GLP-1 Maintenance Dose: What the Research Says

For most of this drug class’s history, there were only two options on the table: stay at the dose that got you here, or stop. The trials were built that way. STEP 4, SURMOUNT-4, the STEP 1 extension - all of them compared continuing at full dose with switching to placebo.

That changed in 2026. Two randomized trials tested middle paths, and the answers were more interesting than either camp expected. Meanwhile, a separate and much noisier conversation about “microdosing” has been running on social media and telehealth sites without any evidence behind it at all.

This article separates the two. It describes what the labels allow and what the trials found: every dosing passage below is a summary of the FDA-approved prescribing information and Instructions for Use, not an instruction to you. It does not tell you what dose to take, and it does not describe a taper schedule - dose decisions belong to the prescriber who wrote your prescription, and you should follow that prescription.

First, what counts as a maintenance dose?

The labels are precise about this, and the distinction they draw is the one most online discussion blurs.

Wegovy (semaglutide). The label, revised June 2026, starts adults at 0.25 mg weekly and escalates every four weeks: 0.5 mg at weeks 5-8, 1 mg at weeks 9-12, 1.7 mg at weeks 13-16, then maintenance from week 17. For weight reduction in adults, the maintenance dosage is either 1.7 mg or 2.4 mg (recommended) once weekly, and the label instructs prescribers to “consider treatment response and tolerability when selecting the maintenance dosage.” For patients who tolerate 2.4 mg for at least four weeks and need additional weight reduction, the dosage may be increased to a maximum of 7.2 mg (marketed as Wegovy HD). The label lists 1.7 mg as an alternative maintenance dosage for weight reduction rather than framing it as a fallback: the explicit “if patients do not tolerate the maintenance dosage of 2.4 mg once weekly, the dosage can be decreased to 1.7 mg once weekly” sentence appears in the label’s MASH section, not in the weight-reduction section. For weight reduction the label simply names both dosages and tells prescribers to weigh response and tolerability [1].

Zepbound (tirzepatide). The label, revised August 2026, starts everyone at 2.5 mg for four weeks and then states in its own words: “The 2.5 mg dosage is for treatment initiation and is not approved as a maintenance dosage.” Recommended maintenance dosages are 5 mg, 10 mg or 15 mg once weekly for weight reduction and long-term maintenance (10 mg or 15 mg for obstructive sleep apnea), with 15 mg as the maximum. The label adds: “Consider treatment response and tolerability when selecting the maintenance dosage. If patients do not tolerate a maintenance dosage, consider a lower maintenance dosage” [2].

So both labels already build in a lower option. Wegovy 1.7 mg and Zepbound 5 mg are approved maintenance doses. Staying at Wegovy 0.25 mg or Zepbound 2.5 mg is something else entirely - those are rungs on a ladder, not a destination, and the Zepbound label says so outright.

What did the first randomized test of a lower maintenance dose find?

SURMOUNT-MAINTAIN, published in The Lancet in May 2026, is the trial this question had been waiting for.

Across 20 US sites, 441 adults with obesity (or overweight with a weight-related condition, excluding type 2 diabetes) took tirzepatide for 60 weeks at the maximum tolerated dose of 10 or 15 mg. Then 378 participants who met the randomization criteria were assigned 3:3:2 to one of three arms for 52 more weeks: continue the maximum tolerated dose, drop to 5 mg, or switch to placebo. From week 84, anyone whose regain exceeded 50% of their loss could receive rescue tirzepatide.

At week 112, the model-based estimate of weight change from baseline was:

Arm	Weight change from baseline	Difference vs placebo
Maximum tolerated dose (10 or 15 mg)	-21.9%	-12.0 points
Tirzepatide 5 mg	-16.6%	-6.6 points
Placebo	-9.9%	—

All comparisons were statistically significant (p<0.0001). Rescue tirzepatide was used by 8% of the maximum-dose arm, 25% of the 5 mg arm and 67% of the placebo arm. Ninety-one percent of randomized participants completed the study, and the most common adverse events were gastrointestinal, mostly mild to moderate and mostly during escalation [3].

The authors’ own conclusion is carefully worded: continuing at the maximum tolerated dose maintained weight reduction and health benefits, while “reducing to 5 mg tirzepatide might provide a valuable alternative to discontinuation, although individuals’ treatment response might vary” [3].

That is the finding in a sentence. A lower dose beat stopping. It did not match staying. And a quarter of the people on the lower dose needed rescue treatment, versus 8% of those who stayed put.

Note the limits: this is tirzepatide only. No equivalent randomized dose-reduction trial exists for semaglutide. Dose de-escalation and re-escalation were not permitted during the maintenance period, so the trial does not tell you what happens if you drop the dose and adjust as you go.

Can switching to a different drug hold the weight off instead?

ATTAIN-MAINTAIN tested a different kind of step-down - from a weekly injection to a daily pill.

The trial took people who had completed the SURMOUNT-5 head-to-head study on injectable tirzepatide (cohort 1, n=205) or injectable semaglutide (cohort 2, n=171) and randomized them to once-daily oral orforglipron or placebo for 52 weeks. Among participants who had reached a weight plateau:

• Cohort 1 (from tirzepatide): 74.7% of weight reduction maintained on orforglipron versus 49.2% on placebo
• Cohort 2 (from semaglutide): 79.3% versus 37.6%

Both differences were highly significant, and all key secondary endpoints were met. Adverse events were mostly mild-to-moderate gastrointestinal effects [4].

The trial’s own stated limitation is the important one: there was no arm that continued the original injectable. So ATTAIN-MAINTAIN shows that an oral GLP-1 holds weight loss far better than stopping. It does not show that it holds it as well as staying on the injection.

Together, these two trials prompted a Diabetes, Obesity and Metabolism commentary in 2026 arguing that obesity pharmacotherapy should borrow the “induction and maintenance” framing used in oncology and rheumatology - a more intensive phase to lose the weight, a possibly less intensive one to hold it [5]. That is a framing to watch. No guideline has adopted it.

What about spacing doses out instead of lowering them?

This is the strategy patients invented and researchers followed.

Mitch Biermann, a Scripps Health physician, told Medscape that the idea came from his patients: “Patients were telling me they switched to taking it every other week… I started recommending it to people who wanted to de-escalate their therapy. There’s no current standard of care on how to de-escalate” [6].

He then studied it. The resulting retrospective case series, published in Obesity in 2026, followed 30 adults in a primary care obesity practice who had plateaued on weekly semaglutide or tirzepatide - plateau defined as less than 5% variation over three months - and moved to reduced-frequency dosing, usually every other week, at their existing dose. Over a mean 36.3 weeks:

• Weight fell from 87.9 kg pre-treatment to 74.1 kg at plateau and further to 72.4 kg on reduced frequency (p<0.01).
• Total body and truncal fat declined; skeletal muscle mass stabilized.
• A1c improved from 5.6% pre-treatment to 5.1% on weekly dosing and stayed there; triglycerides went from 121 to 84.3 mg/dL and stayed stable; mean arterial pressure improved from 90.5 to 84.8 mm Hg and held [7][8].
• Doses were mostly non-maximal, averaging about 7.5 mg for tirzepatide and 1.7 mg for semaglutide [6].
• Schedules varied: 17 patients every other week, six every 10 to 14 days, seven beyond two weeks, with the longest interval six weeks.
• Four of the 30 regained weight and returned to weekly dosing [6].

Biermann’s own interpretation: “you don’t need much of these hormones to maintain weight loss, even though you need a lot of them to reduce weight” [8].

The counterweight came from the session moderator at ObesityWeek 2025, Kimberly Gudzune, chief medical officer of the American Board of Obesity Medicine: “It’s a small study, and we definitely need more data to understand what’s the right approach for different patients long term… If you’re going to try this, there really needs to be very close follow-up because you don’t want folks to lose all the health benefits that they gained” [6].

Harvard obesity specialist Fatima Stanford made a similar point about selection: “individuals who agree to reduce treatment frequency may already be more adherent, more confident in their behaviors or metabolically more responsive.” She also credited the study with reframing the conversation: “Chronic treatment does not necessarily mean maximal weekly dosing forever” [9].

The modeling behind it. Two linked papers by Cengiz, Wu and Lawley built pharmacokinetic-pharmacodynamic models of virtual patients at various dosing intervals. The core finding: reducing frequency does not reduce effect proportionally. Moving semaglutide 2.4 mg from weekly to every two weeks was predicted to take steady-state weight loss from 17% to 12% - halving the drug used while retaining about 72% of the effect. For tirzepatide, weekly 5/10/15 mg predictions of 17%/21%/23% became 12%/16%/18% at every two weeks, retaining roughly 75%. The models also predicted about half the weight loss retained at once-monthly dosing, a prediction the authors flag as especially in need of clinical validation [10][11].

Their conclusion is a request, not a recommendation: “Investigating reduced-frequency dosing in clinical trials, such as one dose every 14 days, is a natural next step to validate this approach” [10].

The case series drew a research-letter write-up in JAMA in April 2026, which is a fair measure of how much interest the idea has attracted relative to how much evidence sits behind it [24].

Be clear about the evidence level here. Thirty patients, retrospectively, no control group, mostly white and privately insured, all of whom had already plateaued and none of whom stopped the drug. Plus computer models. That is a promising signal, not a protocol.

And “microdosing”?

Different thing. Weaker evidence. More marketing.

There is no definition. Amanda Velazquez, director of obesity medicine at Cedars-Sinai, put it directly: “Microdosing is not a medical term. It’s a term that has been popularized on social media… We don’t have any evidence that low-dose, intermittent use of GLP-1s provides meaningful or sustained weight loss or broader health benefits. Even more concerning are the potential risks of inappropriate use” [12].

There is no trial. Every efficacy figure quoted for these drugs - the 15%, the 21%, the cardiovascular and kidney benefits - comes from studies using full, titrated doses. No randomized trial has held people at a deliberately sub-therapeutic dose to measure the outcome.

The dose-response curve runs the wrong way. The phase 2 dose-ranging trial of daily semaglutide showed a clean monotonic relationship: lower doses, less weight loss. In SURMOUNT-1, mean weight change at week 72 was -15.0% at 5 mg, -19.5% at 10 mg and -20.9% at 15 mg - a group-level dose response across the three tested maintenance doses. As Cedars-Sinai’s Karl Korman noted, weight loss at very low doses tends to be modest, “around 5% to 6%,” while “the dramatic results people see with GLP-1s - 15% to 20% body weight loss - happen at higher doses” [12].

The market moved faster than the evidence. STAT News reported in November 2025 that Noom, Found and Hims and Hers all launched compounded microdose GLP-1 programs within three months, with physicians and health services researchers saying there was no robust clinical evidence the drugs work at very small doses. STAT revisited the subject in May 2026 and found nothing had changed: still popular, still unsupported, still no accepted definition of a microdose [13]. A Los Angeles Times report in June 2026 quoted one obesity and neurology specialist estimating that 30 to 40% of her roughly 5,000 patients were microdosing, defining a microdose as the lowest dose that quiets food noise without causing significant weight loss [14].

The risks are concrete. On July 31, 2026, Cleveland Clinic clinicians advised against microdosing, citing dosing errors and side effects such as nausea and diarrhea, and noting it is less effective than long-term full doses [15]. The FDA has separately alerted providers, compounders and patients to overdoses caused by dosing errors with compounded injectable semaglutide in vials, including 10-fold overdoses leading to fainting, dehydration, gallstones, pancreatitis and hospitalization [16]. Through May 31, 2026, FDA had logged more than 1,700 adverse event reports naming compounded GLP-1s, a figure the agency notes is likely an undercount because 503A pharmacies are not required to report [17]. A 2026 review in the Journal of the American Association of Nurse Practitioners frames the clinician’s side of this as a dilemma rather than a settled question: patient anecdotes of benefit at low doses are real and common, and they still do not substitute for trial evidence, particularly once compounding restrictions push people toward unregulated sources [25].

The legitimate cousin. There is a real clinical practice that gets confused with internet microdosing. A January 2025 Diabetes Care commentary described how some prescribers use fractional “click” dosing from multi-dose semaglutide pens to handle severe gastrointestinal side effects, supply shortages, hospital discharge and acute illness. Each such pen delivers roughly 72 unnumbered clicks. The authors are careful: the practice is off-label, not endorsed by the manufacturer, unvalidated by any clinical trial, and requires written instructions, demonstration devices and close monitoring [18]. Cedars-Sinai’s Korman made the same distinction: a lower dose under supervision as part of a titration strategy “is not the same as the microdosing strategies you see trending online” [12].

Three different things, then, and it is worth keeping them apart:

1. A lower approved maintenance dose - Wegovy 1.7 mg, Zepbound 5 mg or 10 mg. In the labels. Supported by SURMOUNT-MAINTAIN for tirzepatide.
2. Clinician-supervised fractional or reduced-frequency dosing - off-label, small evidence base, being actively studied.
3. Internet microdosing protocols - no definition, no trial, no professional endorsement, documented dosing-error risk.

Does a lower dose actually save money?

It depends which drug.

For semaglutide, Novo Nordisk’s self-pay price through NovoCare is the same across Wegovy pen strengths from 0.25 mg through 2.4 mg - $349 per month standard, with $399 for the 7.2 mg high dose [19]. So dropping from 2.4 mg to 1.7 mg does not lower the self-pay bill. Spacing doses out could, if a box of four pens lasts eight weeks instead of four, but that depends on how your prescription is written and filled.

For tirzepatide, Lilly’s self-pay prices do vary by dose: $299 per month for 2.5 mg, $399 for 5 mg and $449 for 7.5 mg through 15 mg, provided the refill happens within 45 days [20]. There, a lower maintenance dose is genuinely cheaper.

That asymmetry is worth knowing, because “a lower dose to save money” is a much better argument for one drug than the other. See this site’s access and pricing coverage for the full picture.

What do the guidelines say?

Two things, and they sit slightly in tension.

In April 2026, The Obesity Society, the Obesity Medicine Association and the Obesity Action Coalition published joint GRADE-based guidance in Obesity. Continuing obesity medications during weight maintenance received a strong recommendation [21]. The statement addresses whether to continue, not at what dose, and it does not endorse any particular de-escalation schedule.

Meanwhile, a 2026 review in Diabetes, Obesity and Metabolism focused specifically on de-escalation notes that recent randomized evidence supports reduced-intensity pharmacological maintenance over abrupt discontinuation, but states that whether structured tapering strategies can successfully facilitate treatment discontinuation “remains unknown and requires prospective evaluation” [22].

The same group published a broader framework in Obesity Pillars in August 2026 arguing that care after incretin-induced weight loss “should move beyond a binary choice between indefinite maximum-dose obesity medication and treatment discontinuation.” The strategies it lays out for study are the ones this article has been describing: staying on drug, monitored dose reduction with pre-agreed criteria for going back up, switching to an oral agent, reduced-frequency dosing, and intermittent rescue treatment. The authors suggest that early changes in appetite and satiety may flag a relapse before the scale does [26]. None of that is a protocol yet. It is a research agenda, and it tells you what the next few years of trials will be about.

So: continue, yes, strongly recommended. Continue at a lower intensity? Now supported by one randomized trial in one drug, with the honest caveat that response varies and a quarter of the lower-dose arm needed rescue.

What should I take from all of this?

• A lower maintenance dose is a labeled option, not an off-label hack. Wegovy lists 1.7 mg. Zepbound lists 5 mg and tells prescribers to consider a lower maintenance dosage for tolerability.
• SURMOUNT-MAINTAIN is the evidence to cite: dropping to tirzepatide 5 mg held meaningfully more weight loss than placebo, meaningfully less than the maximum tolerated dose, and 25% of that arm needed rescue therapy.
• Reduced-frequency dosing is a real research direction with a 30-patient case series and supportive modeling behind it, and no randomized trial. Four of 30 patients in that series had to go back to weekly.
• Microdosing is a marketing term, not a protocol. No definition, no trial, active warnings from Cleveland Clinic and the FDA about dosing errors.
• The right question for your next appointment is probably not “should I stop?” but “is a lower maintenance dose reasonable for me, and how would we monitor it?” That is a conversation the labels support.

Whatever you land on, do not change a dose or a schedule yourself. Every clinician quoted in this article made the same point about close follow-up, and the one published case of a hospitalization in this territory involved someone restarting their old dose after a gap without a plan [23].

Sources

1. Wegovy (semaglutide) US Prescribing Information, revised 06/2026
2. Zepbound (tirzepatide) US Prescribing Information, revised 08/2026
3. Tirzepatide for maintenance of bodyweight reduction in people with obesity in the USA (SURMOUNT-MAINTAIN), The Lancet 2026
4. Orforglipron for maintenance of body weight reduction: ATTAIN-MAINTAIN, Nature Medicine 2026
5. From Induction to Maintenance? Rethinking GLP-1-Based Obesity Pharmacotherapy in the Era of Oral Agents, Diabetes Obes Metab 2026
6. De-Escalating GLP-1s to Every-2-Weeks Maintenance Option, Medscape, November 2025
7. Reduced-Frequency GLP1 Therapy Maintains Weight, Body Composition, and Metabolic Syndrome Improvements: A Case Series, Obesity 2026
8. Weight loss can persist after reducing GLP-1 dosing frequency, Healio, April 2026
9. Some patients keep weight off with fewer GLP-1 injections, study finds, Medical Xpress, March 2026
10. Less frequent dosing of GLP-1 receptor agonists as a viable weight maintenance strategy, Obesity 2025
11. Alternative dosing regimens of GLP-1 receptor agonists may reduce costs and maintain weight loss efficacy, Diabetes Obes Metab 2025
12. Can Microdosing GLP-1s Promote Health and Weight Loss? Cedars-Sinai, May 2026
13. STAT News revisits GLP-1 microdosing, May 2026
14. Los Angeles Times on GLP-1 microdosing, June 2026
15. Microdosing GLP-1s, Cleveland Clinic Health Essentials
16. FDA alerts health care providers, compounders and patients of dosing errors associated with compounded semaglutide injectable products
17. FDA logs more than 1,700 adverse event reports involving compounded semaglutide and tirzepatide, Drug Discovery Trends
18. One Size Does Not Fit All: Understanding Microdosing Semaglutide for Diabetes in Multidose Pens, Diabetes Care 2025
19. NovoCare Wegovy Price Guide
20. Lilly lowers price for Zepbound single-dose vials, Eli Lilly investor release
21. Joint TOS/OMA/OAC Expert Guidance Statement, Obesity 2026
22. Early Weight Regain After GLP-1RA Discontinuation: Mechanisms and Implications for Treatment De-Escalation Strategies, Diabetes Obes Metab 2026
23. Severe Gastrointestinal Intolerance After Resuming Maintenance-Dose Semaglutide Following Treatment Interruption, Cureus 2026
24. Reduced-Frequency GLP-1 Dosing May Sustain Treatment Effects, JAMA, April 2026
25. The “microdosing” dilemma: balancing patient anecdotes with clinical safety amid GLP-1 compounding restrictions, J Am Assoc Nurse Pract, May 2026
26. Adaptive maintenance after incretin-induced weight loss: moving beyond the continue-or-stop paradigm, Obesity Pillars, published online August 2026