What Clinical Trials Show About Microdosing GLP-1 Drugs

The trials that produced the metabolic and cardiovascular results were run at specific labelled doses. In the dose-ranging arms, the effect scaled with the dose. Lower doses delivered smaller effects in dose-ranging arms. At Longevity Benchmark, we read the approved labels and the trial registry records before we write about a dose.

Our search of published medical literature for completed randomized trials of sub-therapeutic "microdose" maintenance regimens identified zero trials with posted or published results. Commercial claims that fractional doses deliver full cardiometabolic outcomes remain unsupported extrapolations from trials that tested standard schedules.

Dose-Response Evidence in Registration Trials

Clinical trial data demonstrate that metabolic and weight reduction outcomes scale directly with GLP-1 receptor agonist dosage.

In the STEP 2 trial (NCT03552757), 1,210 adults with type 2 diabetes and overweight or obesity were randomized to semaglutide 1.0 mg, semaglutide 2.4 mg, or placebo for 68 weeks. Mean weight change was -7.2 percent with 1.0 mg versus -9.9 percent with 2.4 mg and -3.3 percent with placebo. Absolute weight reduction reached -7.1 kg at 1.0 mg and -9.9 kg at 2.4 mg, compared to -3.4 kg on placebo. Waist circumference fell 6.9 cm in the 1.0 mg arm and 9.7 cm in the 2.4 mg arm, versus 4.3 cm on placebo.

Threshold achievements in STEP 2 reflected this separation. A weight reduction of at least 5 percent was reached by 57.1 percent of participants on 1.0 mg versus 68.8 percent on 2.4 mg. A reduction of at least 10 percent was achieved by 28.7 percent on 1.0 mg and 45.6 percent on 2.4 mg, compared to 8.2 percent on placebo. Glycaemic response differed far less between active arms: glycated haemoglobin (HbA1c) fell 1.5 percentage points at 1.0 mg and 1.7 percentage points at 2.4 mg, compared to 0.3 percentage points on placebo. Fasting plasma glucose fell 36.5 mg/dL at 1.0 mg and 37.9 mg/dL at 2.4 mg. Systolic blood pressure fell 3 mmHg at 1.0 mg and 4 mmHg at 2.4 mg.

Tirzepatide demonstrated a step-wise relationship in the SURMOUNT-1 trial (NCT04184622). Across 2,539 adults without diabetes treated for 72 weeks, posted least-squares mean weight change was -2.4 percent on placebo, -16.0 percent at 5 mg, -21.4 percent at 10 mg, and -22.5 percent at 15 mg once weekly. Achieving at least 15 percent weight reduction occurred in 50.24 percent of participants at 5 mg, 73.61 percent at 10 mg, and 78.24 percent at 15 mg, versus 5.98 percent on placebo. Achieving at least 20 percent weight loss reached 31.62 percent at 5 mg, 55.48 percent at 10 mg, and 62.88 percent at 15 mg.

Metabolic parameters tracked this exposure gradient. Waist circumference fell 14.6 cm at 5 mg, 19.4 cm at 10 mg, and 19.9 cm at 15 mg, compared to 3.4 cm on placebo. Fasting glucose fell 7.73 mg/dL, 9.73 mg/dL, and 10.55 mg/dL across the respective doses, while rising 0.86 mg/dL on placebo. Steady-state tirzepatide exposure was dose-proportional, with geometric mean area under the curve values of 88,900 ng·h/mL at 5 mg, 177,000 at 10 mg, and 266,000 at 15 mg. Comparative trial data between these molecules are detailed in our analysis of semaglutide vs tirzepatide.

Approved Label Schedules for Titration and Maintenance

Approved product labels define starting doses solely as temporary initiation steps to mitigate gastrointestinal intolerance.

The Wegovy label, revised June 2026, specifies subcutaneous escalation of 0.25 mg weekly in weeks 1 to 4, 0.5 mg in weeks 5 to 8, 1 mg in weeks 9 to 12, and 1.7 mg in weeks 13 to 16. The label states that the purpose of dose escalation is to "reduce the risk of gastrointestinal adverse reactions." The escalation ends at a maintenance dose. For chronic weight management and cardiovascular risk reduction, the label establishes 2.4 mg once weekly as the recommended maintenance dosage, with 1.7 mg once weekly designated as an alternative maintenance dosage. For noncirrhotic metabolic dysfunction-associated steatohepatitis (MASH), the label specifies 2.4 mg weekly, directing that if not tolerated, dosage should be reduced to 1.7 mg with re-escalation considered.

The Wegovy label also indicates that adults who tolerate 2.4 mg weekly for at least 4 weeks and require additional weight reduction may have their dosage increased to a maximum of 7.2 mg once weekly. The oral Wegovy tablet schedule outlines 1.5 mg daily on days 1 to 30, 4 mg on days 31 to 60, 9 mg on days 61 to 90, and 25 mg daily from day 91 onward as the maintenance dosage. Ozempic follows the same shape. Its label, revised May 2026 directs prescribers to initiate treatment at 0.25 mg weekly for 4 weeks before increasing to 0.5 mg. The Ozempic label states that "The recommended maintenance dosage is 0.5 mg, 1 mg, or 2 mg" once weekly, and identifies 2 mg as the maximum recommended dosage.

Tirzepatide labels maintain an identical separation between initiation and maintenance. The Zepbound label (revised August 2026) states: "The recommended starting dosage of ZEPBOUND for all indications is 2.5 mg injected subcutaneously once weekly for 4 weeks." The label specifies: "The 2.5 mg dosage is for treatment initiation and is not approved as a maintenance dosage." Recommended maintenance dosages for weight reduction are 5 mg, 10 mg, or 15 mg once weekly. The Mounjaro label (revised August 2026) also defines 2.5 mg once weekly as the starting dosage to reduce gastrointestinal adverse reactions, but does not use the explicit non-maintenance exclusion clause found on Zepbound.

Oral semaglutide formulations reinforce that starting doses lack therapeutic efficacy. The Rybelsus label (revised January 2026) mandates 3 mg once daily for 30 days before escalating to 7 mg or 14 mg. The label states that the 3 mg starting dosage "is not effective for glycemic control" and exists only to reduce gastrointestinal events. Tablets of Rybelsus and Ozempic are not substitutable on a milligram-to-milligram basis.

All five labels give the same stated reason for starting low: reducing the risk of gastrointestinal adverse reactions.

Measured Outcomes in Sub-Therapeutic Phase 2 Trial Arms

Phase 2 dose-finding trials evaluated dosages below marketed regimens, showing measurable effects that remained substantially smaller than approved maintenance doses.

In trial NCT02453711, 957 adults with obesity without diabetes were treated for 52 weeks. The trial evaluated once-daily subcutaneous semaglutide at 0.05, 0.1, 0.2, 0.3, and 0.4 mg against liraglutide 3.0 mg daily and pooled placebo. These experimental once-daily doses are not weekly regimens and represent unapproved administration schedules. Mean percent weight reduction scaled with dose. It ran from -5.99 percent at 0.05 mg once-daily through -8.62 percent at 0.1 mg, -11.60 percent at 0.2 mg and -11.17 percent at 0.3 mg, to -13.84 percent at 0.4 mg once-daily under slow escalation. Placebo was -2.29 percent. Achieving at least 5 percent weight reduction occurred in 53.50 percent of participants receiving 0.05 mg once-daily, compared to 82.52 percent at 0.4 mg once-daily and 22.87 percent on placebo.

Sub-therapeutic dosing did not eliminate gastrointestinal intolerance in NCT02453711. These once-daily schedules are not approved regimens. Nausea was reported by 31.1 percent of participants on 0.05 mg once-daily semaglutide versus 17.6 percent on placebo, rising to 48.0 percent at 0.4 mg once-daily. Constipation occurred in 12.6 percent at 0.05 mg once-daily versus 4.4 percent on placebo. Diarrhoea was reported in 19.4 percent at 0.05 mg once-daily versus 11.8 percent on placebo. In non-diabetic participants, HbA1c reduction was minimal across all daily doses, ranging from -0.13 percentage points at 0.05 mg once-daily to -0.29 percentage points at 0.4 mg once-daily, compared to -0.01 percentage points on placebo.

A separate phase 2 trial (NCT03131687) tested a tirzepatide 1 mg once-weekly arm, which sits below the 2.5 mg labelled starting dose, in 318 adults with type 2 diabetes over 26 weeks. In that trial, body weight fell 0.9 kg at tirzepatide 1 mg once weekly versus 0.4 kg on placebo, compared to falls of 4.8 kg at 5 mg, 8.7 kg at 10 mg, and 11.3 kg at 15 mg. Achieving at least 5 percent weight loss was documented in 13.5 percent on 1 mg versus 0 percent on placebo, 47.3 percent at 5 mg, and 70.6 percent at 10 mg. Nausea at 1 mg (3.8 percent) was no higher than placebo (5.9 percent), while reaching 20.0 percent at 5 mg and 39.6 percent at 15 mg.

In a 72-week histological trial (NCT02970942) examining 320 adults with non-alcoholic steatohepatitis, once-daily semaglutide was evaluated at 0.1, 0.2, and 0.4 mg daily against placebo. Steatohepatitis resolution without worsening fibrosis was reached by 40.4 percent at 0.1 mg once-daily, 35.6 percent at 0.2 mg once-daily, and 58.9 percent at 0.4 mg once-daily, compared to 17.2 percent on placebo. Weight fell 4.8 kg at 0.1 mg once-daily, 9.4 kg at 0.2 mg once-daily, and 12.3 kg at 0.4 mg once-daily, versus 1.0 kg on placebo. These daily administrations are unapproved schedules. The approved indication for MASH on the Wegovy label requires 2.4 mg once weekly.

Cardiovascular Outcomes Demonstrated Only at Standard Maintenance Dosages

Major adverse cardiovascular event reductions have been proven exclusively in clinical trials that maintained participants on approved therapeutic dosages.

In the SELECT trial (NCT03574597), 17,604 adults were randomized to semaglutide or placebo. Participants underwent fixed escalation to a maintenance dose of 2.4 mg once weekly. Over the trial duration, the primary composite outcome of cardiovascular death, non-fatal myocardial infarction, or non-fatal stroke occurred in 569 of 8,803 semaglutide participants (6.5 percent) versus 701 of 8,801 placebo participants (8.0 percent). This represented a hazard ratio of 0.80 (95 percent CI, 0.72 to 0.89, p < 0.0001).

Secondary endpoints in SELECT documented 223 cardiovascular deaths on semaglutide 2.4 mg versus 262 on placebo, and 375 all-cause deaths versus 458. The heart failure composite occurred in 300 versus 361 participants. Mean weight reduction at week 104 reached -9.39 percent with semaglutide 2.4 mg versus -0.87 percent on placebo. HbA1c shifted -0.31 percentage points versus +0.01 percentage points. SELECT provides no outcome data for patients maintained on sub-therapeutic doses.

Cardiovascular event reductions in patients with diabetes were examined in the SUSTAIN-6 trial (NCT01720446), which randomized 3,297 participants to once-weekly semaglutide 0.5 mg, 1.0 mg, or placebo. The primary composite outcome occurred in 6.6 percent on pooled semaglutide versus 8.9 percent on pooled placebo (hazard ratio 0.74, 95 percent CI, 0.58 to 0.95). Posted results report this outcome pooled across the 0.5 mg and 1.0 mg arms rather than per individual dose arm. Surrogate metabolic markers showed dose-dependent differences: HbA1c fell 1.09 percentage points at 0.5 mg and 1.41 at 1.0 mg, while body weight fell 3.57 kg at 0.5 mg and 4.88 kg at 1.0 mg.

Comparative efficacy across doses was evaluated in the 40-week SURPASS-2 trial (NCT03987919) in 1,879 adults with type 2 diabetes. Reductions in HbA1c reached 2.09 percentage points with tirzepatide 5 mg, 2.37 at 10 mg, and 2.46 at 15 mg, compared to 1.86 percentage points with semaglutide 1.0 mg weekly. Body weight fell 7.8 kg at tirzepatide 5 mg, 10.3 kg at 10 mg, and 12.4 kg at 15 mg, versus 6.2 kg with semaglutide 1.0 mg. The proportion of patients achieving at least 5 percent weight reduction was 68.55 percent, 82.35 percent, 86.21 percent, and 58.44 percent across the respective groups.

Current Research on Unapproved Low-Dose Maintenance Protocols

A search of ClinicalTrials.gov and PubMed identified no completed randomized controlled trials with posted or published results evaluating "microdose" GLP-1 maintenance regimens.

Only two registered randomized studies address the question directly, and neither has posted results. The first study (NCT07092605), is an Early Phase 1 trial evaluating sublingual and subcutaneous semaglutide 1 mg/mL against placebo over six months. The trial record lists an estimated enrolment of 150 participants and an estimated completion date of November 2025, but no results are posted. Registered primary outcomes track self-reported pain, blood counts, sleep ratios, and heart rate variability rather than body weight, cardiovascular events, or mortality. No trial in this evidence base measured a longevity endpoint at any dose.

The second study (NCT07325500), the REINFORCE trial, is an open-label Phase 2 study with an estimated enrolment of 30 participants living with HIV. Following a 12-week escalation to semaglutide 2.0 mg weekly, participants are randomized to semaglutide 0.5 mg weekly or no semaglutide to track weight regain through week 60. The trial began in February 2026, with an estimated completion date in May 2028. No results are posted.

A PubMed title and abstract search for microdosing combinations with incretin therapies yielded 13 records, of which the GLP-1-relevant items are commentaries, letters, a narrative review, and one retrospective observational cohort. A commentary in the journal Obesity defines microdosing as "an off-label technique where lower or fractional doses are used to improve access, affordability, and/or tolerability." The commentary emphasizes "the paucity of pharmacodynamic, pharmacokinetic, safety, and efficacy studies evaluating this approach" and notes "a void of evidence supporting cardiometabolic benefit with fractional doses."

A letter in Diabetes Care states that "no clinical trials have validated its safety or efficacy" and warns that unapproved dosing "lacks extensive evidence-based support." A 2026 narrative review in Cureus arguing for multisystem benefits acknowledges a "limited direct evidence base for GLP-1 RA microdosing." That review cites zero randomized trials of fractional dosing. It relies instead on meta-analyses, a phase 2 trial in NASH, an observational study, a case report, and preclinical work.

Real-World Evidence and Non-Randomised Cohort Studies

Observational records show substantially smaller weight changes among individuals remaining on initial doses compared to patients completing standard titration.

In a retrospective electronic health record cohort study published in Biology Methods & Protocols, researchers analyzed records across 29 million patients. They identified 814 of 490,072 semaglutide patients (0.17 percent) who maintained 0.25 mg weekly for six months. A further 1,016 of 322,442 tirzepatide patients (0.32 percent) who maintained 2.5 mg weekly. Mean weight change at 6 months was -0.94 percent for sustained semaglutide 0.25 mg and -5.34 percent for sustained tirzepatide 2.5 mg. At 12 months, weight change reached -2.18 percent on 0.25 mg semaglutide and -5.53 percent on 2.5 mg tirzepatide.

In the same study's matched comparison, semaglutide patients maintaining 0.25 mg experienced weight changes between -1.2 percent and -1.4 percent, compared to -5.1 percent to -12.4 percent among patients escalating to higher doses. None of 104 clinical diagnosis codes reached statistical significance between groups. No tolerability signal survived multiple-testing correction (all q > 0.5). The authors stated that lower-dose benefits have not been adequately evaluated in clinical trials and that observational associations must be confirmed prospectively. Reviewing provider notes revealed that only 9.1 percent of patients had explicit microdosing intent documented.

A prospective non-randomised cohort in Japan published in Diabetes, Obesity and Metabolism followed 112 adults with obesity. Participants remained on tirzepatide 2.5 mg weekly (n = 58) or moved to 5 mg (n = 54) based on shared decision-making. Over six months, weight fell 12.3 kg (15.3 percent) in the 2.5 mg group versus 13.0 kg (16.1 percent) in the 5 mg group (p = 0.563). The composite primary outcome was a BMI below 25 kg/m2 or more than 15 percent weight loss. It was met by 62.1 percent on 2.5 mg and 63.0 percent on 5 mg (hazard ratio 0.93, 95 percent CI, 0.58 to 1.49, p = 0.772). The authors explicitly noted three limits. The study was non-randomised, lacked a control group, and was not designed as an equivalence or non-inferiority trial. The null result therefore does not establish therapeutic equivalence.

Smaller observational cohorts describe individual escalation practices. A preliminary pilot study in Slovakia followed 56 adults receiving semaglutide at doses between 0.25 mg and 1.5 mg weekly without escalation once desired weight was reached, reporting 6.45 percent weight loss at 3 months. A single-centre observational study in Shanghai followed 113 participants on semaglutide 0.5 mg or 1.0 mg weekly, reporting 9.9 percent weight loss at 24 weeks without statistically significant differences between the two doses. Both author groups emphasized that their observational designs and small subgroup sizes preclude definitive dose-equivalence conclusions. Detailed escalation timings are discussed in our review of the semaglutide side effects timeline.

Gastrointestinal Tolerability and Dose Escalation Patterns

Adverse event incidence in registration trials concentrated primarily during initial exposure rather than escalating in a direct linear pattern across maintenance doses.

In the SURMOUNT-1 trial, gastrointestinal side effects were reported across all active arms. Nausea occurred in 9.8 percent on placebo, 25.4 percent at tirzepatide 5 mg, 34.1 percent at 10 mg, and 31.9 percent at 15 mg. Diarrhoea was reported by 7.9 percent on placebo, 20.3 percent at 5 mg, 22.5 percent at 10 mg, and 23.7 percent at 15 mg. Vomiting occurred in 1.9 percent on placebo, 9.2 percent at 5 mg, 11.6 percent at 10 mg, and 12.9 percent at 15 mg. Constipation was reported by 5.9 percent on placebo, 17.9 percent at 5 mg, 18.4 percent at 10 mg, and 12.7 percent at 15 mg.

Serious adverse events remained infrequent across all doses in SURMOUNT-1. Cholelithiasis occurred in 0.5 percent on placebo, 0.5 percent at 5 mg, 0.9 percent at 10 mg, and 1.1 percent at 15 mg. Discontinuation attributed to adverse events was low and not dose-ordered, occurring in 6 of 643 on placebo, 4 of 630 at 5 mg, 8 of 636 at 10 mg, and 6 of 630 at 15 mg. Overall trial completion was higher on active therapy than placebo, with 569 participants completing on 5 mg, 566 on 10 mg, and 575 on 15 mg, compared to 509 on placebo.

In the STEP 2 trial, nausea was reported by 31.8 percent of participants receiving semaglutide 1.0 mg versus 33.5 percent on 2.4 mg and 9.2 percent on placebo. Diarrhoea occurred in 21.9 percent at 1.0 mg, 21.3 percent at 2.4 mg, and 11.9 percent on placebo. Vomiting was lower at 1.0 mg (13.4 percent) compared to 2.4 mg (21.3 percent), as was constipation (12.7 percent versus 17.4 percent). In the phase 2 daily trial NCT02453711, 31.1 percent of participants reported nausea at the lowest dose of 0.05 mg once-daily, compared to 17.6 percent on placebo. Further trial safety details appear in our review of semaglutide side effects.

Metabolic Changes and Weight Regain Following Drug Discontinuation

Withdrawal studies demonstrate that cardiometabolic improvements and weight reductions reverse substantially when GLP-1 receptor agonist therapy is stopped.

In the STEP 1 trial extension, researchers followed 327 participants for one year after treatment discontinuation. At week 68 on semaglutide 2.4 mg, mean weight loss had reached -17.3 percent. By week 120, one year off therapy, mean weight loss from baseline stood at -5.6 percent. Participants regained 11.6 percentage points of weight, representing roughly two-thirds of the weight previously lost.

Cardiovascular and glycaemic parameters reverted toward pre-treatment levels during the extension period. Systolic blood pressure shifted from 129 mmHg at baseline to 121 mmHg at week 68, returning to 131 mmHg at week 120. HbA1c moved from 5.7 percent at baseline to 5.2 percent at week 68, returning to 5.6 percent at week 120. Triglycerides shifted from 131.1 mg/dL to 95.7 mg/dL, returning to 122.4 mg/dL. The authors concluded that ongoing treatment is required to maintain health improvements.

Randomized withdrawal trials establish similar trajectories. In STEP 4 (NCT03548987), 803 adults completed a 20-week escalation to semaglutide 2.4 mg and were randomized to continue 2.4 mg or switch to placebo. From week 20 to week 68, those continuing semaglutide lost an additional -8.3 percent, while those switched to placebo regained +6.5 percent. In SURMOUNT-4 (NCT04660643), 670 adults reached a maximum tolerated dose of tirzepatide 10 mg or 15 mg before randomization. Between week 36 and week 88, participants continuing tirzepatide lost an additional -6.7 percent, while those switched to placebo experienced a +14.8 percent weight increase. Discontinuation dynamics are evaluated further in our report on the GLP-1 cessation weight regain trial.

Regulatory Status and Documented Risks of Compounded Formulations

Compounded incretin products are unapproved drugs that carry documented risks of severe overdoses caused by administration and calculation errors.

The Food and Drug Administration (FDA) warns that compounded drugs do not undergo premarket review for safety, effectiveness, or manufacturing quality. In an official alert on compounded semaglutide dosing errors, the agency documented adverse events resulting from patients self-administering incorrect doses and clinicians miscalculating dosages. In most reports, patients received 5 to 20 times the intended dose, leading to severe nausea, vomiting, abdominal pain, dehydration, acute pancreatitis, gallstones, and hospitalizations.

Regulatory warnings also highlight unapproved active ingredients. The FDA stated that salt forms, including semaglutide sodium and semaglutide acetate, represent different active ingredients than approved formulations, noting it is not aware of any lawful basis for their use in compounding. As of May 31, 2026, the agency received 990 adverse event reports associated with compounded semaglutide and over 730 reports associated with compounded tirzepatide, noting that adverse events are likely underreported.

The regulatory framework governing compounded GLP-1 products shifted following supply normalization. The FDA declared the tirzepatide shortage resolved on December 19, 2024, and the semaglutide shortage resolved on February 21, 2025 (FDA policy guidance). Enforcement discretion periods for compounding pharmacies have ended. On April 30, 2026, the FDA proposed to exclude semaglutide, tirzepatide, and liraglutide from the 503B bulk drug substances list (FDA press announcement), reiterating that outsourcing facilities cannot lawfully compound from bulk substances when approved products are available. Quality differences are analyzed in our comparison of compounded vs brand semaglutide.

Device Engineering and Manufacturer Prohibitions on Dose Splitting

Product labelling and biomedical analyses prohibit manipulating delivery pens to administer unauthorized fractional doses.

In a published analysis in Obesity, clinicians examined off-label pen manipulation. The authors noted that approved multi-dose pen solutions should not be transferred to a syringe, used past four doses, or split into smaller amounts. Per labelling, the commentary notes, the device "was neither designed nor studied for these functions." The commentary adds that "Approved tirzepatide doses are not clearly marked," and that the practice carries "a higher probability of dosing errors." A letter in Diabetes Care states that "no clinical trials have validated its safety or efficacy," that "this method lacks extensive evidence-based support," and that the practice "is considered off-label and is not endorsed by the manufacturer."

Unapproved device alterations risk dosing errors that could lead to subtherapeutic effects or increased side effects. Medical literature on cognitive claims is explored in our review of GLP-1 and the brain. Patients evaluating their metabolic health can discuss labelled maintenance regimens with a licensed physician or compare the two in our guide to compounded vs brand tirzepatide.

Related

• Semaglutide vs tirzepatide for weight loss and metabolic health
• Semaglutide side effects timeline and duration
• Compounded vs brand semaglutide evidence comparison
• Compounded tirzepatide vs brand Zepbound and Mounjaro
• Trial tracking weight regain after stopping GLP-1 therapy
• GLP-1 benefits beyond weight loss: every non-weight endpoint that became a labelled indication