Study analysis · JHEP Reports · 2025
Losing weight with a drug might make you gain it back faster — here's why.
Both diet and a weight-loss drug shrink fatty liver the same amount, but stopping the drug triggers weird body signals that could make you regain weight.
Overview
What the study found
The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.
In simple terms
This study is like a fair race between two ways to lose weight: eating less and taking a medicine. It shows that during the race, the medicine helped with blood sugar and fats more than just eating less. But once the race ended, the medicine group started having weird body changes that didn’t happen in the eating group. So we know what happened during the race, but we don’t know if one way is better overall.
What’s the bottom line?
Two ways to lose weight—eating less or taking a drug—both shrink fat in the liver equally. But the drug also helps blood sugar and fat production in the liver while you're taking it. When you stop the drug, those benefits vanish and your body sends out strange signals that might make you gain weight again.
How strong is this study?
This study tried to be fair by randomly picking who got which weight-loss method, which is good. But it only had 29 people, no one was blind to who got what, and it didn’t last long. So while we can trust what happened during the study, we can’t be super sure it would work the same for everyone or for a long time.
40 / 100
- COI disclosure+40/40
- Data availabilitydata not shared
- Code availabilitycode not shared
60 / 100
- Randomization+20/20
- Blindingnot blinded
- Control group+15/15
- Sample size (n=29)+2.7/20
- Follow-up+10/10
100 / 100
77 / 100
- P-values+15/15
- Effect size+20/20
- Confidence intervalsno confidence intervals
- Pre-registration+15/15
Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.
Where it sits
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 545 / 100
Probability of being correct
Participants are randomly assigned to treatment or control groups, minimizing bias. The gold standard for testing whether an intervention causes an effect.
This design can establish causation. Randomization allows causal inference for the comparison between GLP-1RA and lifestyle interventions during the 12-week treatment period. However, the small sample size (n=29), lack of blinding, and short duration limit confidence in long-term or population-level causal claims, especially regarding withdrawal effects.
Major COI
Major conflicts that significantly reduce study credibility
The study was funded by Novo Nordisk, the manufacturer of liraglutide, and the drug was provided by them, raising concerns about potential influence on study design and interpretation, despite no explicit funder control over analysis or publication.
Funders
Conflict Details
Novo Nordisk: Provided liraglutide for the trial and sponsored the study.
The study was sponsored and monitored by the University of Oxford, but Novo Nordisk supplied the investigational drug. No independent data monitoring committee or third-party analysis is mentioned. The trial was registered as an Investigational Medicinal Product trial, indicating industry sponsorship. While authors are not stated to be employed by Novo Nordisk, the close alignment of study design with commercial interests (testing liraglutide vs. lifestyle) and lack of transparency on funder input beyond funding raises significant bias risk.
Key takeaways
- 01
Both groups lost ~5 kg and reduced liver fat by 7.5%.
- 02
The drug group had better blood sugar and lower liver fat production during treatment.
- 03
After stopping, blood sugar worsened and 4 inflammatory proteins (MMP-10, IL10RB, FGF-23, Flt3L) rose only in the drug group.
- 04
Yes—losing weight helps the liver no matter how you do it, but stopping the drug may trigger biological changes that make it harder to keep the weight off.
Surprising findings
- Liraglutide improved glucose control and lowered liver fat production more than diet — but all benefits vanished after stopping, while diet’s effects persisted.Most assume drug effects are stronger and longer-lasting — but here, the drug’s advantages disappeared completely, while diet’s liver improvements stuck, even though participants likely continued some healthy habits.
- Adipose tissue gene expression changed significantly after liraglutide withdrawal — but not after lifestyle withdrawal — despite identical weight loss.We expect lifestyle changes to cause lasting biological shifts — but this study shows the drug, not the diet, triggered unique gene dysregulation after stopping.
Practical takeaways
If you're using a GLP-1 drug for fatty liver or weight loss, plan for long-term maintenance before you stop — don’t assume the benefits will stick.
This study had only 29 participants and no placebo group; lifestyle participants may have continued healthy habits after withdrawal, which could skew results.
medium confidencePrioritize sustainable lifestyle changes over short-term drug use — they produce equal liver benefits without the risk of post-treatment metabolic rebound.
Lifestyle changes are harder to maintain for many people — this doesn’t mean drugs are useless, just that they’re not a permanent fix.
medium confidenceWhy this study matters
Same Weight Loss, Different Biology
Both liraglutide and lifestyle intervention led to identical weight loss (avg. ~5 kg) and reduced liver fat by 7.5%. But only liraglutide improved glucose tolerance and lowered liver fat production (DNL) during treatment — yet all benefits vanished after stopping.
It challenges the idea that drugs are 'better' than diet — showing that while drugs offer temporary metabolic perks, they don't outperform lifestyle changes in the long run and may even backfire.
The Hidden Warning Signs After Stopping the Drug
Twelve weeks after quitting liraglutide, four inflammatory proteins — MMP-10, IL10RB, FGF-23, and Flt3L — spiked in blood, and fat tissue genes became dysregulated. These changes didn’t occur after stopping dieting.
This suggests the drug doesn’t just help you lose weight — it may rewire your body in a way that makes rebound weight gain more likely, like a biological booby trap.
DNL Rebound: Fat Production Comes Back Hard
Hepatic de novo lipogenesis (DNL) — the liver’s process of making new fat — dropped during treatment in both groups, but rebounded after withdrawal in everyone. This means even diet-induced weight loss doesn’t permanently fix fat production.
It proves that weight loss alone doesn’t reprogram your liver — the body fights to restore fat storage, making long-term maintenance harder than we thought.
Want the whole report?
Detailed mode opens the full scientific breakdown — every score component, the methodology, conflicts of interest, the evidence analysis behind each claim, and the raw study data.
Overview
What the study found
The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
Two ways to lose weight—eating less or taking a drug—both shrink fat in the liver equally. But the drug also helps blood sugar and fat production in the liver while you're taking it. When you stop the drug, those benefits vanish and your body sends out strange signals that might make you gain weight again.
Research results
Both groups lost ~5 kg and reduced liver fat by 7.5%. The drug group had better blood sugar and lower liver fat production during treatment. After stopping, blood sugar worsened and 4 inflammatory proteins (MMP-10, IL10RB, FGF-23, Flt3L) rose only in the drug group.
What this means - more context
Yes—losing weight helps the liver no matter how you do it, but stopping the drug may trigger biological changes that make it harder to keep the weight off.
This study compares the effects of liraglutide versus matched lifestyle-induced weight loss on liver fat and metabolic health in adults with MASLD without type 2 diabetes, and assesses changes after treatment withdrawal.
Both liraglutide and lifestyle intervention achieved similar weight loss and reduced liver fat by ~7.5%. Liraglutide improved glucose tolerance, fasting HbA1c, and reduced hepatic DNL more than lifestyle, but these benefits reversed completely after withdrawal. After liraglutide cessation, circulating MMP-10, IL10RB, FGF-23, and Flt3L increased and adipose gene expression became dysregulated—changes not seen after lifestyle withdrawal. DNL rebounded in both groups after treatment ended.
Methods Used
Twenty-nine adults with MASLD without type 2 diabetes were randomized to 12 weeks of liraglutide (1.8 mg/day) or lifestyle intervention (~500 kcal/day deficit). Metabolic phenotyping included liver MRI, DNL measurement via deuterated water, adipose RNA sequencing, circulating proteomics, and stool microbiome analysis, repeated at 12 weeks and again 12 weeks after treatment withdrawal.
Main Finding
Matched weight loss from liraglutide or lifestyle reduced liver fat by ~7.5% similarly; however, liraglutide uniquely improved glucose tolerance and reduced DNL during treatment, with all benefits reversing after withdrawal. Post-withdrawal, liraglutide but not lifestyle caused significant increases in MMP-10, IL10RB, FGF-23, Flt3L, and adipose gene dysregulation.
Confidence Level
Moderate. The study is a well-conducted randomized controlled trial with deep phenotyping and a withdrawal phase, but small sample size (n=29), no placebo arm, and lack of liver biopsy limit generalizability.
Study Flags
Red Flags
- •Small sample size (n=29)
- •No placebo control group
- •Differential adherence post-withdrawal likely (lifestyle advice may have continued)
Surprising Findings
Liraglutide improved glucose control and lowered liver fat production more than diet — but all benefits vanished after stopping, while diet’s effects persisted.
Most assume drug effects are stronger and longer-lasting — but here, the drug’s advantages disappeared completely, while diet’s liver improvements stuck, even though participants likely continued some healthy habits.
Practical Takeaways
If you're using a GLP-1 drug for fatty liver or weight loss, plan for long-term maintenance before you stop — don’t assume the benefits will stick.
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 545 / 100
Probability of being correct
Participants are randomly assigned to treatment or control groups, minimizing bias. The gold standard for testing whether an intervention causes an effect.
Human RCT
Subject
Moderate probability
on the GRADE evidence scale
This study is like a fair race between two ways to lose weight: eating less and taking a medicine. It shows that during the race, the medicine helped with blood sugar and fats more than just eating less. But once the race ended, the medicine group started having weird body changes that didn’t happen in the eating group. So we know what happened during the race, but we don’t know if one way is better overall.
Major conflicts — industry funding with significant control over the research. Reporting score and overall score cap have been reduced.
Score cap reduced to 50% due to major conflicts.
Strengths
- Randomized controlled design
- Direct comparison of two interventions with matched weight loss
- Comprehensive metabolic phenotyping (MRS, RNA-seq, proteomics, microbiome)
Weaknesses
- Very small sample size (n=29)
- No blinding of participants or investigators
- No placebo control group
Methodology
Evidence Keywords
Statistical Reporting
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
Two ways to lose weight—eating less or taking a drug—both shrink fat in the liver equally. But the drug also helps blood sugar and fat production in the liver while you're taking it. When you stop the drug, those benefits vanish and your body sends out strange signals that might make you gain weight again.
Research results
Both groups lost ~5 kg and reduced liver fat by 7.5%. The drug group had better blood sugar and lower liver fat production during treatment. After stopping, blood sugar worsened and 4 inflammatory proteins (MMP-10, IL10RB, FGF-23, Flt3L) rose only in the drug group.
What this means - more context
Yes—losing weight helps the liver no matter how you do it, but stopping the drug may trigger biological changes that make it harder to keep the weight off.
This study compares the effects of liraglutide versus matched lifestyle-induced weight loss on liver fat and metabolic health in adults with MASLD without type 2 diabetes, and assesses changes after treatment withdrawal.
Both liraglutide and lifestyle intervention achieved similar weight loss and reduced liver fat by ~7.5%. Liraglutide improved glucose tolerance, fasting HbA1c, and reduced hepatic DNL more than lifestyle, but these benefits reversed completely after withdrawal. After liraglutide cessation, circulating MMP-10, IL10RB, FGF-23, and Flt3L increased and adipose gene expression became dysregulated—changes not seen after lifestyle withdrawal. DNL rebounded in both groups after treatment ended.
Methods Used
Twenty-nine adults with MASLD without type 2 diabetes were randomized to 12 weeks of liraglutide (1.8 mg/day) or lifestyle intervention (~500 kcal/day deficit). Metabolic phenotyping included liver MRI, DNL measurement via deuterated water, adipose RNA sequencing, circulating proteomics, and stool microbiome analysis, repeated at 12 weeks and again 12 weeks after treatment withdrawal.
Main Finding
Matched weight loss from liraglutide or lifestyle reduced liver fat by ~7.5% similarly; however, liraglutide uniquely improved glucose tolerance and reduced DNL during treatment, with all benefits reversing after withdrawal. Post-withdrawal, liraglutide but not lifestyle caused significant increases in MMP-10, IL10RB, FGF-23, Flt3L, and adipose gene dysregulation.
Confidence Level
Moderate. The study is a well-conducted randomized controlled trial with deep phenotyping and a withdrawal phase, but small sample size (n=29), no placebo arm, and lack of liver biopsy limit generalizability.
Study Flags
Red Flags
- •Small sample size (n=29)
- •No placebo control group
- •Differential adherence post-withdrawal likely (lifestyle advice may have continued)
Surprising Findings
Liraglutide improved glucose control and lowered liver fat production more than diet — but all benefits vanished after stopping, while diet’s effects persisted.
Most assume drug effects are stronger and longer-lasting — but here, the drug’s advantages disappeared completely, while diet’s liver improvements stuck, even though participants likely continued some healthy habits.
Practical Takeaways
If you're using a GLP-1 drug for fatty liver or weight loss, plan for long-term maintenance before you stop — don’t assume the benefits will stick.
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 545 / 100
Probability of being correct
Participants are randomly assigned to treatment or control groups, minimizing bias. The gold standard for testing whether an intervention causes an effect.
Human RCT
Subject
Moderate probability
on the GRADE evidence scale
This study is like a fair race between two ways to lose weight: eating less and taking a medicine. It shows that during the race, the medicine helped with blood sugar and fats more than just eating less. But once the race ended, the medicine group started having weird body changes that didn’t happen in the eating group. So we know what happened during the race, but we don’t know if one way is better overall.
Major conflicts — industry funding with significant control over the research. Reporting score and overall score cap have been reduced.
Score cap reduced to 50% due to major conflicts.
Strengths
- Randomized controlled design
- Direct comparison of two interventions with matched weight loss
- Comprehensive metabolic phenotyping (MRS, RNA-seq, proteomics, microbiome)
Weaknesses
- Very small sample size (n=29)
- No blinding of participants or investigators
- No placebo control group
Methodology
Evidence Keywords
Statistical Reporting
Scoring
How strong is this study?
This study tried to be fair by randomly picking who got which weight-loss method, which is good. But it only had 29 people, no one was blind to who got what, and it didn’t last long. So while we can trust what happened during the study, we can’t be super sure it would work the same for everyone or for a long time.
40 / 100
- COI disclosure+40/40
- Data availabilitydata not shared
- Code availabilitycode not shared
60 / 100
- Randomization+20/20
- Blindingnot blinded
- Control group+15/15
- Sample size (n=29)+2.7/20
- Follow-up+10/10
100 / 100
77 / 100
- P-values+15/15
- Effect size+20/20
- Confidence intervalsno confidence intervals
- Pre-registration+15/15
Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.
Where it sits
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 545 / 100
Probability of being correct
Participants are randomly assigned to treatment or control groups, minimizing bias. The gold standard for testing whether an intervention causes an effect.
This design can establish causation. Randomization allows causal inference for the comparison between GLP-1RA and lifestyle interventions during the 12-week treatment period. However, the small sample size (n=29), lack of blinding, and short duration limit confidence in long-term or population-level causal claims, especially regarding withdrawal effects.
Major COI
Major conflicts that significantly reduce study credibility
The study was funded by Novo Nordisk, the manufacturer of liraglutide, and the drug was provided by them, raising concerns about potential influence on study design and interpretation, despite no explicit funder control over analysis or publication.
Funders
Conflict Details
Novo Nordisk: Provided liraglutide for the trial and sponsored the study.
The study was sponsored and monitored by the University of Oxford, but Novo Nordisk supplied the investigational drug. No independent data monitoring committee or third-party analysis is mentioned. The trial was registered as an Investigational Medicinal Product trial, indicating industry sponsorship. While authors are not stated to be employed by Novo Nordisk, the close alignment of study design with commercial interests (testing liraglutide vs. lifestyle) and lack of transparency on funder input beyond funding raises significant bias risk.
Standing
Who’s using this study?
The videos and claims on this site that lean on this study, and the researchers who wrote it.
1 video from Physionic cite this study, drawing 1 claim from it.
- Strong evidence
At least some randomized or controlled trials support this claim.
Evidence
Authored by
15 researchersIf this is your work, this is how we attribute it on Fit Body Science. Ahmad Moolla is listed as the lead author.