GDF-15

What is GDF-15?

GDF-15 is a divergent member of the TGF-β superfamily, synthesized as a 308-amino-acid precursor and secreted as a disulfide-linked dimer of 112-amino-acid mature subunits. It was identified independently by four groups in the late 1990s under four different names — MIC-1 (macrophage inhibitory cytokine-1), NAG-1 (NSAID-activated gene), PLAB (placental bone morphogenetic protein), and PTGFB (placental TGF-β) — reflecting the diversity of contexts in which it appears. Expression is low under basal conditions in most tissues but strongly induced by virtually every form of cellular stress: hypoxia, oxidative stress, endoplasmic reticulum stress, mitochondrial dysfunction, tissue injury, inflammation, and pregnancy (the placenta is the dominant source during gestation). Circulating GDF-15 rises with age, obesity, type 2 diabetes, cardiovascular disease, chronic kidney disease, and — most dramatically — with advanced cancer and mitochondrial myopathies. The receptor, GFRAL (GDNF family receptor alpha-like), was discovered only in 2017 and is expressed almost exclusively in the area postrema and nucleus of the solitary tract in the hindbrain, where it pairs with the RET co-receptor to trigger the 'emergency feeding circuit' that produces nausea, aversion, and anorexia. This restricted CNS receptor distribution is why GDF-15 — unlike most hormones — acts almost entirely through the brain to suppress food intake, and why GDF-15 biology sits at the intersection of three high-profile clinical problems: metformin's weight-loss effect, hyperemesis gravidarum, and cancer cachexia.

What GDF-15 Is Investigated For

GDF-15 is the rare endogenous peptide whose therapeutic value is in being blocked, not administered. Three high-search-volume stories converge on it. First, the 2024 Fejzo et al. Nature paper (a genetics-plus-biomarker study across multiple cohorts) established fetal GDF-15 production combined with low pre-pregnancy maternal sensitization as the primary cause of hyperemesis gravidarum — the first mechanistic explanation for a condition affecting roughly 1–3% of pregnancies and historically blamed on psychosomatic factors. Second, Coll et al.'s 2020 Nature paper showed that metformin raises circulating GDF-15 and that metformin's weight-loss effect is lost in GDF-15- or GFRAL-knockout mice — the molecular explanation for a decades-old clinical observation. Third, Pfizer's ponsegromab (PF-06946860), a GDF-15-blocking monoclonal antibody, produced a dose-dependent 1.2–2.8 kg weight gain over 12 weeks in a positive Phase 2 trial in cancer cachexia (Groarke et al. NEJM 2024) — currently the most promising cachexia program in late-stage development. GDF-15 agonism for obesity, pursued by multiple companies in the late 2010s, has largely been abandoned because the same receptor that drives food-intake suppression also drives nausea and aversion, and the therapeutic window is narrow. There is no legitimate self-administered use of GDF-15 itself. The page is a biology and pipeline reference, not a dosing guide.

History & Discovery

GDF-15 was identified independently by four research groups in 1997–1998, each arriving at the same protein from a different biological entry point — a convergence that explains the unusual four-name legacy still visible in the literature. Bootcov et al. at St. Vincent's Hospital in Sydney cloned it from a macrophage subtraction library and named it MIC-1 (macrophage inhibitory cytokine-1), noting its TGF-β superfamily divergence and anti-inflammatory activity in macrophages (1997 PNAS). Independently, Baek and Eling identified it from a colorectal cancer cell line upregulated by NSAIDs and named it NAG-1 (NSAID-activated gene-1). Hromas et al. cloned it from placenta as PLAB, and Lawton et al. as PTGFB. Within a few years the community consolidated on the GDF-15 designation (reflecting its TGF-β / GDF subfamily membership), though MIC-1 persists in the cachexia-oncology literature. The next mechanistic breakthrough was connecting GDF-15 elevation to tumor-induced anorexia and cachexia. Johnen et al. (2007 Nature Medicine) showed that tumor-bearing mice with elevated MIC-1/GDF-15 levels developed the classic cachexia syndrome — reduced food intake, fat and muscle loss, weight loss — and that these effects were reversed by GDF-15-blocking antibodies. This was the paper that established GDF-15 as a causal driver of cancer cachexia rather than just a biomarker. The receptor remained elusive for two decades. Then, in a remarkable convergence in 2017, four independent groups simultaneously identified GFRAL (GDNF family receptor alpha-like) as the long-sought GDF-15 receptor. Mullican et al. (NGM Biopharmaceuticals), Yang et al. (Janssen), Emmerson et al. (Eli Lilly), and Hsu et al. (Novo Nordisk/academic collaborators) all published in Nature Medicine or Nature within weeks of each other, each showing that GFRAL is expressed only in the area postrema and nucleus of the solitary tract, that it requires RET as a co-receptor, and that its genetic deletion abrogates GDF-15's anorectic effects. The simultaneous independent identification of the same receptor by four competitors in the same year is among the cleaner validation stories in modern metabolic biology. It also immediately reframed the therapeutic opportunity — a restricted hindbrain receptor meant GDF-15 agonism might avoid many peripheral side effects that had limited other obesity drugs. The late 2010s saw vigorous investment in GDF-15 agonism for obesity. NGM Biopharmaceuticals, Eli Lilly, Johnson & Johnson/Janssen, and Novo Nordisk all advanced long-acting engineered GDF-15 agonists into early clinical development. But the therapeutic window narrowed quickly — the same GFRAL circuit that suppressed food intake also drove nausea and conditioned aversion, and doses producing meaningful weight loss produced unacceptable malaise. Parallel success with GLP-1 and GLP-1/GIP dual agonists further deprioritized the GDF-15 agonist approach. By the early 2020s most GDF-15 agonist programs had been paused, shelved, or wound down. The field pivoted. Pfizer's ponsegromab (PF-06946860), a GDF-15-blocking monoclonal antibody, entered cachexia trials and reported positive Phase 2 results in 2024 (Groarke et al. NEJM), with dose-dependent weight gain and symptom improvement. And the 2024 Fejzo et al. Nature paper, combining GWAS with mass spectrometry, established fetal GDF-15 production as the primary cause of nausea and vomiting of pregnancy — the first mechanistic explanation for hyperemesis gravidarum, a condition historically blamed on psychogenic factors. These two 2024 readouts — cachexia blockade working and pregnancy causality established — repositioned GDF-15 from a failed obesity target to the biology underlying multiple distinct clinical programs.

How It Works

GDF-15 is the hormone your body releases when cells are under stress — injury, inflammation, mitochondrial dysfunction, oxidative damage, and (importantly) pregnancy. It travels to a small area of the brainstem called the area postrema, where a dedicated receptor (GFRAL) tells you: stop eating, feel nauseous, avoid this. That single mechanism explains three separate things: why cancer patients stop eating and waste away, why pregnant women get severe morning sickness (fetal tissue makes a lot of GDF-15), and why metformin causes mild nausea and weight loss. Drugs that block GDF-15 are in late-stage trials for cancer cachexia. Drugs that activate GDF-15 were tried for obesity but had to be dropped because they caused too much nausea.

GDF-15 is synthesized as a 308-amino-acid pre-pro-protein with a signal peptide, an N-terminal propeptide domain, and the 112-amino-acid mature C-terminal domain. It is secreted as a disulfide-linked homodimer of mature subunits — the bioactive form. A significant fraction also circulates as latent pro-GDF-15 or as pro-GDF-15 noncovalently associated with extracellular matrix components, providing a tissue reservoir that can be released by proteolytic activation. The mature dimer binds GFRAL with high affinity; GFRAL in turn recruits the RET tyrosine kinase as an obligate co-receptor, producing downstream ERK, AKT, and PLC-γ signaling in a small population of hindbrain neurons restricted to the area postrema and nucleus of the solitary tract. These GFRAL-positive neurons project to the parabrachial nucleus, central amygdala, and other components of an 'emergency feeding circuit' that integrates nutrient-sensing, visceral malaise, and aversion learning — the same circuitry that mediates food aversion to toxins and chemotherapy-induced nausea. This GFRAL-restricted CNS receptor distribution is biologically unusual. Peripheral tissues express little or no GFRAL, which means circulating GDF-15 acts almost entirely through the brainstem to modulate food intake and energy balance — there are no meaningful direct peripheral effects at physiologic concentrations. This is mechanistically distinct from FGF21 (which acts through β-Klotho/FGFR1c on adipose, liver, and CNS), leptin (hypothalamic ARC), or GLP-1 (pancreas, GI tract, and CNS). The 'single-target, single-site' architecture of GDF-15 is what makes it so amenable to pharmacologic blockade — ponsegromab binds circulating GDF-15 and prevents GFRAL engagement with high specificity. Endogenous induction of GDF-15 is driven by multiple stress-response pathways. The integrated stress response (ISR) via the ATF4-CHOP axis is the best-characterized inducer: mitochondrial dysfunction, ER stress, amino acid imbalance, and oxidative stress all converge on ATF4-driven GDF-15 transcription. Hypoxia induces GDF-15 via HIF. Inflammation induces it via p53 and NF-κB. Metformin induces it via AMPK-independent gut and kidney transcriptional activation. In pregnancy, the placenta (specifically syncytiotrophoblast) is the dominant source, with maternal plasma GDF-15 rising 10–100× baseline. In cancer, tumor tissue produces GDF-15 directly; in mitochondrial disease, impaired oxidative phosphorylation drives sustained ISR activation and chronic GDF-15 elevation.

Evidence Snapshot

Research Gaps & Open Questions

What the current literature has not yet settled about GDF-15:

• 01Whether GDF-15 blockade in cancer cachexia translates to survival benefit, not just weight gain and symptomatic improvement — the Phase 3 ponsegromab program should address this.
• 02Whether GDF-15-blocking agents are safe and effective in hyperemesis gravidarum, particularly given uncertainty about the adaptive roles of fetal-derived GDF-15 during pregnancy.
• 03Why elevated baseline GDF-15 strongly predicts cardiovascular, renal, and all-cause mortality across multiple chronic-disease populations — is this reflecting the underlying disease process, or contributing causally to it?
• 04Whether selective modulation of downstream GFRAL circuits could dissociate anorexia from nausea/aversion — the failure mode of the obesity-agonist programs — and thereby rescue a therapeutic weight-loss approach.
• 05The extent to which GDF-15 mediates other clinically relevant effects of metformin beyond weight loss (e.g., the cancer-preventive signals in observational data) versus effects mediated by AMPK or mitochondrial complex I inhibition.
• 06Whether GDF-15 is a useful therapeutic target in mitochondrial disease (e.g., reducing the chronic anorexia and weight loss of severe mitochondrial myopathies) or whether its elevation in those conditions is purely a biomarker.
• 07The biology of pro-GDF-15 versus mature GDF-15 and whether selective targeting of one form could improve therapeutic specificity.

Forms & Administration

GDF-15 itself is not administered therapeutically. Research-grade recombinant GDF-15 exists for laboratory use. The clinically relevant agents are GDF-15-blocking antibodies: ponsegromab (PF-06946860; Pfizer) is the lead program, administered subcutaneously every 4 weeks in cancer-cachexia trials. Engineered long-acting GDF-15 agonists pursued for obesity (programs at NGM Biopharmaceuticals, Eli Lilly, and others in the late 2010s) reached early clinical testing but have largely been deprioritized because of narrow tolerability windows. GFRAL-antagonist small molecules and peptide antagonists for hyperemesis are in preclinical or very early clinical development. Nothing in the GDF-15 pipeline is available on the compounded-peptide market.

Common Questions

Who GDF-15 Is NOT For

Safety Profile

Legal Status

Regulatory status changes over time. Verify current local rules with a qualified professional.

Myths & Misconceptions