Liraglutide is the molecule the weekly drugs displaced. It reached the market first, it carries three years of placebo-controlled follow-up, and it now sits behind semaglutide and tirzepatide on every weight endpoint that has been measured. The question worth asking is not whether it loses that contest — it does, and by a wide margin — but what is left once it has. How often a drug is injected is a separate argument, taken apart in the dosing interval article.
The comparison that holds the schedule still
One trial isolated the molecule rather than the calendar. A phase 2 dose-ranging study randomized 957 adults with obesity and without diabetes across 71 sites in eight countries, and every arm was a once-daily subcutaneous injection: semaglutide at 0.05, 0.1, 0.2, 0.3 or 0.4 mg, liraglutide 3.0 mg, or matching placebo, for 52 weeks. Estimated mean weight loss was −2.3% on placebo, −7.8% on liraglutide, and −13.8% at semaglutide 0.4 mg, with every semaglutide dose from 0.2 mg upward significantly better than liraglutide. A loss of 10% or more occurred in 37% to 65% of the semaglutide groups against 10% on placebo.[1]
Because both drugs were injected daily on comparable escalation schedules, the gap in that trial is a statement about the two peptides rather than about how often a syringe is used. It also ran before either molecule had its weight-management dose fixed, which is why the semaglutide strengths look unfamiliar. The weekly product that came out of it is covered in the semaglutide article.
What liraglutide produced in its own trials
SCALE randomized 3,731 adults without type 2 diabetes, at a mean baseline weight of 106.2 kg and mean body-mass index of 38.3, to liraglutide 3.0 mg daily or placebo for 56 weeks. Mean weight change was −8.4 ± 7.3 kg against −2.8 ± 6.5 kg, a difference of −5.6 kg (95% CI, −6.0 to −5.1). A loss of at least 5% was reached by 63.2% against 27.1%, and more than 10% by 33.1% against 10.6%. Serious adverse events occurred in 6.2% against 5.0%.[2] The trial reported kilograms rather than percentages, which is worth remembering whenever that figure is set beside a percentage from a different trial.
The extension is the part no weekly agent can match. The same cohort was followed to 160 weeks — a little over three years — in 2,254 participants with prediabetes. Type 2 diabetes was diagnosed on treatment in 2% of the liraglutide group against 6% on placebo, time to onset was 2.7 times longer (95% CI, 1.9 to 3.9), and the hazard ratio was 0.21 (95% CI, 0.13 to 0.34).[3]
Two things sit alongside that. The weight advantage had shrunk: −6.1% against −1.9% at 160 weeks, a difference of 4.3 percentage points, where the 56-week difference had been 5.6 kg on a 106 kg starting weight. And the dataset is half missing — 47% of the liraglutide group and 55% of the placebo group withdrew, and the authors state plainly that withdrawn participants were not followed up.[3] A three-year obesity dataset is worth a great deal, and this one is missing about half of the people it randomized.
The cardiovascular result, and the part of it that did not reach significance
LEADER randomized 9,340 adults with type 2 diabetes at high cardiovascular risk to liraglutide or placebo and followed them a median of 3.8 years. The composite of cardiovascular death, nonfatal myocardial infarction and nonfatal stroke occurred in 13.0% against 14.9% (hazard ratio 0.87; 95% CI, 0.78 to 0.97; P = 0.01 for superiority). Cardiovascular death fell from 6.0% to 4.7% (hazard ratio 0.78; 95% CI, 0.66 to 0.93) and death from any cause from 9.6% to 8.2% (0.85; 95% CI, 0.74 to 0.97).[4]
The same abstract records that the rates of nonfatal myocardial infarction, nonfatal stroke and hospitalization for heart failure were nonsignificantly lower. Two of the three components of the winning composite did not separate on their own; the result was carried by cardiovascular death. It was also produced in people who already had type 2 diabetes and high cardiovascular risk, at the doses used for glycemic control rather than the 3.0 mg weight dose, so it is not evidence about a buyer without diabetes — the distinction drawn in the two-indications article and in the heart article.
The children’s trial, and which quantity actually fell
SCALE Kids is a phase 3a trial in children aged 6 to under 12, a group for which, as its own opening states, no medication is approved for nonmonogenic, nonsyndromic obesity. It assigned 82 children 2:1 to liraglutide 3.0 mg or the maximum tolerated dose, or placebo, plus lifestyle intervention, for 56 weeks. Mean change in body-mass index was −5.8% against +1.6% on placebo, an estimated difference of −7.4 percentage points (95% CI, −11.6 to −3.2; P < 0.001), and 46% against 9% achieved a BMI reduction of at least 5% (adjusted odds ratio 6.3; 95% CI, 1.4 to 28.8).[5]
The confirmatory secondary endpoint is the one that changes the reading. Mean change in body weight was +1.6% on liraglutide and +10.0% on placebo, an estimated difference of −8.4 percentage points (95% CI, −13.4 to −3.3; P = 0.001).[5] Both groups weighed more at the end than at the start. The body-mass index fell because children of that age are still growing taller, so holding weight roughly steady for a year lowers the ratio. That is a real and clinically meaningful result in pediatric obesity, and it is not weight loss, and the two get quoted interchangeably.
The result has not moved the license either. The current Saxenda prescribing information indicates it for adults and for patients aged 12 years and older with a body weight above 60 kg, supported by a 56-week trial in 251 patients aged 12 to 17.[6] Gastrointestinal adverse events in the younger trial ran 80% against 54%, in 82 children.[5] The adolescent picture across the class is in the adolescents article.
The guideline ranks it twice and gets two answers
The American College of Physicians living guideline of April 2026 makes two recommendations in one document. For nonpregnant adults with obesity at a body-mass index of 30 or higher, it names semaglutide and tirzepatide first-line on moderate-certainty evidence, then phentermine-topiramate second, liraglutide third and naltrexone-bupropion fourth, all on low-certainty evidence. For adults with overweight at a body-mass index of 27 to 30 plus type 2 diabetes, dyslipidemia, hypertension, obstructive sleep apnea or cardiovascular disease, the list is semaglutide and tirzepatide first-line and liraglutide second-line.[7]
Liraglutide gains a rank as the population gets lighter, and it does so without any new evidence about liraglutide: phentermine-topiramate simply is not recommended in the lower-BMI population. A rank is a statement about the field standing around a drug, not about the drug.
The pooled evidence is less kind. A 2026 network meta-analysis of 262 trials and 99,791 participants lists six agents with moderate-to-high certainty for weight loss at one year — tirzepatide, cagrilintide-semaglutide, oral semaglutide, orforglipron, subcutaneous semaglutide and phentermine-topiramate — and liraglutide is not among them. It does appear in the group with the highest discontinuation due to adverse events, where risk ratios across the named agents run from 1.9 to 4.2.[8]
What is actually left: the supply
Liraglutide holds one position no other incretin holds, and it is a commercial one. A census of the openFDA drug label index finds current generic liraglutide injection labels from eight companies other than the brand holder — Lupin, Biocon, Meitheal, Teva, Hybio, Hikma, Cipla and NorthStar Rx — alongside the Victoza and Saxenda labels themselves.[9] Every listed label for semaglutide, tirzepatide and dulaglutide in the same index belongs to the brand holder or to a repackager of the brand. Liraglutide is, for now, the only GLP-1 receptor agonist for weight management with a competitive generic market behind it.
That matters because of the size of the gap a generic can close. An analysis of national price databases put the United States 30-day price of liraglutide at $1,418 against an estimated minimum sustainable manufacturing price of $50, and semaglutide at $804 against $40.[10] A 28-fold spread is what generic entry is for. Whether the saving reaches a cash-pay buyer depends on what sellers do with it — the boards track the standing monthly rates in the standing monthly rates, and when the same happens to the newer molecules is the subject of the patent article.
What this comparison settles
On weight, liraglutide loses to semaglutide in the one trial that held the injection schedule constant and lost again in the trial that did not. On tolerability it is in the worst band of a 262-trial network. On guideline standing it is third-line in one population and second-line in the next one down, on low-certainty evidence in both. Against all of that it holds three years of randomized follow-up, a cardiovascular outcome trial with a mortality signal in diabetes, a phase 3a trial in children as young as 6, and eight generic manufacturers.
None of those advantages is an effect size, which is the honest summary: the reason to take liraglutide is access, not potency. And the trial figures above describe FDA-approved liraglutide at labeled doses. Compounded liraglutide is not FDA-approved and is not assessed by the FDA for safety, efficacy or quality before a pharmacy dispenses it, so a compounded preparation does not inherit any of the evidence on this page. How the figures here are established before publication is set out in the methodology.