A mouse that loses 18 percent of its body weight without eating less is a headline almost anyone would click. The more useful question comes after the click: what would this compound have to show before the result meant anything for a person with obesity? This article works through that question using the original paper, not the press releases.
Ask PeptideBlog about any peptide
The short version. TOFA (5-tetradecyloxy-2-furoic acid) is a long-known research chemical, and it is not a peptide. In a 2026 Science Advances paper, oral TOFA reduced body weight and fat mass in obese male mice and appeared to raise energy expenditure, with no change in food intake. Every in vivo result comes from male mice, mostly in groups of 4 to 10. No human trial is registered, and no regulator has approved it. On this site’s evidence scale, that is Tier 3 (animal) and Tier 4 (laboratory and mechanistic) evidence, with no human evidence.
The study in plain numbers
Before interpreting anything, here is what was actually done. These details come from the paper’s full text and methods.[1]
| Item | What the paper reports |
|---|---|
| Paper | Lee JY, Zhu C, Boldridge MA and colleagues, senior author Anders M. Näär. Science Advances 12(34):eaed3119, published 21 August 2026. Based at UC Berkeley with collaborators at UC San Diego, Massachusetts General Hospital and the Helmholtz Center Munich. |
| Species and sex | C57BL/6J mice. All in vivo experiments used males. The authors say follow-up in females is needed. |
| Obesity model | Diet-induced obesity on a 60 percent fat diet, started 7 to 12 weeks before treatment depending on the experiment. Separate liver-disease models were also used. |
| Main weight experiment | Four weeks of oral TOFA versus vehicle, 10 mice per group (Fig. 2). |
| Energy expenditure experiment | A separate cohort treated for one week before metabolic-cage measurement, 7 to 8 mice per group (Fig. 3). |
| PPARα knockout experiment | Two weeks of treatment, 4 to 5 mice per group (Fig. 5). |
| Incretin combination experiment | 24 days, 4 to 5 mice per group, with TOFA, semaglutide, tirzepatide or each combination (Fig. 7). A separate rebound study used 5 mice per group. |
| Source limits | This review is based on the full text in PubMed Central. Supplementary figures and tables (referred to below as fig. S and table S) were not reviewed, so claims that depend on them are flagged. |
| Human data | None. |
What TOFA is, and what it is not
TOFA is a small molecule that resembles a long-chain fatty acid. Inside cells it is converted into a form (TOFyl-CoA) that blocks acetyl-CoA carboxylase, usually shortened to ACC. It has been known to biochemists for decades as a tool for blocking fat synthesis. What is new is the claim that it also activates two fuel-sensing receptors called PPARα and PPARδ, and that the combination has striking effects in obese mice.[1]
Two clarifications save confusion. First, TOFA is not a peptide, and it is not related to the injectable incretin drugs it was compared with. Second, “TOFA” is also a common abbreviation for tofacitinib, an unrelated approved immune drug. Searches that mix the two return mismatched results.
How it is proposed to work
The plain-English version
Think of a cell as a fuel depot with two jobs: building fat and burning it. ACC is a gatekeeper enzyme for building fat. Block it and the cell makes less new fat. PPARs work in the opposite direction: they are switches that turn on genes for burning fat. TOFA is described as doing both at once, closing the construction gate while opening the furnace.
The technical version
The authors showed TOFA and TOFyl-CoA inhibit ACC1 and ACC2 in biochemical assays, with the CoA form the more potent. In liver cells, TOFA raised oxygen consumption in a mitochondrial stress test. Liver RNA sequencing in treated mice showed a PPAR-type gene program, and binding experiments suggested TOFA interacts directly with the ligand-binding domains of PPARα and PPARδ.[1] The sequencing data are deposited as GEO series GSE302447.[6]
The PPAR half of the story matters because the authors note that ACC inhibition alone has not reduced weight in earlier work and tends to raise circulating triglycerides. In whole-body PPARα knockout mice (Fig. 5), TOFA still lowered body weight, but its triglyceride-lowering effect was lost (p = 0.0251 in wild-type mice versus p = 0.9995, not significant, in knockouts), and the cholesterol effect was diminished. Liver gene data in the knockouts still showed a PPAR signature, which the authors read as compensation by another PPAR family member (fig. S7). That points to PPARα being needed for the lipid effect while other PPAR activity may carry some of the weight effect. It is a clue about mechanism, not a full account, and the knockout groups had only 4 to 5 mice.[1]
The authors also benchmarked TOFA against firsocostat (a clinical ACC inhibitor) and elafibranor (an approved PPARα/δ agonist), alone and combined (fig. S8). By their account, firsocostat raised body weight and triglycerides, elafibranor cut weight less than TOFA and did little for liver fat, and the pair together did not match TOFA. That supports the idea that TOFA is doing something neither mechanism does alone. It does not tell us which extra property is responsible, and the supplementary panels were not reviewed for this article.
What each finding does and does not show
This table is the core of the article. It separates what was measured from what people infer from it.
| Finding | Supporting experiment | Reasonable interpretation | Limitation |
|---|---|---|---|
| About 18% lower body weight with unchanged food intake | Four weeks of oral TOFA in diet-induced obese male mice, 10 per group (Fig. 2). | Food intake did not explain the weight loss. | One strain, one sex, one diet model. Early dose was reduced partway to avoid exceeding humane weight-loss limits, so the lower dose is the sustained one. |
| About 18% higher energy expenditure | A separate one-week cohort in metabolic cages, 7 to 8 per group, at room temperature and thermoneutrality (Fig. 3). | Treated mice burned more energy per day than expected for their body composition. | A different number from the weight figure, from a different cohort. Short exposure, small groups. |
| Smaller energy-expenditure rise in the cold (about 4%) | Same cohort at 4 degrees C (Fig. 3). | The effect is probably not just heat production in brown fat. | Does not identify which tissue is responsible. |
| No fecal calorie difference | Bomb calorimetry of feces (fig. S4A, not viewed). | The drug was not simply blocking absorption of food energy. | Does not rule out other effects on gut physiology. |
| Fat mass fell, lean mass change not significant | EchoMRI body composition (Figs. 2 and 3). | Weight lost was mostly fat by this measure. | “Not significant” is not “preserved.” See the muscle section. |
| Lower liver fat, triglycerides and cholesterol | Diet-induced obesity and choline-deficient liver-disease models. | Metabolic benefits extended beyond body weight in these models. | Mouse liver disease models only partly resemble human steatohepatitis. |
| PPARα needed for the triglyceride effect | PPARα knockout mice, 4 to 5 per group (Fig. 5). | PPARα activity contributes to lipid effects. | Small groups, global knockout, short treatment. |
| Combination with semaglutide or tirzepatide beat each alone | 24-day study, 4 to 5 per group (Fig. 7). | The mechanisms appear complementary in mice. | Small, short, male only, and not proof of human benefit. Details below. |
Two different “18 percents”
Coverage often merges two numbers that sit in different experiments. The roughly 18 percent drop in body weight came from the four-week study (Fig. 2, 10 mice per group). The roughly 18 percent rise in energy expenditure came from a separate cohort treated for one week before cage measurement (Fig. 3, 7 to 8 mice per group, a different diet duration and dosing schedule). They are not one result about the same animals, so the second cannot be read as a measured explanation of the first. They are two pieces of one hypothesis, measured at different moments.
The baseline is also worth stating. The text reports an average 18% decrease in body weight at the end of the course, and the accessible text does not make clear whether that is measured against vehicle-treated mice or against the animals’ own starting weights. Either way, mouse and human metabolism differ so much that percentages do not transfer.
The muscle-preservation claim
The abstract says TOFA works “without affecting food intake or muscle mass.” The data behind that are more modest than the wording.
- What was measured. EchoMRI estimates fat mass and lean mass. Lean mass is everything that is not fat or bone mineral, including skeletal muscle, organs and water. It is a proxy for muscle, not a measurement of it.
- What the statistics say. The Fig. 2 legend lists three EchoMRI p-values: 1.099E-05, 0.053 (marked not significant) and 6.2E-06. The extracted legend does not label which value belongs to which measure. The text says fat mass fell significantly with no significant change in lean mass, so 0.053 is most plausibly the lean-mass comparison. The figure itself is the place to confirm this. In the separate one-week cohort (Fig. 3), the legend also lists a non-significant value (p = 0.8), and again the labels are not spelled out.
- Not significant is not preserved. A p of 0.053 with 10 mice per group means the study could not tell a small lean-mass change from none. Showing preservation would need a pre-set equivalence margin, a larger sample or direct muscle measures.
- What is missing. The text reviewed here has no muscle weights, fiber cross-sectional area or muscle histology. The methods describe a treadmill run-to-exhaustion test, but its result does not appear in the main text. The supplement may contain more, and it was not reviewed.
- Muscle gene data exist, but they are not a size measure. RNA sequencing of gastrocnemius muscle showed upregulated PPAR signaling and reduced inflammatory pathways (Fig. 4, fig. S6). That describes gene activity, not how much muscle there is or how well it works.
- Comparison to incretins. The paper’s discussion says incretin drugs are associated with muscle loss, and speculates that a combination could help. In the mouse combination study, the text says weight changes were solely fat loss with no significant lean-mass loss, and no comparison is shown in which incretin-only mice lost lean mass in that experiment. So the data do not show TOFA protecting muscle relative to those drugs.
A fair reading: lean mass did not fall significantly in these experiments, which is encouraging. It is a starting hypothesis for muscle sparing, not a demonstrated property.
Human trials of other muscle-preserving drugs are summarized in our review of what trials can and cannot show about body composition.
The energy-expenditure result, carefully read
Energy expenditure was measured by indirect calorimetry in metabolic cages (Oxymax-CLAMS), which converts oxygen consumed and carbon dioxide produced into heat. Mice were studied at room temperature (23 degrees C), thermoneutrality (30 degrees C) and cold (4 degrees C). The relevant data are in Fig. 3, with ANCOVA details in table S1 and fig. S4, which were not reviewed.[1]
- Timing. The cage measurements were taken when body weights between the groups had already reached a significant difference. Because size and composition affect energy use, the authors adjusted with ANCOVA, using lean mass and fat mass as covariates. That is the appropriate approach and a good defense against a trivial explanation. It does not remove the chance that treated animals differed physiologically for other reasons. The text also notes a significant contribution by lean mass in the model, which is a reminder that the adjustment itself depends on accurate body-composition data.
- The temperature pattern. The increase was about 18% at room temperature and thermoneutrality and about 4% in the cold. The Fig. 3 legend lists adjusted p-values of 0.0004, 0.0379 and 0.1462 (not significant), which appear to match the three temperatures, though the extracted legend does not label them. A large rise where heat-making demand is low argues against ordinary brown-fat thermogenesis as the whole story. Body temperature did not differ between groups (p = 0.971).
- Mechanism remains open. The authors point to PPAR-driven genes, FGF21, and higher expression of uncoupling-related genes in liver and muscle. These are gene-expression associations, not demonstrations that any one pathway causes the extra energy use. They also note that UCP1-linked thermogenesis is unlikely to explain it fully.
The result is a real signal in a small cohort. It needs replication in independent laboratories, in females and with longer exposure before anyone calls it robust.
Combination with semaglutide and tirzepatide
Fig. 7 reports TOFA alone, semaglutide alone, tirzepatide alone, and each incretin combined with TOFA for 24 days in obese male mice, 4 to 5 per group. The paper describes the incretin doses as minimal efficacious doses and the TOFA dose as sub-optimal. The single treatments each produced about 10% lower weight than controls, and the combinations produced significantly more.[1]
That is interesting, but the limits matter:
- Small groups, male only, and a short 24-day window.
- The paper calls the result “more than additive.” No formal statistical interaction test appears in the text, so that wording goes beyond a comparison of group means.
- Mice given tirzepatide plus TOFA lost weight fast enough that their doses were halved by day 10, which complicates comparison with the other groups.
- The rebound experiment used different doses. In it, semaglutide was given at 10 nmol/kg/day, twenty times the 0.5 nmol/kg/day used in the combination study, for three weeks, with 5 mice per group and five weeks of follow-up. The authors report that TOFA-treated mice held their end-of-treatment weight longer while semaglutide-treated mice regained quickly. Because the semaglutide dose and schedule differ from the combination study and the two drugs were not matched for weight lost, this does not show TOFA is better at keeping weight off.
- Combined weight loss in mice says nothing about nausea, gastrointestinal effects or drug interactions in people.
Animal combination findings generate hypotheses. They are not established human synergy. For how the approved incretin drugs are positioned, see our GLP-1 and obesity peptide guide.
What TOFA would need to demonstrate
This is the assessment the opening asked for. Each row is a gap between a mouse result and a human treatment, and what would begin to close it.
| Question | Why it matters | What would count as an answer |
|---|---|---|
| Does it replicate independently? | Most striking single-lab results shrink on repeat. | Independent laboratories reproducing the weight, fat mass and energy findings, including in females and other strains. |
| Is it safe in formal toxicology? | The paper itself calls safety at the doses used “incompletely characterized.” Older rat data are reassuring but old. | Regulated studies in two species, with attention to liver, heart, kidney, reproductive and bone effects, and to receptor-class concerns. |
| What exposure does it reach? | A mouse dose does not translate to a human dose, and the authors’ own scaling is an estimate. | Pharmacokinetic data: blood levels, half-life, metabolism, and what drives them, in animals and then humans. |
| Does it work in people? | Mouse metabolism differs from human metabolism in speed and fat handling. | Registered, controlled human trials, starting with safety and tolerability, then efficacy against placebo. |
| Is the weight lost actually fat? | Weight alone can be fat, muscle, or water. | Body composition by DXA or MRI in people, with a prespecified lean-mass endpoint. |
| Does muscle function hold up? | Lean mass is a proxy. Function is what patients feel. | Grip strength, chair-stand and gait measures, plus muscle biopsies or imaging in early studies. |
| Is the weight loss durable and the benefit meaningful? | Short rodent studies do not show long-term maintenance. | Trials of a year or more with clinical endpoints, not just weight. |
| Does adding it to an incretin help in people? | Mouse combination data were small and short. | A controlled trial of the combination against the incretin alone, with tolerability reported. |
As of 7 October 2026, a ClinicalTrials.gov search for TOFA returned no registered trials, and no regulatory approval was found.[5] A search can miss trials registered under other names, so absence of results is informative but not final. Because the compound is an old one, the authors report previously published rat studies with high tolerability, which is a reason for interest and not a substitute for modern toxicology.[1]
Who is involved, and what to read
The paper lists an author affiliated with ReRx Therapeutics, and press coverage describes that company as working to develop TOFA.[2] The paper’s formal competing-interest statement was not visible in the text reviewed, so readers should check it on the journal page. Company involvement does not invalidate results. It is a reason to want independent replication, which is also on the list above.
Frequently asked questions
Why does a non-peptide compound appear on a peptide site?
TOFA is a small fatty-acid-like molecule, not a peptide. It is covered here because it was tested alongside injectable incretin drugs and targets the same obesity problem from a different angle.
Has TOFA been tested in people with obesity?
No registered human trial and no published human data were found as of 7 October 2026.
Does the paper prove TOFA protects muscle?
No. Lean mass did not change significantly in these experiments, but the comparison was close (p = 0.053 in the main study), the groups were small, and direct muscle measurements do not appear in the text reviewed.
How were the semaglutide and tirzepatide doses chosen?
The methods describe them as minimal or similar efficacious doses guided by earlier literature, given daily because the drugs clear faster in mice. The aim was to leave room to see a combination effect, so the experiment says little about TOFA versus the drugs at full strength.
Why does the energy-expenditure finding matter?
Approved incretin drugs mainly reduce appetite. A compound that raised energy use without changing food intake would act on the other side of the energy balance. That is the rationale, and it is still a hypothesis from one small mouse cohort.
Is there a human dose?
No established or reliably sourced dosing information was identified. This article does not give or imply one, and mouse amounts should not be converted into human use.
What is the biggest weakness of the evidence?
Small, short, male-only mouse experiments from one research program, with a company affiliation among the authors and no independent replication yet. Each is fixable, which is why replication tops the list of needs.
Could the PPAR activity cause problems?
Unknown, and exactly what formal toxicology and early human safety trials exist to find out. The authors argue TOFA’s partial activity and lack of PPARγ activity may help, but that is a hypothesis too.
Bottom line
TOFA is an interesting pre-clinical idea: block fat construction, switch on fat burning, and see weight fall without a change in appetite. The mouse data support that as a hypothesis worth testing. They do not show a human treatment, a muscle-protecting drug, or something better than approved options. What would change the picture is independent replication, serious toxicology, measured exposure, and controlled human trials with body composition and functional endpoints. Until then, the honest label is promising mouse study.
References
- Lee JY, Zhu C, Boldridge MA, et al. A multi-functional oral small molecule targeting energy and lipid metabolism to treat obesity and related metabolic disorders. Science Advances. 2026;12(34):eaed3119. DOI | PubMed | PubMed Central. Primary source.
- UC Berkeley News. A promising new weight loss and diabetes treatment helps burn fat while keeping muscle. 21 August 2026. news.berkeley.edu. Institutional press release.
- ScienceAlert. Experimental drug makes obese mice burn fat without losing muscle or eating less. sciencealert.com. Press coverage.
- SciTechDaily. Scientists discover new compound that burns fat without killing your appetite. scitechdaily.com. Press coverage.
- ClinicalTrials.gov. Search for TOFA / 5-(tetradecyloxy)-2-furoic acid, run 7 October 2026. clinicaltrials.gov. Government registry.
- NCBI Gene Expression Omnibus, series GSE302447 (liver RNA sequencing). ncbi.nlm.nih.gov/geo. Data repository.
Educational use only. This article summarizes published research and is not medical advice. TOFA is an unapproved investigational compound. Mouse results do not establish safety or benefit in people, and nothing here is a recommendation to use any substance. Talk with a qualified, licensed clinician about personal health decisions.
Research verified: 7 October 2026. Sources checked: the primary paper’s full text (PubMed Central), press coverage and ClinicalTrials.gov. Supplementary materials were not reviewed.
