Medical disclaimer: This article is for informational and educational purposes only and does not constitute medical advice. GLP-1 receptor agonists are prescription medications. Decisions about starting, stopping, tapering, or maintaining therapy should be made in consultation with a qualified healthcare professional who knows your full clinical picture.
The question patients ask their GP after twelve months on semaglutide or tirzepatide is rarely "is it working?" The scale answers that. The question, increasingly, is "what happens when I stop?"
It has become one of the loudest research and clinical conversations of 2026. The short version: stopping abruptly without a structured lifestyle scaffold produces predictable regain, and the rate, depth, and trajectory depend on factors that are at least partly modifiable. This article surveys what is known, what is debated, and what practical "off-ramp" frameworks have emerged — written for an Australian audience navigating a system where, as of mid-2026, neither semaglutide nor tirzepatide is PBS-reimbursed for weight management.
The STEP 1 trial (Wilding et al., 2021, New England Journal of Medicine) established the headline figure: mean weight loss of approximately 14.9% over 68 weeks at 2.4 mg semaglutide weekly, versus 2.4% on placebo. What put the post-stop question on the map was the extension analysis.
In the STEP 1 trial extension (Wilding et al., 2022, Diabetes, Obesity and Metabolism), participants who completed the 68-week intervention and then stopped were followed for a further year. By week 120, they had regained approximately two-thirds of the weight lost. Cardiometabolic improvements (waist circumference, HbA1c, blood pressure, lipid markers) also reverted, in many cases close to baseline.
The SURMOUNT-4 trial (Aronne et al., 2024, JAMA) reproduced the pattern with tirzepatide. After a 36-week lead-in, those randomised to placebo regained roughly 14% of body weight over 52 weeks, ending about 5% below starting weight — well short of where they had been at the end of the lead-in. The regain curve in both trials is steepest in the first three to six months and then flattens, consistent with rapid restoration of appetite signalling rather than slow lifestyle drift.
Why is the post-GLP-1 regain so consistent? Several overlapping mechanisms drive it.
Ghrelin rebound. The orexigenic gut hormone is suppressed during active therapy and rebounds — sometimes above pre-treatment levels — within weeks of stopping. Subjective hunger ratings track it closely.
Leptin and the set-point hypothesis. Weight loss reduces leptin output from adipose tissue. The hypothalamus reads low leptin as a defended-weight emergency, amplifying hunger and down-regulating satiety. The body appears to defend a recent peak weight more vigorously than a recent trough, especially when the trough was achieved pharmacologically.
Appetite-hormone reset. Peptide YY, cholecystokinin, and GLP-1 itself shift during weight loss in ways that favour hunger over fullness. These shifts persist after stopping — documented since Sumithran et al. (2011) on post-diet hormonal changes and broadly extended to the GLP-1 era.
Adaptive thermogenesis. Resting energy expenditure falls beyond what the new body weight predicts. Some is reversed with refeeding; some persists.
Lean mass loss. And this is where the GLP-1 story becomes particularly tricky.
DEXA substudies from STEP and SURMOUNT consistently show that a meaningful fraction of weight lost on GLP-1 therapy is lean mass rather than fat — most analyses sit in the 25–40% lean-mass range (Heymsfield et al., 2023).
This matters for three reasons. Skeletal muscle is the largest site of insulin-stimulated glucose disposal. Muscle is metabolically active, so losing it lowers resting metabolic rate and narrows the daily caloric "room for error" that determines whether a maintenance diet holds. And lean mass loss is functionally consequential — strength, balance, and physical reserve decline with it, slowly reversible in the 50-plus cohort.
Someone arriving at the end of GLP-1 therapy with most of their muscle intact has a structurally easier maintenance phase. Strategies that protect lean mass during therapy — covered in our Ozempic muscle loss prevention protocol — are the same strategies that make the off-ramp viable.
There is no head-to-head randomised trial of tapered versus abrupt discontinuation of semaglutide or tirzepatide for weight management as of 2026. The case for a taper is mechanistic.
A stepped withdrawal — halving the weekly dose for four to eight weeks, then extending the dosing interval from weekly to fortnightly, then fortnightly to monthly — flattens the hormonal cliff and gives lifestyle scaffolding time to take over. Published case-series and protocol papers generally describe a 12- to 24-week off-ramp rather than a single discontinuation event.
The case for cold stop is pragmatic: supply shortages, cost, side effects, planned pregnancy, or surgery. The off-ramp is then dictated externally, and the lifestyle scaffold has to do the work a slower taper would share.
Either way, the structure of the off-ramp matters less than what surrounds it. A clean taper into an empty lifestyle environment regains weight at roughly the same rate as a cold stop into the same environment. The difference shows up when protein intake, resistance training, and sleep are in place.
A growing real-world pattern — described in case series but not yet supported by large randomised trials — is "maintenance dosing": continuing a GLP-1 agonist indefinitely at the lowest dose that holds weight. For semaglutide this often means dropping from 2.4 mg weekly to 0.5 or 1.0 mg weekly; some users report holding weight at <0.5 mg weekly. For tirzepatide, 2.5 or 5 mg weekly plays a similar role.
The rationale is that obesity is chronic, and intermittent pharmacotherapy for a chronic relapsing condition is structurally similar to how hypertension or hyperlipidaemia are managed. The SURMOUNT-4 design — continuation versus placebo after a lead-in — is the closest the trial literature has come to formally testing this, and the continuation arm clearly did better.
Maintenance dosing is not without complications. Long-term safety data beyond five years are still accumulating, cost is significant given the absence of PBS coverage, and gastrointestinal side effects do not disappear entirely.
Protein floor. The most robust recommendation across reviews is roughly 1.6 g per kg of lean body mass per day, across at least three meals with 30–40 g per serving. This supports muscle protein synthesis, blunts hunger, and raises diet-induced thermogenesis modestly. The Wadden et al. (2021) STEP 3 analysis — semaglutide plus intensive behavioural therapy with a structured high-protein diet — produced better body composition outcomes than standard STEP 1. For depth, see our piece on optimal protein targets for weight loss.
Fibre. 35–40 g per day, with a meaningful soluble fraction from oats, psyllium, legumes, chia, and fruit, supports satiety, slows gastric emptying, and feeds short-chain fatty acid production. SCFAs stimulate endogenous GLP-1 from L-cells — a natural backstop to the pharmacological GLP-1 that has been withdrawn.
Volumetric, satiety-dense foods. Leafy vegetables, brassicas, broths, berries, lean protein, and eggs give the gastric stretch signals a smaller calorie-dense plate does not. Patients coming off GLP-1 therapy frequently report higher food volume tolerance than during therapy.
Time-restricted eating. TRE with 8–10 hour feeding windows shows modest maintenance benefits and improved glycaemic markers. Liu et al. (2022, NEJM) suggests TRE is roughly equivalent to continuous calorie restriction when calories are matched — useful as one tool, not a standalone solution.
If there is a single intervention with the strongest case for the post-GLP-1 phase, it is structured resistance training. Two to four sessions per week, progressive in load, covering squat, hinge, push, pull, and carry, protect lean mass during therapy and rebuild it after. Aerobic exercise has its own merits but does not preserve muscle the way resistance work does. For broader context, see our piece on the science of body recomposition.
Short sleep (under six hours) and disrupted circadian timing both raise ghrelin, lower leptin, and bias food choices toward energy-dense items. Sleep restriction during weight maintenance roughly doubles the difficulty of holding a new weight, yet sleep is consistently the variable patients compromise first. A protected 7- to 8-hour window, consistent timing across weekdays and weekends, morning light, and a low-stimulus evening cost nothing — and arguably matter as much as the resistance training that gets all the attention.
Endogenous GLP-1 is secreted from intestinal L-cells in response to protein, fat, fermentable fibre, and bile-acid signalling. Dietary patterns that reliably nudge secretion upward: protein-leading meals (stronger GLP-1 and CCK release than starting with carbohydrate); soluble and fermentable fibre (beta-glucans, inulin, resistant starch, psyllium — feeding SCFA-producing bacteria); live-culture fermented foods (kefir, kimchi, sauerkraut, yoghurt); and bitter-flavour exposure (gut T2R receptors appear to modulate GLP-1 release — speculative but biologically plausible).
These stimuli will not replicate a 2.4 mg semaglutide dose. They do support the natural appetite-regulation system the medication has been carrying.
As of mid-2026, neither semaglutide nor tirzepatide is PBS-listed for weight management in Australia. Semaglutide is PBS-listed for type 2 diabetes under specific criteria; weight-management use is paid out of pocket.
For many Australians, the off-ramp is imposed by cost rather than chosen. A monthly out-of-pocket spend of several hundred dollars is sustainable for some and unsustainable for others, and households often hit a budget wall after twelve to eighteen months. A meaningful share of patients are stopping not because they have reached a clinical endpoint but because the next box is unaffordable. Episodic supply shortages through 2024–2026 have added forced tapers and switches. For broader policy context, see our cross-site companion piece on GLP-1 medication shortage policy. Plans must be robust to forced interruptions, not just planned ones.
Signals that point toward reintroducing GLP-1 therapy after a period off include a steady upward weight trajectory over three to six months (stronger than a single holiday-weight reading), reversion of cardiometabolic markers (HbA1c, waist circumference, fasting glucose, blood pressure) toward baseline despite stable lifestyle anchors, return of pre-treatment hunger, and recapture of a defined threshold — for example, regaining more than 5% from the post-treatment low.
Restarting at a low dose and titrating up is generally better tolerated than reverting directly to the prior maintenance dose; the side-effect profile after a break behaves more like an early initiation than a continuation. Side-effect strategies during titration are covered in our GLP-1 side effects management guide.
Drawing the threads together:
There is no single move that replaces the pharmacology; there is a set of overlapping interventions that, together, narrow the regain curve.
The post-GLP-1 phase is not the absence of treatment but a different kind of treatment, in which the medication is replaced by a stack of behavioural and nutritional levers whose combined effect is smaller, slower, and more demanding — but achievable. For background on how the medications themselves work, see our explainer on how GLP-1 agonists work.
A research-backed explainer on orforglipron (Foundayo) — the first oral small-molecule GLP-1 receptor agonist approved for weight loss, why no-restriction dosing matters, what the ATTAIN trials showed, and how it compares to injectables and oral semaglutide.
A detailed breakdown of the TRIUMPH-1 Phase 3 trial results for retatrutide — the triple-hormone agonist that achieved up to 28.3% mean weight loss at 80 weeks and ~30.3% in a higher-BMI subgroup at 104 weeks.
Cagrilintide is a once-weekly long-acting amylin analog, and Cagrisema combines it with semaglutide. This guide covers the REDEFINE Phase 3 trial results, how amylin co-agonism works, why the headline weight loss disappointed market expectations despite being clinically significant, and what the Australian regulatory path looks like.