Biotech & Longevity · Wednesday, 22 July 2026

01 · Briefing · what happened

Why evolution never weeded out the genes that make us sick when we're old

Biotech & Longevity 4 min 80 sources

A new analysis argues the diseases of old age persist because the genes behind them helped us reproduce young — as a wave of billion-dollar deals, late-stage failures, and a safer Alzheimer's drug reshaped the week.

Key takeaways

New genetic evidence supports an old idea: we suffer diseases of old age because the genes behind them helped us reproduce young, so evolution never removed them.
Three billion-dollar-plus acquisitions landed in two days, even as a string of drugs — for osteoarthritis, sickle cell, an eye disease, and chronic cough — quietly failed or were dropped.
A next-generation Alzheimer's drug's advance is a smaller downside, not a bigger effect: lower doses and fewer dangerous brain bleeds.

The biggest idea in biotech this week wasn’t a drug or a deal. It was an old question, asked with new data: why does evolution let us fall apart at all? If natural selection weeds out harmful genes, why are heart disease, cancer, and dementia written into almost every ageing body?

The genes evolution never fixed

A new analysis in Nature Ecology & Evolution offers a hard-nosed answer. The same gene variants that raise your risk of disease later in life are, on average, tied to more children earlier in life [23]. Evolution “sees” a gene by whether it helps you reproduce. A variant that gives you an edge at 25 and a tumour at 65 gets passed on anyway — the harm arrives long after the copies are made. So the disease genes were never under pressure to leave. Some were actively kept.

This idea, called antagonistic pleiotropy, has been around since the 1950s. What’s new is the scale of the genetic evidence lining up behind it. It reframes ageing not as a machine wearing out, but as a bill that always comes due after the part of life evolution was counting.

Two other papers this week circled the same theme. Restoring a thymus-gland hormone called thymulin in old mice calmed the low-grade inflammation of ageing and made cancer immunotherapy work better [25] — though, as ever, a result in mice is a hypothesis about humans, not a treatment. And zoo elephants, it turns out, are living markedly longer than they used to, challenging the assumption that captivity shortens their lives [38]. Longevity, across species, looks less fixed than we thought.

A week of billion-dollar consolidation

While the scientists asked why we age, the dealmakers spent. Three acquisitions worth more than a billion dollars each landed in 48 hours.

Novartis agreed to buy the British biotech Myricx for up to $1.5 billion [13], ending its chief executive’s long public reluctance to enter the crowded field of antibody-drug conjugates — cancer drugs that use an antibody as a guided missile to carry a toxin straight to a tumour cell. Tempus, a cancer-data company, bought the genomics specialist Personalis for $1.5 billion [15][65]. And Samsung Biologics launched a $1.8 billion all-cash bid for PolyPeptide, a maker of the peptide ingredients behind drugs like the GLP-1 weight-loss shots [72].

The money is real but uneven. British biotechs raised £2.05 billion ($2.7 billion) in venture funding last quarter, a five-year high — yet most of that was a single £1.6 billion round for one Alphabet-backed AI-drug company, and firms are still holding off on public listings, blaming geopolitical nerves [9]. Cash is flowing to a few big bets, not the whole field.

A brutal week for the pipeline

For every deal, a drug died. Kolon’s cell-based gene therapy for knee osteoarthritis — the wear-and-tear joint disease that afflicts hundreds of millions — failed its phase 3 trial [14], the large, final test against a placebo that decides approval. A year after injection, patients were no better than those given a dummy. Osteoarthritis is a graveyard for drugs; several companies have crashed on the same rock.

Agios dropped its sickle-cell disease programme after underwhelming mid-stage data [33]. Johnson & Johnson axed a gene therapy for an advanced eye disease after a key readout [54]. GSK ended its chronic-cough treatment [49]. None of these were safety scandals — they were drugs that simply didn’t clear the bar, quietly shelved. Most drugs that enter human testing never reach a patient, and this week made the attrition visible.

A safer swing at Alzheimer’s

The one clinical bright spot came from Alzheimer’s. The first generation of anti-amyloid antibodies — drugs that clear the sticky plaque linked to the disease — worked modestly but carried a real danger: brain swelling and bleeding in some patients. A next-generation antibody uses a more efficient system to cross into the brain, so it can be given at lower doses with fewer of those serious bleeds [8]. The advance here isn’t a bigger effect. It’s a smaller downside — often what actually decides whether a drug reaches people.

The quiet, hopeful one

The under-covered story closes on the hardest cases. The Broad Institute, Boston Children’s Hospital, and the Jackson Laboratory launched a joint effort to build gene therapies for rare diseases [59] — the thousands of conditions, most of them affecting children, that are too uncommon for any one company to justify developing a drug for alone. Pooling the science is how those patients get a shot at all. It’s early, and building a therapy is not the same as proving one works. But it points at the diseases the market usually leaves behind.

02 · Lesson · why it matters

What a scorekeeper can't see, it never charges for

Evolution only counts a gene's effect up to the age you reproduce — so any harm that lands after that was never on the bill, and never had to go.

The question hiding in the headline

A briefing about deals and dead ends opened with a stranger question: why do we get sick when we’re old at all? If nature removes what hurts us, why does almost every ageing body carry the seeds of cancer, heart disease, and dementia?

The answer this week’s science offered is not about wear. It’s about a scorecard. And once you see it, you find the same shape running through boardrooms, elections, and your own choices.

The window that does the counting

Natural selection is a scorekeeper with one metric: how many surviving children a gene helps you make. That’s it. A gene that helps you grow strong at twenty, win a mate, and raise young gets copied forward. A gene that harms you at seventy does not get removed — because by seventy, the copying is long done.

So there is a window. It runs from birth to the end of your reproductive years. Inside that window, every effect is counted. Outside it, nothing is. A gene that gives you an edge at twenty-five and a tumour at sixty-five reads, on the scorecard, as pure gain. The tumour lands after the window closes. It was never charged.

That is the new evidence: the gene variants that raise disease risk late in life are, on average, the ones tied to more children early on. Evolution didn’t fail to fix our old-age diseases. It kept the genes that cause them, because inside its window they looked like winners.

The cost didn’t vanish — it moved

Here is the part that matters. The harm was never cancelled. It was moved to a place the scorekeeper doesn’t look.

This is different from a gene being weak or rare. These variants are common precisely because they scored well. The disease they carry is not a flaw the system missed. It’s the exact price of the thing the system rewarded — collected decades later, out of sight of the only clock that mattered.

Ageing, in this light, isn’t the body breaking. It’s the bill for a young body arriving in an old one. The strong metabolism that fed your twenties helps starve your sixties. The cell growth that healed you fast can, later, grow the wrong thing. Same gene. The benefit came inside the window; the cost waited outside it.

Now notice what this pattern is, stripped of biology. A system optimises for a number measured over a bounded stretch of time. Anything whose cost lands outside that stretch does not enter the number. So the system will happily accumulate that cost — will even reward the thing that creates it — because the harm is, to its scorecard, invisible.

A company judged on this quarter’s earnings will cut the maintenance, the training, the research whose bill comes in three years. Those cuts raise the score. The plant that fails, the talent that leaves, the pipeline that empties — all land after the window the quarter counts.

A politician judged at the next election will take the debt, the deferred repair, the slow damage that surfaces two terms later. An election cycle is a reproductive window. What ripens after it was never on the ballot.

A product warranty ends at twelve months, so the part is built to last thirteen. The score is “returns inside the window.” Everything after is free.

None of these require a villain. In each case someone is simply doing well on the scorecard they’re handed. The rot isn’t chosen. It’s what the measure can’t see.

Who’s inside this

You are, twice over. You live in a body tuned for a window that closed in your thirties — every ache of ageing is a cost evolution posted to an address it stopped reading. That is worth sitting with, because it quietly reframes decline. The diseases of age are not, mostly, your failures. They are the predictable second half of a deal your genes made before you were born, on terms no one consulted you about.

And you are the scorekeeper, too. Every choice you make is scored over some window — this month’s budget, this year’s health, the deadline in front of you. Whatever falls past the edge of that window is, to you right now, weightless. The younger self spending a body it won’t own for decades is running the exact same math evolution ran. The gap between what your scorecard sees and what your whole life will pay is not a moral failing. It’s the default setting of any mind that has to decide before the results come in.

What the whole looks like

The longevity scientists chasing these genes aren’t fixing a bug. They’re arguing with a scorekeeper that was never trying to keep you alive past your reproductive prime — it was indifferent to it. That’s a humbler and more accurate picture than “curing ageing.” You cannot see past your own window any more than evolution could see past its. The most any of us can do is widen it a little — count one more year, one more consequence, one more person the cost will reach — knowing the edge never disappears. It just moves.

03 · Lab · your turn

The Scorekeeper's Window

Play selection: keep gene variants scored only up to reproduction, then reveal the whole-life bill, and feel how widening the window — not changing the genes — is what turns winners into losses.

04 · Hope · carry this

The same science that explains why we age is now, for the first time, being used to argue with it — and a field that once only described the bill has started, gene by gene, learning to read a little further past the window evolution stopped counting.