Zepbound Maintenance Dose After Goal Weight: SURMOUNT-4 Evidence Review

At a glance

• The Zepbound label does not define a maintenance dose. Section 2 of the prescribing information names three maintenance doses — 5 mg, 10 mg, and 15 mg weekly — and instructs the prescriber to use the highest-tolerated dose^[4]. There is no tapering schedule and no post-goal protocol.
• SURMOUNT-4 is the only randomized withdrawal trial. 670 adults lost a mean 20.9% over a 36-week tirzepatide lead-in. Randomized to continue tirzepatide or switch to placebo for 52 more weeks: the continued arm lost an additional 5.5%; the placebo arm regained 14.0% — about two-thirds of what they had lost^[2].
• STEP-4 found the same pattern for semaglutide. After a 20-week semaglutide 2.4 mg lead-in (mean −10.6%), continuing patients lost an additional 7.9% over 48 weeks while patients switched to placebo regained 6.9%^[3]. The class behavior is consistent.
• Step-down dosing is not trial-tested. No published randomized trial compares 15 mg maintenance to a stepped-down 10 mg, 7.5 mg, or 5 mg post-goal arm. Step-downs happen in practice but the evidence base is clinical experience, not RCT data.
• The cost gap between tiers is real. LillyDirect Self Pay vials are roughly $349/mo for 2.5–5 mg, $499/mo for 7.5–10 mg, and $599/mo for 12.5–15 mg^[7]. A step-down can save $2,400–$3,000 per year if it holds weight stable.
• Obesity is a chronic disease, per the AMA (2013) and AACE/ACE (2016). The framing matters: the question is not when do I stop the drug, but what is the lowest effective long-term dose^[5][6].

What does "maintenance dose" actually mean on Zepbound?

Most patients reading this article use the phrase maintenance dose to mean "the dose I take after I hit my goal weight." The Zepbound FDA labeling uses the phrase differently. In Section 2 of the prescribing information, maintenance doses are the three approved ongoing doses — 5 mg, 10 mg, or 15 mg once weekly — that the patient holds after the titration ramp (2.5 mg for 4 weeks, then 5 mg, then 2.5 mg steps at 4-week intervals to a maintenance dose)^[4]. The label instructs the prescriber to use the highest tolerated dose to achieve treatment goals; it does not define a separate post-goal maintenance dose. There is no taper protocol, no step-down algorithm, and no goal-weight stop rule in the Zepbound label.

This matters because patient and prescriber expectations often diverge. Patients arrive at goal weight expecting a structured wind-down. The label expects continued therapy. The two pivotal questions — do I stay at my current dose? and do I stop entirely? — are not addressed by the FDA label. Both have to be answered from the SURMOUNT-4 withdrawal data, cross-class STEP-4 evidence, the chronic-disease framing, and individual prescriber judgment.

For the broader decision tree of what to do at goal — stay, space, microdose, or taper — see our life-after-GLP-1 decision hub. This article focuses specifically on the Zepbound dose question.

SURMOUNT-4: what happens when you stop

SURMOUNT-4^[2] is the only randomized withdrawal trial of tirzepatide for obesity. 670 adults with BMI ≥ 30 (or ≥ 27 with a weight-related comorbidity), without type 2 diabetes, completed a 36-week open-label lead-in on tirzepatide titrated to their maximum tolerated dose (10 or 15 mg). The mean weight loss at the end of lead-in was −20.9%. Patients were then randomized 1:1 to continue tirzepatide or switch to placebo for an additional 52 weeks while continuing diet and physical activity counseling.

The placebo arm regained about two-thirds of lost weight within one year of withdrawal. Importantly, regain was not back to baseline — the discontinued arm held a net −9.9% versus pre-treatment, meaningfully better than diet-and-physical-activity counseling alone (placebo arm in SURMOUNT-1 reached −3.1% at 72 weeks)^[1]. Stopping is not equivalent to never starting. But for a patient whose goal weight is at −20%, returning to −10% within a year typically means crossing back into overweight or obese BMI ranges, with the metabolic consequences that follow.

For the broader regain dynamics across the GLP-1 class, see our GLP-1 stopping and rebound evidence review. SURMOUNT-4 and STEP-4 are the two highest-quality datasets and both point the same direction.

STEP-4: the semaglutide analog confirms the class behavior

STEP-4^[3] tested the same continue-vs-withdraw design with semaglutide 2.4 mg. 803 adults with BMI ≥ 30 (or ≥ 27 with a comorbidity), without diabetes, completed a 20-week run-in titrating to the 2.4 mg maintenance dose and reached mean −10.6% body weight loss. Patients were then randomized 2:1 to continue semaglutide or switch to placebo for 48 more weeks.

• Continued semaglutide: additional −7.9% TBWL over weeks 20–68 — net −17.4% from baseline.
• Switched to placebo: +6.9% body weight regained over weeks 20–68 — net −5.0% from baseline (about two-thirds of lost weight returned).

The proportional regain — about two-thirds of lost weight within a year — is the same on both molecules. This is not an idiosyncrasy of tirzepatide or semaglutide individually. It is the GLP-1 class behavior: the drug suppresses appetite and slows gastric emptying for as long as it is dosed; once it is gone, the biology that produced obesity in the first place is still there. For the cross-class comparison with the head-to-head SURMOUNT-5 data, see our tirzepatide vs semaglutide review.

Step-down vs stay-at-max: the evidence is thin

SURMOUNT-4 tested only continue at the maintenance dose versus complete withdrawal. There is no published randomized trial comparing 15 mg maintenance to a stepped-down 10 mg, 7.5 mg, or 5 mg post-goal arm. That gap matters: the most common real-world question is not "should I stop?" but "can I go from 15 mg back to 7.5 mg and hold my weight?" The honest answer is that the evidence base for step-down dosing is clinical experience and dose-response extrapolation, not RCT data.

What we can say from the SURMOUNT-1 dose-response^[1]:

• 5 mg: mean −15.0% TBWL at 72 weeks. The lowest approved maintenance dose still produced clinically meaningful weight loss.
• 10 mg: mean −19.5% TBWL.
• 15 mg: mean −20.9% TBWL.

The 5 mg-to-15 mg gradient at the population level is roughly linear-but-compressed: each 5 mg increment adds 4–5 percentage points to the lead-in TBWL, with the top step (10→15 mg) adding only about 1.4 points. The reverse inference is what step-down dosing relies on: if 5 mg was sufficient to achieve −15% from baseline, it is plausible that 5 mg can maintain a similar level post-goal. Plausible is not proven; SURMOUNT-4 did not test it. The closest data point is that any continued tirzepatide dose in SURMOUNT-4 prevented regain — the trial allowed patients to maintain whatever dose they had escalated to during lead-in (10 or 15 mg), not a stepped-down dose.

In practice, prescribers who step down typically do it for one of three reasons:

1. Side-effect tolerance. A patient who tolerated 15 mg through the loss phase but has persistent GI symptoms may benefit from dropping to 10 mg or 7.5 mg post-goal. Symptom relief is the primary endpoint; weight stability is a secondary check.
2. Cost. The LillyDirect Self Pay tier structure creates real annual savings at lower doses (see the cost section below).
3. Insurance step-edit pressure. Some payers reauthorize ongoing coverage only at lower doses or require documentation of a maintenance attempt at a lower dose before re-approving 15 mg.

Prescribers who keep patients at the highest-tolerated dose typically cite SURMOUNT-4 as the strongest available evidence — the trial was designed around the highest tolerated dose, and any step-down is an extrapolation. Neither approach is wrong; both are clinical judgments outside the FDA label.

Why obesity treatment is chronic, not a "course"

This is the most important section of the article because it reframes the entire question. Patients commonly ask when can I stop the medication? — the same way they would ask about a course of antibiotics or a steroid taper. That framing is wrong, and reframing it explicitly changes the answer.

The American Medical Association formally recognized obesity as a disease in June 2013 (Resolution 420, Policy H-440.842)^[5]. The American Association of Clinical Endocrinologists (AACE) and the American College of Endocrinology have published comprehensive clinical practice guidelines treating obesity as a chronic disease since 2014, with the 2016 update^[6] stating explicitly that pharmacotherapy for obesity is intended to be a chronic long-term therapy, analogous to the way hypertension is managed with long-term antihypertensives or hypothyroidism with long-term levothyroxine.

The biology supports this. The mechanism by which tirzepatide produces weight loss — GIP and GLP-1 receptor agonism slowing gastric emptying, reducing food intake, and signaling satiety — is active for as long as the drug is dosed. When the drug is withdrawn, the receptor signal stops and the biology that produced the original obese phenotype reasserts. SURMOUNT-4 documented this mechanistically: appetite returned, caloric intake increased, and weight regain followed within weeks of the placebo switch^[2].

The clinical analogies are intentional and load-bearing:

• A patient does not finish a course of lisinopril after blood pressure normalizes. They stay on it because stopping returns blood pressure to baseline. Obesity therapy works the same way.
• A patient does not finish a course of metformin or insulin after blood sugar normalizes. They titrate to the lowest effective dose and continue indefinitely.
• A patient does not finish a course of statin therapy after LDL cholesterol normalizes. Discontinuation reverses the gains.

The question that flows from this framing is not when do I stop Zepbound? — it is what is the lowest effective long-term dose, and how do I keep continuous access to it?

Cost trade-offs at each Zepbound dose tier

The 2026 LillyDirect Self Pay program for Zepbound single-dose vials is tiered by strength^[7]. The dollar gap between tiers is meaningful and is one of the legitimate reasons a step-down may make sense for cost-stable patients who are post-goal.

A patient who steps from 15 mg down to 7.5 mg post-goal saves about $1,200/year; a patient who steps to 5 mg saves about $3,000/year. If insurance covers the drug at any dose, the calculation collapses and the clinical question dominates. For patients on self-pay or with high deductibles, the dose tier is a real component of the decision. For a personalized estimate, our GLP-1 savings calculator walks through the math at each tier.

Compounded tirzepatide ($200–$300/mo range in 2026) is the cheaper but regulatorily grey-zone option — the FDA enforcement-discretion period for tirzepatide compounding ended in October 2024, and ongoing 503A activity sits outside the FDA-approved supply chain. The brand-name LillyDirect channel is the cleaner option for post-goal maintenance if cost permits. For an alternatives view if Zepbound itself is the constraint, see our Wegovy and Zepbound alternatives review.

How prescribers handle the goal-weight decision in 2026

There is no single "right" answer because the prescribing information leaves the decision to the prescriber. Across the obesity-medicine specialists tracking this question in 2026, four post-goal patterns recur:

Two cross-cutting principles show up in essentially every clinical framework:

1. Reassess every visit. Weight, blood pressure, A1c, side effects, lipid panel. A step-down that holds for 6 months is a success; a step-down that lets weight drift back 1–2% is the signal to step back up.
2. Never "just stop" without a plan. SURMOUNT-4 documented that "simply stop" produces regain of about two-thirds of lost weight within a year. If full discontinuation is the goal — pregnancy planning, surgical preparation, intolerable side effect, patient preference — build a structured bridge with intensified lifestyle support, frequent follow-up, and a low threshold to restart. For the broader stop/taper discussion, see our GLP-1 stopping evidence review.

Patients at goal should be tracking the brand under their drug page — Zepbound — for updated FDA labeling and supply information that affects long-term maintenance decisions.

Verdict

There is no FDA-defined post-goal maintenance protocol for Zepbound. The pivotal randomized evidence — SURMOUNT-4 for tirzepatide and STEP-4 for semaglutide — shows that complete withdrawal produces regain of about two-thirds of lost weight within a year. Continued therapy at the highest tolerated dose produces additional weight loss and zero regain. Step-down dosing happens in clinical practice and is biologically plausible from the SURMOUNT-1 dose-response (5 mg still produces −15% TBWL) but has not been tested in a randomized trial. Obesity is a chronic disease per the AMA (2013) and AACE/ACE (2016); the right framing of the question is what is the lowest effective long-term dose, not when can I stop.

For most patients in the first 12–18 months post-goal, staying at the highest-tolerated dose is the trial-supported default. Cost, side-effect tolerance, and insurance step edits are legitimate reasons to consider a step-down with close follow-up. Stopping cold without a structured bridge plan is the option with the strongest evidence against it.