Translation map · v1 · 28 Aug 2026

Distance to the Patient

Every route from cell reprogramming to a treatment, ordered by how close it actually is to a human being — and where a topical wound programme could enter.

Prepared for E. Feoli-TufiPolímeros y Servicios S.A.Scope: chemical & pharmacological routes

The governing constraint

Chemical reprogramming is not self-sustaining. Cells require continuous compound exposure for weeks; withdraw early and they relax back. The only state that maintains itself after withdrawal is full pluripotency — the one state you can never induce inside a person, because it forms teratomas.

Everything below follows from that. The chemistry is a manufacturing process, not a therapy. Which means the whole field splits into two routes that have almost nothing in common except vocabulary.

Route A

Ex vivo — the factory

Reprogram cells outside the body under controlled chemistry, differentiate them, transplant the product. Nothing self-sustaining about the chemistry; what persists is the graft. Regulatory class: advanced cell therapy. Cost structure: per-patient manufacture, cleanroom, months of lead time.

Bench / preclinicalIn trialsFiled with regulatorReached patientsDiscontinued

Autologous CiPSC-derived islets, type 1 diabetes

Chemically induced pluripotent stem cells → islet organoids · Peking University / Deng lab

In patients

BenchPreclinPh 1Ph 2/3FiledPatients

DeliveryTransplant under abdominal anterior rectus sheath

ResultInsulin independence, first reported case, Sept 2024

Regulatory classCell therapy (ATMP)

Advantage of chemical routeNo viral vector, no transgene, defined compounds

This is the field’s real clinical achievement and it is genuine. But read what it is: a bespoke factory. The chemistry ran in a dish for weeks under full control, then stopped. Nothing was reprogrammed inside the patient.

Human chemical reprogramming platform

Multi-compound cocktails via an intermediate plastic state · Guan 2022; Liuyang 2023

Bench

BenchPreclinPh 1Ph 2/3FiledPatients

Exposure requiredWeeks of continuous compound, staged protocol

Key barrier identifiedJNK / pro-inflammatory signalling — inhibition indispensable

Intermediate stateDedifferentiation the authors liken to axolotl limb regeneration

FeedsThe factory route above

The finding worth carrying forward is not the cocktail. It is that inflammation is the brake, not the driver — established cell-autonomously, in a dish, with no immune system present.

Route B

In situ — drug applied to tissue

Apply an agent to the tissue itself, locally, for a defined time, then withdraw. This is the route that could ever be a product rather than a procedure. No small-molecule reprogramming agent has entered a clinical trial anywhere. The programmes below are what actually occupies this lane.

ER-100 — partial reprogramming in the eye

OCT4 / SOX2 / KLF4 (OSK), gene therapy · Life Biosciences · NCT07290244

Phase 1

BenchPreclinPh 1Ph 2/3FiledPatients

IndicationOpen-angle glaucoma; NAION

First patient dosed9 June 2026

EndpointsSafety and tolerability; visual function secondary

Regulatory classGene therapy — not a small molecule

The first-in-human partial reprogramming trial exists, and it is transcription factors delivered by vector — the heavy, expensive, irreversible modality. It is the benchmark to be measured against, not a template to copy.

PY-60 topical gel — pharmacological YAP activation

Annexin A2 binder, CAS 2765218-56-0 · Scripps / Calibr · Grzelak 2023

Preclinical

BenchPreclinPh 1Ph 2/3FiledPatients

RouteTopical gel, 0.15% w/w, every other day

ModelsYucatán mini pig wounds; human skin equivalents; ex vivo human skin

Beat the approved comparatorYes — Regranex (becaplermin) failed to move partial-thickness wounds

ReversibleFully, within 20 days of withdrawal

The closest thing in this map to your own product space, and the subject of the section below. Note the stage honestly: preclinical. No human trial.

Two-compound rejuvenation cocktail

Tranylcypromine + RepSox · Schoenfeldt 2025 · derived from the 7-compound reprogramming set

In vitro

BenchPreclinPh 1Ph 2/3FiledPatients

CellsAged human dermal fibroblasts (donors 56, 70, 83) and keratinocytes

Exposure6 days, continuous

ReadoutsγH2AX down, senescence markers down, H3K9me3 / H3K27me3 restored

OrganismC. elegans median lifespan 19 → 27 d (+42%)

Two drugs, six days, and the exact cell types that make up skin. This is the shape of a topical formulation — but it has never been applied to a wound, to an animal, or to intact skin. The gap between this and a product is entirely unaddressed.

Lorecivivint — Wnt modulation for cartilage

Intra-articular CLK/DYRK1A inhibitor · Biosplice

NDA filed

BenchPreclinPh 1Ph 2/3FiledPatients

IndicationModerate-to-severe knee osteoarthritis

NDA submitted6 January 2026

NotableJoint-structure preservation in Phase 3 — rare in OA

Regulatory classSmall molecule, local injection

Proof that a locally delivered pathway-modulating small molecule with a structural claim can reach an NDA. It is the precedent, not the mechanism.

FX-322 — the failure to study

CHIR99021 + valproic acid, intratympanic · Frequency Therapeutics

Discontinued

BenchPreclinPh 1Ph 2/3FiledPatients

IntentExpand Lgr5+ cochlear progenitors, regrow hair cells

CompoundsTwo of the seven chemical-reprogramming compounds, exactly

OutcomePhase 2 missed twice; programme killed, February 2023

ConsequenceCompany abandoned hearing, cut 55% of staff

Correct tissue, plausible mechanism, clean preclinical story, right compounds — and no patient benefit. Any translation claim in this field has to survive being held next to FX-322. It is the single most useful data point on this page.

Expansion

PY-60: the topical case, read properly

Correction to my earlier note

I cited “Regeneration and rejuvenation of skin by a topical YAP activator” (PNAS 2023) as the primary paper. It is not. That is a Commentary by James F. Martin — PNAS 120(32):e2309991120 — with no abstract and one figure. The primary research article is Grzelak et al., PNAS 2023;120(28):e2305085120, from Scripps Research / Calibr.

I also wrote that it “shifts cutaneous wound repair toward regeneration rather than scar.” That claim is not supported by the paper. Scar was not characterised: no Engrailed-1 lineage work, no myofibroblast markers, no quantified collagen organisation. Masson’s trichrome was performed on pig wounds for collagen deposition, but no scar outcome is reported. What the paper actually demonstrates is accelerated re-epithelialisation and regranulation. The word “regenerative” is theirs; the anti-scar reading was mine, and it was wrong.

What the compound is

PY-60 is not a LATS or Hippo-kinase inhibitor, which is how most secondary sources describe this class. Its molecular target is Annexin A2 (K_d ≈ 1.4 µM; cellular EC₅₀ for YAP activation ≈ 1.6 µM). ANXA2 normally shepherds YAP to the plasma membrane, where YAP receives inhibitory phosphorylation. PY-60 binds ANXA2 and blocks that trafficking, so YAP escapes membrane-localised inactivation and accumulates in the nucleus. It acts upstream of, and parallel to, the kinase cassette — on YAP’s location, not its phosphorylation state directly.

What was actually done

Model	Regimen	Result
Mouse, intact skin (PD only — no wound model)	10 mg/mL gel, once daily × 10 d	~4× epidermal thickness. Solvent formulations (acetone, DMSO) gave only ~2×.
Yucatán mini pig, intact skin	0.15 wt% once daily × 10 d	Epidermal thickness ~40 µm → ~60 µm (≈ +50%)
Yucatán mini pig wounds — 3 × 3 cm, partial and full thickness, 4 animals	0.15 wt% every other day over 8 d	Markedly accelerated closure of both wound types. Final epidermal thickness ~2× that of Regranex.
Regranex comparator (becaplermin, 0.01% rhPDGF)	Same pig wounds	No effect in partial-thickness wounds. Modest acceleration in full-thickness only.
Human skin equivalents (EpiDermFT), 3 mm full-thickness	0.2–0.5 µM, 72 h	Accelerated closure; outperformed 1% FBS and 10 ng/mL rhPDGF
Ex vivo human skin (abdominoplasty)	7 d	>2× epidermal thickness; increased rete peg density

The two facts that matter commercially

The vehicle was worth as much as the molecule. Same compound, same dose, same duration: the gel produced roughly twice the effect of solvent formulations. The gel is propylene glycol 26% / Labrasol ALF 17.5% / PEG 400 46.5% / Povidone K-90 8.5%, with PY-60 at 0.15% or 1.5% w/w. A pharmacology lab published that formulation in a methods section and moved on. It is not their competence and they did not optimise it.

The approved comparator failed. Becaplermin is the reference growth-factor product for chronic wounds and it did nothing to partial-thickness pig wounds in this study. The bar in this indication is low, and it is low because delivery to a wound bed is unsolved, not because the biology is.

Reversibility — the safety argument and the therapeutic ceiling

Mice treated for 10 days and then left alone for 20 were statistically indistinguishable from vehicle controls on epidermal thickness, Ki-67 positivity and dermal nuclei. Systemic exposure is negligible (t_½ = 0.46 h, rapid hepatic clearance, no YAP activation in internal organs). No fibroblast cytotoxicity to 100 µM. No phototoxicity under solar irradiation.

The authors offer reversibility as their answer to the obvious objection — YAP is an overgrowth node. Note where the cancer framing actually comes from: Martin’s commentary raises squamous carcinoma, citing Debaugnies 2018, that sustained active YAP causes skin cancer in mice. The primary paper contains no SCC statement and no carcinogenicity, long-term, or dysplasia-histopathology study was performed. Attribute that risk discussion correctly if you write about it.

And read reversibility the other way too: it confirms the governing constraint at the top of this page. PY-60 does not start a programme that runs. It holds a proliferative state open while the drug is present, and the state collapses when it is withdrawn. That is exactly the safety feature and exactly the therapeutic limit.

Position

Where a topical wound programme enters

The translatable unit is not “reprogramming.” It is a short, local, withdrawable chemical modulation of a repair programme the tissue already has. Four propositions, ordered by how cheaply each can be killed.

01

The bottleneck is delivery, and delivery is a device problem

A pharmacology lab’s throwaway gel doubled its own compound’s effect. Exposure control — concentration held at the wound bed, for a defined window, then removed — is formulation and dressing work. That is your competence and a stem-cell lab’s blind spot. It is also the cheapest place in this whole map to generate a defensible position.

02

A wound bed is the only tissue where the window is physically controllable

Every partial-reprogramming protocol that works imposes closure from outside — a timer, a cycle, a withdrawal. In the eye or the pancreas that timer has to be engineered into a vector. On a wound you apply, hold and remove. The requirement your own framework identifies — open the window, then close it — is solved by a dressing rather than by biology.

03

Test the brake, not the accelerator

Human chemical reprogramming found JNK / pro-inflammatory signalling to be the barrier whose removal permits plasticity — cell-autonomously, with no immune system in the dish. Your own wound reading converges: germ-free skin closes faster and scarless, L. reuteri accelerates closure with fewer neutrophils, and the shared variable across both is inflammatory load rather than microbial presence.

Lowering local inflammatory tone and letting the endogenous programme run is a cheaper hypothesis to test and a far lighter regulatory object than pushing cells uphill with a proliferative agonist. It also avoids inheriting YAP’s oncogene problem.

04

Whatever you choose has to survive FX-322

Correct target tissue, correct compounds, plausible mechanism, clean preclinical data, two failed Phase 2 trials and a dead company. Before committing, state in one sentence what makes your candidate different from FX-322 — and if the answer is only “different tissue,” it is not an answer.

Audit

Established, inferred, and unverified

Split explicitly, so nothing on this page enters a manuscript at the wrong confidence.

Established

• PY-60 targets Annexin A2; compound identity, K_d, EC₅₀, gel composition and dosing all verified against the article’s own text.
• Pig epidermal thickening ~40 → ~60 µm; Regranex null in partial-thickness wounds; >2× epidermal thickness and increased rete pegs in human explants.
• Full reversal of PY-60 effects 20 days after withdrawal; PK and phototoxicity findings.
• CiPSC-islet transplantation with insulin independence; ER-100 first dose 9 June 2026; lorecivivint NDA 6 January 2026; FX-322 discontinuation.
• JNK inhibition as the indispensable barrier-removal step in human chemical reprogramming.

Inferred

• That formulation, not molecule, is the exploitable gap — my reading of the gel-vs-solvent difference, not a claim the authors make.
• That a wound bed’s physical exposure control substitutes for an engineered timer — an argument, not a result.
• That the inflammatory-brake finding in a dish transfers to a wound bed with an intact immune system. Untested. This is the load-bearing assumption in proposition 03.

Not established

• Any anti-scar effect of PY-60. Not measured.
• Hair follicle neogenesis. Not measured — it appears only as a discussion aside about YAP in follicle biology.
• Numeric wound-closure kinetics. Presented graphically only; do not cite a closure percentage without reading the figures.
• Carcinogenicity or long-term safety of topical YAP activation. No such study exists.
• Any induction of paligenosis, or of reprogramming of any kind, in a human tissue in vivo. Nothing on this page demonstrates it.
• Mouse strain and sex for the PY-60 pharmacodynamic experiments; per-arm wound allocation; the n of the pig PD study — all ambiguous in the source.
• Page range for Liuyang 2023 (Cell Stem Cell) — verify before citing.

Sources

Primary record

1. Grzelak EM, Elshan NGRD, Shao S, Bulos ML, Joseph SB, Chatterjee AK, Chen JJ, Nguyên-Trân V, Schultz PG, Bollong MJ. Pharmacological YAP activation promotes regenerative repair of cutaneous wounds. PNAS 2023;120(28):e2305085120. PMID 37399395. primary
2. Martin JF. Regeneration and rejuvenation of skin by a topical YAP activator. PNAS 2023;120(32):e2309991120. PMID 37494422. commentary — not primary
3. Shao S, et al. YAP-dependent proliferation by a small molecule targeting annexin A2. Nat Chem Biol 2021;17:767–775. compound origin
4. Guan J, et al. Chemical reprogramming of human somatic cells to pluripotent stem cells. Nature 2022;605(7909):325–331. PMID 35418683.
5. Hou P, et al. Pluripotent stem cells induced from mouse somatic cells by small-molecule compounds. Science 2013;341(6146):651–654.
6. Liuyang S, et al. Highly efficient and rapid generation of human pluripotent stem cells by chemical reprogramming. Cell Stem Cell 2023. PMID 36944335. pages unverified
7. Wang S, et al. Transplantation of chemically induced pluripotent stem-cell-derived islets under abdominal anterior rectus sheath in a type 1 diabetes patient. Cell 2024. PMID 39326417.
8. Schoenfeldt L, et al. Chemical reprogramming ameliorates cellular hallmarks of aging and extends lifespan. EMBO Mol Med 2025;17(8):2071–2094. PMID 40588563.
9. Debaugnies M, et al. YAP and TAZ are essential for basal and squamous cell carcinoma initiation. eLife 2018;7:e33304. cited by commentary for SCC risk
10. Life Biosciences. ER-100 Phase 1, optic neuropathies. ClinicalTrials.gov NCT07290244. First patient dosed 9 June 2026.

Compiled 28 August 2026 · Claims split established / inferred / not established above · Verify all figure-level numbers to primary before embedding in a manuscript