotes and small series, and the authors say so (Beck 2026). A program that quotes "up to 20 percent weight loss" for its microdose plan is quoting SURMOUNT-1's 15 mg arm for a product that delivers a tenth of it.

The specific risks

Measurement error grows as the dose shrinks

The brand pens deliver a fixed dose; a microdose comes from a vial and a U-100 insulin syringe, on which 100 units equal 1 mL (units guide). The smaller the dose, the smaller the volume, and the less room for error. At example concentrations:

This is an illustration of why the arithmetic fails, not a dosing chart; the concentrations are examples, and only the number on your own label counts. A 2-unit dose is the smallest mark on a 0.3 mL syringe and a hairline on a 1 mL syringe. FDA's July 2024 alert on compounded semaglutide described exactly this failure mode: a patient told to draw 5 units drew 50, prescribers converted milligrams to units incorrectly by 5 to 10 times, and patients reached hospital with severe vomiting, hypoglycemia and dehydration after doses 5 to 20 times what was intended (FDA dosing-error alert 2024). The tirzepatide dosage chart has the full units table; run every new vial through the peptide calculator.

Click counting

Dialing an Ozempic pen by clicks produces a dose that appears nowhere on the pharmacy label. The Diabetes Care authors who published a click-counting handout list the consequences themselves: dosing errors "could lead to subtherapeutic effects or increased side effects," the pen's 56-day beyond-use date runs out before a stretched pen does, and the practice is unsuitable for anyone at risk of confusion (Komé 2025). A prescription label that says 0.5 mg while the patient takes 18 clicks is also a hospital admission waiting to misread the chart.

Paying for a sub-therapeutic dose

The 1 mg arm lost under a kilogram in six months. A microdose program charges a monthly fee for a product whose expected effect is somewhere between that and nothing, while the nausea, the injection and the legal exposure are all still present. The commentary literature calls these "hazardous false economies" (Trainer 2026).

Weight regain

Anyone microdosing to "maintain" after a full-dose course is running an experiment whose control arms are known: continuing the dose held the loss, stopping it returned 14 percent within a year on tirzepatide and two-thirds of the loss on semaglutide (Aronne 2024) (Wilding 2022). The microdose arm has never been run.

Legal status after the shortages, and the 2026 warning letters

Semaglutide and tirzepatide are approved drugs; the microdose vial is not. Since FDA declared the tirzepatide shortage resolved on December 19, 2024 and the semaglutide shortage on February 21, 2025, a compounded version is lawful only for an individual patient whose documented clinical need the approved product cannot meet, and FDA has said that price or commercial availability alone is not such a need (FDA GLP-1 concerns page). "The patient needs a smaller dose than the pens deliver" is the argument microdose programs make for themselves; the brief report that examined it concludes the legitimate role for compounding is narrow, for documented allergies, not for a marketed low-dose plan (Trainer 2026). In 2026 FDA sent warning letters to 30 telehealth companies on March 3 and 25 more by June 16 over marketing that implied compounded GLP-1s were equivalent to or approved like the brands (FDA 2026); the legal tracker has every date and the May 2026 proposal to remove both drugs from the 503B list. A smaller amount of a copy is still a copy.

Where the vial came from

A secret-shopper study of 75 weight-loss clinics and medical spas in West Virginia and Oklahoma, run from August to October 2025 after the shortages ended, identified 23 compounding facilities supplying them; 4 of the 21 that could be checked were not licensed for sterile compounding, one had received multiple FDA warning letters since 2023, and three had faced state discipline (DiStefano 2026). A microdose drawn from an unverified vial is an unknown fraction of an unknown amount.

Who should not microdose

• Anyone the labels exclude at any dose: a personal or family history of medullary thyroid carcinoma or MEN 2, pregnancy or a planned pregnancy, prior serious hypersensitivity to the drug. The boxed warning does not scale with the dose (Wegovy label) (Zepbound label).
• People outside the trial population, meaning a BMI below 27, who are using a microdose to lose "vanity weight." Every efficacy and safety number above was earned in adults with obesity; none of it transfers.
• Anyone who cannot do the units arithmetic, read a half-unit mark, or keep a vial refrigerated, and anyone who would be counting clicks with impaired vision, tremor or memory (Komé 2025).
• People on insulin or sulfonylureas, where an unlabeled dose confuses the hypoglycemia management the labels require.
• Anyone whose plan is to stop. The regain trials apply at every dose.
• Anyone buying "research-grade" vials without a prescription, a pharmacy or a concentration they can trust; FDA's 2026 letters rejected the research-use defense, as the tracker records.
• Anyone doing it only to save money. At the lower doses, brand Zepbound vials cost $299 to $399 and Wegovy $199 to $349 a month through the manufacturer programs in October 2026 (cost guide); a cheap microdose vial buys an untested dose of an unverified product.

What a clinician-managed low-dose approach actually looks like

There is a legitimate version of "I cannot tolerate the standard climb," and it lives entirely inside the labels.

1. Slower titration. Both labels allow a step to be delayed when it is not tolerated; four weeks at each dose is a minimum, not a schedule. A clinician can hold 0.5 mg of semaglutide or 5 mg of tirzepatide for as long as it takes.
2. Tolerability-driven dose holds. The Zepbound label tells prescribers to weigh response and tolerability in choosing a maintenance dose and to consider a lower one when a higher one is not tolerated; the Wegovy label offers 1.7 mg as a maintenance dose for weight reduction (Zepbound label) (Wegovy label). These were trial arms with measured outcomes.
3. Switching drugs, not fractions. Someone who cannot tolerate one drug may tolerate the other at its lowest maintenance dose; the comparison hub covers the trade-offs.
4. A written dose in milligrams, a concentration, and the units, whether the product is a pen or a vial, with the pharmacist walking through the syringe.
5. A maintenance plan before goal weight, which in 2026 means continuing a labeled maintenance dose, because that is the only strategy with a trial behind it.

Where a clinician does individualize below the label, for instance during a transition between drugs, the published guidance from the clinicians who do it is to document the exact dose and click count in the record and on the pharmacy label, counsel verbally and in writing, select patients carefully, and monitor closely, while acknowledging that no trial supports the regimen (Komé 2025). That is a clinician taking responsibility for an unstudied decision, which is different from a subscription page selling one.

Deciding: match the reason to the evidence

Bottom line

Microdosing GLP-1s is a practice without a trial. The lowest doses ever tested, 1 mg of tirzepatide in diabetes and 0.05 mg of semaglutide a day in obesity, did less than the label doses and still caused side effects; everything below them is unmeasured, and everything stretched beyond a week is a lower dose with a different name. The labels already give a clinician two honest tools for people who struggle with the standard schedule, a slower climb and the lowest approved maintenance dose, and both come with trial data. The microdose vial comes with a syringe measured in single units, a product nobody verified, a legal status that depends on a written clinical need, and a monthly fee for an effect that may be close to nothing. If you want less drug, ask for the smallest dose that was studied, not the smallest one that was sold.

Frequently asked questions

+−What is microdosing tirzepatide?

Taking less tirzepatide than the Zepbound label's 2.5 mg starting dose (often 0.5 to 1.25 mg a week), staying on 2.5 mg for months instead of four weeks, or stretching the weekly injection to every 10 to 14 days. A 2026 commentary defines it as 25 to 50 percent of the manufacturer's starting dose or an extended interval. It is done from a compounded vial drawn up in insulin-syringe units, because the brand single-dose pens cannot deliver a fraction. No trial has tested any of these regimens.

+−Does microdosing tirzepatide work?

Nobody knows, because it has not been studied. The only dose below 2.5 mg ever trialed was 1 mg once weekly in a 26-week phase 2 study in type 2 diabetes, where HbA1c fell by about one point and weight by under a kilogram, against 11.3 kg at 15 mg; 23 percent of the 1 mg group still had gastrointestinal side effects. In the phase 3 obesity trial, 5 mg produced 15.0 percent weight loss and 15 mg 20.9 percent, so the effect shrinks as the dose falls. Where between 0 and 5 mg it disappears in people without diabetes has never been measured.

+−What are the benefits of microdosing tirzepatide?

The honest answer is 'possibly less nausea, at the cost of less effect, with no trial to size either'. Gastrointestinal side effects are dose-related (23 percent at 1 mg, 66 percent at 15 mg in the phase 2 trial), so a smaller dose does produce fewer of them. A 2026 pharmacotherapy commentary summarizes the early reports as reduced nausea, better adherence and modest weight loss, and calls high-quality evidence 'essentially absent'. Claims about metabolic resets, inflammation or longevity at microdoses have no human data at all.

+−Is there a microdosing tirzepatide chart?

Charts circulate online, but none is based on a study, and that is the point: a chart implies someone measured what each step does, and for doses below 2.5 mg nobody has. The only tirzepatide chart with evidence behind it is the Zepbound label schedule, which starts at 2.5 mg for four weeks and climbs in 2.5 mg steps at least four weeks apart to a maintenance dose of 5, 10 or 15 mg. Our tirzepatide dosage chart reproduces it with the units arithmetic for compounded vials.

+−What is the lowest dose of Zepbound?

2.5 mg once weekly is the lowest dose on the label, and the label says it is for starting treatment only and is not approved as a maintenance dose. The lowest approved maintenance dose is 5 mg; the label tells prescribers to consider a lower maintenance dose if a higher one is not tolerated. For semaglutide, the lowest approved maintenance dose is 1.7 mg of Wegovy for weight reduction and 0.5 mg of Ozempic for type 2 diabetes.

+−What is the Zepbound maintenance dose after goal weight?

The label has no such dose. Zepbound's indication is to reduce weight and maintain the reduction on the same maintenance dose of 5, 10 or 15 mg. The only trial of what happens after goal, SURMOUNT-4, kept people on their 10 or 15 mg dose; those switched to placebo regained 14 percent of their body weight within a year, while those who continued lost a further 5.5 percent. Whether a lower dose or a longer interval can hold the weight has not been tested.

+−Can you microdose semaglutide?

People do, usually by counting clicks on an Ozempic multidose pen or drawing a fraction of a dose from a compounded vial; Wegovy pens are fixed-dose and cannot be split. The manufacturer does not endorse click counting, and a 2025 Diabetes Care commentary from the pharmacists who described the practice says no clinical trial has validated its safety or efficacy. The lowest semaglutide dose ever tested for weight loss was 0.05 mg a day in a phase 2 trial, which produced 6.0 percent weight loss at 52 weeks against 2.3 percent on placebo; nothing below that has been studied.

+−Is microdosing GLP-1 safe, and is it legal?

The drug's risks do not disappear at a low dose: the thyroid, pancreatitis and pregnancy warnings on the labels apply at any dose. The added risk is measurement. Microdoses are 2 to 10 units on an insulin syringe, where FDA has documented patients drawing 5 to 20 times the intended dose. Legally, the drugs themselves are approved; a compounded microdose vial is a compounded copy, which since the shortages ended is lawful only for a patient with a documented need the brand cannot meet. In 2026 FDA sent 55 warning letters to telehealth companies over how compounded GLP-1s were marketed.

Sources

1. LabelNovo Nordisk. Wegovy (semaglutide) injection and tablets: prescribing information, revised June 2026 (DailyMed setid ee06186f-2aa3-4990-a760-757579d8f77b). (2026)
2. LabelEli Lilly. Zepbound (tirzepatide) injection, for subcutaneous use: prescribing information, revised August 2026 (DailyMed setid 487cd7e7-434c-4925-99fa-aa80b1cc776b). (2026)
3. StudyFrias JP et al. Efficacy and safety of LY3298176, a novel dual GIP and GLP-1 receptor agonist, in patients with type 2 diabetes: a randomised, placebo-controlled and active comparator-controlled phase 2 trial. Lancet 2018;392:2180-93. (2018)
4. StudyJastreboff AM et al. Tirzepatide once weekly for the treatment of obesity (SURMOUNT-1). N Engl J Med 2022;387:205-16. (2022)
5. StudyO'Neil PM et al. Efficacy and safety of semaglutide compared with liraglutide and placebo for weight loss in patients with obesity: a randomised, double-blind, placebo and active controlled, dose-ranging, phase 2 trial. Lancet 2018;392:637-49. (2018)
6. StudyDavies M et al. Semaglutide 2.4 mg once a week in adults with overweight or obesity, and type 2 diabetes (STEP 2): a randomised, double-blind, double-dummy, placebo-controlled, phase 3 trial. Lancet 2021;397:971-84. (2021)
7. StudyAronne LJ et al. Continued treatment with tirzepatide for maintenance of weight reduction in adults with obesity: the SURMOUNT-4 randomized clinical trial. JAMA 2024;331:38-48. (2024)
8. StudyWilding JPH et al. Weight regain and cardiometabolic effects after withdrawal of semaglutide: the STEP 1 trial extension. Diabetes Obes Metab 2022;24:1553-64. (2022)
9. ReviewBeck H, Clements JN. Considerations and challenges in microdosing of GLP-1-based receptor agonists (commentary). Expert Opin Pharmacother 2026, online October 5. (2026)
10. ReviewTrainer N. The 'microdosing' dilemma: balancing patient anecdotes with clinical safety amid GLP-1 compounding restrictions (brief report). J Am Assoc Nurse Pract 2026, online May 26. (2026)
11. SourceKomé AM et al. One size does not fit all: understanding microdosing semaglutide for diabetes in multidose pens (commentary). Diabetes Care 2025;48:e25-e27. (2025)
12. StudyDiStefano MJ et al. Postshortage compounded GLP-1 RA market in 2 states with potentially high demand (secret-shopper study). JAMA Health Forum 2026;7:e262207. (2026)
13. RegulatoryFDA. FDA alerts health care providers, compounders and patients of dosing errors associated with compounded injectable semaglutide products. July 26, 2024 (examples as summarized by Patient Care Online, July 30, 2024). (2024)
14. RegulatoryFDA. FDA's concerns with unapproved GLP-1 drugs used for weight loss (compounding after the shortages, salt forms, adverse event reports; content current as of October 1, 2026).
15. RegulatoryFDA. FDA warns 30 telehealth companies against illegal marketing of compounded GLP-1s. Press announcement, March 3, 2026 (25 further letters made public June 16, 2026). (2026)