FGF21: The Metabolic Hormone from Your Liver
Liver-Derived Peptide Hormones
3.3 year lifespan
Mice overexpressing FGF21 on a high-fat diet lived up to 3.3 years, resisted weight gain, and showed improved insulin sensitivity and reduced liver fat.
Zhang et al., Cell Metabolism, 2025
Zhang et al., Cell Metabolism, 2025
View as imageMost hormones that regulate metabolism originate from the pancreas, the gut, or the brain. FGF21 comes from the liver. This 181-amino-acid protein, classified as an atypical fibroblast growth factor because it lacks the heparin-binding domain that anchors other FGFs to local tissues, circulates freely in the bloodstream and acts on distant organs including adipose tissue, the brain, and the pancreas. Its effects read like a wish list for metabolic health: increased fatty acid oxidation, improved insulin sensitivity, reduced triglycerides, decreased hepatic fat, and in mouse models, extended lifespan.
The pharmaceutical industry has noticed. At least three FGF21 analogs are in clinical trials: pegozafermin (89bio), efruxifermin (Akero Therapeutics), and several dual or triple agonists that combine FGF21 activity with GLP-1 or glucagon receptor activation. The initial clinical results, particularly in metabolic dysfunction-associated steatohepatitis (MASH, formerly NASH), have been among the most promising in hepatology in a decade. For a broader look at the signaling molecules your liver produces, see our article on hepatokines.
Key Takeaways
- FGF21-overexpressing mice on a high-fat diet lived up to 3.3 years, with reduced liver steatosis, lower inflammatory immune cells, and decreased lipotoxic ceramides in adipose tissue (Zhang et al., Cell Metabolism, 2025)
- A GLP-1/FGF21 dual agonist (HEC88473) reduced hepatic fat content by up to 50% and improved lipid profiles in patients with MASLD and type 2 diabetes in a randomized placebo-controlled trial (Xiang et al., 2025)
- Efruxifermin combined with a GLP-1 receptor agonist achieved MASH resolution in 50% of patients and fibrosis improvement in 47% over 24 weeks (Harrison et al., 2025)
- FGF21 requires beta-klotho as a co-receptor, limiting its direct action to tissues expressing this protein: adipose tissue, liver, brain, and pancreas
- Circulating FGF21 levels rise paradoxically in obesity and type 2 diabetes, suggesting FGF21 resistance analogous to insulin resistance
- The combination of FGF21 analogs with GLP-1 receptor agonists shows additive effects beyond either drug alone in preclinical and early clinical testing (Norlin et al., 2025)
What FGF21 Is and Where It Comes From
FGF21 belongs to the FGF19 subfamily of fibroblast growth factors, a group that includes FGF19 and FGF23. Unlike classical FGFs, which act locally as paracrine growth factors, FGF19 subfamily members function as endocrine hormones, traveling through the bloodstream to act on distant tissues. What makes them endocrine rather than paracrine is the absence of a heparin-binding domain: they do not stick to the extracellular matrix at their site of production and instead enter systemic circulation.
The liver is the primary source of circulating FGF21, though smaller amounts are produced by adipose tissue, skeletal muscle, and the pancreas. Hepatic FGF21 expression is regulated by peroxisome proliferator-activated receptor alpha (PPARa), a transcription factor activated during fasting, ketogenesis, and high-fat feeding. This means FGF21 levels rise in metabolic states where the liver is processing large quantities of fatty acids, serving as a signal that the liver's fat-handling capacity is being stressed.
FGF21 signals through FGF receptor 1c (FGFR1c) in complex with beta-klotho (KLB), a single-pass transmembrane protein that functions as a co-receptor. Beta-klotho expression is restricted to specific tissues: white and brown adipose tissue, the liver, the pancreas, and specific brain regions including the hypothalamus and hindbrain. This co-receptor requirement provides tissue selectivity: even though FGFR1c is widely expressed, FGF21 can only activate signaling in tissues that co-express beta-klotho.
Lee et al. (2025) reviewed the molecular pathways connecting liver and adipose tissue in MASLD, highlighting FGF21's role as a key mediator of hepatic-adipose crosstalk.[1] When the liver produces excess FGF21 in response to fat overload, the hormone acts on adipose tissue to increase glucose uptake, enhance fatty acid oxidation, and promote the browning of white adipose tissue, which increases energy expenditure.
The brain effects of FGF21 are increasingly recognized. FGF21 acts on the hypothalamus and hindbrain through beta-klotho-expressing neurons to suppress sugar preference, reduce alcohol intake, and modulate macronutrient selection. In mice, FGF21 signaling in the brain reduces the preference for sweet tastes and suppresses alcohol consumption, effects that are independent of its peripheral metabolic actions. This central nervous system activity positions FGF21 as more than a metabolic hormone; it is a behavioral regulator that links liver metabolic state to dietary choice.
FGF21's effects on the pancreas are less well characterized but include protection of beta cells from lipotoxicity and glucolipotoxicity-induced apoptosis. In animal models, FGF21 preserves beta cell mass and function under metabolic stress, suggesting a protective role against the progressive beta cell failure that characterizes type 2 diabetes.
The FGF21 Paradox: High Levels in Metabolic Disease
One of the most confusing aspects of FGF21 biology is that circulating levels are elevated, not reduced, in obesity, type 2 diabetes, and MASLD. If FGF21 improves metabolic health, why do people with metabolic disease have more of it?
The current explanation parallels insulin resistance. Just as type 2 diabetes involves high insulin levels that fail to lower blood glucose (because target tissues have become resistant to insulin signaling), metabolic disease appears to involve FGF21 resistance. The mechanisms of this resistance are not fully resolved but include downregulation of beta-klotho expression in adipose tissue, impaired FGFR1c signaling, and increased FGF21 degradation by the protease fibroblast activation protein (FAP).
This resistance model explains why pharmacological FGF21 analogs can work even when endogenous levels are already high. The analogs are engineered for longer half-lives (endogenous FGF21 has a half-life of only 1-2 hours), resistance to FAP-mediated degradation, and enhanced receptor binding. They overwhelm the resistance by delivering supraphysiological receptor activation, similar to how insulin therapy works in advanced type 2 diabetes despite high endogenous insulin levels.
Srivastava et al. (2018) identified FGF21 analogs as one of the most promising classes of anti-obesity drugs on the horizon, noting their ability to improve multiple metabolic parameters simultaneously rather than targeting a single pathway.[2]
Clinical Trials: FGF21 Analogs in MASH
The clinical development of FGF21 analogs has advanced rapidly, driven primarily by the unmet need in MASH/NASH treatment. No approved therapy existed for NASH until resmetirom (Rezdiffra) in 2024, and the field remains wide open. The most consistent clinical effect of FGF21 analogs across trials is reduction of liver fat and improvement of dyslipidemia, particularly triglyceride lowering. The weight loss effects have been more variable and generally more modest than GLP-1 receptor agonists.
Efruxifermin + GLP-1 Receptor Agonist
Harrison et al. (2025) reported results from a trial combining efruxifermin (an Fc-FGF21 fusion protein) with a GLP-1 receptor agonist in patients with NASH/MASH and liver fibrosis (F1-F3).[3] After 24 weeks, 50% of patients achieved MASH resolution (the primary endpoint in regulatory trials), and 47% showed fibrosis improvement by at least one stage. These rates substantially exceeded what either drug class achieves alone.
The combination works because efruxifermin and GLP-1 agonists operate through complementary mechanisms. GLP-1 agonists reduce appetite, slow gastric emptying, and improve insulin secretion. FGF21 acts directly on hepatocytes to reduce de novo lipogenesis and on adipose tissue to reduce free fatty acid flux to the liver. Together, they address both the caloric input and the hepatic fat-handling capacity.
Newsome et al. (2025) reviewed the therapeutic landscape in MASH, placing FGF21 analogs among the most promising approaches alongside thyroid hormone receptor beta agonists and GLP-1-based therapies.[4]
GLP-1/FGF21 Dual Agonists
Rather than combining two separate drugs, several companies are developing single molecules that activate both GLP-1 and FGF21 pathways. Xiang et al. (2025) reported results from a randomized, double-blind, placebo-controlled trial of HEC88473, a GLP-1/FGF21 dual agonist, in patients with MASLD and type 2 diabetes.[5] The drug reduced hepatic fat content by up to 50% and markedly improved lipid profiles, with a favorable safety profile. Zhang et al. (2025) reported the first-in-human pharmacokinetic and pharmacodynamic data for another FGF21 analog (HEC96719), demonstrating dose-dependent metabolic effects with acceptable tolerability.[6]
Norlin et al. (2025) demonstrated in a mouse model that the combination of zalfermin (an FGF21 analog) and semaglutide (a GLP-1 agonist) produced additive therapeutic effects on body weight, liver fat, and metabolic parameters, exceeding what either agent achieved alone.[7] This preclinical validation supports the clinical strategy of combining FGF21 with GLP-1 receptor agonists. The additive effects are not simply dose-stacking: the two agents target different nodes in metabolic regulation. GLP-1 agonists primarily reduce caloric intake and improve postprandial glucose handling. FGF21 primarily increases energy expenditure, reduces hepatic lipogenesis, and improves lipid clearance. The combination addresses both sides of the energy balance equation simultaneously.
The pace of drug development in this space is extraordinary. Between 2023 and 2026, at least five different FGF21 analogs or FGF21-containing combination molecules entered clinical testing. The competitive landscape reflects pharmaceutical industry conviction that FGF21 biology is therapeutically actionable, though the commercial viability will depend on demonstrating benefits beyond what GLP-1 agonists achieve alone.
The Liver Disease Pipeline
Newsome et al. (2023) reviewed how incretins, including GLP-1 receptor agonists, dual and triple agonists, interact with FGF21 signaling in the liver.[8] FGF21 analogs exert direct effects on hepatocytes that GLP-1 agonists do not: they reduce hepatic de novo lipogenesis, increase fatty acid beta-oxidation, and suppress the inflammatory pathways (including NLRP3 inflammasome activation) that drive progression from simple steatosis to steatohepatitis and fibrosis.
Tacke et al. (2023) placed FGF21 within the broader landscape of anti-inflammatory and antifibrotic targets for NASH, noting that FGF21 is one of the few targets that addresses both inflammation and fibrosis simultaneously.[9]
FGF21 Beyond Liver Disease
Obesity and Weight Management
FGF21 analogs produce modest weight loss in clinical trials, typically 2-5% of body weight over 12-24 weeks. This is less than GLP-1 agonists (which can achieve 15-20% weight loss), but the metabolic improvements with FGF21 are disproportionately large relative to the weight change. Triglycerides drop 40-60%, and insulin sensitivity improves substantially even before significant weight loss occurs.[2]
Enyew et al. (2024) reviewed innovative glucagon-based therapies for obesity, noting that FGF21's effects on energy expenditure and lipid metabolism complement the appetite-suppressing effects of GLP-1 agonists.[10] The emerging paradigm is not FGF21 as a standalone obesity drug, but as a component of combination therapies that address multiple aspects of metabolic dysfunction.
Beji et al. (2026) argued that GLP-1 alone is insufficient for optimal metabolic outcomes and that glucagon receptor co-activation (which stimulates hepatic FGF21 production) may fill the energy expenditure gap that GLP-1 agonists leave open.[11]
Alcohol and Substance Use
FGF21 has an unexpected connection to alcohol metabolism. The liver produces FGF21 in response to alcohol exposure, and FGF21 acts on the brain to suppress alcohol preference in animal models. Sailer et al. (2026) reported that changes in FGF21 levels are associated with alcohol consumption patterns and smoking cessation in humans, suggesting FGF21 may play a role in the behavioral response to these substances.[12]
Zhang et al. (2025) reviewed pharmacotherapies to prevent alcohol relapse in alcohol-associated liver disease, identifying FGF21 agonists as a potential dual-purpose therapy: treating both the liver damage from alcohol and the behavioral drive to continue drinking.[13] This dual hepatoprotective and behavioral effect is unique among liver disease therapies. If validated in human clinical trials, FGF21 agonists could address one of the central challenges in treating alcohol-associated liver disease: patients who cannot stop drinking continue to damage their livers regardless of how effectively the liver disease itself is treated. A drug that simultaneously reduces liver inflammation and decreases the drive to consume alcohol would change the treatment paradigm.
Longevity
The longevity data comes from transgenic mouse studies. FGF21-overexpressing mice fed a high-fat diet lived up to 3.3 years (compared to a typical lifespan of 2-2.5 years for wild-type mice on the same diet). These mice resisted weight gain, maintained insulin sensitivity, showed reduced liver steatosis, and had lower levels of inflammatory immune cells and lipotoxic ceramides in visceral adipose tissue. The longevity benefit was mediated through metabolic improvements rather than growth suppression, distinguishing FGF21 from other longevity interventions like caloric restriction that extend lifespan partly by reducing growth hormone/IGF-1 signaling. For a related perspective on growth factor signaling, see our article on IGF-1 and the liver.
Whether FGF21 extends human lifespan is unknown and may remain so for decades. But the mechanism, improving metabolic health without the trade-offs of growth suppression, is biologically plausible for human relevance.
FGF21 and the Incretin Revolution
The relationship between FGF21 and GLP-1 is bidirectional. GLP-1 receptor agonists stimulate hepatic FGF21 production. Deshmukh et al. (2024) showed that sotagliflozin (an SGLT inhibitor) modulated dyslipidemia in mice partly through effects on FGF21 signaling, demonstrating that existing metabolic drugs already engage FGF21 pathways indirectly.[14]
Guo et al. (2025) found that pre-pregnancy GLP-1 receptor agonist use improved maternal lipid metabolism during pregnancy specifically through liver-secreted FGF21, providing evidence that GLP-1 drugs exert some of their metabolic benefits via FGF21 as a downstream mediator.[15]
This has practical implications for drug development. If GLP-1 agonists already partially work through FGF21 induction, then adding an exogenous FGF21 analog should amplify effects that GLP-1 therapy initiates but cannot maximize. The clinical data from the efruxifermin + GLP-1 agonist trial supports this logic. It also raises the question of whether some of the metabolic benefits attributed to GLP-1 agonists in clinical practice, particularly the improvements in liver fat and triglycerides that occur rapidly before significant weight loss, are partly mediated by FGF21 induction rather than direct GLP-1 receptor activation. If so, patients who are FGF21-resistant (due to low beta-klotho expression or high FAP activity) may respond less well to GLP-1 therapy, and FGF21 analog co-administration could rescue this subpopulation.
What Remains Unknown
FGF21's clinical promise is real, but several gaps persist. The long-term safety profile of FGF21 analogs is not established. Concerns include potential effects on bone metabolism (FGF21 has been linked to bone loss in some animal models through suppression of osteoblast activity and enhancement of osteoclast differentiation), cardiovascular safety (though early data is reassuring, with some trials showing improved cardiovascular risk markers), and the theoretical risk that chronic FGF21 receptor activation could desensitize the system over time, similar to how chronic GLP-1 agonist use can lead to receptor desensitization in some tissues. The bone concern is particularly relevant given that many target patients (obese, diabetic) already have altered bone metabolism.
The optimal patient population is not defined. FGF21 analogs show the strongest effects in patients with MASH and metabolic syndrome, but whether they benefit patients with isolated obesity, type 2 diabetes without liver disease, or other metabolic conditions is unclear. The severe hypertriglyceridemia population is another potential target, as FGF21 analogs produce dramatic triglyceride reductions (40-60%) that could prevent acute pancreatitis in patients with triglyceride levels above 500 mg/dL.
The mechanism of FGF21 resistance is incompletely understood. If FAP-mediated degradation is a primary driver, then combining FGF21 analogs with FAP inhibitors could enhance efficacy. If beta-klotho downregulation is the bottleneck, then strategies to upregulate beta-klotho expression could restore endogenous FGF21 sensitivity without requiring pharmacological analogs.
The interaction between FGF21 and the immune system is an active area of investigation. FGF21 has been shown to suppress NLRP3 inflammasome activation in the liver, reduce pro-inflammatory macrophage polarization in adipose tissue, and decrease circulating inflammatory markers. Whether these anti-inflammatory effects are primary actions of FGF21 or secondary to its metabolic improvements is debated. The answer matters for clinical application: if FGF21's anti-inflammatory effects are independent of its metabolic actions, they would represent an additional therapeutic dimension.
The delivery challenge remains. FGF21 analogs currently require subcutaneous injection, typically weekly or every two weeks. Rao et al. (2025) demonstrated that neutrophil-membrane-coated nanoparticles could enhance FGF21 delivery and therapeutic efficacy for myocardial infarction in a preclinical model, suggesting advanced delivery systems could improve tissue targeting and reduce dosing frequency.[16]
The Bottom Line
FGF21 is a liver-derived hormone that improves fat metabolism, insulin sensitivity, and liver health through direct action on adipose tissue, brain, and pancreas via the beta-klotho co-receptor. FGF21 analogs like efruxifermin and pegozafermin have shown strong results in MASH clinical trials, particularly when combined with GLP-1 receptor agonists. The combination achieves MASH resolution rates of 50% and fibrosis improvement of 47% over 24 weeks. FGF21's effects extend beyond the liver to obesity, alcohol preference, and possibly longevity, though the clinical evidence outside liver disease is early-stage.