KY-19382

KY19382; 5,6-dichloroindirubin-3'-methoxime — an indirubin-3'-monoxime (I3O) analog that activates Wnt/beta-catenin signaling by blocking the CXXC5–Dishevelled interaction. From Kang-Yell Choi's group; sibling molecule KY19334 comes from the same program and is NOT the same compound

KY-19382 (KY19382; 5,6-dichloroindirubin-3'-methoxime — an indirubin-3'-monoxime (I3O) analog that activates Wnt/beta-catenin signaling by blocking the CXXC5–Dishevelled interaction. From Kang-Yell Choi's group; sibling molecule KY19334 comes from the same program and is NOT the same compound) is a healing & recovery research compound. A brake release, not an accelerator — which is the whole argument for it. CXXC5 is a negative-feedback regulator of Wnt/beta-catenin signaling: in the cytosol it binds Dishevelled (Dvl) and blocks Dvl's job inside the Wnt signalosome, so the pathway stays damped. KY-19382 is a small molecule that inhibits that CXXC5–Dvl protein-protein interaction. Freed Dvl restores signalosome function, GSK-3beta-mediated phosphorylation of beta-catenin falls, beta-catenin escapes beta-TrCP-dependent ubiquitination and proteasomal degradation, accumulates, enters the nucleus and partners with TCF/LEF to transcribe Wnt target genes. In hair that matters because dermal papilla Wnt/beta-catenin activity is what drives anagen: in human dermal papilla cells KY-19382 raised beta-catenin, alkaline phosphatase and the proliferation marker PCNA. Chemically it is 5,6-dichloroindirubin-3'-methoxime, an analog of indirubin-3'-monoxime. CXXC5 is also the node through which DHT suppresses follicles via PGD2, which places this mechanism directly downstream of the androgen problem rather than beside it.

How KY-19382 works

A brake release, not an accelerator — which is the whole argument for it. CXXC5 is a negative-feedback regulator of Wnt/beta-catenin signaling: in the cytosol it binds Dishevelled (Dvl) and blocks Dvl's job inside the Wnt signalosome, so the pathway stays damped. KY-19382 is a small molecule that inhibits that CXXC5–Dvl protein-protein interaction. Freed Dvl restores signalosome function, GSK-3beta-mediated phosphorylation of beta-catenin falls, beta-catenin escapes beta-TrCP-dependent ubiquitination and proteasomal degradation, accumulates, enters the nucleus and partners with TCF/LEF to transcribe Wnt target genes. In hair that matters because dermal papilla Wnt/beta-catenin activity is what drives anagen: in human dermal papilla cells KY-19382 raised beta-catenin, alkaline phosphatase and the proliferation marker PCNA. Chemically it is 5,6-dichloroindirubin-3'-methoxime, an analog of indirubin-3'-monoxime. CXXC5 is also the node through which DHT suppresses follicles via PGD2, which places this mechanism directly downstream of the androgen problem rather than beside it.

Proposed benefits

Researched as a Wnt/beta-catenin pathway activator acting through blockade of the CXXC5–Dishevelled interaction, studied for hair regrowth and hair follicle neogenesis, cutaneous wound healing, and dermal matrix and collagen deposition.

Where to get KY-19382

The evidence for KY-19382

Graded by what exists behind each claim.

✅ Clinically validated

• No human trial of KY-19382 exists, and no pharmacokinetic study in any species. The molecule has a substantial preclinical record and zero human exposures in the published literature.
• The closest thing to human data is human tissue, not a human. KY-19382 increased hair length in cultured human hair follicles and raised beta-catenin, alkaline phosphatase and PCNA in human dermal papilla cells (Ryu 2021); it drove migration of human keratinocytes and dermal fibroblasts with increased Collagen I, Keratin 14, PCNA and alpha-smooth muscle actin (Yoon 2023); and it restored the length of human hair follicles shortened under high-glucose culture (Ryu 2022). Explanted follicles and cultured cells are a genuinely stronger starting point than rodent-only data, and they are still not a person. This tier stays empty on purpose.

📊 Correlative data

• No human association, observational or uncontrolled data on KY-19382 either. What does exist at the human level is target validation rather than compound evidence: CXXC5 is overexpressed, with Wnt/beta-catenin signaling and its wound-healing and angiogenesis target genes suppressed, in wound tissue from diabetic foot ulcer patients (Kim 2023), and CDK1 induction with Wnt/beta-catenin activation was found in human cutaneous squamous cell carcinoma samples against normal samples (Lee 2025). So the pathway this molecule acts on is demonstrably dysregulated in human skin disease in both directions. That is a reason the target is real. It is not evidence about the bottle.

🧪 Theoretical / extrapolated

• The mechanism is a protein-protein interaction block, which is why it reads as restorative rather than driving. CXXC5 is a negative regulator of Wnt/beta-catenin that works by binding Dishevelled; KY-19382 was identified as a Wnt/beta-catenin activator via inhibition of the CXXC5–Dvl interaction (Ryu 2021). Lifting a brake has a ceiling set by how much upstream Wnt tone the tissue actually has. That distinction carries most of the weight in this page's safety section.
• Three separate in vivo hair models, which is more than most compounds here manage. In C57BL/6 mice KY-19382 induced hair regrowth, and promoted generation of de novo hair follicles as shown by wound-induced hair follicle neogenesis (WIHN) and hair patch assays (Ryu 2021). Follicle neogenesis is a different and more demanding claim than lengthening existing hairs, and it is the one that makes this mechanism interesting rather than incremental.
• It puts the mechanism downstream of the androgen problem instead of alongside it. CXXC5 was characterized as the mediator of PGD2-induced hair loss, with that hair loss reversed by *Cxxc5* knockout or by PTD-DBM, a peptide that activates Wnt/beta-catenin by interfering with CXXC5's Dvl-binding function — and CXXC5 also mediates DHT-induced hair loss via PGD2, with DHT-induced loss alleviated by inhibiting both GSK-3beta and CXXC5 (Ryu 2023). If that chain holds, this is a lever on androgenetic hair loss that does not require touching androgens.
• Wound repair, with the specific proteins named. In human keratinocytes and dermal fibroblasts KY-19382 increased migration with raised beta-catenin, phalloidin, Keratin 14, PCNA, Collagen I and alpha-SMA, without significant cytotoxicity; in a murine cutaneous wound model it raised Collagen I, Keratin 14, PCNA and stem cell markers and accelerated re-epithelialization and neo-epidermis formation with collagen deposition at an early stage (Yoon 2023). Note that this is injured skin, where the barrier that would otherwise govern delivery is absent.
• It works where Wnt is suppressed, which is the cleanest test of a restorative mechanism. In *db/db* mice and in high-fat-diet plus streptozotocin diabetic mice, hair regrowth and WIHN were reduced through suppression of Wnt/beta-catenin; KY-19382 restored both (Ryu 2022). A brake-release mechanism should do most for tissue whose brake is most engaged, and that is what was observed.
• The finding that complicates the proliferation worry, from the developers' own hands. (Lee 2025) opens by conceding that Wnt is hard to drug because of *its concerning role in cancer*, then reports that KY19382 and KY19334 INHIBITED Wnt/beta-catenin signaling in human cutaneous squamous cell carcinoma cells, with suppressed CDK1 expression, and that the two molecules attenuated two-stage mouse skin carcinogenesis. The direction of effect on Wnt inverted in a tumor context. That is the opposite of what a naive reading of 'Wnt activator' predicts, it is the most important safety datum this compound has, and it is the developer's own work rather than an independent replication.

How to read these tiers: they say how much human evidence exists, not how well something works — and ✗ flags harm, never a disappointing trial. How the evidence tiers work →

What KY-19382 actually does

The important thing about this molecule is the kind of intervention it is, not the pathway it touches. Every compound that raises Wnt/beta-catenin signaling invites the same objection — that Wnt drives proliferation and proliferation drives tumors — and the answer depends entirely on where in the pathway you push. KY-19382 does not mimic a Wnt ligand, does not inhibit the destruction complex directly, and does not stabilize beta-catenin by mutation. It blocks a negative-feedback protein from doing its job. That is a ceiling-limited intervention, and the ceiling is how much upstream Wnt tone the tissue already has.

The target, named precisely. CXXC5 — CXXC-type zinc finger protein 5 — is a negative regulator of Wnt/beta-catenin signaling Ryu 2023. Its relevant activity is cytosolic rather than nuclear: it binds Dishevelled (Dvl) and occupies the interaction surface Dvl needs in order to function inside the Wnt signalosome at the membrane. Hold Dvl and the pathway stays damped, whatever the ligand situation. Ryu 2021 identified KY-19382 as a Wnt/beta-catenin activator specifically “via inhibition of the interaction between CXXC-type zinc finger protein 5 (CXXC5) and dishevelled (Dvl)”, and the companion work on the sibling molecule names the same mechanism as inhibiting the cytosolic function of CXXC5 Kim 2023.

Substrate and product, through the canonical pathway. Released Dvl restores signalosome assembly, which suppresses the destruction complex — Axin, APC and GSK-3beta. GSK-3beta's substrate here is beta-catenin, which it phosphorylates on N-terminal serine and threonine residues; the product of that phosphorylation is a beta-catenin recognized by the beta-TrCP E3 ubiquitin ligase, polyubiquitinated and destroyed in the proteasome. Take that phosphorylation away and cytosolic beta-catenin accumulates, translocates to the nucleus and displaces repressors on TCF/LEF transcription factors. Every downstream observation in this file — alkaline phosphatase, PCNA, Collagen I, Keratin 14 — is that transcriptional switch being read out in a different cell type.

Why hair is the obvious target rather than an opportunistic one. Ryu 2021 states the premise directly: promotion of hair regeneration “heavily depends on the activation of Wnt/beta-catenin signaling in the hair follicle, including dermal papilla”. The dermal papilla is the mesenchymal signaling center that instructs the follicle to enter anagen, and in human dermal papilla cells KY-19382 raised beta-catenin, alkaline phosphatase and PCNA — ALP being a marker of dermal papilla inductive capacity rather than a generic readout. So the compound engages the cell type whose Wnt activity actually sets the cycle.

The chemistry, and the family resemblance that matters. KY-19382 is 5,6-dichloroindirubin-3'-methoxime Yoon 2023, described as one of the newly synthesized analogs of indirubin-3'-monoxime (I3O) Ryu 2021. Indirubin is the planar bis-indole pigment from Danggui Longhui Wan; the 3'-oxime and methoxime modifications exist because the parent is almost insoluble. Working from that name, the formula is C₁₇H₁₁Cl₂N₃O₂ and the molecular weight about 360 g/mol — arithmetic done here rather than read off a paper, and it matters in the PK section. The family resemblance is also a pharmacological loose end. The best-characterized activity of indirubins as a class is inhibition of GSK-3beta and the CDKs — and GSK-3beta inhibition is itself a way to activate Wnt, far more bluntly and systemically than a protein-protein interaction block. Ryu 2023 explicitly reports DHT-induced hair loss being alleviated by inhibiting both GSK-3beta and CXXC5 functions, so both levers are in this program's field of view. Whether KY-19382 itself retains meaningful GSK-3beta activity is not stated in any abstract read for this page, and it is the single most useful thing a reader could learn next.

Do not confuse KY19382 with KY19334. They are different molecules from the same program acting on the same target. Kim 2023 is a KY19334 paper; Lee 2025 tests both. Where this page cites the sibling, it says so.

Cell, rodent, human — and where it stops

This chain is genuinely longer than most in the Vault, and it is worth walking rung by rung because the rungs are unusually well populated — right up to the one that is missing.

Human cells. Human dermal papilla cells: KY-19382 activated Wnt/beta-catenin and raised alkaline phosphatase and PCNA Ryu 2021. Human keratinocytes and dermal fibroblasts: enhanced migration with raised beta-catenin, phalloidin, Keratin 14, PCNA, Collagen I and alpha-SMA, and the paper states this occurred “without causing significant cytotoxicity” Yoon 2023. Two different human cell lineages, two different target-gene signatures, one mechanism.

Organ culture, which is the rung most compounds in this catalog skip entirely. KY-19382 increased hair length in ex vivo-cultured mouse vibrissa AND cultured human hair follicles Ryu 2021, and restored the length of human hair follicles shortened under high-glucose conditions Ryu 2022. An explanted human follicle is a complete mini-organ with its own dermal papilla, matrix and sheath, and a length response in it is a real functional result in human tissue. It is still not a person: there is no immune system, no systemic hormonal context, no barrier to cross, and the drug is in the medium.

Mouse, in three model designs rather than one. In C57BL/6 mice: hair regrowth after depilation, wound-induced hair follicle neogenesis (WIHN), and hair patch assays, analyzed by immunoblotting, ALP and immunohistochemistry Ryu 2021. WIHN is the demanding one — it scores new follicles forming in healing skin, not existing follicles lengthening. Separately, in db/db and high-fat-diet plus streptozotocin diabetic mice, where anagen was delayed and both regrowth and WIHN were reduced through suppressed Wnt signaling, KY-19382 restored both Ryu 2022. And in a murine cutaneous wound model it accelerated re-epithelialization with collagen deposition and stem cell activation Yoon 2023.

Mouse, in the direction nobody expected. Lee 2025 took KY19382 and KY19334 into two-stage mouse skin carcinogenesis and reported that the molecules attenuated it, alongside inhibition of Wnt/beta-catenin signaling and CDK1 suppression in human cutaneous squamous cell carcinoma cells. A Wnt activator reducing skin tumors in a chemical carcinogenesis model is a result worth sitting with, and it is discussed properly in the safety section.

Human target validation, without the compound. CXXC5 is overexpressed with Wnt/beta-catenin and its wound-healing and angiogenesis target genes suppressed in wound tissue from diabetic foot ulcer patients Kim 2023, and CDK1 induction with pathway activation appears in human cSCC patient samples Lee 2025. The target is dysregulated in human skin disease in both directions, which is the best kind of argument for a target and no kind of argument for a product.

Humans given KY-19382 — zero. No trial, no case series, no exposure. The three specific obstacles to the chain completing. First, barrier. Nearly every in vivo result comes from depilated, wounded or otherwise disrupted skin; the product is applied to intact skin, and no permeation study exists for this molecule through anything. Second, the species where hair is concerned. Mouse hair cycles in synchronized waves across the whole dorsum, which is what makes depilation-and-regrowth a clean assay; human scalp follicles cycle asynchronously over years, so a mouse regrowth result compresses a timescale that does not compress in people. Third, one lab. The compound literature is overwhelmingly Kang-Yell Choi's group: every one of these papers carries Choi KY as an author Ryu 2021 Yoon 2023 Ryu 2022 Ryu 2023 Kim 2023 Lee 2025. The work is published in real journals and is internally consistent across six papers and multiple model systems; it is also, as a body, unreplicated by anyone with no stake in it.

KY-19382 pharmacokinetics — how much of it actually gets in

No pharmacokinetic study of KY-19382 has been published in any species, by any route. No absorption fraction, no C_max, no half-life, no clearance pathway, no skin-permeation measurement. What follows bounds the exposure from structure, because that is the only honest substitute.

Size is in this compound's favor, which is unusual for the category. From the chemical name in Yoon 2023, 5,6-dichloroindirubin-3'-methoxime works out to C₁₇H₁₁Cl₂N₃O₂, about 360 g/mol. Passive permeation of intact stratum corneum falls off steeply above roughly 500 Da, so 360 sits comfortably inside the range where a topical route is mechanically plausible — and the two chlorines raise lipophilicity, which favors partitioning into the lipid-rich barrier. On molecular properties alone this is a better transdermal candidate than most things sold as one.

Solubility is the countervailing problem, and for indirubins it is the defining one. The parent scaffold is a flat, rigid bis-indole whose molecules stack; indirubin is notoriously close to insoluble in water, and the entire reason the 3'-oxime and methoxime derivatives exist is to break that stacking and make the class formulable at all. Skin flux is driven by thermodynamic activity in the vehicle, not by concentration on a label: a compound dissolved in a strong organic solvent can have high concentration and low driving force, and it can precipitate out on the skin surface as the volatile fraction evaporates, at which point delivery stops regardless of what the bottle said. DA states 5 mg/mL and does not name the vehicle. For a molecule of this class that omission is not a detail — it is the variable that decides whether anything is delivered.

What happens if it is absorbed, reasoned from structure. The methoxime is an O-methylated oxime, a group liable to hydrolysis back toward the ketone and to CYP-mediated demethylation; the indole nitrogens and the aromatic system are the usual substrate for oxidation followed by phase II conjugation. A chlorinated planar aromatic of this shape is also lipophilic enough to distribute into tissue rather than stay in plasma, which argues for a long apparent terminal half-life driven by redistribution rather than by slow metabolism. None of that is measured; it is the shape of the answer rather than the answer.

The one route fact the published record does establish, and it flatters the product. The wound paper's in vivo work is a murine cutaneous wound model Yoon 2023 — injured skin, where the stratum corneum that governs delivery is absent by design. Accelerated re-epithelialization in an open wound demonstrates that the compound reaches viable epidermis when the barrier is gone. It demonstrates nothing about intact scalp. The hair models in Ryu 2021 include depilation-based regrowth and wound-induced neogenesis, both of which also involve a disrupted or recently disrupted barrier. So the strongest efficacy data in this file comes from preparations where delivery was least of an obstacle, and that is the specific gap between the literature and a bottle applied to closed skin.

The arithmetic that is actually available. 5 mg/mL × 30 mL = 150 mg total content, DA's own figure. Because no permeation study exists, no fraction of an applied volume can be converted into a tissue or plasma concentration. That is why this card carries no dose rather than a cautious one.

What would have to be true, and how you would know it was not

Four predictions. The first is the product's actual claim, the third cuts against it, and each names an instrument and a window.

1. Terminal hair density, by standardized macrophotography and trichoscopy, at baseline, 16 weeks and 24 weeks. This is the claim, and the window is not negotiable: human scalp follicles cycle asynchronously over years, anagen induction needs a cycle to express itself as visible hair, and nothing read before about four months means anything. The measurement is a fixed-position photograph of a clipped, tattooed or otherwise reproducibly located target area with counts of terminal versus non-terminal hairs per cm² — the same methodology used in registration trials for the licensed hair drugs, and entirely available outside a laboratory. The falsification is clean: if terminal density is flat at 24 weeks under consistent application, then either the mechanism does not transfer from mouse to human scalp or the molecule is not crossing intact stratum corneum from this vehicle, and the PK section explains why the second is the better bet.

2. The anagen fraction should move before the hair count does. The mechanism is anagen induction via dermal papilla Wnt signaling Ryu 2021, so the earlier readout is the anagen:telogen ratio on a phototrichogram, which detects the cycle shift weeks before it becomes length. Baseline and 8–12 weeks. If the ratio shifts and density later does not, the mechanism engaged and something downstream limited it; if neither moves, delivery is the suspect. Those two outcomes call for different next steps, which is why it is worth measuring both.

3. THE PREDICTION THAT CUTS AGAINST THE PRODUCT, and it is a dose-response shape rather than a toxicity. Maurya 2024 expressed constitutively active beta-catenin in keratinocytes under doxycycline control. Short induction in telogen produced enlarged pilosebaceous units expanding into the dermis — the desired direction. Prolonged induction produced significant thickening and folding of the epidermis, increased keratinocyte proliferation in the basal layer and the outer root sheath, numerous hyperplastic cysts, and the authors' own conclusion that this “likely explains the decrease in hair density observed”. Sustained maximal pathway activation reduced hair. So the prediction is non-monotonic: the mechanism has an optimum, and past it the visible signs are epidermal — scaling, thickening, follicular plugging, sebaceous prominence — rather than more hair. Anyone responding to a disappointing 16-week photograph by applying more is moving toward the wrong arm of that curve. Two honest caveats, both important: a transgene producing undegradable beta-catenin is a far stronger and less regulated signal than a negative-feedback blockade can generate, and in that model the phenotype reversed on withdrawal, which is the mitigation as well as the warning.

4. On intact skin the effect should be markedly weaker than the wound literature implies, and that is testable in one experiment. The strongest in vivo results come from wounded or depilated skin Yoon 2023 Ryu 2021, where the barrier is absent or disrupted. The prediction is a substantially smaller response on closed skin, and the cheap test is a within-subject split-scalp design — same person, same vehicle, treated and vehicle-only target areas, photographed on the schedule in prediction 1. A within-subject control removes the single biggest confound in all hair anecdote, which is that people change three things at once and attribute the result to the new bottle.

What nobody has tested yet

Five experiments nobody has published. The first is embarrassing in its cheapness given what this product costs.

1. Does KY-19382 cross intact human skin, from any vehicle? A Franz diffusion cell with excised human skin, the solution applied at its stated 5 mg/mL, receptor fluid sampled across 24 hours, with a quantified vehicle. Every efficacy result in this file was generated in depilated, wounded or explanted tissue; no permeation measurement through intact skin exists for this molecule. This is a routine formulation study and it is the load-bearing unknown for anything sold as a topical.

2. Any pharmacokinetics at all, in any species, by any route. Plasma and skin concentrations against time after a stated topical application, with the parent and its likely demethylated and hydrolyzed metabolites measured. Never published. Without it, nobody can say whether a topical application is a local intervention or a systemic one — and for a molecule whose scaffold family inhibits GSK-3beta and the CDKs, that is not a bookkeeping question.

3. Does KY-19382 itself inhibit GSK-3beta, and how much? Its parent I3O is a well-characterized GSK-3beta inhibitor; the program's own work reports DHT-induced hair loss being relieved by inhibiting both GSK-3beta and CXXC5 Ryu 2023. No abstract read for this page states a GSK-3beta IC₅₀ for KY-19382. This matters because the two mechanisms have different safety shapes: a CXXC5–Dvl blockade is ceiling-limited by upstream Wnt tone, while systemic GSK-3beta inhibition activates Wnt everywhere at once and hits many other substrates. A single kinase panel would settle which molecule this actually is.

4. Whether the indirubin scaffold's aryl hydrocarbon receptor activity survives in this analog. Planar bis-indoles are the classic geometry for AhR engagement, and AhR drives CYP1A induction and therefore the clearance of other drugs. I could not verify a citation for indirubin-AhR binding within this session, so this is recorded as a question to check rather than as a fact — but it is a specific, cheap, unasked question about a compound applied daily by people who take other things.

5. Repeat-application dermal safety, and the tumor question asked by someone without a stake. There is no chronic dermal toxicity study, no NOAEL, no photocarcinogenicity study for KY-19382 — which matters unusually much here, because the intended application site is sun-exposed scalp skin and UV damage is the initiating event for the tumors Wnt signaling is implicated in. Lee 2025 is genuinely reassuring on carcinogenesis and it is the developers' own two-stage model; the same group authors essentially the entire compound literature. An independent chronic-application study on UV-exposed skin is the experiment that would move this compound's safety from argued to established, and nobody has run it.

KY-19382 — its own safety story, not its class's

The safety question for a Wnt/beta-catenin activator is proliferation, and it deserves a real answer rather than either a dismissal or a scare. The class-level fact is not in dispute: aberrant Wnt/beta-catenin activation contributes significantly to tumor initiation and progression, and this is the reason the pathway has been so hard to drug Song 2024. The developers of this molecule open their own cancer paper by conceding exactly that point Lee 2025. So the objection is legitimate and the honest work is in figuring out how much of it applies to this intervention.

The mechanistic answer, and it is the strongest thing on this page. KY-19382 does not stabilize beta-catenin, mimic a Wnt ligand or mutate a destruction-complex component. It blocks CXXC5, a negative-feedback regulator, from holding Dishevelled Ryu 2021. Releasing a brake cannot produce more signal than the tissue's upstream Wnt tone can support, which is a structurally different exposure from a constitutively active mutant that signals without limit. The diabetic-mouse result is the cleanest evidence that this is how it behaves in practice: the compound did most where Wnt was most suppressed Ryu 2022, which is what a restorative mechanism looks like and not what an indiscriminate agonist looks like.

The empirical answer, which went the opposite way to the prediction. Lee 2025 tested KY19382 and KY19334 in human cutaneous squamous cell carcinoma cells and found they INHIBITED Wnt/beta-catenin signaling there, with suppressed CDK1 expression, and that both molecules attenuated two-stage mouse skin carcinogenesis. The direction of effect inverted in a tumor context, apparently because the tumor's cellular landscape is different from normal skin's. That is the single most important safety datum this compound has and it points the reassuring way. Two limits on how far to carry it: it is the developing group's own work, unreplicated by anyone independent, and context-dependent bidirectionality is harder to predict than a fixed direction, not easier — a molecule whose sign flips with cellular context is a molecule whose behavior in your tissue is an assumption.

The prediction that does cut against the product, from a model with no connection to this compound. Maurya 2024 drove constitutively active beta-catenin in keratinocytes. Brief induction enlarged pilosebaceous units; prolonged induction thickened and folded the epidermis, raised keratinocyte proliferation in the basal layer and outer root sheath, produced numerous hyperplastic cysts, and decreased hair density. The overshoot phenotype is not cancer — it is hyperplasia, and it reversed when the transgene was switched off. Read carefully, that is the most useful safety sentence available: the failure mode of too much pathway activation in skin is a visible, hyperplastic, reversible change, and the earliest sign of it is epidermal rather than follicular. Mitigation: watch the skin, not the blood. Scaling, thickening, follicular plugging or new sebaceous prominence at the application site are the signal to stop and let it reverse, and they will appear before anything more serious does.

There is no blood test for this, and pretending otherwise would be worse than saying so. No routine marker reports on Wnt/beta-catenin activity in skin, and a normal CBC and CMP is equally consistent with the compound working perfectly and with it never leaving the stratum corneum. The appropriate surveillance is dermatological: a baseline look at the application area, standardized photographs on the schedule in the prediction section, and a low threshold for having a new or changing lesion examined — particularly on scalp skin, which is among the most sun-damaged on the body and so has the most pre-existing damage for a proliferative stimulus to act on.

Where the mechanism argues for not applying it — these follow from the biology, not from caution. Existing skin malignancy, actinic keratoses or field change in the application area, and a personal history of non-melanoma skin cancer: a proliferative stimulus onto already-transformed keratinocytes is the scenario the reassuring carcinogenesis data least clearly covers, since that study modeled prevention rather than established field damage. Pregnancy — Wnt/beta-catenin is a core developmental patterning pathway, there is no reproductive toxicology for this molecule, and no permeation study bounds systemic exposure. Broken or freshly wounded skin — not because it is harmful there, but because that is the condition under which the published work shows real delivery Yoon 2023, and therefore the condition under which exposure is highest and least predictable.

The scaffold's loose end, stated as a loose end. The parent indirubin's best-known activity is inhibition of GSK-3beta and the CDKs, and no abstract read for this page reports whether KY-19382 retains it. If it does, and if absorption occurs, the exposure is not a local protein-protein interaction block but systemic kinase inhibition with a far broader substrate list — at which point a CBC becomes meaningful for marrow-proliferative effects. Nothing here establishes that; it is named because one kinase panel would answer it.

What is genuinely favorable, so this reads as an assessment rather than a warning. No significant cytotoxicity in human keratinocytes and fibroblasts Yoon 2023. Six papers across four model systems pointing the same way. The carcinogenesis result went the good direction in the one model where it was tested, and the overshoot phenotype in the nearest worst-case model was reversible. This is a better-characterized compound than most of what sits beside it. It has also never been given to a person, has no pharmacokinetics in any species, no published dose, no disclosed vehicle, and an evidence base authored almost entirely by the group that developed it.

Sources read for this page

• Ryu YC, Lee DH, Shim J, Park J, Kim YR, Choi S, Bak SS, Sung YK, Lee SH, Choi KY. KY19382, a novel activator of Wnt/beta-catenin signalling, promotes hair regrowth and hair follicle neogenesis. British Journal of Pharmacology 2021;178(12):2533-2546 · PMID 33751552
• Yoon M, Kim E, Seo SH, Kim GU, Choi KY. KY19382 Accelerates Cutaneous Wound Healing via Activation of the Wnt/beta-Catenin Signaling Pathway. International Journal of Molecular Sciences 2023;24(14):11742 · PMID 37511501
• Ryu YC, Kim YR, Park J, Choi S, Kim GU, Kim E, Hwang Y, Kim H, Bak SS, Lee JE, Sung YK, Han G, Lee SH, Choi KY. Wnt/beta-catenin signaling activator restores hair regeneration suppressed by diabetes mellitus. BMB Reports 2022;55(11):559-564 · PMID 36016500
• Ryu YC, Park J, Kim YR, Choi S, Kim GU, Kim E, Hwang Y, Kim H, Han G, Lee SH, Choi KY. CXXC5 Mediates DHT-Induced Androgenetic Alopecia via PGD2. Cells 2023;12(4):555 · PMID 36831222
• Kim E, Seo SH, Hwang Y, Ryu YC, Kim H, Lee KM, Lee JW, Park KH, Choi KY. Inhibiting the cytosolic function of CXXC5 accelerates diabetic wound healing by enhancing angiogenesis and skin repair. Experimental & Molecular Medicine 2023 (published 1 August 2023) · PMID 37524876
• Lee SH, Kang MJ, Roh MR, Choi KY. Novel small molecules downregulate CDK1 expression and inhibit Wnt/beta-catenin signaling in cutaneous squamous cell carcinoma by targeting its distinct tumor-specific cellular landscape. Experimental & Molecular Medicine 2025 (published 1 September 2025) · PMID 40887499
• Song P, Gao Z, Bao Y, Chen L, Huang Y, Liu Y, Dong Q, Wei X. Wnt/beta-catenin signaling pathway in carcinogenesis and cancer therapy. Journal of Hematology & Oncology 2024;17(1):46 · PMID 38886806
• Maurya VK, Ying Y, Lydon JP. A Mouse Model for Conditional Expression of Activated beta-Catenin in Epidermal Keratinocytes. Transgenic Research 2024 (published 7 August 2024) · PMID 39110314

KY-19382 — interference & stacking

Bloodwork to run alongside KY-19382

Baseline first, then again at 8–12 weeks.

The Inflammation Deep Dive panel covers these in one order — 10 markers, $248.35 with the discount applied.

Check results you already have → · All 103 markers A–Z

KY-19382 — frequently asked questions

What KY-19382 is used for

KY-19382 appears under 2 goals in the goal router.

Related Healing & Recovery compounds

← Explore the full Protocol Vault

↑ Back to on this page