The dose ladder gets all the attention at the start of GLP-1 treatment, and almost none at the point it actually matters — the dose you stay on. There is a widespread assumption that the top of the ladder is the destination and anything below it is an unfinished course. Both the labels and the trial data say otherwise: more than one maintenance dose is approved for each drug, the highest one buys less than most people expect, and it is not free.
The label already allows more than one answer
Start with the regulatory text, because it is more permissive than the folklore. For weight reduction in adults, the Wegovy label states that the maintenance dosage is either 1.7 mg or 2.4 mg once weekly, with 2.4 mg recommended — and it instructs that if a patient does not tolerate 2.4 mg, the dosage can be decreased to 1.7 mg, with re-escalation considered later.[1]
Zepbound goes further. Its recommended maintenance dosage for weight reduction and long-term weight maintenance is 5 mg, 10 mg or 15 mg once weekly — three approved destinations, not one. The label adds two instructions that rarely make it into a clinic conversation: consider treatment response and tolerability when selecting the maintenance dosage, and if a patient does not tolerate a maintenance dosage, consider a lower one. It also notes that the 2.5 mg starting dose is for initiation and is not approved as a maintenance dosage.[2]
What each step up actually buys
SURMOUNT-1 is the cleanest dose-response evidence in the class, because it randomized 2,539 adults to three different tirzepatide doses and a placebo for 72 weeks. Mean weight change at week 72 was −15.0% at 5 mg, −19.5% at 10 mg and −20.9% at 15 mg, against −3.1% with placebo.[3]
Read those as steps rather than endpoints and the shape changes. Going from 5 mg to 10 mg bought 4.5 percentage points of additional weight loss. Going from 10 mg to 15 mg — a 50% larger dose — bought 1.4. The proportion reaching at least 5% weight reduction was 85%, 89% and 91% across the three doses: the lowest approved maintenance dose already got the large majority of participants to a clinically meaningful result.
| Tirzepatide dose | Mean weight change at 72 weeks | Gained over the dose below | Discontinued for adverse events |
|---|---|---|---|
| Placebo | −3.1% | — | 2.6% |
| 5 mg weekly | −15.0% | — | 4.3% |
| 10 mg weekly | −19.5% | 4.5 points | 7.1% |
| 15 mg weekly | −20.9% | 1.4 points | 6.2% |
The tolerability column is worth sitting with, because it does not behave the way the weight column does. Discontinuation for adverse events was 4.3% at 5 mg, 7.1% at 10 mg and 6.2% at 15 mg, against 2.6% on placebo.[3] It does not climb neatly with dose — which is consistent with the broader pattern that gastrointestinal events cluster around dose increases rather than at any particular steady dose, a timeline covered in GLP-1 gastrointestinal side effects.
The semaglutide picture
Semaglutide’s ladder was studied differently. STEP 2 directly compared the 2.4 mg weight-management dose against the 1.0 mg dose approved for diabetes, in adults with type 2 diabetes: mean body-weight change at week 68 was −9.6% with 2.4 mg versus −3.4% with placebo, and 68.8% of the 2.4 mg group reached at least 5% weight reduction against 28.5% on placebo.[4] The trial was designed to show the higher dose was superior, and it did — which is the argument for titrating up when it is tolerated.
STEP 4 supplies the other half. Its 20-week run-in reached 2.4 mg in 89.0% of participants, and those who continued at that dose lost a further 7.9% over 48 weeks while those switched to placebo gained 6.9%.[5]Maintenance at a real dose is what held the loss. The comparison that does not exist is the one between a full maintenance dose and a deliberately reduced one.
The question the trials have not answered
Everything above describes doses that were escalated to a target and then held. No randomized trial has tested stepping a stable patient down to a lower maintenance dose and following what happens to weight. The withdrawal trials — covered in what the withdrawal trials show about stopping a GLP-1 — compared a full dose against nothing at all, which is a different experiment and a much blunter one.
That gap is where a lot of marketing lives. Programs offering deliberately small doses are increasingly common, and the evidence behind them is thinner than the pitch; we assess it separately in the evidence on microdosing GLP-1s. The defensible reading of the current data is narrower and more useful: the label permits a lower maintenance dose, the dose-response curve flattens near the top, and the difference between the last two rungs is small enough that tolerability and cost can legitimately decide it.
The honest bottom line
The top of the dose ladder is a recommendation, not a requirement. Both labels approve more than one maintenance dose and both tell prescribers to weigh response and tolerability in choosing between them. The trial data support that flexibility: in SURMOUNT-1 the lowest approved maintenance dose already delivered −15.0% and got 85% of participants past 5%, while the top dose added 1.4 points over the one below it. What no one can yet tell you from evidence is whether stepping down after you have reached your goal holds that goal — that trial has not been run.