Five Drugs. One Molecule. The Quintuple Agonist That Beat Tirzepatide Just Showed Up.
A team in Munich strapped lanifibranor onto a GLP-1/GIP backbone, dosed it 6,898 times lower than the standalone, and outperformed semaglutide and tirzepatide-equivalent in mice. Tirzepatide was Act Two. This is what Act Three looks like.
A team in Munich strapped lanifibranor onto a GLP-1/GIP backbone, dosed it 6,898 times lower than the standalone, and outperformed semaglutide and tirzepatide-equivalent in mice.
The numbers, at a glance
What GLP-1–GIP–Lani Did In Mice
- **6,898×** lower lanifibranor dose vs. the dose required for hepatic effect as a standalone (10 nmol/kg vs 30 mg/kg ≈ 68.98 µmol/kg)
- **2.63×** greater placebo-corrected weight loss than the GLP-1–lanifibranor mono-incretin conjugate at day 14
- Outperformed **semaglutide**, **GLP-1/GIP coagonism** (the tirzepatide-class backbone), **lanifibranor alone**, AND **the cocktail of GLP-1/GIP plus lanifibranor given together**
- **5,411** differentially expressed genes in liver vs. 913 (GLP-1/GIP) and 57 (lanifibranor alone)
- **8,060** differentially expressed genes in white adipose tissue vs. 264 and 108 respectively
- Effects **abolished** in dual-incretin-receptor knockout mice; blunted with single-receptor knockouts or PPARδ blockade
- **No** heart hypertrophy, anemia, fluid retention, or creatinine change at 26-39 days
- **Improved** ejection fraction, stroke volume, and cardiac output vs. vehicle controls
- Did **not** cross an in vitro human blood-brain barrier model, yet activated POMC neurons in the hypothalamus
That is the headline. Now the substance.
* * *
1\. This Is Not a Sudden Reveal. It Is a Fifteen-Year Master Plan.
Most coverage of this paper, including the inevitable wave of takes that will land in the next forty-eight hours, will treat it as a surprise. A new molecule, dropping out of the German metabolic research apparatus, beating the best obesity drugs we have. _Where did this come from?_
It came from a lineage that started in 2013.
Brian Finan, then at the Helmholtz Institute for Diabetes and Obesity Research and the Indiana University laboratory of Richard DiMarchi, published a paper in _Science Translational Medicine_ describing the first unimolecular dual incretin agonist: a single peptide that activated both GLP-1R and GIPR with balanced potency. ( Finan et al., 2013, PMID 24174327.) The molecule was a proof of concept. It outperformed pure GLP-1 monotherapy in obese rodents and primates. The pharmacological argument, that combined incretin co-agonism would unlock weight-loss effect sizes a single receptor could not reach, was made for the first time in that paper.
Eli Lilly was paying attention. Within a decade, that pharmacological argument became tirzepatide, which became Mounjaro, which became Zepbound, which generated billions of dollars in revenue and last year became the first weight-loss drug to outperform semaglutide head-to-head in a Phase 3 trial. SURMOUNT-5 reported a mean weight loss of -20.2% on tirzepatide versus -13.7% on semaglutide over 72 weeks in 751 adults with obesity. ( Aronne et al., 2025, NEJM, PMID 40353578.) That number reset the field's expectations for what a single drug could do.
In 2022, the same intellectual circle, this time led by Carmelo Quarta at the same Helmholtz institute, published in _Nature Metabolism_ the first paper to extend the unimolecular logic into a different direction entirely. Instead of stacking another receptor onto the incretin backbone, they covalently attached a small-molecule nuclear receptor agonist, tesaglitazar, a PPARα/γ dual agonist, to a GLP-1 peptide. The peptide delivered the small molecule into cells expressing the GLP-1 receptor. The result was a conjugate that outperformed either component alone or a free combination of the two in obese mice. ( Quarta et al., 2022, Nature Metabolism, PMID 35995995.) That paper introduced a concept that the metabolism field had borrowed from oncology and given a new name: the Peptide-Drug Conjugate, or PDC.
GLP-1–GIP–Lani is the third major iteration of that lineage. It does two things the Quarta conjugate did not. It uses the GLP-1/GIP dual incretin backbone (the same MAR709 chassis behind tirzepatide) instead of GLP-1 alone, which expands the targeted-cell population to include cells that express GIPR (most importantly, mature adipocytes). And it swaps tesaglitazar for lanifibranor, broadening the nuclear-receptor payload from PPARα/γ to pan-PPAR (α, γ, and δ).
That is the master plan in three sentences. Build a delivery vehicle that hits more cell types. Load it with a payload that hits more transcriptional programs. Get more done with less off-target dose.
If you want to understand why this paper is, in the field's own terms, a big deal, you have to understand that it is the third move in a sequence whose first move shipped a $20 billion drug. (For the broader landscape this paper lands into, see our Ozempic Era Is Ending piece on how the GLP-1 monoculture is fragmenting into specialized successors, and our tirzepatide vs. semaglutide guide for the head-to-head clinical context.)
* * *
2\. What It Actually Is. The Oncology Analogy.
Skip this section if you already know what an antibody-drug conjugate is. Read it carefully if you don't.
For two decades, the most exciting category of cancer drugs has been the antibody-drug conjugate, or ADC. The premise is simple. You have a cytotoxic chemical that kills any cell it touches. You have an antibody that binds, with surgical precision, only to a protein expressed on cancer cells. You chemically link the chemical to the antibody. You inject it into the patient. The antibody finds the tumor. The cancer cell internalizes the whole conjugate. The chemical is released inside the cell. The cell dies. The rest of the body is largely spared.
ADCs are not perfect. Linker stability matters. Off-target uptake happens. Bystander effects exist. But the field has now produced more than a dozen FDA-approved ADCs, and the design philosophy, _use precision targeting to deliver an otherwise-too-toxic payload directly to the right cells_, has reshaped oncology.
GLP-1–GIP–Lani applies the same philosophy to metabolism.
The peptide backbone is a balanced GLP-1R/GIPR co-agonist (the MAR709 design lineage that became tirzepatide). It binds the two incretin receptors with picomolar affinity. The receptors are expressed in pancreatic β-cells, mature adipocytes, certain hypothalamic neurons, hindbrain nuclei, and a few other tissues directly involved in metabolism. When the peptide binds, the receptor is internalized, and the cell takes the entire conjugate inside.
Lanifibranor is the payload. It is a small-molecule pan-PPAR agonist developed by the French biotech Inventiva, currently in Phase 3 trials for metabolic dysfunction-associated steatohepatitis (MASH, formerly NASH). PPARs (peroxisome proliferator-activated receptors) are nuclear receptors. They sit inside cells, bind ligands, and regulate the transcription of large gene programs governing lipid metabolism, glucose handling, and inflammation. Lanifibranor activates all three PPAR isoforms (α, γ, and δ), each of which controls a partially overlapping but distinct set of genes.
The design move is the covalent tether. Lanifibranor is chemically bonded to the peptide. The peptide cannot dock at GLP-1R or GIPR without dragging the lanifibranor with it. Once internalized, the lanifibranor is liberated inside the cell and free to bind its nuclear receptors. The PPAR program activates. The cell's metabolism shifts.
Cells that do not express GLP-1R or GIPR, in principle, do not see the lanifibranor at all. The dose-limiting toxicities of free lanifibranor (weight gain, edema, anemia, fluid retention) are the consequences of saturating PPAR receptors throughout the body, including in tissues where activation is unwanted. If you only release the molecule inside cells that express the right delivery receptors, you potentially escape the side-effect ceiling that has historically capped what nuclear-receptor agonists can do.
That is the theory. Now the data, and the part where the theory has to confront some inconvenient details.
* * *
3\. The 6,898× Number and Why It Is the Whole Story
Lanifibranor's Phase 2b NATIVE trial used oral doses of 800 mg and 1,200 mg per day. Preclinically, the dose required to produce hepatic metabolic effects in mice is roughly 30 mg per kilogram per day, which works out to about 68.98 micromoles per kilogram. ( Francque et al., 2021, NEJM, PMID 34670042.)
GLP-1–GIP–Lani in this paper was dosed at 10 nanomoles per kilogram per day.
Divide. The conjugate delivers lanifibranor at 6,898 times less of the standard preclinical dose. The molar math is real. (Quick translation for general readers: nanomoles and micromoles count molecules. The conjugate delivers about seven thousand times fewer lanifibranor molecules per kilogram of mouse, per day, than the dose required to produce the same nuclear-receptor effects when lanifibranor is given as a standalone drug.)
If the conjugate were merely diluted lanifibranor, this dose would do nothing. The paper's central claim is that it is not merely diluted lanifibranor. It is _targeted_ lanifibranor. Each molecule that arrives in the body has a delivery vehicle that pulls it specifically into cells where the nuclear receptors will produce the metabolic outcomes you want, and away from cells where saturating PPAR activation produces the toxicities you don't.
The reason this matters is that it changes what the molecule _is_. A 6,898-fold dose reduction is not an incremental improvement. It is a different drug. The free-lanifibranor side effects (peripheral edema in 14 of roughly 200 NATIVE participants, mean weight gain of 2.4-2.7 kg, anemia) all arise from systemic PPAR saturation. If the conjugate avoids systemic saturation by orders of magnitude, it has, in principle, escaped the side-effect ceiling that has limited the entire pan-PPAR class for two decades.
Whether it actually escapes that ceiling in humans is, of course, the entire question. The conjugate is preclinical. There has been no first-in-human study. The longest in vivo experiment in this paper ran for 39 days. PPAR drugs have a long history of looking clean in rodents and showing fluid retention, heart failure, or fracture risk in human trials lasting months or years. (Tesaglitazar, the precursor payload in the Quarta 2022 conjugate, failed in human trials for renal toxicity. Muraglitazar failed for cardiovascular events. Rosiglitazone survived but was rebuked for heart-failure signals in the late 2000s.) The PPAR graveyard is real. This conjugate has dodged an early bullet, not won a war.
* * *
4\. The Mouse Data, In Layers
Most papers in this space report that a new molecule beats one comparator. This paper beats four.
**Versus vehicle.** Diet-induced obese mice given GLP-1–GIP–Lani at 10 nmol/kg/day for 12 days lost roughly 25% of body weight compared to vehicle controls. Vehicle mice gained.
**Versus semaglutide.** At the same molar dose and the same duration, GLP-1–GIP–Lani further decreased body weight, food intake, and blood glucose with statistically and clinically meaningful margins. The semaglutide arm produced the kind of weight loss curve that would dominate any other paper. In this paper, it is the floor.
**Versus the GLP-1/GIP dual co-agonist alone.** This is the tirzepatide-equivalent comparator. The MAR709 backbone _without_ the lanifibranor payload produced strong weight loss and glycemic control on its own. The conjugate beat it. By 14 days, placebo-corrected weight loss was 2.63× greater than the GLP-1–lanifibranor (mono-incretin) conjugate, and substantially greater than the GLP-1/GIP backbone alone, with further improvements in glucose tolerance, insulin sensitivity (measured by hyperinsulinaemic-euglycaemic clamp), and suppression of endogenous glucose production.
**Versus the cocktail.** This is the most important comparison in the paper, and the one the authors were clearest about wanting to make. They co-administered free lanifibranor alongside the GLP-1/GIP peptide as separate injections at the same molar doses. Same drugs, same total exposure, just unbonded. The conjugate outperformed the cocktail on weight loss, fat mass loss, food intake, glucose tolerance, insulin sensitivity, and fasting glucose. It did so without affecting lean mass beyond the cocktail.
That last comparison is the proof of the targeted-delivery thesis. If you give the same three drugs at the same total doses but as separate molecules, you do not get the same effect as giving them as a single conjugated molecule. The only mechanistic explanation that fits the data is that intracellular co-localization, the convergence of both incretin signaling and PPAR signaling within the same cells at the same time, produces synergy that systemic polypharmacy cannot match.
> **Genetic Matching Hypotheses, GLP-1–GIP–Lani.** The 23andMe Research Institute paper published in _Nature_ on April 8, 2026 (Su et al., n ≈ 28,000 GLP-1RA users) identified a missense variant in the GLP-1 receptor gene, _rs10305420_, associated with an additional -0.76 kg of weight loss per copy of the effect allele on GLP-1 monotherapy. Variants in _GIPR_ were specifically tied to nausea and vomiting in tirzepatide users. If GLP-1–GIP–Lani depends on both incretin receptors for delivery, both loci will likely modulate response. The patients who do best on this molecule, if it ever reaches the clinic, may be the ones whose GLP-1R and GIPR variants both fall on the responder side. Our peptide pharmacogenomics guide walks through what's known about each variant.
* * *
5\. The Mechanism Paradoxes Nobody Will Talk About
Two things in this paper do not, on first read, make sense. They make sense once you understand the underlying biology, and any honest piece on this molecule has to walk through them.
Paradox One: The Liver Doesn't Have the Receptors
Hepatocytes, the dominant cell population in the liver, do not meaningfully express GLP-1R or GIPR. This is well-established. ( Müller et al., 2019, Mol Metab, PMID 32067739.) If GLP-1–GIP–Lani only releases its lanifibranor payload inside cells that express the incretin receptors, the liver should be invisible to the molecule's PPAR activity.
And yet the liver in treated mice shows 5,411 differentially expressed genes versus vehicle, compared to 913 with the GLP-1/GIP backbone alone and only 57 with free lanifibranor. The hepatic transcriptome is being remodeled at scale. The drug is having profound effects on a tissue it cannot, by its own design, directly target.
There are three possible explanations. Each is at least partly true.
The first is _indirect metabolic spillover_. White adipose tissue (which does express GIPR on mature adipocytes) is being heavily treated. eWAT in this paper showed 8,060 DEGs versus 264 with the dual backbone alone. When adipose tissue rewires its lipid handling, the liver receives different free fatty acid flux, different inflammatory signals, different adipokine output. A drug that profoundly remodels adipose can profoundly affect the liver without ever touching a hepatocyte.
The second is _neural cross-talk_. The hindbrain and hypothalamic nuclei targeted by the conjugate (more on this below) regulate hepatic glucose production via vagal and sympathetic outflow. The paper's data on suppressed endogenous glucose production and reduced expression of _Pcx_ and _Pepck1_ (the master regulators of hepatic gluconeogenesis) is consistent with central nervous system mediation of liver effects.
The third is _linker leakage_. No covalent peptide-drug conjugate is perfectly stable in circulation. Some fraction of lanifibranor is liberated from the peptide before the conjugate reaches its target cells, and that free lanifibranor will bind PPARs anywhere it finds them, including in the liver. If 1% of the dose leaks, that is still vastly less than the standalone Lani dose, but it is not zero. The conjugate is a smart bomb in design and a slightly leaky one in practice.
The honest answer is that all three are happening and the paper does not, and probably cannot, fully decompose them. If you read coverage that calls this molecule a "smart bomb" without addressing the indirect-effects question, you are reading a press release.
Paradox Two: The Brain Without Crossing the Brain
GLP-1–GIP–Lani does not cross the blood-brain barrier. The paper confirms this directly using an in vitro human BBB model derived from CD34+ stem cells co-cultured with brain pericytes. Under those conditions, neither the conjugate nor the free GLP-1/GIP backbone crossed the barrier in measurable amounts.
And yet the conjugate fires POMC (pro-opiomelanocortin) neurons in the arcuate nucleus of the hypothalamus harder than the GLP-1/GIP backbone alone, as measured by both in vivo fiber photometry in _Pomc-cre_ mice and ex vivo whole-cell patch recordings in _Pomc-GFP_ mice. POMC neuron activation suppresses appetite. The behavioral data (further reduced food intake) is consistent with the electrophysiology.
How does a drug that doesn't cross the BBB activate neurons that live inside the BBB?
The arcuate nucleus is not a normal brain region. It sits adjacent to the median eminence, one of the circumventricular organs where the blood-brain barrier is, by anatomical design, leaky. POMC neurons and AgRP neurons in the arcuate extend dendrites that physically project into the median eminence to sample circulating hormones, glucose, and now, evidently, large peptide drug conjugates. ( Müller et al., 2019, Mol Metab, PMID 32067739 reviews this anatomy in the context of GLP-1 action.) This is the same loophole liraglutide and semaglutide use. They are too large to cross the BBB through normal endothelial routes, but they can be sensed by neurons that have evolved specifically to detect things in the blood.
The fact that GLP-1–GIP–Lani activates POMC neurons more than its non-conjugated backbone, despite identical incretin-receptor signaling on cells that lack PPAR machinery, suggests the lanifibranor payload is doing something inside arcuate neurons themselves. That is genuinely interesting. It also raises a question the paper does not fully resolve: which PPAR isoform is doing the central work, and what are the long-term consequences of chronic PPAR activation in hypothalamic appetite circuits?
This is the kind of mechanism question that takes the next three papers to answer, not this one.
* * *
6\. The Knockouts. The Mechanism Test the Paper Aces.
A drug's claim to a mechanism is only as good as the experiments designed to disprove it. This paper runs the right experiments.
If GLP-1–GIP–Lani really works through GLP-1R, GIPR, and PPARδ, then breaking each of those receptors should diminish the drug's effect.
In _Vglut2-Cre; Glp1r_-floxed mice (selectively missing GLP-1R from glutamatergic neurons), the conjugate's weight loss was significantly impaired. In _Gipr_-knockout mice, weight loss was diminished. With pharmacological blockade of PPARδ using GSK3787, the glucose-lowering effect of the conjugate was abolished, although weight loss was preserved (suggesting weight effects route through PPARδ-independent mechanisms while glucose effects depend on it).
The cleanest test is the dual-incretin-receptor knockout, or DIR-KO. In mice lacking both GLP-1R and GIPR throughout the body, GLP-1–GIP–Lani had _no detectable effect_ on body weight, food intake, glucose tolerance, fasting glucose, or insulin. The molecule that out-performed every comparator in wild-type mice became inert when its delivery receptors were removed.
This is the experimental gold standard. The drug works through the receptors it claims to work through, and it stops working when you take those receptors away.
* * *
7\. Cardiac, Inflammatory, and the "No Edema" Reveal (Heavily Qualified)
This is the section where it is easiest to overclaim, so the qualifications come first.
The PPAR drug class has a brutal cardiovascular and renal history in humans. Rosiglitazone (a PPARγ agonist) survived the 2000s but was sharply restricted after Steven Nissen's meta-analysis flagged ischemic events. Muraglitazar (PPARα/γ dual) was rejected after CV signals in late-stage trials. Tesaglitazar failed for renal toxicity. Free lanifibranor in NATIVE caused mean weight gain and peripheral edema in 14 of approximately 200 patients. Pan-PPAR agonism is not a cardiometabolically clean intervention in humans, no matter how clean it looks at 39 days in a mouse.
With that established, here is what the paper reports.
GLP-1–GIP–Lani caused no heart hypertrophy at 14-39 days, no anemia, no fluid retention (haematocrit and total body fluid unchanged), no creatinine change, and no histological alterations in liver, muscle, heart, eWAT, or kidney. It _improved_ ejection fraction, fractional shortening, stroke volume, and cardiac output versus vehicle controls. Hepatic triglycerides decreased. Plasma AST and ALT decreased. Anti-inflammatory gene programs in liver and skeletal muscle were robustly induced.
Read straight, this is a very clean short-term cardiometabolic profile. Read in context, it is a 39-day mouse experiment in a drug class that has failed humans on six- to twenty-four-month timelines. Both readings are correct. The cleanness is real. The translation is unproven. The fluid-retention question is not whether this conjugate causes it in mice (it doesn't); it is whether the systemic exposure floor in humans, after dose escalation, accidental linker leakage, and long-term cumulative effects, stays below the threshold that triggers the historical PPAR liabilities.
The conjugate is a hypothesis that targeted delivery breaks the historical PPAR ceiling. The hypothesis is alive. It is not yet tested in humans.
* * *
8\. Genetics: Who Would Actually Respond
The 23andMe Research Institute published a paper in _Nature_ on April 8, 2026 (Su et al.) reporting the largest genome-wide analysis of GLP-1 receptor agonist response to date, drawn from approximately 28,000 self-reported users of weight-loss drugs in the 23andMe research cohort.
Two findings are immediately relevant to GLP-1–GIP–Lani.
The first is a missense variant in _GLP1R_ (rs10305420) associated with an additional -0.76 kg of weight loss per copy of the effect allele. The variant sits in the receptor itself. People who carry it lose more weight on GLP-1 monotherapy, presumably because their receptor binds the drug differently or signals differently downstream.
The second is a cluster of variants in _GIPR_ tied to gastrointestinal side effects (nausea, vomiting) specifically in tirzepatide users. The GIP receptor variants did not predict efficacy as cleanly as the GLP-1R variant did, but they predicted tolerability.
GLP-1–GIP–Lani is, by design, a delivery system that depends on both receptors. If you carry the favorable _GLP1R_ variant, your incretin receptors bind the conjugate more effectively, more conjugate gets internalized, more lanifibranor reaches the inside of your cells, and (in principle) the drug works better. If you carry an unfavorable _GIPR_ tolerability variant, you may experience more nausea on this drug than on a GLP-1-only conjugate.
This is the broader thesis that the genetics-and-peptides intersection has been building for years. The first generation of weight-loss drugs was prescribed without regard to genotype because the drugs themselves were monolithic. The second generation, the dual and triple agonists, was prescribed without regard to genotype because the regulatory infrastructure had not caught up. The third generation, the targeted-delivery conjugates, is being designed in a world where genotype-stratified response data exists. It would be malpractice not to use it.
This is also, transparently, what we built TPL Genetics for. If a molecule like GLP-1–GIP–Lani ever reaches the clinic, the question of whether it works for _you_ will not be answered by the trial. The trial will tell you the average. Your _GLP1R_, _GIPR_, _PPARG_, and _PPARD_ genotypes will tell you whether the average is your average. (See also our pieces on why most peptides won't work for you and the DNA-driven peptide response landscape.)
Curious which weight-loss drugs your genetics actually support? TPL Genetics reads your _GLP1R_, _GIPR_, _PPARG_, _PPARD_, and _FTO_ variants and tells you which class is most likely to work for you. The kit is $250 and the report is built on the same evidence base our team uses for clinical research.
* * *
9\. What We Don't Know
Mandatory section. There is more here than there is in most papers' worth of "we don't know" lists.
**This is mice.** Not non-human primates, not patients, not even cynomolgus monkeys. Diet-induced obese C57BL/6J mice. Mouse incretin biology overlaps with human biology but is not identical. Mouse GIPR signaling on adipocytes is meaningfully different from human GIPR signaling, and the GIP-translation debate (whether GIPR agonism, GIPR antagonism, or some context-dependent mix produces the desired human metabolic effects) is genuinely unresolved in the field.
**The longest experiment was 39 days.** PPAR drugs have failed in humans on six-month, twelve-month, and twenty-four-month timelines. Cardiovascular events. Renal events. Bone fractures. Fluid retention. None of these would necessarily appear in 39 days of mouse dosing. The clean safety profile is encouraging. It is not predictive.
**There is no first-in-human data.** No Phase 1 study has been initiated, registered on ClinicalTrials.gov, or announced. The closest commercial owner of this lineage is Novo Nordisk (which funded the peptide work and employs key collaborators), but no company has publicly claimed development rights.
**There is no NASH or MASH histology.** The paper reports decreased hepatic triglycerides, decreased AST/ALT, and induction of anti-inflammatory hepatic gene programs. These are encouraging surrogates. They are not biopsy-confirmed reversal of fibrosis or steatohepatitis. Lanifibranor's existing Phase 3 program in MASH was designed around histological endpoints. The conjugate's MASH efficacy in humans is an open question.
**Pancreatic effects are partially characterized.** The paper documents in vitro insulin secretion in isolated mouse islets and acute (7-hour) proteomic effects in pancreas (778 proteins induced by the conjugate vs. 304 by the dual backbone). Chronic in vivo pancreatic histology (β-cell proliferation, islet morphology over weeks of dosing) is less developed than the liver and adipose data.
**Conditioned taste avoidance was higher.** A single bolus of GLP-1–GIP–Lani induced slightly more conditioned taste avoidance in lean wild-type mice than the GLP-1/GIP backbone alone. CTA is a rodent proxy for nausea. The paper notes this and shows that the effect is GLP-1R-dependent (absent in glutamatergic _Glp1r_ knockouts). In translation, this could mean modestly more nausea in humans on this conjugate than on tirzepatide. The history of GIPR co-agonism has been that GIPR activation actually _reduces_ the emetic effects of GLP-1R agonism. Whether the lanifibranor payload offsets that benefit is unknown.
**Manufacturing and immunogenicity.** Covalent peptide-drug conjugates are harder to manufacture, characterize, and stabilize than unconjugated peptides. They are also potentially more immunogenic. Neither is a deal-breaker, but both add development risk and cost.
**Long-term central effects of PPAR activation are unknown.** If lanifibranor is being released inside hypothalamic neurons that express GLP-1R, what does chronic PPAR activation do to those neurons over years? POMC neuron health, plasticity, and aging are non-trivial questions. The paper does not answer them.
* * *
10\. The Two Real Skeptics (and Why They're Both Half-Right)
The shallow skeptics are easy to dispatch. "It's mice." Yes, and the same lab's mouse work on dual incretin agonism in 2013 became the $20 billion drug tirzepatide. Mouse data from this group has a real-world track record. The lineage matters.
The serious skeptics are harder.
**Skeptic One: The PPAR Veteran.** "You're going to tell me a pan-PPAR agonist is safe based on a 39-day mouse study? Have you read the rosiglitazone meta-analysis? Have you looked at what muraglitazar did in PROACTIVE? The PPAR class has been graveyarded by trials that looked exactly like this one and turned out to cause heart failure or fractures or fluid retention or all three after a year of human exposure."
The PPAR veteran is right that this exact pattern has burned the field before. They are also missing a structural change. Every previous PPAR drug was a free, systemically distributed small molecule. GLP-1–GIP–Lani is the first PPAR-class drug to attempt receptor-targeted delivery at a 6,898-fold dose reduction. The PPAR class's failures were largely failures of dose-related, off-target, systemic exposure. Targeted delivery is the first design intervention that addresses the actual mechanism of those failures rather than tweaking the molecule's selectivity. The veteran's caution is appropriate. The veteran's certainty is not.
**Skeptic Two: The Cocktail Pragmatist.** "Why do we need a complex covalent conjugate when you could just give a tirzepatide injection alongside an oral lanifibranor pill? The pharmacology is the same. The supply chain is simpler. The regulatory path is easier. You're solving a problem that doesn't exist."
The cocktail pragmatist is right that the simplest path is always to pile drugs on top of each other. They are wrong about whether the pharmacology is the same. The paper directly addresses this objection by giving the cocktail (free lanifibranor plus the GLP-1/GIP peptide) at matched doses to a parallel arm. The cocktail produced less weight loss, less food intake suppression, less glucose improvement, and less insulin sensitization than the conjugate. _Same drugs, same doses, separate molecules: not the same effect._
The mechanistic interpretation is that intracellular co-localization of incretin signaling and PPAR signaling within the same cells produces synergy that systemic polypharmacy cannot match. If you accept the data, the cocktail is not equivalent. If you don't accept the data, you are arguing with the experiment.
There is a deeper version of this skeptic's argument that the paper does not fully resolve. The conjugate's superiority over the cocktail in mice does not guarantee its superiority over the cocktail in humans, where pharmacokinetics, distribution, clearance, and tissue-specific receptor expression all differ. A serious head-to-head Phase 2 study of the conjugate versus a cocktail of approved tirzepatide plus approved lanifibranor (assuming both reach the market) would settle this. Until then, the proof is mouse-deep.
* * *
11\. The Quintuple Regulatory Nightmare
A PPAR triple agonist on its own is a hard regulatory submission. A GLP-1/GIP dual agonist on its own took years and a full Phase 3 program. A unimolecular drug that activates five receptors at once does not have a clean regulatory path.
The FDA's current framework for combination products generally requires that each component contributes meaningfully to the combined effect, that the safety profile of the combination has been characterized, and that the combination's risk-benefit ratio is superior to its individual components or to standard of care. GLP-1–GIP–Lani will, if it ever enters human trials, have to demonstrate all three. The experimental architecture for proving which of five mechanistic components is doing what is genuinely difficult. Single-receptor and single-isoform knockouts in mice are tractable. Equivalent experiments in humans (or even in larger animals) are not.
The cancer-risk question is also non-trivial. Pan-PPAR activation has been associated with bladder cancer (PPARγ specifically) in long-term rodent and human exposure data for some agents in this class. Whether targeted delivery at orders-of-magnitude lower exposure escapes that risk is unknown and will require multi-year carcinogenicity studies in two species.
None of this means the molecule won't reach the clinic. It means the clinic is a long way off, and the path goes through hurdles the field hasn't built infrastructure for.
The closing reframe of the Ozempic Era piece was that the next decade of obesity medicine would not be a magnitude race; it would be a fragmentation into specialized successors, each optimized for a specific axis of patient response. GLP-1–GIP–Lani is what one corner of that fragmentation looks like. It is the first molecule to argue that the next axis is not _which receptors_ you hit, but _how cleanly_ you can hit them. (See also our retatrutide guide, which covers the magnitude axis, and Brenipatide, the once-monthly drug after retatrutide, which covers the convenience axis.)
The next-generation question is no longer "Can we get more weight loss?" Tirzepatide cleared 20%. Retatrutide cleared 28%. Bariatric surgery clears 30-35%. The pharmacology has caught the surgery. The next question, the one this paper plants the first flag in, is "Can we do it without the metabolic and side-effect baggage that has historically been the price of admission?"
A 39-day mouse study is not the answer. It is the first move in the right direction.
* * *
12\. The Lineage, In One Paragraph
If you want to remember one thing from this article: the same lab in Munich, working with the same chemistry team in Indianapolis, has now produced three foundational moves in unimolecular metabolic polypharmacology over fifteen years. Finan 2013 introduced the dual incretin co-agonist concept that became tirzepatide. Quarta 2022 introduced the peptide-drug conjugate concept that turned an incretin into a delivery vehicle for a nuclear-receptor agonist. Liskiewicz 2026 unifies both, producing a quintuple agonist that depends on dual-incretin delivery to release a pan-PPAR payload exclusively inside cells that express the right receptors. This is not a sudden breakthrough. It is a slow, deliberate, fifteen-year program by people who knew exactly what they were building and have shipped one $20 billion drug already. The next one, if it gets that far, will not be a surprise to anyone who was paying attention.
* * *
Conflicts of interest, in a box
**Funding and conflicts.** This work was funded by the European Research Council (ERC-CoG 101044445), the German Research Foundation (DFG TRR296, TRR152, SFB1123, GRK 2816/1), the German Center for Diabetes Research (DZD), and NIH grant R01DK133140 (to R.J. Seeley). Senior author Timo D. Müller holds Eli Lilly stock and receives Novo Nordisk research funding. Co-author Brian Finan is a current employee of Eli Lilly. Co-authors Jonathan D. Douros, Stephanie A. Mowery, and Patrick J. Knerr are former Novo Nordisk employees and current shareholders/co-founders of Volari Therapeutics. Richard D. DiMarchi is co-inventor on intellectual property owned by Indiana University and licensed to Novo Nordisk. The MAR709 backbone, the Quarta 2022 GLP-1-tesaglitazar conjugate, and now GLP-1-GIP-Lani all share intellectual lineage with peptides developed within or in collaboration with Novo Nordisk's Indianapolis research center. Readers should weigh the comparator choices (the paper compares the conjugate favorably against semaglutide, a Novo product, and against a tirzepatide-equivalent backbone, a Lilly product) in light of these affiliations. None of this invalidates the data. All of it is context the paper itself discloses.
* * *
FAQ
#### When will GLP-1–GIP–Lani be available to patients?
It won't be available for years, possibly more than a decade, if it ever reaches the market at all. The paper reports preclinical (mouse) data only. There has been no first-in-human study, no Phase 1 trial registered, and no announced commercial development partner. A typical timeline from a Nature preclinical paper of this caliber to FDA approval, for drug classes with established regulatory paths, is seven to twelve years. For a quintuple agonist with a novel peptide-drug conjugate design and a PPAR-class history of late-stage human failures, the path is longer and more uncertain.
#### Can I get this drug through a compounding pharmacy?
No. GLP-1–GIP–Lani has never been synthesized at scale outside the academic and industry labs that developed it. There is no legitimate compounding pharmacy producing this molecule. If anyone offers to sell you "quintuple agonist" or "GLP-1-GIP-Lani" online, they are selling you something that almost certainly is not what they claim, and even if it were, no one knows the human pharmacokinetics, dosing, or safety profile. Stick to compounds with established human safety data and verifiable sourcing. Our where to buy directory lists vetted providers for the compounds that are actually in legitimate clinical use.
#### What is lanifibranor and why does it matter?
Lanifibranor is a small-molecule pan-PPAR agonist (activates PPARα, γ, and δ) developed by the French biotech Inventiva. It is currently in Phase 3 trials for metabolic dysfunction-associated steatohepatitis (MASH, formerly NASH). In the Phase 2b NATIVE trial, lanifibranor showed meaningful improvement in liver fibrosis but caused mean weight gain of 2.4-2.7 kg, peripheral edema, and anemia. In GLP-1-GIP-Lani, lanifibranor is delivered at roughly 7,000-fold lower systemic exposure via covalent conjugation to a peptide that targets it specifically into incretin-receptor-expressing cells, in principle escaping the dose-related side effects.
#### Does this kill tirzepatide?
No. Tirzepatide is an FDA-approved drug with established human safety, efficacy, and supply chain, and it remains the most effective approved obesity drug on the market. ( SURMOUNT-5: -20.2% vs. -13.7% on semaglutide.) GLP-1-GIP-Lani is a preclinical research molecule. The right way to think about it is as a possible successor in the fifteen-year lineage that produced tirzepatide, not as a current alternative to it.
#### Why does targeted delivery matter so much?
Because nuclear-receptor drugs have historically been limited by what happens when you saturate them everywhere in the body. Lanifibranor at the dose required to fix the liver causes weight gain and fluid retention because PPARs are everywhere. If you can deliver the same drug only into the cells where you want PPAR activation (the metabolically-active cells expressing GLP-1R or GIPR) and leave the rest of the body's PPARs untouched, you have a chance of breaking the side-effect ceiling that has limited the entire class for two decades. That is the thesis being tested. Whether it survives translation to humans is the open question.
#### What does my genetics tell me about whether this would work for me?
If GLP-1-GIP-Lani ever reaches the clinic, the relevant variants are likely to include rs10305420 in _GLP1R_ (-0.76 kg additional weight loss per allele on GLP-1 monotherapy, per the 23andMe Nature paper from April 8, 2026), variants in _GIPR_ tied to nausea and tolerability, and possibly variants in _PPARG_ and _PPARD_ that affect nuclear receptor signaling. We test all of these in our genetics report and walk through the interpretation in our pharmacogenomics guide.
#### Is this just an academic curiosity, or could it actually become a drug?
It could actually become a drug. The lineage matters. The same lab and chemistry team produced the conceptual basis for tirzepatide thirteen years ago. The Quarta 2022 GLP-1-tesaglitazar conjugate established the peptide-drug-conjugate platform. The infrastructure (Helmholtz Munich, Indiana Biosciences, Novo Nordisk Research Center Indianapolis) has shipped real drugs before. The remaining uncertainty is in safety translation, regulatory path, and whether the molecule's superiority over the cocktail of approved drugs survives in humans.
#### What should I do with this information today?
If you are on a GLP-1, stay on it; the data here changes nothing about the standard of care. If you are considering one, the choice between semaglutide, tirzepatide, and (when available) retatrutide should be made with your physician based on your goals, tolerability, and (increasingly) your genetics. If you are interested in where the field is heading, this paper is one of three or four signposts published in the last six months that point in the same direction: the next decade of metabolic medicine is going to be about precision delivery, multi-mechanism polypharmacology, and genotype-stratified prescribing. Read our Ozempic Era piece for the broader fragmentation thesis. Don't buy quintuple-agonist anything from any vendor. It does not exist as a legitimate product.
* * *
**Educational use only.** This article describes a preclinical mouse study published in _Nature_. GLP-1-GIP-Lani is not approved for human use, has not been tested in human clinical trials, and is not available through any legitimate pharmacy, compounding facility, or research-chemical supplier. Nothing in this article constitutes medical advice or a recommendation to use any product. Discuss any decisions about weight-loss medications with a qualified physician who knows your medical history. The science discussed here is current as of April 29, 2026 and may be superseded by subsequent peer-reviewed work.
DNA Kit
Want to know which peptides match your genetics?
Our $250 DNA kit analyzes 111 SNPs across 15 pharmacogenes plus peptide-response markers. Get a personalized report on which peptides fit your biology.
Free PDF Guide
Want the complete picture?
Download our free 56-page Essential Guide to Peptides. Covers the science, safety, and how to pick a legit provider.
Get the Free Guide
No spam. Unsubscribe anytime. We respect your privacy.
Ready to Find a Provider?
Browse our directory of verified peptide providers.
Related Articles
**The Ozempic Era Is Ending. Here's What Replaces It.** \\
\\
Five molecules. Five different ways of beating Ozempic. And the reason "best peptide" is about to stop being the right question. **Tirzepatide vs Semaglutide: 15.7% vs 9.9% Weight Loss. Which Wins?** \\
\\
Direct efficacy data, side effect profiles, and a practical decision framework **Retatrutide Guide 2026: 28.7% Weight Loss, New Safety Signal** \\
\\
Eli Lilly's triple agonist just posted the best weight loss data in pharmaceutical history. A new safety signal emerged. And your genetics may decide if you respond. **Retatrutide Isn't the Endgame. Brenipatide Is.** \\
\\
Eli Lilly just quietly moved a once-monthly GLP-1/GIP agonist into Phase 3 for alcoholism and bipolar disorder. It is in Phase 2 for schizophrenia, opioid addiction, smoking, and asthma. This is not an obesity drug. It is the first serious bet that incretins are central nervous system medicine. **Semaglutide 2026: Ozempic vs Wegovy vs Rybelsus Guide** \\
\\
The complete breakdown of brands, dosing, and your options in 2026 **GLP-1 Drugs Cut Heart Attacks 20%. Weight Loss Was Just the Start.** \\
\\
Cardiovascular protection, addiction treatment, and neuroprotection data from 400+ ongoing trials **Your GLP-1 Receptor Gene May Explain Why Ozempic Didn't Work (2026)** \\
\\
The science behind genetic variation in peptide response, and why one-size-fits-all dosing is ending
This website is for informational purposes only and does not constitute medical advice. Consult a licensed physician before using any peptides. Provider listings do not constitute endorsements. None of the statements on this site have been evaluated by the FDA.