Discovery 2004
Kolonin, Saha, Pasqualini, Arap. Ablating the vasculature that feeds fat, via a homing peptide fused to a pro-apoptotic payload, reverses obesity in mice.
Adipotide is the compound that every modern selective-fat discussion eventually circles back to — a peptide that killed adipose-tissue blood vessels in mice and monkeys, shed up to a third of body fat in four weeks in primates, and then quietly disappeared after a four-patient human trial that never read out. A decade later it is still sold as a research peptide and there is exactly one publicly named human case of dialysis-grade kidney failure on it. This page compresses the full evidence picture.
Adipotide is best described as an abandoned clinical candidate with a genuinely interesting preclinical story and no usable human readout — not an experimental option currently in play. "Research use only" marketing creates the illusion of continuing development; there isn't any.
Mechanism is orthogonal to both incretin (GLP-1) appetite suppression and adrenergic (yohimbine / aminophylline) lipolysis. Primate data suggested preferential hit on enlarged visceral adipocytes — exactly the fat compartment that drives cardiometabolic risk. That's the genuine steelman.
Bostin Loyd, competitive bodybuilder, publicly attributed his stage-5 chronic kidney failure to adipotide self-use at roughly 5 mg/day — about 5× typical forum doses. He died in 2022 of aortic dissection with documented end-stage renal disease as background. Polypharmacy confound is real; it's still the worst named outcome.
Despite a decade of research-peptide availability, the forum record contains almost no convincing efficacy logs. Loud vendor marketing, quiet user base, one catastrophic outcome. That shape is itself the finding.
This page was AI-assembled from the local research claim files and primarily AI fact-checked against cited sources. It is a synthesis of public evidence, not personal testing and not clinical advice.
Kolonin, Saha, Pasqualini, Arap. Ablating the vasculature that feeds fat, via a homing peptide fused to a pro-apoptotic payload, reverses obesity in mice.
Obese rhesus macaques lost ~11% body weight and ~39% fat in 28 days at 0.43 mg/kg/day SC, with dose-dependent renal tubular injury flagged in the same paper.
NCT01262664 dosed its first patient in 2012, enrolled a total of four patients over six and a half years, terminated in January 2019 "per PI's request," and never posted results.
No Phase 2, no reformulation, no veterinary program. The Arap/Pasqualini lab continued basic-science prohibitin work; the drug itself went nowhere.
The compound is a chimeric peptide in two halves: a homing domain that addresses adipose-tissue capillary endothelium, and a payload that destroys the mitochondria of whatever cell it enters. Killing the vasculature, not the fat cell directly, is the point.
The sequence CKGGRAKDC-GG-D(KLAKLAK)2 joins two independent pieces of peptide biology.
CKGGRAKDC is a short homing peptide discovered by phage display. Its original advertised target was prohibitin (PHB) on the luminal surface of white-adipose-tissue blood vessels. The current mechanistic consensus — following Salameh 2016, Daquinag 2021, and Gao 2022 — is that the real address is a PHB / ANXA2 / CD36 complex regulating bidirectional long-chain fatty-acid transport across adipose capillary endothelium. ANXA2 scaffolds PHB's surface localization; CD36 is the transporter the complex regulates. This revision matters: the peptide is not binding a pristine single-target marker, it is disrupting a multi-protein transport assembly that exists, at lower density, in any tissue with the same machinery.
D(KLAKLAK)2 is a standard pro-apoptotic amphipathic peptide that is benign in extracellular fluid but destroys mitochondrial membranes once inside a cell. Paired with any internalizing ligand it produces targeted apoptosis.
When a vessel endothelial cell in white adipose tissue binds the homing peptide and internalizes the chimera, its mitochondria rupture, the cell dies, the capillary regresses, and the fat lobule downstream of that capillary loses its blood supply and shrinks. Unlike lipolysis-based interventions, which empty existing adipocytes without reducing their count, adipotide reduces the capillary-perfused adipose mass. Whether the fat cells themselves die or simply become non-functional is less important than the fact that the support infrastructure is removed.
The Kim 2010 paper (an independent replication in C57BL/6 mice and Long-Evans rats) attributed roughly 76% of the observed weight loss to reduced food intake rather than direct adipose ablation. This supports the Criscione critique that adipotide may be largely an anorectic with vascular-ablation as a secondary mechanism.
Kim et al. 2010, DiabetesAppetite drugs (GLP-1, amylin) shrink fat by reducing calorie intake. Lipolytic signals (yohimbine, aminophylline, catecholamines) empty existing adipocytes without killing them. Mechanical methods (lipo, cryo, HIFU) remove or apoptose fat cells but cannot reach visceral fat and do not scale well.
Adipotide is the only demonstrated mechanism that could, in principle, reduce visceral fat by destroying its blood supply from the inside — which is why it keeps circulating even after the program was abandoned.
Two landmark papers carry almost all the weight: the 2004 mouse discovery and the 2011 obese-monkey translational study. Both were strongly positive on efficacy. Both also contained the renal signal that eventually ended the clinical program.
The original paper, "Reversal of obesity by targeted ablation of adipose tissue," introduced both the homing peptide and the chimera. Leptin-deficient ob/ob mice and diet-induced-obese C57BL/6 mice on a 4-week SC course lost on the order of 30% body weight with marked white-adipose-tissue involution, while lean wild-type mice on the same protocol showed no weight loss. That lean-vs-obese selectivity is the original grounding of the "vascular-target-is-adipose-selective" claim.
The mouse work established: (a) the homing peptide reaches adipose vasculature selectively, (b) the payload only kills cells after internalization, (c) effect is reversible on withdrawal (fat regrows), and (d) food intake declined in parallel with weight, which the authors attributed to signal loss from shrinking adipose rather than direct appetite suppression.
Reference: Nature Medicine 2004, Kolonin, Saha, Pasqualini, Arap (PMID 15133506).
The strongest independent preclinical paper on adipotide. C57BL/6 mice at 3 mg/kg SC, Long-Evans rats at 1.5 mg/kg SC, 27-day dosing. Weight loss and fat loss were confirmed. Crucially, Kim et al. attributed roughly 76% of the weight loss to reduced food intake, not to direct adipose ablation. POMC expression dropped but third-ventricle CNS injection produced no signal and no hypothalamic cell death — i.e. the anorectic effect was driven by something downstream of adipose signalling rather than by direct CNS action.
This paper is the scientific backbone of the Criscione critique. If roughly three-quarters of adipotide's weight loss is anorectic, the "selective vascular ablation" story is at best incomplete, and most of the proof-of-concept evidence for targeted adipocyte destruction needs to be re-read with that discount applied.
Reference: Diabetes 2010, Kim et al. (PMID 20103704).
This is the paper the forum culture is actually referencing when it quotes "11% body weight / 39% fat in 28 days." Design: a dose-finding cohort of 4 obese rhesus monkeys (testing 0.10 / 0.25 / 0.43 / 0.75 mg/kg) selected 0.43 mg/kg/day SC × 28 days; the efficacy cohort was n=10 treated vs n=5 control at that dose.
Efficacy. Treated monkeys lost 7.4-14.7% body weight versus -3.5 to +3.3% in controls. DEXA total body fat fell 38.7% vs 14.8%. MRI abdominal WAT was 17.5% lower at end of dosing and 27.0% lower during recovery — fat loss continued past the last dose. Insulin AUC fell 36.2%; the insulinogenic index rose 48.5% in treated vs 33.8% in controls (p=0.006). Lean control monkeys at the same 0.43 mg/kg dose lost no weight, consistent with the obesity-selectivity claim.
Renal signal. Dose-dependent proximal tubular degeneration with single-cell necrosis on histology, classified "minimal-to-mild." Serum creatinine rose mildly-to-moderately at doses >0.25 mg/kg; BUN was not concordantly elevated. Urine showed mild-to-marked glucosuria, mild-to-moderate proteinuria, and increased transitional/renal epithelial cells. Urinary markers normalized within 28 days post-dosing — but three of the treated animals retained "minimal tubular degeneration" on post-recovery histology, which is the detail the Criscione critique centered on and which vendors routinely elide when they repeat "reversible in monkeys."
The practical translation from this paper is therefore sharper than the marketing copy suggests. The efficacy was real, the target was real, the therapeutic index was narrow, and histological recovery was not complete in a third of animals. Everything in the human program after 2011 should be interpreted against the fact that this renal signal was known before the first human was dosed.
References: Science Translational Medicine 2011 (Barnhart et al., PMID 22072637); Criscione 2012 commentary (PMID 22539771); Barnhart reply (scitranslmed.3003893).
No chronic-dosing data in any species. No repeat-cycle data answering whether the second round of treatment preserves the effect or loses the vascular target. No long-term rebound architecture — primate fat regrew but distribution after regrowth was not deeply characterized. No head-to-head against incretin mimetics. No dog or cat work beyond press speculation in 2012. The preclinical story is unusually clean for a single high-impact paper and unusually thin as a development package.
One registered trial, four participants over six and a half years, no posted results, no peer-reviewed readout. Everything "human" about adipotide today is self-experimenter data.
The trial's termination reason in the ClinicalTrials.gov record is procedural ("per PI's request") and carries no specific mechanistic statement. Secondary sources routinely assert that the termination was driven by human nephrotoxicity matching the primate finding. That assertion is consistent with the documented preclinical signal and with the stalled single-cohort enrolment, but is not confirmed by any primary MD Anderson or Arrowhead disclosure. The defensible summary is: the program was abandoned for unspecified reasons consistent with a dose-limiting safety finding, not "human nephrotoxicity forced termination."
Arrowhead subsequently pivoted entirely to RNA-interference therapeutics. Its current obesity pipeline (ARO-INHBE, ARO-ALK7) silences adipocyte-regulating genes and is mechanistically unrelated to adipotide. Ablaris Therapeutics appears inactive as a going concern.
References: clinicaltrials.gov/NCT01262664; Arrowhead 2012 first-patient release; Arrowhead current pipeline.
Everything below is low-evidence by construction — forum logs, press coverage, self-report. The reason to take it seriously is that the forum record is the only non-simian human data that exists, and one of its data points is a named dialysis case.
Loyd publicly attributed his 2020 diagnosis of stage-5 chronic kidney disease to adipotide self-use, reporting dosing around 5 mg/day — roughly 5× the protocol anchored on the primate scaling — injected near adipose depots on the kidneys and chest. He described blood in semen as an early warning. He died in 2022 of aortic dissection with documented end-stage renal disease as a background finding in the autopsy.
The polypharmacy confound is significant. Loyd had a multi-decade history of anabolic steroid use, high-dose insulin, SARMs, and diuretics; any of these alone can damage kidneys, and the combination is routinely implicated in bodybuilder renal failure. Loyd's own causal attribution should be weighted accordingly.
What it is: the only publicly named human reaching dialysis after an adipotide cycle. What it is not: controlled evidence that adipotide at any dose causes kidney failure in otherwise-healthy users. The honest causal range is "rapid, large, possibly irreversible kidney injury at supraphysiological self-dosing in a heavily co-medicated adult."
References: Fitness Volt 2020 kidney-failure coverage; Muscle Insider obituary; Fitness Volt autopsy summary.
The most detailed public self-log. Subject 116 kg, dosing 50 mg/day SC at the primate-scaled high end. Reported ~3.5 kg weight loss over roughly two weeks. Documented adverse events: severe injection-site burning and lumps from day 1, cloudy urine from day 2, unquenchable thirst from day 5, a severe hypoglycemic episode on day 10, strong food aversion (especially to fat-rich foods), nausea. The subject discontinued early with the self-assessment that the result was "not worth all the pain and drama."
No labs were reported. The cloudy urine, polydipsia, and the hypoglycemic episode are qualitatively consistent with the primate-study renal tubular pattern plus insulin-sensitivity shift, but are not sufficient evidence for any specific mechanism at human doses.
Reference: Shaman-Australis forum log.
Across MESO-Rx, eroids, iSARMS, MuscleMecca, UK-Muscle, ExcelMale, The Iron Den, AnabolicSteroidForums, and the peptide-focused Reddit communities, the aggregate picture is striking for what it does not contain. One calibration note: adipotide discussion on Reddit is largely unreachable via standard web search due to Reddit's post-2023 indexing changes, so the forum base rates here come mainly from non-Reddit bodybuilder/biohacker boards. This does not appear to change the conclusion — the secondary Reddit aggregators that do surface repeat the same "2-5 mg/day is the dangerous upper band, remember Bostin Loyd" pattern without producing detailed cycle logs.
Despite a decade of research-peptide availability, there is no community of satisfied adipotide users generating convincing before/after logs. Most threads describe either (a) people never actually running it, (b) people aborting early due to injection-site reactions, or (c) repeating the primate-paper headline number as if it were a human result. The density of "I'm thinking about running this" posts is much higher than the density of completed-cycle reports.
This asymmetry — loud vendor marketing, quiet user base, one catastrophic named outcome — is itself the informative finding. Compared to any popular bodybuilding peptide (BPC-157, TB-500, tesamorelin, GHRP-2/6) the signal-to-noise ratio of adipotide user experience is dramatically worse.
Thread samples: MESO-Rx adipotide thread; eroids adipotide discussion.
For calibration only, not as a recommendation. These are the numbers that appear in the underground literature; they anchor on the primate paper, not on any human safety envelope.
| Source | Dose | Route / duration | Notes |
|---|---|---|---|
| Primate-scaled (most commonly cited) | 0.43 mg/kg/day (~30 mg/day for 70 kg) | SC × 28 days | Direct translation of Barnhart 2011. Dominant forum protocol. |
| Conservative ramp | 0.1 → 0.25 → 0.43 mg/kg/day | SC × 4–6 weeks | Attempt to reduce acute renal exposure; no controlled data. |
| Phase 1 human starting dose | ~0.03 mg/kg/day | SC × 28 days | An order of magnitude below primate-scaled; rarely cited in forums. |
| Bostin Loyd (self-reported) | ~5 mg/day total (~0.05 mg/kg for 100 kg) | SC, adipose-proximal injections | Loyd described this as roughly 5× what others ran; reached stage-5 CKD. |
| Shaman-Australis log | 50 mg/day (~0.43 mg/kg for 116 kg) | SC | Close to primate-scaled; aborted early. |
Adipotide is openly sold "for research use only" by at least ten active peptide vendors as of 2026. A secondary layer of affiliate-driven "peptide information" sites translates the primate protocol into human-scaled dosing guides and funnels traffic to the sellers.
Sites that publish "dosing guides" or "complete guides" to adipotide and carry affiliate links to the sellers:
Recurring claims across the commerce layer:
No vendor has, or claims to have, human efficacy data.
This matters because "research use only" creates the impression of continuing development. It is not a fair impression.
The active successor literature is not "adipotide 2.0." It is the homing peptide CKGGRAKDC repurposed as a delivery address, grafted onto nanoparticles carrying non-cytotoxic payloads. The D(KLAKLAK)2 pro-apoptotic head that caused the primate and human-dose-limiting toxicity has been quietly abandoned by the translational field.
Representative work in this thread:
The Arap/Pasqualini lab at Rutgers CINJ no longer lists obesity or adipose targeting among funded programs — current themes are lung cancer, melanoma, prostate cancer, SARS-CoV-2 vaccines, and antibody discovery. The Kolonin group at UTHealth Houston continues basic PHB/ANXA2/CD36 biology but has published no adipotide-successor therapeutic. Ablaris Therapeutics appears dormant. Arrowhead Pharmaceuticals pivoted wholesale to RNA-interference (ARO-INHBE, ARO-ALK7 in obesity).
No companion-animal (dog, cat) adipotide program materialized despite 2012-era press speculation. Companion-animal obesity drug development has moved entirely to incretin-class agents — for example Okava's MEOW-1 GLP-1 implant for cats.
Net: adipotide the drug is dormant. The homing peptide is alive and well as a nanoparticle-delivery vector for unrelated adipose-targeting payloads. Any eventual "next adipotide" is most likely to come from that vector, not from a redesigned D(KLAKLAK)2 chimera.
A compact version of the risk model. Start with what the primate paper showed, layer on what human self-experimenters have added, then read the modern peptide market through that lens.
Dose-dependent proximal renal tubular injury in obese rhesus macaques at efficacious doses. Incompletely reversible on recovery in the Criscione reinterpretation. This was known before any human was dosed.
One publicly named human (Bostin Loyd) reached stage-5 chronic kidney disease on high-dose self-administration and later died. Polypharmacy confound applies but does not eliminate the signal.
No human liver, cardiac, or long-term-endocrine data exists. No chronic-dosing information in any species. No reliable identity or purity standard for "research use" vial contents.
Severe injection-site burning and lumps are the dominant adverse event in existing forum logs. No antidote or rescue pathway once the pro-apoptotic payload is internalized.
Not approved anywhere. Not scheduled. The FDA has issued generic warning letters to peptide sellers marketing unapproved injectables; adipotide is not specifically named in those actions but sits squarely inside the category.
Adipose-vascular-ablation remains the only demonstrated pathway to potentially reduce visceral fat by removing its blood supply from the luminal side. This is a genuine reason the idea keeps returning, even if the specific compound should not.
The practical reading for the main post.
Adipotide should appear in the main writeup as a footnote, not as a candidate option. The mechanism is the interesting part, and it is worth two or three sentences; everything else is either disclaimer or vendor-ecosystem context.
The common framing "adipotide is experimental and awaiting Phase 1 readout" is wrong in two directions. It is not experimental — it is abandoned. There is no Phase 1 readout coming — the trial enrolled four patients over six and a half years, terminated, and never posted results. Any writeup that implies continuing development is repeating a 2012 status that has not been true for more than a decade.
The common framing "adipotide is dangerous but might work" is also incomplete. The first part is supported by primate data and a publicly named dialysis case. The second part has no human confirmation. A decade of underground research-peptide availability has not produced the convincing efficacy logs that exist for almost every other popular bodybuilding peptide.
If a single sentence has to carry the topic: "Adipotide is a 2011-vintage anti-obesity peptide that worked in monkeys, never produced a human efficacy readout, damaged kidneys at every dose level that worked, and is still sold as a research peptide to consumers who routinely do not finish a cycle — including at least one named bodybuilder who ended up on dialysis."
Starting points if you want to audit or extend this page.