Liraglutide research findings - The Echo Compendium

Liraglutide MARKETED

Specific groups of people Decrease

High evidence

In the two pivotal double-blind RCTs, liraglutide added to insulin in type-1 diabetes produced a small, dose-dependent placebo-adjusted HbA1c reduction (about -0.2% at...

Mathieu C, Zinman B, Hemmingsson JU, et al. Efficacy and Safety of Liraglutide Added to I... Source

Full findingIn the two pivotal double-blind RCTs, liraglutide added to insulin in type-1 diabetes produced a small, dose-dependent placebo-adjusted HbA1c reduction (about -0.2% at the top dose over 52 weeks), significant only at the higher doses - a modest glycaemic benefit.

PopulationLiraglutide pivotal double-blind T1D RCTs (Mathieu et al., ADJUNCT ONE, 52wk, n=1,398, 3:1, NCT01836523; corroborated by Ahren et al., ADJUNCT TWO, 26wk, n=835, NCT02098395).

Fundingindustry - Novo Nordisk (sponsor; NCT01836523)

Comparatorsplacebo (insulin background)

Liraglutide MARKETED

Specific groups of people Decrease

High evidence

Liraglutide added to insulin in type-1 diabetes produced significant weight loss at every dose (about -5kg at the top dose) and a modest reduction in total insulin...

Mathieu C, Zinman B, Hemmingsson JU, et al. Efficacy and Safety of Liraglutide Added to I... Source

Full findingLiraglutide added to insulin in type-1 diabetes produced significant weight loss at every dose (about -5kg at the top dose) and a modest reduction in total insulin requirement - the metabolic benefits beyond glycaemia.

PopulationLiraglutide pivotal double-blind T1D RCTs (Mathieu ADJUNCT ONE, NCT01836523; Ahren ADJUNCT TWO, NCT02098395).

Fundingindustry - Novo Nordisk (sponsor; NCT01836523)

Comparatorsplacebo (insulin background)

Liraglutide MARKETED

Specific groups of people Increase

Moderate evidence

Liraglutide reversed obesity-related functional hypogonadism by restoring the HPG axis, unlike exogenous testosterone which suppresses it.

Liraglutide vs testosterone for obesity hypogonadism (open-label RCT, n=30), Endocr Conne... Source

PopulationPeer-reviewed primary (women's-health wave)

Fundingacademic / investigator - University Medical Centre Ljubljana grant P3-0298

Liraglutide MARKETED

Specific groups of people No change

Moderate evidence

Heavier (overweight/obese) type-1 patients did NOT get greater benefit from liraglutide: in a pooled post-hoc, effects on HbA1c, weight and insulin did not differ by...

Dejgaard TF, von Scholten BJ, Christiansen E, et al. Efficacy and safety of liraglutide i... Source

Full findingHeavier (overweight/obese) type-1 patients did NOT get greater benefit from liraglutide: in a pooled post-hoc, effects on HbA1c, weight and insulin did not differ by baseline BMI - the 'double diabetes' rationale for targeting obese T1D is not supported by this subgroup.

PopulationPooled ADJUNCT ONE/TWO post-hoc subgroup analysis (Dejgaard et al., 2021; PMC9292057).

Fundingindustry+academic (Novo Nordisk sponsor; Steno Diabetes Center Copenhagen)

Comparatorsplacebo (insulin background)

Liraglutide MARKETED

Specific groups of people Decrease

Moderate evidence

A pooled post-hoc of the two pivotal trials shows the placebo-adjusted HbA1c reduction peaked near -0.30/-0.35% at week 26 then waned by week 52 - so the larger...

Dejgaard TF, von Scholten BJ, Christiansen E, et al. Efficacy and safety of liraglutide i... Source

Full findingA pooled post-hoc of the two pivotal trials shows the placebo-adjusted HbA1c reduction peaked near -0.30/-0.35% at week 26 then waned by week 52 - so the larger week-26 figure should not be headlined over the smaller 52-week primary endpoint.

PopulationPooled ADJUNCT ONE/TWO post-hoc analysis (Dejgaard et al., 2021; PMC9292057).

Fundingindustry+academic (Novo Nordisk sponsor; Steno Diabetes Center Copenhagen)

Comparatorsplacebo (insulin background)

Liraglutide MARKETED

Heart failure Mixed

High evidence

LIVE: in stable chronic HFrEF on optimal therapy, liraglutide did NOT change LV systolic function but raised heart rate and was associated with more serious cardiac...

Jorsal A et al. (LIVE), Eur J Heart Fail, 2017;19:69-77 Source

Full findingLIVE: in stable chronic HFrEF on optimal therapy, liraglutide did NOT change LV systolic function but raised heart rate and was associated with more serious cardiac adverse events (safety caution).

PopulationN=241, LVEF <=45% (mean ~34%), clinically stable, with/without diabetes; 24 weeks; investigator-initiated RCT vs placebo.

Fundinginvestigator-initiated (academic; not manufacturer-run)

Comparatorsplacebo

Liraglutide MARKETED

Heart failure No change

Moderate evidence

FIGHT: in patients recently hospitalised for acute HFrEF, liraglutide did NOT improve post-hospitalisation clinical stability (null), with numerically more deaths and...

Margulies KB et al. (FIGHT), JAMA, 2016;316:500-508 Source

Full findingFIGHT: in patients recently hospitalised for acute HFrEF, liraglutide did NOT improve post-hospitalisation clinical stability (null), with numerically more deaths and rehospitalisations.

PopulationN=300, established HFrEF (median LVEF 25%), recently hospitalised, 59% T2D; 180 days; phase 2 RCT vs placebo.

Fundingacademic-NHLBI (study drug supplied by Novo Nordisk)

Comparatorsplacebo

Liraglutide MARKETED

Heart failure Decrease

Moderate evidence

In a LIVE substudy (231 HFrEF patients), liraglutide reduced natriuretic peptides only in the type-2-diabetes subgroup (NT-proBNP -25%, MR-proANP -27% vs placebo) with...

Nielsen R et al., Diabetes Obes Metab 2020;22(11):2141-2150 (LIVE natriuretic-peptide sub... Source

Full findingIn a LIVE substudy (231 HFrEF patients), liraglutide reduced natriuretic peptides only in the type-2-diabetes subgroup (NT-proBNP -25%, MR-proANP -27% vs placebo) with no change in non-diabetic patients (significant T2D interaction), indicating any neurohormonal benefit in HFrEF was confined to the diabetic subset.

PopulationLIVE substudy, 231 chronic HFrEF patients (LVEF <=45%), with and without T2D, 24 wk; NCT01472640

Fundinginvestigator-initiated (academic; not manufacturer-run)

Comparatorsplacebo

Liraglutide MARKETED

Heart failure No change

Moderate evidence

In a LIVE PET sub-study (36 non-diabetic HFrEF patients), liraglutide did NOT change myocardial glucose uptake, myocardial blood flow or myocardial flow reserve versus...

Nielsen R et al., J Nucl Cardiol 2019;26(2):585-597 (LIVE PET substudy) Source

Full findingIn a LIVE PET sub-study (36 non-diabetic HFrEF patients), liraglutide did NOT change myocardial glucose uptake, myocardial blood flow or myocardial flow reserve versus placebo despite lowering weight and HbA1c - so the increased cardiac-event signal with liraglutide in chronic HF does not appear to be mediated by changes in myocardial perfusion or substrate uptake.

PopulationLIVE PET substudy, 36 non-diabetic HFrEF patients (LVEF <=45%); NCT01472640

Fundinginvestigator-initiated (academic; not manufacturer-run)

Comparatorsplacebo

Liraglutide MARKETED

Heart failure No change

Low evidence

A systematic review of the two isolating GLP-1RA HFrEF RCTs (FIGHT and LIVE, combined n=541, 24-week liraglutide) found no benefit and an adverse-leaning signal - no...

Hamad F et al., Curr Diabetes Rev 2021;17(3):280-292 (systematic review FIGHT+LIVE) Source

Full findingA systematic review of the two isolating GLP-1RA HFrEF RCTs (FIGHT and LIVE, combined n=541, 24-week liraglutide) found no benefit and an adverse-leaning signal - no between-group difference in FIGHT's global rank/deaths/HF rehospitalisation and no LVEF change in LIVE - concluding liraglutide use in HFrEF was associated with increased risk of HF-related outcomes.

Populationsystematic review of FIGHT + LIVE (two GLP-1RA HFrEF RCTs), combined n=541

Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)

Comparatorsplacebo

Liraglutide MARKETED

safety-other No change

High evidence

Rodent C-cell tumour/calcitonin signal does not translate: human sequential calcitonin screening over up to 2 yr in >5000 liraglutide/control subjects showed no...

Hegedus L et al., J Clin Endocrinol Metab 2011 Source

Full findingRodent C-cell tumour/calcitonin signal does not translate: human sequential calcitonin screening over up to 2 yr in >5000 liraglutide/control subjects showed no consistent dose/time-dependent rise and no between-group difference.

Population>5000 subjects with T2D or non-diabetic obesity; calcitonin at 3-month intervals up to 2 yr

Fundingindustry - Novo Nordisk (trial sponsor)

Comparatorscontrol therapy

Liraglutide MARKETED

safety-other Not directional

High evidence

The pivotal-trial investigators themselves concluded that, despite the metabolic benefits, the increased hypoglycaemia and hyperglycaemia-with-ketosis limit the...

Mathieu C, Zinman B, Hemmingsson JU, et al. Efficacy and Safety of Liraglutide Added to I... Source

Full findingThe pivotal-trial investigators themselves concluded that, despite the metabolic benefits, the increased hypoglycaemia and hyperglycaemia-with-ketosis limit the clinical use of liraglutide as an insulin adjunct in T1D - the honest net verdict.

PopulationLiraglutide pivotal double-blind T1D RCTs (Mathieu ADJUNCT ONE; Ahren ADJUNCT TWO) - authors' own conclusions.

Fundingindustry - Novo Nordisk (sponsor; NCT01836523)

Comparatorsplacebo (insulin background)

Liraglutide MARKETED

safety-other Increase

High evidence

Liraglutide added to insulin significantly increased the rate of symptomatic hypoglycaemia in type-1 diabetes - a disease-specific harm that accompanies the modest...

Ahren B, Hirsch IB, Pieber TR, et al. Efficacy and Safety of Liraglutide Added to Capped ... Source

Full findingLiraglutide added to insulin significantly increased the rate of symptomatic hypoglycaemia in type-1 diabetes - a disease-specific harm that accompanies the modest efficacy.

PopulationLiraglutide pivotal double-blind T1D RCTs (Ahren ADJUNCT TWO, 26wk, NCT02098395; corroborated by Mathieu ADJUNCT ONE, NCT01836523).

Fundingindustry - Novo Nordisk (sponsor; NCT02098395)

Comparatorsplacebo (insulin background)

Liraglutide MARKETED

safety-other Increase

High evidence

Both pivotal trials showed a dose-dependent increase in hyperglycaemia-with-ketosis at the highest liraglutide dose (event rate roughly doubled) - the real,...

Mathieu C, Zinman B, Hemmingsson JU, et al. Efficacy and Safety of Liraglutide Added to I... Source

Full findingBoth pivotal trials showed a dose-dependent increase in hyperglycaemia-with-ketosis at the highest liraglutide dose (event rate roughly doubled) - the real, disease-specific DKA-adjacent safety signal that, with the hypoglycaemia, limits clinical use.

PopulationLiraglutide pivotal double-blind T1D RCTs (Mathieu ADJUNCT ONE, NCT01836523; Ahren ADJUNCT TWO, NCT02098395).

Fundingindustry - Novo Nordisk (sponsor; NCT01836523)

Comparatorsplacebo (insulin background)

Liraglutide MARKETED

safety-other Decrease

Very low evidence

In an HFE-knockout haemochromatosis mouse model, liraglutide lowered both serum and hepatic iron through a hepcidin-independent mechanism (Hamp not differentially...

Bozadjieva-Kramer N et al. Liraglutide lowers iron in HFE-knockout mice independent of he... Source

Full findingIn an HFE-knockout haemochromatosis mouse model, liraglutide lowered both serum and hepatic iron through a hepcidin-independent mechanism (Hamp not differentially expressed), arguing for a genuine iron-handling effect rather than only an acute-phase change.

PopulationHFE-knockout (HFE-KO) haemochromatosis mice; hepatic and serum iron, Hamp/hepcidin expression measured. Murine model; no human read-across.

Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)

Comparatorsvehicle-treated HFE-KO mice

Liraglutide MARKETED

appetite-intake Decrease

Moderate evidence

Liraglutide (1.8 and 3.0 mg, 5 wk) reduced ad libitum energy intake at a test lunch by approximately 16% vs placebo in obese non-diabetic adults, increased...

van Can J et al., Int J Obes (Lond) 2014;38(6):784-793 Source

Full findingLiraglutide (1.8 and 3.0 mg, 5 wk) reduced ad libitum energy intake at a test lunch by approximately 16% vs placebo in obese non-diabetic adults, increased postprandial satiety/fullness and reduced hunger and prospective food consumption; authors concluded weight loss is appetite/intake-mediated, not expenditure-driven. 5-h gastric-emptying AUC was equivalent to placebo, but 1-h gastric emptying was 23% lower (3.0 mg, P=0.007) — an acute early-phase delay that does not persist.

PopulationObese non-diabetic adults, 5 weeks (NCT00978393)

Fundingindustry - Novo Nordisk

Comparatorsplacebo

Liraglutide MARKETED

appetite-intake Decrease

Low evidence

Liraglutide (escalated to 1.8 mg, 17 days) decreased activation of the parietal cortex (and, secondarily, insula and putamen reward regions) in response to highly...

Farr OM et al., Diabetologia 2016;59(5):954-965 Source

Full findingLiraglutide (escalated to 1.8 mg, 17 days) decreased activation of the parietal cortex (and, secondarily, insula and putamen reward regions) in response to highly desirable food images vs placebo in T2D, providing human fMRI evidence for a central/reward-system contribution to GLP-1R-mediated reduced food intake.

PopulationT2D, crossover RCT, n=18 analysed, 17 days

Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)

Comparatorsplacebo

Liraglutide MARKETED

appetite-intake Decrease

Low evidence

Liraglutide 1.8 mg reduced CNS (insula, putamen) responses to food pictures and enhanced the satiating effect of meals (putamen, amygdala) vs insulin glargine after 10...

Ten Kulve JS et al., Diabetes Care 2016;39(2):214-221 Source

Full findingLiraglutide 1.8 mg reduced CNS (insula, putamen) responses to food pictures and enhanced the satiating effect of meals (putamen, amygdala) vs insulin glargine after 10 days, but these CNS differences were NO LONGER present after 12 weeks, suggesting the central food-cue/reward effect contributes to the induction of weight loss but not necessarily its maintenance.

PopulationT2D, n=20, active-comparator RCT (insulin glargine)

Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)

Comparatorsinsulin glargine

Liraglutide MARKETED

appetite-intake Decrease

Very low evidence

The acute GLP-1R-mediated gastric-emptying delay is subject to tachyphylaxis under sustained exposure whereas the appetite/body-weight effect is not: in rats,...

Jelsing J et al., Diabetes Obes Metab 2012;14(6):531-538 Source

Full findingThe acute GLP-1R-mediated gastric-emptying delay is subject to tachyphylaxis under sustained exposure whereas the appetite/body-weight effect is not: in rats, liraglutide's gastric-emptying inhibition was markedly diminished after 14 days while body-weight reduction persisted, indicating brain appetite signalling (not gastric emptying) is the main mechanism of liraglutide-induced weight loss.

PopulationRats (mechanistic)

Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)

Comparatorsvehicle; exenatide (short-acting comparator)

Liraglutide MARKETED

body-composition Decrease

High evidence

In a randomised DXA study, liraglutide reduced bone density more than exercise at the same weight loss, and resistance exercise prevented the loss.

Liraglutide vs exercise on bone density (RCT DXA secondary analysis), JAMA Netw Open 2024 Source

PopulationPeer-reviewed primary (women's-health wave)

Fundingacademic / non-commercial (Novo Nordisk Foundation + Danish academic bodies; drug supplied in-kind; disclosed)

Liraglutide MARKETED

body-composition Decrease

High evidence

Phase 4 MRI RCT (high-CV-risk overweight/obesity, no diabetes): liraglutide 3.0 mg significantly lowered visceral adipose tissue over 40 weeks vs placebo (primary...

Neeland IJ et al., Lancet Diabetes Endocrinol, 2021 Source

Full findingPhase 4 MRI RCT (high-CV-risk overweight/obesity, no diabetes): liraglutide 3.0 mg significantly lowered visceral adipose tissue over 40 weeks vs placebo (primary endpoint VAT% by MRI).

Populationn=185 randomised (liraglutide 92, placebo 93); 128 analysed; 92% female, 37% Black; mean BMI 37.7; 40 weeks + 500 kcal-deficit diet; RCT (NCT03038620)

Fundingacademic-led (University of Texas Southwestern) with Novo Nordisk collaborator

Comparatorsplacebo

Liraglutide MARKETED

body-composition Decrease

Moderate evidence

Pre-specified MRI secondary analysis: liraglutide 3.0 mg reduced thigh-muscle fat (myosteatosis) and adverse muscle composition vs placebo - a muscle-QUALITY...

Pandey A et al., J Cachexia Sarcopenia Muscle, 2024 Source

Full findingPre-specified MRI secondary analysis: liraglutide 3.0 mg reduced thigh-muscle fat (myosteatosis) and adverse muscle composition vs placebo - a muscle-QUALITY improvement distinct from muscle quantity.

Populationn=128 with follow-up MRI (liraglutide 73, placebo 55); 92.2% women, 36.7% Black; median 36 wk; same trial as NCT03038620

Fundingindustry-Novo Nordisk

Comparatorsplacebo

Thin or bounded evidence here means uncertainty remains visible. It is not evidence that an effect has been ruled out.

Liraglutide MARKETED

pharmacology No change

Very low evidence

In high-fat-diet rodents, liraglutide reproducibly changes gut-microbiota composition, and in one study the drug's effect on the microbiota was statistically separable...

Wang L, Li P, Tang Z, Yan X, Feng B. Structural modulation of the gut microbiota and the ... Source

Full findingIn high-fat-diet rodents, liraglutide reproducibly changes gut-microbiota composition, and in one study the drug's effect on the microbiota was statistically separable (by ridge regression) from the effect of reduced food intake - but this is a compositional shift only, with no test of whether the shift causes any metabolic benefit.

PopulationHigh-fat-diet C57BL/6 mice (Wang 2016 Sci Rep, liraglutide vs saxagliptin/control; 16S rRNA), corroborated by Wistar and Goto-Kakizaki rats (Zhao 2018 Front Endocrinol; 16S).

Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)

Comparatorssaxagliptin; saline/control

Liraglutide MARKETED

pharmacology No change

Very low evidence

The one study that does show causation (a germ-free/antibiotic faecal-transplant experiment in mice) transferred an improvement in INSULIN SECRETION and gut immunity -...

Charpentier J, Briand F, Lelouvier B, et al. (Burcelin R, senior). Liraglutide targets th... Source

Full findingThe one study that does show causation (a germ-free/antibiotic faecal-transplant experiment in mice) transferred an improvement in INSULIN SECRETION and gut immunity - not weight loss - was specific to liraglutide (exendin-4 did not reproduce it), and has no human equivalent; it does not show the microbiome mediates GLP-1-RA weight loss.

PopulationGerm-free + antibiotic-treated diet-induced-dysmetabolic mice; faecal microbiota transfer from liraglutide-treated vs control mice; Exendin-4 as a non-reproducing comparator (Charpentier/Burcelin 2021 Acta Diabetol).

Fundingacademic/independent (INSERM/I2MC Toulouse; Burcelin lab)

Comparatorsvehicle; Exendin-4; germ-free recipients of control microbiota

Thin or bounded evidence here means uncertainty remains visible. It is not evidence that an effect has been ruled out.

Liraglutide MARKETED

Heart and circulation Decrease

High evidence

In T2D at high CV risk, liraglutide reduced MACE versus placebo; unlike SUSTAIN-6, CV death and all-cause death were also significantly reduced.

Marso SP et al. (LEADER), NEJM, 2016 Source

PopulationLEADER: N=9340; T2D high CV risk; double-blind placebo-controlled RCT; median 3.8 years.

Fundingindustry-Novo Nordisk + academic-NIH (co-funded)

Comparatorsplacebo

Thin or bounded evidence here means uncertainty remains visible. It is not evidence that an effect has been ruled out.

Liraglutide MARKETED

renal Decrease

Moderate evidence

LEADER prespecified secondary renal composite reduced with liraglutide versus placebo (driven by new macroalbuminuria).

Mann JFE et al. (LEADER renal), NEJM, 2017;377:839-848 Source

PopulationLEADER: N=9,340 T2D, high CV risk; median 3.84 years; RCT.

Fundingindustry-Novo Nordisk + academic-NIH (co-funded)

Comparatorsplacebo

Thin or bounded evidence here means uncertainty remains visible. It is not evidence that an effect has been ruled out.

Liraglutide MARKETED

Heart rate Increase

Very low evidence

Liraglutide raised sinoatrial firing rate by a DIRECT action on the node in isolated (denervated) rabbit Langendorff hearts and murine SAN myocytes: the effect...

Jons C et al., Life Sci 2021;282:119815 Source

Full findingLiraglutide raised sinoatrial firing rate by a DIRECT action on the node in isolated (denervated) rabbit Langendorff hearts and murine SAN myocytes: the effect survived beta-blockade (propranolol) but was abolished by funny-current (I_f) blockade with ivabradine, isolating an I_f-mediated direct-SAN component independent of autonomic reflexes.

Populationisolated denervated rabbit Langendorff hearts and isolated murine sinoatrial-node myocytes

Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)

Comparatorspropranolol; ivabradine

Thin or bounded evidence here means uncertainty remains visible. It is not evidence that an effect has been ruled out.

Liraglutide MARKETED

glycaemic No change

Moderate evidence

FDA Victoza Clinical Pharmacology review: in a stepwise hypoglycaemic clamp, a single liraglutide dose did NOT impair the glucagon counter-regulatory response...

FDA Victoza NDA 022341 Clinical Pharmacology (OCP) Review (scanned/OCR) Source

Full findingFDA Victoza Clinical Pharmacology review: in a stepwise hypoglycaemic clamp, a single liraglutide dose did NOT impair the glucagon counter-regulatory response (glucagon rose ~1.5-fold with falling glucose in both arms). This is the GLP-1-monoagonist baseline for the retatrutide GCGR-arm hypoglycaemia-counter-regulation question: pure GLP-1 does not impair counter-regulation; whether chronic GCGR agonism plus incretin insulin secretion does is retatrutide-specific and open.

PopulationStepwise hypoglycaemic clamp, single-dose liraglutide vs comparator (FDA Victoza NDA 022341 ClinPharm/OCP review, counter-regulation section, Fig 10)

Fundingindustry - Novo Nordisk (applicant/manufacturer; FDA Clinical Pharmacology/OCP review of Novo's NDA 022341 submission)

Comparatorsplacebo/comparator arm

Thin or bounded evidence here means uncertainty remains visible. It is not evidence that an effect has been ruled out.

Liraglutide MARKETED

Stomach and gut Increase

High evidence

GI AEs more frequent than placebo, largely mild-to-moderate, transient, in escalation; AE-related discontinuation higher than placebo.

Pi-Sunyer X et al., N Engl J Med 2015 (SCALE) Source

PopulationSCALE Obesity and Prediabetes: 3731 adults BMI>=30 (or >=27+comorbidity); liraglutide 3.0 mg daily vs placebo

Fundingindustry - Novo Nordisk (disclosed in publication)

Comparatorsplacebo

Thin or bounded evidence here means uncertainty remains visible. It is not evidence that an effect has been ruled out.

Liraglutide MARKETED

Brain and mood Mixed

Moderate evidence

NEURODEGENERATION (Alzheimer's): the ELAD phase-2b trial of liraglutide in mild-moderate Alzheimer's MISSED its primary endpoint (cerebral glucose metabolism), with a...

Edison P, Femminella GD, Ritchie C et al., Nat Med, 2025 (ELAD) Source

Full findingNEURODEGENERATION (Alzheimer's): the ELAD phase-2b trial of liraglutide in mild-moderate Alzheimer's MISSED its primary endpoint (cerebral glucose metabolism), with a surviving positive SECONDARY cognitive signal (ADAS-Executive) on an unadjusted p-value among several null secondaries. A missed-primary trial with a hypothesis-generating cognitive secondary, not a clean positive.

PopulationELAD (NCT01843075): N=204 mild-moderate AD, non-diabetic; multicentre double-blind RCT; daily s.c. liraglutide vs placebo, 52 weeks.

Fundingacademic / non-commercial (ELAD; Imperial College / MRC + charities; disclosed)

Comparatorsplacebo

Thin or bounded evidence here means uncertainty remains visible. It is not evidence that an effect has been ruled out.

Liraglutide MARKETED

weight-loss Decrease

High evidence

Once-daily subcutaneous liraglutide 3.0 mg produced a mean weight loss of -8.0% (-8.4 kg) vs -2.6% (-2.8 kg) placebo at week 56; 63.2% achieved >=5% weight loss vs...

Pi-Sunyer X et al., NEJM 2015 (SCALE Obesity and Prediabetes) Source

Full findingOnce-daily subcutaneous liraglutide 3.0 mg produced a mean weight loss of -8.0% (-8.4 kg) vs -2.6% (-2.8 kg) placebo at week 56; 63.2% achieved >=5% weight loss vs 27.1% placebo.

PopulationSCALE Obesity and Prediabetes: 3731 adults without T2D, BMI >=30 (or >=27 with comorbidity), 56 weeks, randomised 2:1 double-blind vs placebo

Fundingindustry-Novo Nordisk

Comparatorsplacebo

Thin or bounded evidence here means uncertainty remains visible. It is not evidence that an effect has been ruled out.

Liraglutide MARKETED

weight-maintenance-regain Decrease

Moderate evidence

SCALE Maintenance: adults who first lost >=5% on a low-calorie-diet run-in were randomised to liraglutide 3.0 mg or placebo to test MAINTENANCE of diet-induced loss....

Wadden TA, Hollander P, Klein S et al., Int J Obes (Lond), 2013 (SCALE Maintenance) Source

Full findingSCALE Maintenance: adults who first lost >=5% on a low-calorie-diet run-in were randomised to liraglutide 3.0 mg or placebo to test MAINTENANCE of diet-induced loss. Liraglutide held and added to the loss while placebo did not. SCOPE CAVEAT: this tests maintenance-vs-placebo on a background of diet-induced loss, NOT withdrawal of established drug responders, so it is mechanistically maintenance-vs-placebo, not active-drug-discontinuation regain.

PopulationSCALE Maintenance (NCT00781937): N=422; obese/overweight adults without diabetes who lost >=5% on a low-calorie diet; double-blind RCT, 56 weeks.

Fundingindustry - Novo Nordisk

Comparatorsplacebo

Thin or bounded evidence here means uncertainty remains visible. It is not evidence that an effect has been ruled out.

Liraglutide MARKETED

Liver and MASH Decrease

Moderate evidence

LEAN was the historical GLP-1-mono proof-of-concept for histological NASH resolution: liraglutide 1.8 mg/day increased resolution of definite NASH without worsening...

Armstrong MJ, Gaunt P et al. (LEAN). Lancet 2016;387:679-690. Source

Full findingLEAN was the historical GLP-1-mono proof-of-concept for histological NASH resolution: liraglutide 1.8 mg/day increased resolution of definite NASH without worsening fibrosis vs placebo and reduced the proportion with fibrosis PROGRESSION. Small (n=52 randomised, ~45 biopsied), single-group (A'Hern) design. The fibrosis result is LESS PROGRESSION, distinct from active >=1-stage IMPROVEMENT. Resolution is a SURROGATE; no hard liver-outcome data.

PopulationLEAN (NCT01237119), phase 2 double-blind RCT, 4 UK centres. n=52 (26/26); biopsy NASH; 48 wk; analysed among ~45 end-of-treatment biopsies. A'Hern single-group design.

Fundingacademic / non-commercial (LEAN; investigator-initiated NIHR/Wellcome; disclosed)

Comparatorsplacebo

Thin or bounded evidence here means uncertainty remains visible. It is not evidence that an effect has been ruled out.

Liraglutide MARKETED

Sleep apnoea Decrease

High evidence

SCALE Sleep Apnea was the historical GLP-1 mono-agonist OSA RCT and proof-of-concept that an incretin lowers AHI: liraglutide 3.0 mg, adjunct to diet/exercise, reduced...

Blackman A, Foster GD, Zammit G et al. Effect of liraglutide 3.0 mg in individuals with o... Source

Full findingSCALE Sleep Apnea was the historical GLP-1 mono-agonist OSA RCT and proof-of-concept that an incretin lowers AHI: liraglutide 3.0 mg, adjunct to diet/exercise, reduced AHI more than placebo at 32 weeks in obese non-diabetic adults with moderate-to-severe OSA unwilling/unable to use CPAP. The effect is modest and parallels modest (~4% placebo-subtracted) weight loss.

PopulationSCALE Sleep Apnea (NCT01557166); N=359 (liraglutide 180, placebo 179); obese (mean BMI 39.1) non-diabetic, baseline mean AHI 49.2 events/hr, 67% severe; CPAP-refusing; 32-wk double-blind RCT, both arms on a 500 kcal/day deficit + exercise.

Fundingindustry - Novo Nordisk

Comparatorsplacebo

Thin or bounded evidence here means uncertainty remains visible. It is not evidence that an effect has been ruled out.

Liraglutide MARKETED

insulin-beta-cell Increase

Low evidence

GLP-1R agonism improves hepatic insulin sensitivity at the clamp and tracer level: liraglutide (12 wk, biopsy-proven NASH) increased suppression of endogenous glucose...

Armstrong MJ et al., J Hepatol 2016;64(2):399-408 Source

Full findingGLP-1R agonism improves hepatic insulin sensitivity at the clamp and tracer level: liraglutide (12 wk, biopsy-proven NASH) increased suppression of endogenous glucose production under low-dose insulin and improved adipose insulin sensitivity, using paired hyperinsulinaemic-euglycaemic clamps with stable-isotope tracers.

PopulationBiopsy-proven NASH, n=14, 12 weeks

Fundingacademic/independent analysis (manufacturer not study sponsor; per-study COI not individually audited)

Comparatorsplacebo