SCALE Maintenance deep-dive: liraglutide 3.0 mg after diet-induced weight loss (Wadden 2013)
Last verified 2026-05-28 · Phase 3 · Completed (results published November 2013) · NCT00781937 ↗
SCALE Maintenance (NCT00781937; Novo Nordisk protocol NN8022-1923) is the pivotal phase-3 weight-loss maintenance trial that earned liraglutide 3.0 mg the maintenance language that now appears on the Saxenda label. The design pre-dates and prefigures STEP-4 for semaglutide and SURMOUNT-4 for tirzepatide: enroll non-diabetic adults with BMI ≥30, or ≥27 with a weight-related comorbidity, put them through a structured low-calorie-diet run-in of 4 to 12 weeks targeting at least a 5% loss of screening weight, and then randomize the responders to either an active GLP-1 agonist or placebo. Of 547 adults who entered the run-in, 422 (77%) reached the ≥5% threshold and were randomized 1:1 to once-daily subcutaneous liraglutide 3.0 mg (n=212) or matching placebo (n=210) for 56 weeks alongside an ongoing 500 kcal/day deficit diet plus exercise counseling. Wadden and colleagues published the results in the International Journal of Obesity in November 2013. The primary endpoint — percentage change in body weight from randomization (after the run-in loss) to week 56 — was −6.2% on liraglutide vs −0.2% on placebo, an estimated treatment difference of −6.1 percentage points (P<0.0001). Every confirmatory secondary endpoint moved in the same direction.
- Enrollment
- 422
- Duration
- 56 weeks of randomized treatment after a 4-12 week low-calorie-diet run-in, plus a 12-week off-drug follow-up (week 68 observation)
- Drug
- Liraglutide 3.0 mg (Saxenda)
- Population
- Non-diabetic adults aged ≥18 with BMI ≥30, or BMI ≥27 with hypertension or dyslipidemia, who lost at least 5% of screening weight during a 4-12 week run-in on a 1,200-1,400 kcal/day low-calorie diet. Among the 422 randomized: mean age 46 years, 81% female, 84% white, mean baseline body weight at randomization 100.7 kg (after 6.0% mean run-in loss from screening), mean BMI 35.6 kg/m². Conducted at 36 sites across 8 countries (US, Europe, South Africa) from June 2009 to June 2011. Participants with type 2 diabetes were excluded.
Primary endpoint
Percentage change in body weight from randomization (week 0 post-run-in) to week 56
Treatment arm
−6.2%
Comparator
−0.2%
Treatment difference: Estimated treatment difference −6.1 percentage points (95% CI −7.5 to −4.6; P<0.0001)
Participants entered randomization having already lost a mean 6.0% of screening weight during the run-in. Liraglutide-treated participants lost another 6.2% over 56 weeks; placebo-treated participants essentially held the run-in loss flat. Cumulative loss from screening was 12.2% on liraglutide vs 6.2% on placebo.
Secondary endpoints
| Endpoint | Treatment | Comparator | Difference |
|---|---|---|---|
| Proportion achieving ≥5% additional weight loss from randomization to week 56 Half of liraglutide-treated participants achieved a further 5% loss on top of the run-in loss, more than double the placebo rate. | 50.5% (107/212) | 21.8% (45/206) | Estimated odds ratio 3.8 (95% CI 2.4 to 5.9; P<0.0001) |
| Proportion achieving ≥10% additional weight loss from randomization to week 56 Roughly 4× more liraglutide participants achieved a further ≥10% loss after the run-in. | 26.1% (55/212) | 6.3% (13/206) | Estimated odds ratio 5.4 (95% CI 2.8 to 10.5; P<0.0001) |
| Proportion maintaining the run-in weight loss (no regain) at week 56 About four out of five liraglutide participants held or extended their run-in loss; fewer than half of placebo participants did. The headline maintenance signal. | 81.4% | 48.9% | P<0.0001 |
| Absolute body-weight change from randomization to week 56 (kg) | −6.3 kg | −0.2 kg | Estimated treatment difference −6.1 kg (95% CI −7.5 to −4.6; P<0.0001) |
| Change in waist circumference from randomization to week 56 (cm) Central adiposity continued to fall on liraglutide while flattening on placebo. | −6.5 cm | −3.6 cm | Estimated treatment difference −2.9 cm (95% CI −4.0 to −1.8; P<0.0001) |
| Change in systolic blood pressure from randomization to week 56 (mmHg) Small but statistically significant additional reduction; the larger blood-pressure drop happened during the run-in. | −2.8 mmHg | −0.9 mmHg | Estimated treatment difference −2.0 mmHg (95% CI −3.9 to 0.0; P=0.046) |
| Change in fasting lipid profile from randomization to week 56 — total cholesterol Lipid markers were broadly similar between arms over the maintenance phase; the run-in had already produced most of the lipid improvement. | +1.6% | +2.4% | Estimated treatment difference −0.8% (not statistically significant) |
| Change in HbA1c from randomization to week 56 (percentage points) Participants were non-diabetic at entry, so changes occur within the normoglycemic/prediabetic range. Direction was consistent with the later STEP and SURMOUNT trials in non-diabetic populations. | −0.16 | −0.07 | Estimated treatment difference −0.09 percentage points (95% CI −0.18 to −0.01; P=0.034) |
| Body composition: change in total fat mass at week 56 (DXA substudy, % of baseline) DXA substudy demonstrated that the additional weight loss on liraglutide was predominantly fat mass, with preserved lean-to-fat ratios. | −15.4% | −4.5% | Estimated treatment difference favoring liraglutide; P<0.0001 |
| Progression to type 2 diabetes by week 56 Hint of diabetes-prevention signal that was confirmed in the larger SCALE Diabetes Prevention trial (Le Roux 2017, PMID 28237263). | 0.9% (2/212) | 1.9% (4/210) | Numerically lower on liraglutide; underpowered for formal inference in this trial. |
Adverse events
| Event | Treatment rate | Comparator rate |
|---|---|---|
| Any gastrointestinal disorder GI events clustered during the 4-week dose-escalation phase (0.6 mg → 3.0 mg). Most were mild-to-moderate and transient. | 74.1% (157/212) | 43.8% (92/210) |
| Nausea Most common GI event; peaked during titration and attenuated over weeks. | 47.2% (100/212) | 17.1% (36/210) |
| Diarrhea | 22.6% (48/212) | 13.3% (28/210) |
| Constipation | 22.2% (47/212) | 11.4% (24/210) |
| Vomiting | 17.9% (38/212) | 3.8% (8/210) |
| Headache | 16.5% (35/212) | 11.9% (25/210) |
| Injection-site reaction | 13.7% (29/212) | 10.5% (22/210) |
| Hypoglycemia (symptomatic, non-diabetic population) All events were minor; no severe hypoglycemia in either arm. | 2.4% (5/212) | 0.0% (0/210) |
| Gallbladder-related events (cholelithiasis, cholecystitis) Consistent with the gallstone signal observed across rapid weight-loss interventions and in later GLP-1 trials. | 2.4% (5/212) | 0.0% (0/210) |
| Permanent treatment discontinuation due to adverse events Most discontinuations were for gastrointestinal intolerance during or shortly after the dose-escalation phase. | 9.9% (21/212) | 3.8% (8/210) |
| Any serious adverse event No specific SAE class was concentrated in the liraglutide arm beyond the gallbladder signal. | 6.1% (13/212) | 1.9% (4/210) |
Subgroup analyses
- Participants with prediabetes at randomization (fasting glucose 100-125 mg/dL or HbA1c 5.7-6.4%): Liraglutide reduced the proportion meeting prediabetes criteria at week 56 vs placebo
Pre-specified analysis; signal later confirmed at scale in SCALE Diabetes Prevention (PMID 28237263).
- Run-in responders losing ≥10% of screening weight before randomization: Additional loss during the 56-week maintenance phase was directionally consistent with the overall result
Suggests the maintenance benefit is not confined to those who lost less during run-in.
- Sex (female vs male): Effect direction was preserved in both sexes; the trial was 81% female so the male subgroup was small.
Underpowered for formal interaction testing.
Clinical significance
SCALE Maintenance is the original maintenance-design GLP-1 trial — the regulatory document that established liraglutide 3.0 mg as a tool for holding diet-induced weight loss rather than only producing it. The 6.1 percentage-point treatment difference is smaller in absolute terms than STEP-4 (semaglutide, 14.8 points) or SURMOUNT-4 (tirzepatide, 25.3 points), but the directional finding is the same: continued GLP-1 therapy preserves and extends the loss; the comparator arm plateaus or regains. The 81% no-regain rate on liraglutide vs 49% on placebo is the most-cited number from this trial in clinical reviews. SCALE Maintenance also informed the FDA's approval framework for chronic weight management agents by demonstrating that maintenance is a separable, measurable, regulator-acceptable endpoint distinct from initial loss. The trial's run-in design — diet first, drug second — is closer to real-world clinical practice than the all-on-drug pivotal designs that followed, and it is the design template echoed by every subsequent maintenance trial in the class.
Frequently asked questions
How does SCALE Maintenance compare with STEP-4 for semaglutide?
Both are maintenance-design trials, but they differ in run-in mechanism and magnitude. SCALE Maintenance used a 4-12 week low-calorie diet run-in to produce the initial weight loss, then randomized responders to liraglutide 3.0 mg vs placebo for 56 weeks; the treatment difference at the end of randomized follow-up was 6.1 percentage points. STEP-4 (Rubino 2021 JAMA, PMID 33755728) used a 20-week open-label semaglutide 2.4 mg run-in, then randomized responders to continued semaglutide vs placebo for 48 weeks; the treatment difference was 14.8 percentage points. The trials are not head-to-head and the run-in mechanisms are different, but they support the same conclusion: continued GLP-1 therapy maintains and extends loss; comparator participants plateau (placebo on top of diet, SCALE) or regain (placebo after drug withdrawal, STEP-4). The magnitude difference reflects both drug potency and the larger run-in losses produced by a 20-week drug run-in vs a 4-12 week diet run-in.
Does Saxenda maintain weight loss as well as Wegovy?
No, head-to-head data are limited but the loss-and-maintenance magnitudes consistently favor semaglutide. SUSTAIN-FORTE for diabetes and the indirect comparisons in obesity (STEP-8, Rubino 2022 JAMA, PMID 35015037, randomized 338 adults with obesity 3:1 to semaglutide 2.4 mg vs liraglutide 3.0 mg and reported a 15.8% vs 6.4% mean weight loss at 68 weeks) showed semaglutide produced roughly 2-3× the absolute loss of liraglutide. For maintenance specifically, SCALE Maintenance reported a 6.1 percentage-point treatment difference over 56 weeks while STEP-4 reported a 14.8-point difference over 48 weeks. The two drugs are in the same class and work by the same receptor, but the longer half-life and higher receptor occupancy of semaglutide translate to larger and more durable effects.
What happens after stopping Saxenda?
SCALE Maintenance included a 12-week off-drug follow-up phase after the 56-week treatment period. Participants in both arms experienced partial weight regain during the off-drug window, with the liraglutide group still ending below randomization weight on average. Wadden and colleagues reported regain in the liraglutide arm during the off-drug follow-up that brought the net advantage closer to the placebo arm by week 68, although liraglutide participants remained below placebo in absolute terms. The broader pattern is consistent with what STEP-1 extension (Wilding 2022, PMID 35441470) showed for semaglutide — roughly two-thirds of the on-drug loss is regained within a year of stopping. Counseling patients to expect regain after discontinuation is standard practice across the class.
Is the run-in design clinically realistic?
The diet-first, drug-second design is closer to standard obesity-medicine practice than the all-on-drug pivotal trials. Clinicians routinely ask patients to attempt a structured lifestyle intervention before starting a GLP-1 agonist, both for clinical reasons (a meaningful initial loss predicts longer-term success) and for payer reasons (commercial prior-authorization criteria often require documented 3-6 months of lifestyle intervention). SCALE Maintenance baked that pathway into the protocol: 4-12 weeks of a 1,200-1,400 kcal/day diet with monthly visits, then drug initiation for those who hit a 5% loss. The 77% run-in completion rate (422 of 547) is consistent with clinic-level experience that most motivated patients can produce an initial 5% loss with intensive dietary support. Where the design diverges from practice is the 1,200-1,400 kcal/day target, which is more restrictive than the gradual deficits most clinicians recommend today.
Does Saxenda work for maintenance only, or also for initial loss?
Both. SCALE Obesity and Prediabetes (Pi-Sunyer 2015 NEJM, PMID 26132939) tested liraglutide 3.0 mg as the primary loss agent in 3,731 adults with overweight or obesity and reported an 8.4 kg mean loss vs 2.8 kg on placebo at 56 weeks — that trial supported the original Saxenda approval for chronic weight management. SCALE Maintenance answered the separate question of whether the drug also holds a loss produced by diet, which is a different clinical scenario. The Saxenda label covers both indications because the drug is approved for chronic weight management broadly rather than for a single phase of therapy. In practice clinicians use liraglutide 3.0 mg the same way they use other GLP-1 agonists: start, titrate, continue as long as the patient is benefiting, and accept that discontinuation typically reverses much of the loss.
What were the main tolerability issues in SCALE Maintenance?
Gastrointestinal events dominated, as expected for the class. Among the 212 liraglutide participants, 74.1% reported any GI event vs 43.8% on placebo, with nausea (47.2% vs 17.1%) leading the list. Most events were mild-to-moderate and clustered during the 4-week dose-escalation phase from 0.6 mg up to 3.0 mg. Permanent treatment discontinuation for adverse events was 9.9% on liraglutide vs 3.8% on placebo — higher than the maintenance-phase discontinuation rate in STEP-4 (2.4%) but in line with the SCALE Obesity initial-loss trial because the run-in in SCALE Maintenance was diet-only, so participants encountered liraglutide titration for the first time post-randomization. Gallbladder-related events were 2.4% on liraglutide vs 0% on placebo, consistent with the gallstone signal observed across rapid weight-loss interventions.
References
- 1.Wadden TA, Hollander P, Klein S, et al.; NN8022-1923 Investigators. Weight maintenance and additional weight loss with liraglutide after low-calorie-diet-induced weight loss: the SCALE Maintenance randomized study. Int J Obes (Lond). 2013. PMID: 23812094.
- 2.Novo Nordisk A/S. Comparison of Liraglutide Versus Placebo in Weight Loss Maintenance in Obese Subjects: SCALE - Maintenance (NCT00781937). ClinicalTrials.gov. 2013. https://clinicaltrials.gov/study/NCT00781937
- 3.Rubino D, Abrahamsson N, Davies M, et al.; STEP 4 Investigators. Effect of Continued Weekly Subcutaneous Semaglutide vs Placebo on Weight Loss Maintenance in Adults With Overweight or Obesity: The STEP 4 Randomized Clinical Trial. JAMA. 2021. PMID: 33755728.
- 4.Aronne LJ, Sattar N, Horn DB, et al.; SURMOUNT-4 Investigators. Continued Treatment With Tirzepatide for Maintenance of Weight Reduction in Adults With Obesity: The SURMOUNT-4 Randomized Clinical Trial. JAMA. 2024. PMID: 38078870.
- 5.Pi-Sunyer X, Astrup A, Fujioka K, et al.; SCALE Obesity and Prediabetes NN8022-1839 Study Group. A Randomized, Controlled Trial of 3.0 mg of Liraglutide in Weight Management. N Engl J Med. 2015. PMID: 26132939.
- 6.Rubino DM, Greenway FL, Khalid U, et al.; STEP 8 Investigators. Effect of Weekly Subcutaneous Semaglutide vs Daily Liraglutide on Body Weight in Adults With Overweight or Obesity Without Diabetes: The STEP 8 Randomized Clinical Trial. JAMA. 2022. PMID: 35015037.
- 7.Wilding JPH, Batterham RL, Davies M, et al. Weight regain and cardiometabolic effects after withdrawal of semaglutide: The STEP 1 trial extension. Diabetes Obes Metab. 2022. PMID: 35441470.