The Immortality Industry

On 28 January 2026 the United States Food and Drug Administration cleared an investigational new drug application filed by Life Biosciences for a gene therapy called ER-100. The decision was procedural and the document is unremarkable. What it authorised is not. For the first time, a therapy whose stated purpose is to partially reset the biological age of human cells was permitted to be tested in people.

The first patients will be treated in the eye, for glaucoma and for a condition called non-arteritic anterior ischaemic optic neuropathy, and the first safety data are expected in late 2026 or early 2027. The choice of organ is not incidental, and the reason for it explains a great deal about the industry that has formed around this idea.

The oldest wish

The desire of the human being to attain the characteristics of a God is in the very DNA of existence, which lives through us, behind a cluster of concepts, waiting for the moment to reach magnificence and splendour. Life being life itself.

Every civilisation has written this down. What is new is not the wish but the balance sheet behind it, and the fact that it now has a regulatory file number.

The money

Altos Labs has raised about $5.56 billion. That is, as far as can be established, the largest sum ever assembled around a single biological hypothesis, which is that the ageing of a cell is written in reversible marks rather than in permanent damage. Calico, founded by Alphabet in 2013, has had at least $3.5 billion committed, and its first drug candidate failed its trial in January 2025. Retro Biosciences has gone from a $180 million seed to more than $1.18 billion raised at a $1.8 billion pre-money valuation. NewLimit more than doubled its funding when Eli Lilly took a direct equity position in October 2025, the first time a top-five pharmaceutical company had bought into a cellular reprogramming asset directly.

Across the sector, roughly $4.8 billion has been raised since 2021, at a current rate somewhere between $200 million and $300 million a quarter. Alongside the private money sits a different kind: Hevolution, the Saudi foundation established to treat ageing as a process rather than a set of diseases, has put $101 million behind the XPRIZE Healthspan competition together with the prize organisation. That contest drew 602 teams from 58 countries, has narrowed to 20 finalists, and from 2026 to 2029 will run clinical trials of up to a year each, in adults between 50 and 90, with the goal of restoring muscle, cognition and immune function by at least ten years.

Read the list and a pattern appears immediately. This is not a pharmaceutical industry in the ordinary sense, where money follows a validated mechanism. It is capital arriving ahead of the science, in the hope of creating the conditions that would let the science be tested at all.

What these companies actually do

The field is organised around a framework published in 2013 and revised a decade later, which enumerates the hallmarks of ageing: genomic instability, telomere attrition, epigenetic alteration, loss of proteostasis, mitochondrial dysfunction, cellular senescence and several others. It converts a vague process into a list of targets, and each target has drawn its own companies. Grouped by what they are physically attempting, rather than by who owns them, there are seven lines of attack.

Resetting the cell

This is where the money is, and it rests on a discovery made in 2006. Four genes, now called the Yamanaka factors, can return an adult cell to an embryonic state; the finding won a Nobel Prize and created the entire field of induced stem cells. Doing that to a living body would be fatal, because a cell that forgets what it is becomes a tumour. The bet of this generation of companies is that the factors can be applied partially and briefly, erasing age-associated marks while the cell keeps its identity.

Altos Labs assembled the field's founding figures: Shinya Yamanaka, who discovered the factors, as senior scientific adviser; Juan Carlos Izpisúa Belmonte, who demonstrated partial reprogramming in mice; and Steve Horvath, who built the epigenetic clock that the field uses to measure itself. It runs institutes in San Francisco, in Cambridge in England and in San Diego. What it has produced so far is basic science and instruments rather than a clinical candidate, which is what $5.56 billion buys when a field is still deciding what is true.

Life Biosciences, cofounded by the Harvard geneticist David Sinclair, holds the clearance described at the top of this piece. ER-100 uses three of the four factors, the combination known as OSK, deliberately leaving out the one most associated with cancer, and delivers them to the eye in a viral vector for a limited period. It is the only authorised human trial of rejuvenation in existence.

NewLimit, founded by the Coinbase chief executive Brian Armstrong, takes a narrower route: rather than the whole organism, specific cell types, beginning with liver cells and T cells of the immune system. Its method is industrial screening of transcription factor combinations with machine learning to choose among them. Eli Lilly's equity position followed.

Retro Biosciences, funded by Sam Altman, runs three programmes: autophagy, which is the process by which a cell clears its own debris, plasma-inspired therapies, and reprogramming. Its lead asset, RTR242, is an oral autophagy enhancer aimed at Alzheimer's disease, and by May 2026 the company had dosed its first patient in its first human trial. Retro also produced the clearest example yet of the two subjects of this newspaper meeting: OpenAI built it a small model specialised for protein engineering, called GPT-4b micro, which designed new variants of two Yamanaka factors. The redesigned proteins, named RetroSOX and RetroKLF, were reported to raise the expression of reprogramming markers roughly fiftyfold against the standard versions, with early indications of improved DNA damage repair.

Shift Bioscience attacks the same problem from the cheapest possible direction. Instead of testing combinations in the laboratory, it simulates cells computationally and searches for genes that rejuvenate without causing the cell to lose its function, then tests only the survivors.

Removing the cells that will not die

As tissue ages, some cells stop dividing and refuse to die, emitting signals that damage everything around them. Clearing them selectively has produced striking results in animals. Unity Biotechnology carried the idea furthest in humans with foselutoclax, an inhibitor injected directly into the eye, and reported durable improvements in vision from a single dose in a peer-reviewed journal in April 2025. In September 2025 its board approved the company's dissolution.

Imitating hunger

The oldest and least glamorous line. Restricting calories extends life in nearly every species tested, and two drugs are thought to mimic parts of that effect: rapamycin, which suppresses a growth-signalling pathway, and metformin, a diabetes drug of sixty years' standing. This is where most of the public's attention sits and where the human evidence is thinnest, which is examined below.

Mining human data for targets

Rather than beginning with a mechanism, some companies begin with people. BioAge Labs works from longitudinal cohorts, blood samples taken from individuals followed for decades, and looks for the molecular differences between those who aged well and those who did not. Calico, Alphabet's subsidiary, has spent more than a decade on the basic biology, including species that age unusually, and produced one clinical candidate, which failed.

Letting software design the molecule

Insilico Medicine runs generative chemistry end to end, from target identification to candidate. Isomorphic Labs, spun out of DeepMind, applies protein structure prediction to drug design. Shift's virtual cells and Retro's engineered factors belong to the same category. The common claim is not that software understands ageing, but that it shortens the search between a hypothesis and a molecule worth testing.

Replacing the part instead of repairing it

The commonest way for a modern person to die of ageing is organ failure, and the shortage of transplantable organs is absolute rather than economic. United Therapeutics is running EXPAND, the first regulated clinical xenotransplantation programme, using a kidney from a pig carrying ten gene edits, in patients with end-stage renal disease unlikely to receive a human organ; the first transplant under that protocol took place at NYU Langone Health on 3 November 2025. eGenesis is running a three-patient pilot at Massachusetts General Hospital with a kidney carrying 69 edits. More than seventy patients are expected to receive life-sustaining pig kidneys within the coming year.

This is the part of the industry that has dates, protocols, patients and numbers. It reverses nothing and claims nothing. It replaces what failed, which is what medicine has always done.

Changing the rules

The last group does not make a therapy at all. Hevolution funds the science on the premise that ageing is a treatable process. XPRIZE Healthspan exists to force into being the measurement that the field lacks. And one company has gone further than any of them, by a route nobody expected.

The crack in the wall is a dog

Loyal makes longevity drugs for dogs. Its candidate LOY-002 is a daily pill for senior animals, and its purpose, stated plainly on the label it is seeking, is to extend lifespan.

Three technical sections must be accepted before a veterinary drug can reach the market in the United States. Loyal has two of them. The regulator accepted its reasonable expectation of effectiveness, and in January 2026 accepted its target animal safety package, which is the first known safety acceptance for a lifespan extension drug of any kind. What remains is the manufacturing review, after which the company intends to apply through a conditional pathway built for therapies addressing unmet needs. Acceptance is not approval and the distinction matters. But if it completes, LOY-002 becomes the first drug ever approved by a regulator for the extension of lifespan itself, in any species, and it could reach the market within the year.

That is the significant fact buried in this industry, and it has almost nothing to do with the billions. A regulator has agreed to evaluate lifespan as an outcome. The precedent is being set in a veterinary office rather than a hospital, for an animal that ages seven times faster than its owner and therefore produces an answer in years rather than decades, which is precisely why it got there first.

The two walls

Neither of the obstacles that have held this field back is biological.

The first is measurement. To run a trial you need an endpoint, and the natural endpoint for an ageing therapy is death, which arrives too late to be useful to anyone designing a study. The field's answer has been the epigenetic clock, a statistical model that reads chemical marks on DNA and returns an estimate of biological age. The clocks correlate well with chronological age, which is precisely the problem: a measure that tracks the calendar does not prove that moving the measure moves the outcome. No biomarker of biological age has been validated and accepted by a regulator. This is not a side issue, it is the central one, and it is explicitly what the XPRIZE competition has set out to force into existence.

The second wall is regulatory, and it is simply stated: ageing is not an approved indication. A regulator licenses treatments for diseases, and growing old is not classified as one. Every programme in this industry must therefore present itself as a treatment for something else, and the choice of that something else shapes the science. It is why the first authorised rejuvenation trial treats two eye conditions. The eye is small, so a dose is small; it is partly shielded from the immune system; it can be imaged directly; and one eye can be treated while the other serves as a control inside the same patient. It is, in regulatory terms, the cheapest place in the body to find out whether an idea works.

What the evidence actually shows

The honest summary is that the field has one authorised human trial of rejuvenation, which has not yet reported, and a thin record elsewhere.

The most famous study in longevity medicine is the one that has never run. TAME, a trial of metformin in three thousand people aged 65 to 79, designed around a composite endpoint of cardiovascular events, cancer, dementia and death, has been planned for more than a decade and as of March 2026 had still not enrolled a patient, for lack of funding. It has produced no results because it has produced no participants, and any claim that it has shown metformin to slow ageing is false.

Rapamycin has human data, and it is modest. The PEARL trial, a 48-week randomised, placebo-controlled study of intermittent low doses in healthy adults, found significant improvements in lean tissue mass and in self-reported pain among women taking the higher dose, and no significant change in visceral fat. That is a real result, obtained properly, and it is a long way from what the word longevity implies in a press release.

The most instructive case remains Unity. The biology worked and the business did not, which is the lesson the sector has been slowest to absorb: proof of mechanism is not a company, and a treatment that cannot find a market large enough to repay a decade of trials will not reach anyone regardless of whether it works.

The consumer layer

Underneath the laboratories sits a market that is not waiting for evidence. Blueprint, the company built around Bryan Johnson's protocol, raised $60 million in October 2025 and in February 2026 announced a programme priced at a million dollars a year with three places. Longevity clinics, which sell testing, infusions and monitoring, are a market of roughly $4.2 billion. At the far end, cryonics preserves between 500 and 650 people worldwide, at around $220,000 for the most comprehensive arrangement, on the proposition that a future technology will be able to do something that no present technology can.

These are not the same industry as the one holding the regulatory file, and the distance between them is the distance between a hypothesis in trial and a product on sale.

What would have to be true

Three things, and none of them is a discovery.

A biomarker of biological age that a regulator will accept as an endpoint, because without one every trial takes a lifetime and no one will fund it. A disease indication broad enough that a therapy which happens to slow ageing can be approved and paid for, which is the wall the dogs are currently walking through. And a distinction that the field itself often blurs: the difference between extending healthspan, the number of years lived in good function, and extending maximum lifespan, the hard ceiling of the species. Over the last century medicine moved the average enormously and left the ceiling where it was. The longest confirmed human life ended at 122, and it did so in 1997.

Everything described here, the five billion dollars, the three Yamanaka factors, the pig kidneys, the eye with the better vision and the pill for old dogs, is an attempt to find out whether that ceiling is a law or merely a fact that no one has yet had the tools to change.

Because if as a race you have the opportunity to reach the maximum possible without hitting other people's interests, and respecting the destructive protocols of others, you can do it, right? After all, this is a game of protocols cohabiting, and as long as they respect themselves, everything makes sense. So immortality is the new path to be reached publicly for humanity, represented by all the figures we have seen, and the young promises will surely appear. I imagine the protocols must be respected by them too, in order to cohabit in peace and harmony between protocols.

Sources

Regulatory actions. The clearance of Life Biosciences' investigational new drug application for ER-100 on 28 January 2026, first reported by MIT Technology Review and Fortune. Loyal's announcement of the acceptance of its target animal safety package for LOY-002 (13 January 2026), with the company's own account of what the milestone means here.

Published results. Unity Biotechnology's phase 2 results for foselutoclax (company release), published in a peer-reviewed journal in April 2025. The PEARL trial of intermittent low-dose rapamycin (paper). The first clinical renal xenotransplantation study and its protocol (paper).

Programmes and announcements. XPRIZE Healthspan, its structure, finalists and timetable (competition page). OpenAI's account of the model built for Retro Biosciences and the engineered reprogramming factors (OpenAI).

Press reports, identified as such in the text. Sector funding totals, company valuations and the status of the TAME trial are drawn from trade and financial press; where a figure comes from a press account rather than a filing or a published paper, the text says so.