Whole Organ Regeneration

Two projects share a name and fail at different walls. One builds an organ outside the body — an organoid, a printed scaffold, a gene-edited pig kidney — and implants it. The other induces an organ that has already failed to rebuild itself where it stands, from the patient's own tissue. This brief is about the second. The distinction decides which problems you inherit: the manufactured route pays at immunology and vascular plumbing, the in-situ route pays at cell-cycle control and at the mammalian scar. Lab-Grown Organs holds the manufactured route and xenotransplantation; Limb Regeneration holds appendages, which carry a proximodistal axis and a joint series and are a different problem again.

Established Mammals have exactly one documented case of stem-cell-based epimorphic regeneration of a whole organ, and it is the deer antler. Everything else claimed for mammalian regeneration falls into three weaker categories. Established Compensatory growth that restores mass but not architecture, which is the liver. Established A developmental window that shuts in the first week of life, which is the neonatal mouse heart. Established And cell replacement at a rate orders of magnitude below what one injury destroys, which is the adult human heart. Each is real; none is an organ regrowing.

Established Two walls run under the subject. The first is the fibrotic default: adult mammalian injury resolves to scar rather than to restored pattern, and every intervention here is an argument about overriding that default without producing a tumour. Established The second is diffusion — oxygen and nutrients reach roughly 200 to 400 micrometres from a perfused surface — which bounds regenerating tissue exactly as it bounds an organoid in a dish. New parenchyma that outgrows its own perfusion dies whether it was printed or regrown.

Frontier The exotic question is whether the mammalian regenerative programme is present-but-suppressed or genuinely absent, and this brief's answer is that the question is not currently the binding one. Handwave The binding link is the tumour ledger: nobody has measured neoplasia incidence in reprogrammed or de-differentiated tissue in a large animal over a meaningful interval. Established It is unmeasured because no funded programme needs the answer — the organ shortage that would pay for it is being served by replacement organs that are nearer, cheaper and already in trials.

Established The human liver does not regenerate. It compensates, incompletely, and it is still short four years later. Aoki and colleagues followed 33 living liver donors: 23 right hepatectomies removing about 42 per cent of liver volume, and 10 left hepatectomies removing about 37 per cent. Established Right-hepatectomy donors recovered 51 per cent of preoperative volume at one week, 64 per cent at one month, 74 per cent at three months and 81 per cent at one year; left-hepatectomy donors ran 69, 76, 79 and 84 per cent over the same intervals. Established The authors state plainly that liver volume was still not restored to the preoperative value at four years. Established What grows is the remaining lobes, larger; the missing lobe is not rebuilt and lobar architecture is not restored. Frontier Whether to call this hyperplasia or hypertrophy is a real terminological dispute — the rodent literature counts cell divisions, the human volumetric data measure only volume — and it does not change the endpoint. The process restores mass, not pattern, and it does not restore all of the mass.

Established The one mammalian organ that truly regenerates does it by pairing hypergrowth with hyperactive tumour suppression and then throwing the organ away. Deer antler reaches 2 centimetres of growth per day, the fastest documented growth of any animal tissue, and a stag lays down on the order of 10 kilograms of antler in a year. Established Three stem-cell populations are characterised: antlerogenic periosteal cells, pedicle periosteal cells, and reserve mesenchymal cells. Established The same lineage shows strong positive selection on p53 cofactor and p53 regulator genes, and highly efficient apoptosis in the reserve mesenchyme. Speculative Deer, in other words, solved the cancer problem first and got the regeneration as a consequence, so the transferable lesson is about tumour suppression rather than about growth. Established The organ is also cast annually, which means the animal never has to keep a hypergrowth programme running inside a body it intends to occupy for fifteen years.

Established The heart's deficit is a window and a rate, not an absolute incapacity. Porrello and colleagues showed that hearts of one-day-old mice regenerate after partial surgical resection with minimal scarring, that the capacity is lost by seven days of age, and that lineage tracing attributes the majority of new tissue to pre-existing cardiomyocytes rather than to a resident stem cell; echocardiography at two months showed normal function. Frontier The window itself is widely reproduced, but the apex-resection assay specifically has a disputed replication record: at least two groups published partial or failed replications in 2014, and a published comment rebutted them. This brief does not adjudicate that dispute, and any summary presenting the neonatal result without it is presenting half a literature. Established On the adult side, human cardiomyocyte renewal is real and continuous, established by carbon-14 bomb-pulse dating; the annual turnover percentages in circulation come from a primary this brief's source pass could not open, so they are not quoted here. Established The asymmetry survives without them: a large myocardial infarction destroys cardiomyocytes on the order of a billion within hours, and no renewal rate compatible with a stable adult heart meets that.

Established The zebrafish heart is what makes cardiac regeneration look tractable at all, and it is worth being precise about which zebrafish result is being invoked. Zebrafish regenerate resected ventricular myocardium without permanent scarring, from dedifferentiated existing cardiomyocytes rather than from a stem-cell reserve. Frontier The commonly quoted extent and time course trace to the 2002 founding paper by Poss, Wilson and Keating, which this brief's source pass could not open, so no percentage and no day count are printed here. Frontier What matters more is a methodological dispute the popular account omits: resection and cryoinjury are different models, cryoinjury produces a transient scar that is subsequently resolved, and cryoinjury is the one that resembles a human infarct. Speculative A field that cites the resection model and treats the cryoinjury model as a technicality is quietly citing the easier experiment.

Established The kidney resists for a structural reason that has nothing to do with cell division: it stops making nephrons before you are born. Human nephrogenesis normally ceases around 36 weeks of gestation, and in preterm infants it can persist for some weeks after birth and then stops permanently. Frontier The specific figures — the postnatal persistence interval, and the roughly ten-fold between-individual range in glomerular number — sit in primaries this pass could not open, and are given here as shape rather than as numbers. Established The structural consequence is not in dispute: an adult kidney that loses nephrons responds by hypertrophy and hyperfiltration of the survivors, a compensation that degrades the survivors. Speculative There is no postnatal nephrogenic niche to reactivate, which makes the kidney the organ where the present-but-suppressed hypothesis is hardest even to state. It is not obvious what would be un-suppressed.

Established The lung resists because of what it is: a gas-exchange interface, not a mass of tissue. An in-situ programme for the lung has to restore register between two branching trees, airway and vascular, across a surface whose area is the organ's entire function. Established The diffusion wall that merely constrains other regenerating tissues is here the working principle of the organ. Established What adult mammalian lung does after significant injury is fibrose, and pulmonary fibrosis is the clearest clinical demonstration that the mammalian repair default is not simply inadequate but actively destructive of function. Speculative Of the four organs in this brief the lung is the one where no serious in-situ proposal exists at all, which is a finding rather than an omission.

Established Where regeneration does exist in mammals it is finite, and the exhaustion is systemic. Mouse digit tips regenerate through four cycles of repeated amputation and fail on the fifth. Established Excess regenerated bone volume declines monotonically — 57 per cent, then 43, then 19, then 13, then failure — and fifth-cycle digits carried 65.7 per cent fewer osteoblasts. Frontier The result nobody expected is that adjacent, un-amputated digits also showed impaired regeneration, implying a depletable resource that is not local to the injury. Speculative If that is right, the intervention target is not to add progenitor cells but to raise or reallocate a systemic budget, and nobody has named the quantity being budgeted.

Established Two human results define the honest ceiling, and neither of them is regeneration. Vertex's zimislecel, stem-cell-derived islets given as a single infusion, produced insulin independence at twelve months in 10 of 12 patients with at least a year of follow-up, a mean 92 per cent reduction in daily insulin dose, HbA1c below 7 per cent and time-in-range above 70 per cent in all 12, and freedom from severe hypoglycaemia from day 90 — under chronic immunosuppression, with two deaths adjudicated as unrelated. Established That is organ function restored without the organ, and the immunosuppression is the entire remaining problem. Established The second ceiling is the 2006 Wake Forest bladder cohort: seven children aged 4 to 19 with poor bladder function from spina bifida, autologous cells expanded onto a biodegradable scaffold, seven to eight weeks from biopsy to implantation, follow-up beyond seven years, preserved kidneys and resolved incontinence. Established A bladder is hollow, thin-walled and metabolically undemanding, and the host perfuses it by ingrowth. It is the ceiling of what worked, not a template for a solid organ.

Frontier The most-cited human organ-regeneration result is a nine-person uncontrolled study and should be read as one. The TRIIM trial gave nine healthy white men aged 51 to 65 recombinant human growth hormone with DHEA and metformin; MRI showed thymic fat replaced by tissue in seven of nine, and the mean epigenetic-age estimate fell 2.5 years, with no control group. Established Wolfgang Wagner, quoted in Nature's news coverage, said the results “are not rock solid because the study is very small and not well controlled”. Established The correct calibration for cell therapy in a human organ system has the same shape: an iPSC-derived neural progenitor trial in subacute cervical spinal cord injury enrolled four men aged 26 to 66, gave each two million cells, and reported two of four improving AIS grade, a median 13-point motor score improvement at 52 weeks, no tumour formation and no graft-related serious adverse events across two to four years. Established That is a real result, honestly reported, with four participants and no control arm.

Established The market that would pay for all of this is large, specific and pointed elsewhere. As of 30 July 2026 the US national transplant waiting list held 109,817 people, 12 people died each day waiting, more than 49,000 transplants were performed in 2025, and a new candidate was added every seven minutes, with kidneys the most-needed organ. Established The largest sustained public funder of tissue-restoring regenerative medicine is the US military: the AFIRM consortium led by the Wake Forest Institute for Regenerative Medicine received $40 million over five years in its 2023/24 round, and the two prior consortia since 2008 produced more than 20 clinical studies across craniofacial, extremity, genitourinary, skin and wound, on-demand blood, and cellular therapies for trauma. Established Whole-organ in-situ regeneration is on none of those lists, and it is on nobody else's either.

Frontier The most interesting live minority programme in the field is bioelectric, and its strongest single result is a frog. A silicone cap holding a drug-loaded silk gel — the BioDome — delivered a five-drug cocktail to an adult Xenopus amputation site for 24 hours only, and over the following 18 months the animals grew a limb with bone, neurons and vasculature, used it to swim and responded to touch. Several toes formed without underlying bone. Speculative A 24-hour exposure producing an 18-month morphogenetic outcome is either the most important result in regenerative biology or an artefact, and as of this pass no group outside the originating laboratories has published a replication. Handwave The institutional release does not state the sample size, which is the first thing an independent reader needs.

Frontier Pattern-control circuits are being read out in invertebrates at a level mammals have nothing comparable to. In Hydra the head organiser is a self-limiting Wnt3/beta-catenin/Sp5 feedback loop: Sp5 is induced by Wnt signalling and represses Wnt3. Established Silencing Sp5 by RNA interference produced ectopic heads in 50 per cent of uncut animals one day after the second round and 100 per cent two days after the third; the ectopic heads fed and expressed neuronal markers, so they were functional rather than malformations. Speculative The open question is whether such a circuit exists in mammalian tissue in any usable form, or whether it is simply what development looks like with no scar response competing against it.

Frontier The textbook premise that regenerators have a dedicated stem-cell compartment and we do not has just weakened. In the early-branching catenulid flatworm Stenostomum brevipharyngium, whole-body regeneration is supported by multiple division-competent populations, some resident inside differentiated tissues and organs, with no classical neoblast compartment. Frontier The authors argue neoblast-like cells may have evolved convergently rather than causing regenerative capacity. Speculative That cuts both ways for mammals: it removes a clean explanation for why we cannot regenerate, and raises the possibility that division-competent cells already resident in human organs are a sufficient substrate given a different signalling environment.

Frontier Reprogramming an organ rather than reseeding it is now a delivery question as much as a biology question. Cyclic, short-duration OSKM induction extended life in progeroid mice by 30 per cent without producing cancers and improved pancreas and muscle regeneration after injury in wild-type animals; the senior author's own estimate at publication was that clinical trials were up to ten years away. Frontier The delivery half moved separately: Capstan's CPTX2309, a CD8-targeted lipid nanoparticle delivering anti-CD19 CAR messenger RNA to T cells inside the body, entered a phase 1 trial in healthy volunteers on 11 June 2025, avoiding both lymphodepletion and genomic integration. Speculative That is a proof of concept that a cell population can be reprogrammed inside a person with an injection. Handwave It is not a proof that a reprogrammed fraction of a solid organ produces a coherent tissue rather than a mosaic, and that question has no published answer at all.

Frontier The organ does not have to be regenerated where it was, and that is in a human trial. LyGenesis engrafts donor hepatocytes into mesenteric lymph nodes by endoscopic ultrasound, using the node as an already-vascularised bioreactor in which miniature ectopic livers grow; the programme is in phase 2a in end-stage liver disease and raised a $19 million Series A-2 in October 2023, with pancreas, kidney and thymus in the pipeline. Speculative The move sidesteps the scaffold problem and the vascularisation wall by refusing to build vasculature at all. Frontier What it does not sidestep is whether ectopic tissue reaches a functionally sufficient mass, whether the niche fibroses, and whether the absence of native architecture is tolerable — and those are being tested rather than assumed.

Frontier Chimeric organogenesis has produced human-cell-dominated kidney structures inside pig embryos, at an efficiency almost nobody quotes. Human iPSCs injected into SIX1/SALL1-deficient cloned pig embryos yielded metanephric tubules with 40 to 60 per cent human-derived cells and minimal human contribution to non-target organs. Established The denominator is 1,820 embryos transferred, yielding two chimeric fetuses at embryonic day 25 and three more at day 28 — five in total, about 0.27 per cent, at a developmental stage far short of a functioning organ.

Frontier The most consequential open question in the competing route is whether the industry has been optimising the wrong variable. A pig kidney carrying a single edit — GGTA1 knockout, nothing else, no CD40 blockade — supported haemodynamic and electrolyte stability and dialysis independence for a pre-planned 61 days in a brain-dead human recipient. Established An antibody-mediated rejection episode on postoperative day 33 was completely reversed with plasma exchange, C3/C3b inhibition and anti-thymocyte globulin. Frontier If one edit buys two months of function with treatable rejection, the marginal value of edits two through ten is an open empirical question that the commercial programmes have no incentive to ask. Frontier The residual barrier the study identifies is not antibody at all, but pre-existing xenoreactive T cell clones that expanded and acquired effector profiles.

The workback plan for a regrown human organ — a working left ventricle's worth of myocardium after infarction, or a functioning nephron population — runs through eight links. They are given in order, with the binding one named.

Speculative Link 1, competence. Show that adult human cardiomyocytes or nephron progenitors can be driven into a proliferative, patterning-competent state and returned to a normal state afterwards; the measurement is the fraction re-entering the cell cycle multiplied by the fraction exiting without transformation. Frontier Cyclic partial reprogramming does something like this in mice and there is no human organ data. Frontier Link 2, the fibrotic default. A quantitative scar-versus-regenerate assay in a large mammal: percentage of infarct area occupied by collagen at 90 days, with and without intervention, in a neonatal pig heart. Not done at scale.

Handwave Link 3, the tumour ledger, is the binding link. Nobody has prospectively measured neoplasia incidence in reprogrammed or de-differentiated tissue over two years or more in a large animal. Established The deer's answer to that problem was p53-pathway amplification; ours is unknown. Frontier The strongest thing in the literature is a report of no tumours in a short-lived progeroid mouse, which is the weakest possible version of the test — an animal that dies before a tumour would have time to appear. Speculative Until this link carries a number, every downstream claim about in-situ regeneration is a claim about a therapy whose principal risk has never been quantified.

Frontier Link 4, delivery. Measure the distribution of a reprogramming payload across a human-sized organ: what fraction of target cells receive a dose inside the therapeutic window. Targeted lipid nanoparticles are the best current answer and are only now in first-in-human testing. Established Link 5, perfusion. Show that regenerating parenchyma recruits capillaries faster than it outgrows the 200 to 400 micrometre diffusion limit, measured as capillary density and oxygen tension inside the growing region. Speculative This has not been done in a mammalian regeneration context and is shared with Lab-Grown Organs.

Speculative Link 6, function. Ejection fraction or glomerular filtration rate at twelve months against a control arm, not histology. Speculative Link 7, regulation. An agreed endpoint set and comparator for the proposition that a patient's own organ regrew; no precedent exists. Established Link 8, money. No funded programme has this as its stated objective.

Frontier Which link actually binds? Link 3, with link 8 as the reason link 3 is not being worked on. Established Links 4 and 5 are hard but have well-capitalised industrial programmes pointed at them for other reasons. Speculative Link 3 has no programme because the product it would enable is one nobody is currently buying. Speculative The honest conclusion is that whole-organ in-situ regeneration is not primarily blocked by biology being hard; it is blocked by biology being hard in a direction that no funded programme needs to go.

Frontier The first dependency is a replication, not a discovery. The neonatal-mouse apex-resection assay has a contested record that the field never settled. Speculative A properly powered multi-laboratory replication would either establish the foundational mammalian result or remove it, and it would cost a rounding error against what is currently spent building replacements. Speculative The second is the p21-null mouse literature, which is the direct test of the present-but-suppressed hypothesis and whose tumour-risk arm is precisely the part never reported alongside the healing phenotype.

Speculative The second class of dependency is a measurement nobody has defined. The finite-regeneration result implies a depletable systemic quantity, and identifying it — metabolic, immune, or progenitor-pool — requires a candidate assay that tracks the decline from 57 per cent to zero across five amputation cycles. Handwave Nothing in the literature nominates one. Frontier A third dependency is comparative: the axolotl genome is 32 gigabase-pairs, roughly ten times human, and lacks Pax3 outright, with Pax7 having assumed its functions. Speculative Any dependency framed as finding the mammalian orthologue has to survive the fact that the reference regenerator differs from us in genome architecture rather than in a switch.

Established The third dependency is infrastructural and slightly embarrassing. A substantial fraction of this subject's load-bearing primaries — the founding zebrafish paper, the cardiomyocyte turnover measurement, the partial-reprogramming paper, the clinic census, the misconduct verdict that defines the field's worst failure — sit behind publisher blocks a research pass cannot open. This brief writes around each of them and says so at the point of use. Speculative A field whose canonical numbers are quoted secondhand for years is a field where a misattribution propagates through a peer-reviewed review without anyone checking, which is exactly what section 14 documents.

Frontier The cheapest decisive experiment is the scar assay, and it is not being run. Take neonatal and juvenile pigs, produce a controlled infarct, and measure the percentage of infarct area occupied by collagen at 90 days with and without a candidate intervention, alongside ejection fraction. Speculative A large-mammal dose-response curve for the closing of the regenerative window would tell the field whether the mouse window is a mouse fact.

Handwave The experiment that would change the funding picture is a two-year tumour count. Reprogram or de-differentiate a defined tissue compartment in a large animal and follow neoplasia incidence prospectively for at least two years against a matched control. Speculative It is expensive, slow, and produces a number that could kill the field, which is a reasonable description of why it has not been done and not a reasonable justification.

Frontier Three measurements would convert hand-waves into variables. First, payload distribution: label a targeted lipid nanoparticle payload and quantify, across a human-sized organ, the fraction of target cells receiving a dose inside the therapeutic window. Frontier Second, perfusion kinetics: capillary density and oxygen tension inside a growing regenerate, tested against the 200 to 400 micrometre limit. Handwave Third, the mosaic question: reprogram a defined fraction of cells in a solid organ and measure whether the tissue behaves as one organ or as a patchwork.

Speculative Two replications are worth more than any new result. An external group should attempt the 24-hour BioDome protocol with a stated sample size, because a whole hypothesis currently rests on one laboratory's 18-month endpoint. Speculative And the p21-null regeneration phenotype should be re-run with a tumour registry attached, because the suppressed-programme hypothesis stands or falls on whether relaxing the brake costs more in neoplasia than it returns in repair.

Speculative One experiment nobody proposes. The systemic-exhaustion result implies a shared budget; a parabiosis or serial-transplant design testing whether the depleted state is transferable would either name the resource or rule out the framing. Handwave It is a small experiment with a large consequence and it is absent from every roadmap this brief's source pass examined.

Frontier The engineering here is measurement and delivery, not construction, which is the sharpest single difference from the manufactured-organ route. Nothing in this brief needs a bioreactor that grows a heart. It needs a way to put a defined molecular instruction into a defined fraction of cells inside a solid organ, hold it for a defined interval, and stop.

Frontier Three engineering objects already exist in prototype. The wearable bioreactor — a sealed silicone chamber holding a drug-loaded gel against a wound for 24 hours — is the only device in the field that has produced a limb-scale morphogenetic outcome in a tetrapod. Frontier Cell-type-targeted lipid nanoparticles are in first-in-human trials for a different indication and are the current best answer to organ-wide delivery. Frontier Endoscopic ultrasound delivery into a lymph node is in phase 2a and shows that a delivery route can substitute for the construction problem entirely.

Speculative What does not exist is the instrumentation to know whether any of it worked. There is no clinical assay for the proposition that a tissue is currently in a regenerative state, no in-vivo readout of cell-cycle fraction inside a human organ, and no imaging modality resolving capillary density inside a growing region at the scale the diffusion limit requires. Handwave The closed-loop version — dose, measure the regenerative state, adjust — is the obvious architecture and is entirely notional, because the measurement half does not exist at any dose.

Established The manufacturing constraint differs in kind from the manufactured-organ route. An in-situ therapy is a batch of a molecule or a vector, not a patient-specific organ, which removes the per-patient bioreactor problem that dominates Lab-Grown Organs. That is the strongest structural argument in this route's favour and it is rarely made: if it worked, it would scale like a drug rather than like a transplant.

Established The nearest neighbour is the competing route, and it is winning. Lab-Grown Organs holds organoids, bioprinting, decellularised scaffolds and xenotransplantation — everything built outside the body. Speculative From a patient's point of view a regenerated kidney and a gene-edited pig kidney are the same product category: restored organ function without a human donor. Frontier The difference is which wall you pay at — immunology for the xenograft, a 0.27 per cent embryo yield for the chimera, or biology we do not have for regeneration.

Frontier Three technologies feed this brief directly. Nanomedicine supplies the targeted delivery vehicles on which link 4 depends. Frontier Bioelectric Medicine holds the ion-channel and gap-junction programme whose organ-scale application is the exotic end of this subject. Frontier Cellular Rejuvenation holds partial reprogramming, whose dosing window is the window this brief needs and whose tumour ledger is the same unpaid bill.

Established Two more sit one step out. Limb Regeneration shares the fibrotic default and the finite-regeneration result but adds a proximodistal axis and a joint series that organ regeneration does not have to solve. Frontier Longevity Therapies shares the trial-design pathology on display in the TRIIM study: small, uncontrolled, surrogate-endpoint studies read as demonstrations. Speculative Organ preservation and machine perfusion sit adjacent in a less obvious way, because they are the technologies that decide how long a failing organ can be held while something else is tried.

Speculative One adjacency is institutional rather than technical. The FDA's roadmap to reduce animal testing will fund organoid and organ-chip maturation at scale over the next decade — as models, not as therapies. Established The nearest large pot of money in this neighbourhood is pointed away from the goal, and that is a structural fact about the field rather than an accident of one agency's priorities.

Established The regulatory hinge in this cluster turned in February 2025, and it turned for the competing route. The FDA cleared the first pig-kidney clinical trial — six patients initially, expandable to 50 — ending an era in which every human xenotransplant was a compassionate-use or expanded-access single case. Established The United Therapeutics EXPAND trial performed its first transplant on 3 November 2025 at NYU Langone: multicentre, open-label, a phaseless 1/2/3 design intended to support a licence application, six transplants across two centres, twelve weeks between the first and the second, 24-week endpoints covering survival, graft survival, GFR, quality of life, proteinuria and zoonotic infection, with an independent monitoring committee gating each cohort. Established eGenesis received IND clearance for EGEN-2784 on 8 September 2025 for a combined phase 1/2/3 study in dialysis-dependent waitlisted patients aged 50 and over, also with a 24-week primary endpoint.

Speculative Nothing equivalent exists for a therapy whose product is an instruction rather than a substance. A regeneration therapy has no comparator regulators have seen before, no agreed endpoint for the claim that an organ regrew as distinct from a functional surrogate, and a principal risk — neoplasia at some latency — that falls outside the follow-up window of any standard design. Handwave That is not a gap in the rules; it is the absence of a product class.

Established The funding institutions are legible and none of them is buying this. AFIRM's $40 million over five years covers six areas and organ regrowth is not one of them. Established The FDA's non-animal-methods roadmap — a combined animal and human toxicity database within three years, acceptance of new-approach-method data alongside animal studies, primate testing for monoclonal antibodies cut from six months to three where the data support it, and a new-methods default at three to five years — funds organoids as models. Established The commercial money is in replacements.

Established The vacuum is already occupied by people selling the promise. A 2016 census identified 187 US companies operating 351 clinics marketing direct-to-consumer stem-cell interventions for aging, diabetes, neurological and orthopaedic conditions at costs of tens of thousands of dollars, with no supporting trial evidence for the advertised indications. Speculative An unregulated market that precedes the science is not a side effect of slow regulation; it is what fills the space where a legitimate product class has not been defined.

Frontier The consent problem here is unusual because the principal risk arrives late. A therapy that pushes cells back into the cycle carries a neoplasia risk with a latency exceeding any plausible trial follow-up, offered for indications — heart failure, chronic kidney disease — where the alternative is not immediate death. Consent to an unquantified cancer risk in exchange for an unquantified functional gain is a harder document to write than consent to a transplant.

Established The research model that produced this cluster's best data raises its own question. The 61-day single-edit kidney study and several other results in this brief were performed in brain-dead human decedents maintained on support for pre-planned observation periods. Frontier The scientific value is high, and consent is given by families for a study whose subject cannot benefit. Speculative This is a genuinely new research category and it has grown faster than the ethical literature covering it.

Speculative The competing route has an animal-industrial dependency this one does not. Xenotransplantation at clinical scale requires designated-pathogen-free donor herds bred, housed and slaughtered to a transplant-grade standard, permanently. Speculative A world that solves organ failure by regeneration needs no herds; a world that solves it by xenografts needs them indefinitely, and that accounting rarely appears next to the survival figures.

Handwave Chimeric organogenesis compresses the most ethics into the fewest sentences. Producing human-cell-dominated organs inside pig embryos worked at roughly 0.27 per cent efficiency, and the authors themselves note that knocking out brain and germline developmental genes in the human donor cells may be necessary for the approach to proceed. Speculative That is a research programme whose stated method is to prevent human cells from contributing to a pig's brain or gametes, and it deserves to be read out loud before it is normalised. Established Meanwhile the direct-to-consumer clinic market sells regeneration today to people who cannot evaluate it, which is the concrete, present-tense harm in this subject.

Speculative If in-situ regeneration worked, the transplant system would not be improved — it would become unnecessary. A waiting list of 109,817 people, 12 deaths a day, an allocation bureaucracy, a donation ethic and a lifelong immunosuppression market all exist because organs are scarce objects that must be moved between bodies. Speculative A therapy that makes an organ rebuild itself removes the object, and with it the scarcity, the allocation problem and the immunosuppression.

Frontier The likelier world is the replacement world, and it has a different shape. On current trajectories organ failure becomes survivable through manufactured and xenogeneic organs, which implies a permanent immunosuppressed population, a permanent pathogen-free herd infrastructure, and a per-patient cost structure that scales like surgery rather than like a drug. Speculative That world is reachable and is being built. It is also the world in which the regeneration route stays unfunded, because the demand it would have served has already been met.

Speculative The demographic consequence is the one worth stating plainly. Organ failure is a major terminal pathway in aging populations, and converting it from terminal to episodic would change the shape of late life more than any single longevity intervention now in trials — without touching the aging process at all. Frontier That is why Longevity Therapies and this brief are more closely coupled than either field's rhetoric admits. Handwave A civilisation able to regrow organs on demand would also have to decide, for the first time, how many times a body is entitled to do it — a question the finite-regeneration result suggests biology may answer before policy does.

These horizons track results and trial registrations rather than programme announcements, and each is written so it can be checked against the record when it arrives:

• 10 yr: Frontier Expect the xenotransplant trials to report: both current programmes carry 24-week primary endpoints, and by the mid-2030s the record will show whether any human recipient has passed one year on a pig organ, which nobody has done. Frontier Expect the single-edit question to be settled empirically, because it is cheap: a decedent series comparing minimally-edited against ten-edit kidneys resolves it in about a year. Speculative Expect partial reprogramming to reach a first-in-human trial in a localised indication — eye, skin, or one muscle group — rather than in a solid organ. Speculative Expect no demonstration of in-situ regeneration of a human solid organ, and expect the headline results to keep coming from invertebrates and neonatal rodents.
• 25 yr: Speculative If the tumour ledger is measured in a large animal and the number is tolerable, the plausible sequence is a localised cardiac or renal reprogramming trial with a functional endpoint, then slow expansion of the treated fraction. Speculative If it is measured and the number is not tolerable, the correct inference is that the suppressed-programme hypothesis was right about the suppression and wrong about the price. Frontier Either way, expect ectopic organogenesis — growing tissue somewhere permissive rather than where the organ was — to stay ahead of in-situ regeneration, because it is already in phase 2a and does not require overriding the scar. Frontier Expect the replacement route to have consolidated: a licensed xenogeneic or manufactured organ product inside this window is more likely than not, and it sets the comparator any regeneration therapy must beat.
• 50 yr: Speculative Restoring a defined organ function in place, in humans, is credible in this window only if it arrives through the reprogramming and delivery stack rather than through anything resembling a blastema. Speculative The organ that yields first will be the one with the simplest architecture and the clearest functional endpoint — endocrine tissue, then liver, then kidney, with heart and lung last. Handwave A regenerated organ that restores architecture rather than only function — lobar structure, nephron number, alveolar interface — has no plausible route in this window and no programme working towards it.
• 100 / 250+ yr: Handwave On these horizons the question stops being whether a human organ can regrow and becomes whether the mammalian tradeoff can be renegotiated: a body with deer-grade tumour suppression and an inducible growth programme is a coherent thing to describe, and it is not a therapy, it is a different animal. Handwave Nothing in current comparative genomics supports a small-number-of-loci route to that state, and a brief that puts a date on it is doing the handwave rather than reporting one.

• Depends on No brief on this map produces the result this one waits on. The binding link is a scientific unknown — the neoplasia incidence of reprogrammed tissue in a large animal over years — and no other brief is positioned to produce it, so no typed depends-on edge is claimed. The nearest suppliers are Nanomedicine, which supplies targeted delivery, and Cellular Rejuvenation, which supplies the reprogramming dose-response. Both are shared bottlenecks rather than dependencies: neither is generating the tumour data this brief needs, and all three are waiting on the same measurement.
• Requires (not on this map) Two non-brief constraints are recorded, and only one of them belongs to this route. A regeneration therapy is an instruction rather than a substance: it has no comparator regulators have seen, no agreed endpoint distinct from a functional surrogate, and a principal risk whose latency exceeds any standard follow-up window, so the product class has to be created before a trial can be designed. The second constraint belongs to the competitor, and is recorded because the competitor decides whether this route is ever funded: clinical xenotransplantation runs on designated-pathogen-free donor herds bred and housed to transplant-grade standard in perpetuity, and that supply chain is being built now. A regeneration therapy would need no herds, which is its strongest long-run argument and, today, precisely why nobody is paying for it.
• Enables If in-situ regeneration of a solid organ were demonstrated, the first thing it would change is not on this map — it would collapse the transplant allocation system and the immunosuppression market that sit around it. Within the map, it would remove the demand case that currently drives Lab-Grown Organs, and it would supply Limb Regeneration with the one thing that brief lacks: a demonstrated method for overriding the mammalian fibrotic default in a living adult. No typed enabling edge is claimed, because each consequence is conditional on a result that does not exist, and a typed edge would assert a dependency the record does not support.
• Adjacent Comparative regeneration biology and the invertebrate model systems; developmental genetics; bioelectricity and ion-channel pharmacology; transplant immunology, which supplies every human data point about what a failed organ replacement looks like; oncology, which owns the risk this brief cannot price; organ preservation and machine perfusion; and lipid-nanoparticle formulation science. Within this map: Lab-Grown Organs, Limb Regeneration, Bioelectric Medicine, Cellular Rejuvenation, Nanomedicine and Longevity Therapies.

Established “The liver regenerates.” It compensates. Living donors reach about 81 per cent of preoperative volume at one year and are still short at four, without restored lobar architecture. This is the claim most often offered as the existence proof for human organ regeneration, and it proves something else: mass restoration by growth of what remains.

Established “The heart cannot regenerate at all.” Human cardiomyocytes turn over throughout adult life, and neonatal mice regenerate a resected ventricular apex outright until postnatal day seven. The deficit is a rate and a window rather than an absolute, which is a more useful thing to know, because rates and windows have mechanisms.

Frontier “A dedicated stem-cell compartment is what regenerators have and we lack.” An early-branching flatworm regenerates its whole body with no canonical neoblasts, using division-competent cells resident inside differentiated tissues. The neoblast is a good story rather than a general explanation, and the misconception matters because it points intervention at the wrong target.

Established “The TRIIM trial reversed human aging and regrew the thymus.” Nine men, no control group, thymic fat replacement on MRI in seven of nine, a mean 2.5-year shift in an epigenetic-clock estimate, and a peer quoted in the coverage calling the results not rock solid because the study was very small and not well controlled. The result may be real; the study cannot establish it.

Established “Regenerative capacity, once switched on, is unlimited.” Mouse digit tips regenerate four times and then stop, with a monotonic decline in regenerated bone volume, a 65.7 per cent osteoblast deficit at the failing cycle, and impairment spreading to neighbouring un-amputated digits. Speculative This is the most under-appreciated result in the mammalian regeneration literature, because it converts the goal from a switch into a budget.

Established “Deer antlers prove mammals can regrow organs, so we could too.” Antler regeneration is coupled to annual disposal of the organ and to strong positive selection on p53-pathway genes. The transferable lesson is about cancer suppression rather than growth, and it cuts against the optimistic reading rather than supporting it.

Established “Lab-grown bladders proved we can grow organs; hearts are just bigger.” The 2006 cohort implanted hollow, thin-walled, low-metabolic-demand structures the host perfuses by ingrowth, and solid organs fail at vascularisation, which is a different problem rather than a larger version of the same one. Established Relatedly: a 2016 census found 187 US companies operating 351 clinics selling unproven stem-cell interventions at tens of thousands of dollars a course, counts quoted here from a university report because the primary could not be opened in this pass.

Frontier “Pig organs need about ten gene edits to work.” A GGTA1-knockout-only pig kidney — one edit, no CD40 blockade, no other modification — supported dialysis independence for a pre-planned 61 days in a human decedent, and its day-33 antibody-mediated rejection was completely reversed with plasma exchange, C3/C3b inhibition and anti-thymocyte globulin. Frontier The marginal contribution of edits two through ten is not established, and the parties best placed to test it are the ones selling ten-edit animals. Established Note also that the preclinical benchmark does not lead the clinical record: ten-edit kidneys in six baboons gave a median survival of 261 days and a maximum of 337, against a human record of 271 days — an unusual ordering for a preclinical-to-clinical comparison, and one the field has not explained.

Established “The 271-day xenotransplant record belongs to the NYU Langone group and involved 69 gene edits.” It does not, and it did not. This brief's source pass found the record misattributed four separate ways in a single 2025 peer-reviewed review, which tabulates the case as NYU Langone / Montgomery, 69 gene-edited, eGenesis. Established The 271-day patient is Tim Andrews, transplanted at Massachusetts General Hospital by the Riella and Kawai team on 25 January 2025, with the graft removed on 23 October 2025 for declining function; he returned to dialysis and remains listed. Established The NYU Langone case is a different patient, Towana Looney, who received a ten-edit United Therapeutics kidney and reached 130 days. Established And the 69-edit figure conflates two kinds of change: eGenesis describes EGEN-2784 as three glycan-antigen knockouts plus seven human transgenes plus genome-wide inactivation of endogenous retroviruses — roughly ten functional edits, plus dozens of retroviral-locus disruptions. Repeating “69 gene edits” without that decomposition quotes a number whose denominator is a different kind of thing.

Established “A high-prestige journal will have the date right.” A Nature Biotechnology news item on the same case, DOI 10.1038/s41587-025-02912-5, dates the transplant to February 2024 and the removal to the end of October 2025 — an interval of about twenty months, inconsistent with the same item's own reference to eight months and with the 271-day figure reported everywhere else. Established The arithmetic settles it: 23 October 2025 minus 271 days is 25 January 2025, which is the date in the company's own release. Speculative The object lesson is worth more than the correction. In a field whose load-bearing primaries sit behind publisher blocks, a company release and a subtraction beat a journal news desk, and errors of this kind propagate into review tables that later readers treat as the record.

Frontier “Chimeric pig-human organs are nearly here.” The headline number is 40 to 60 per cent human-derived cells in metanephric tubules. Established The number nobody quotes is the denominator: 1,820 embryos transferred to produce five chimeric fetuses at embryonic days 25 and 28 — about 0.27 per cent, at a developmental stage far short of a functioning organ.

Speculative “Mammalian regeneration is simply suppressed, and one brake releases it.” The hypothesis deserves its steelman: the machinery is retained and held down by tumour-suppression and immune programmes selected for in large, long-lived, cancer-prone animals, with deer as the proof, having relaxed the growth constraint only after strengthening p53. Speculative What would have to be true is that a defined set of loci gates the programme, and that relaxing them does not produce tumours at a rate that dominates the benefit. Frontier The direct test is the p21-null mouse literature, which is contested and whose primary this pass could not open. Handwave The one-brake version specifically is unsupported: the reference regenerating vertebrate carries a 32-gigabase genome, roughly ten times human, and lacks Pax3 entirely with Pax7 having assumed its functions. That is not a mammal with a switch flipped; it is a differently built animal, and an argument that leans on the axolotl analogy without that number is doing the handwave rather than reporting one.

Speculative “The bioelectric route rewrites pattern without touching DNA.” The programme claims that resting membrane potential distributions encode target morphology and can be rewritten by manipulating gap junctions and ion channels; the exhibits include gap-junction blockade reportedly producing head shapes characteristic of other planarian species in genetically wild-type animals, voltage control of eye patterning in Xenopus, and proton-pump-dependent induction of tail regeneration. Frontier The frog and flatworm results are striking and come overwhelmingly from one laboratory group. Handwave The mammalian translation is unpublished, and the load-bearing planarian paper could not be opened in this pass; it is cited from the laboratory's own verified publication index, without figures.

Speculative “The immune system is the block, and it is modifiable.” The minority position holds that regeneration-competent animals mount a qualitatively different injury response, that macrophage depletion abolishes salamander limb regeneration, and — the extraordinary version — that enhanced wound closure in one mouse strain is transferable to non-healer mice by faecal microbiome transplantation. Speculative If the last of those held up, a substantial part of the mammalian deficit would be environmental and modifiable within a human lifetime — the most actionable hypothesis in this brief. Handwave It is a single report this pass could not open, its replication status is unknown, no figure from it is quoted here, and it is named because a claim that size should be checked rather than left out.

Frontier “The field's failures were clinical bad luck.” The defining catastrophe of tissue engineering — the synthetic and decellularised trachea implants — ended in an adjudicated finding of research misconduct and retracted papers, not merely in dead patients; this brief's source pass could not open the primary journal statement of that verdict, so no patient counts are given here. Speculative The shape of the failure is the transferable part, and it was epistemic before it was surgical: the first-in-human step is where a field's evidentiary standards get tested, and this field has failed that test once already.