Biotech & Longevity

Two people died of measles in Pennsylvania. The US national count still says zero. →

The new head of the US disease agency ordered her staff to challenge a state's finding, breaking a rule that has stood for decades. The agency has no investigators at either bedside, so its review can only add the deaths or leave them out.

Denmark offered half a million adults an RSV shot at random, then counted the hospital beds →

A trial of 511,535 people found the vaccine cut hospital stays from one winter virus by 70%. It also found that total hospital stays and total deaths barely moved.

Eight drugs had failed them. A cancer therapy left three of six needing none. →

Doctors in Berlin gave six people with severe rheumatoid arthritis a treatment designed for blood cancer. Every one improved, and three ended up on no arthritis medicine at all. Plus measles deaths in Pennsylvania, a one-shot cholesterol edit, and the day's other biotech news.

A heart drug failed because most patients were already on one that works →

AstraZeneca and Ionis explained a 1,400-person failure on Friday. Eight in ten volunteers were already taking a rival drug, and there was little harm left to prevent.

A patch of muscle under the skin, attached to nothing, did much of what exercise does →

Mice given a self-contracting graft of their own muscle cells gained lean mass, bone density, grip and stamina - and the graft moves nothing. Plus 37 more stories from the beat's day.

The first drug to lengthen life in pancreatic cancer is approved, at $39,800 a month →

The FDA cleared Rasonque six and a half months early for a cancer that had almost nothing to offer. Two thousand people were already taking it free.

A gene therapy halted over spots no child can feel, in a week medicine spent looking harder →

The FDA has paused Regenxbio's Hunter syndrome gene therapy after extra scans found small masses on the spines of five treated boys. None of them has symptoms, and nobody has ever routinely scanned the spines of untreated children with this disease - so there is nothing to read the finding against.

The first blind test of AI antibody design, and the week biology checked its own tools →

Twenty-nine organisations committed 511 AI-designed antibodies before a lab made a single one. On the ranking task all but one model lost to picking clones at random - and days later a sleuth found 54 ageing papers built on an antibody that binds the wrong organism.

The immune cell that multiplies after 100, in a week spent at the far edge of life →

A Japanese team found killer immune cells growing, not shrinking, in people past 110 - a week when longevity science, ageing clocks and the rarest-disease drugs all came down to how few people you can actually study.

The gene therapy that fixes a child's liver and leaves the family history alone →

A one-time infusion reaches children with an inherited sugar disorder, two children die in gene-editing trials abroad, and a new study finds a mother's age leaves a mark that is not written in DNA.

A cancer vaccine finally clears a phase 3, in a week of four approvals →

Merck and Moderna say a personalised mRNA shot kept melanoma from coming back in a 1,137-patient trial. Four FDA clearances landed alongside it, two gene-therapy deaths surfaced in China, and a quieter thread ran through the week: the handedness of a molecule.

One eye chart, a billion dollars: the week the yardstick decided everything →

EyePoint lost nearly a billion dollars in an hour because a vision drug missed on the measure the trial had picked, while winning on the one patients complain about. It was that kind of week: an approval built on a brand-new stand-in for remission, a Duchenne filing rewritten around a different body part, and one company that measured the emergencies themselves.

The switch, not the gene: a week spent on what a cell is allowed to read →

A company raised $90m for a drug that silences a gene without cutting DNA, and four journals reported on cells locked into the wrong state. Plus a big lung-cancer failure and two undisclosed deaths in China.

The deadliest Ebola outbreak on record, and a vaccine built for the wrong virus →

Congo's outbreak is on course to pass 2014-16 as the worst ever, against a species with no vaccine and no drug. Elsewhere: engineered cancer-fighting cells still alive ten years on, the first whole-body treatment tested for vitiligo, and blood immune cells found moving into the ageing brain.

A new class of cancer drug wins approval by changing a protein's shape →

The FDA cleared iberdomide for multiple myeloma on 13 August, the first of a class that works by reshaping a protein's surface rather than blocking it. The same week brought a map of what else that trick could reach, a cystic fibrosis pill that failed to fix a broken fold, and fresh work on the folds that go wrong in the brain.

One dose, twelve weeks: an LSD-based drug clears a second phase 3 for anxiety →

A 214-patient trial found a single psychoactive session still beating placebo on anxiety scores three months later - the first real movement in anxiety drugs since 2007. Much of the rest of the week was an argument about the opposite question: how long a drug should be taken, and what a body does to itself while it waits.

A vaccine order lands as measles hits a 35-year high, and the math of the crowd →

Trump ordered fewer childhood shots and the MMR vaccine split into three, as US measles reached its worst level since 1991 - a week where the science of collective protection ran straight into policy.

A deafness gene therapy reaches a child, in a week ruled by the delivery problem →

Getting a working drug to its target, past the body's own barriers, is often harder than inventing it - and this week's biotech news kept proving the point.

A blood test that hunts 50 cancers reaches the FDA, and the math that decides what a positive means →

US regulators will review Grail's Galleri, a single-draw blood test for more than 50 cancers, on September 23. It is 99.6% accurate at clearing the healthy, yet a big UK trial found it did not catch cancer earlier overall, because when a disease is rare, even a near-perfect test throws off false alarms.

The weight-loss pill arrives, and it rewrites how often you dose →

Britain became the first country in Europe to clear a once-daily obesity pill that rivals the weekly injections, as a busy week added a narcolepsy drug, the first mRNA flu shot, and a self-copying melanoma virus.

The gene that decides who suffers side effects from antidepressants →

A 114,000-person study found one liver-enzyme gene predicts who gets sick on common antidepressants, sharpening a bigger truth this week: the same gene can help one person and harm another depending on the setting it sits in.

A diabetes drug keeps protecting organs it was never aimed at →

New pooled trial data shows semaglutide guards the kidney, on top of the heart and the waistline. It is the reach of a drug that hits one receptor the body keeps in many places at once.

A living cancer drug: the FDA clears an engineered virus that copies itself inside tumors →

After two rejections, regulators approved Replimune's Tudriqev, a virus reengineered to multiply inside melanoma and burst it from within - part of a two-approval week that also cleared the first mRNA flu shot.

A vaccine for inherited cancer risk, and a week that redrew who's really at risk →

Britain began the first trial of an mRNA vaccine to prevent cancer in people born with Lynch syndrome, while two other studies reshaped who counts as genetically at risk.

The muscle pills that hid a hormone, and the invisible line every drug walks →

A US peptide company was jailed for selling unapproved muscle pills secretly laced with testosterone, pushing one buyer into psychosis. It is the loudest sign yet of a market that sells drugs with no dose you can trust.

An AI that read 24 million papers points to a new Parkinson's target →

New tools this week all attacked biotech's oldest problem - the space of possible molecules and targets is astronomically large, so the whole game is narrowing the search.

A first drug for the disease where the body attacks its own nerves →

The immune system dominated biotech this week - drugs to calm it in autoimmune disease, and drugs to wake it against cancer it had missed.

A heart drug lowered inflammation exactly as planned - and failed anyway →

Novo Nordisk's ziltivekimab hit its biological target but did not protect the heart, and it raised serious infections - a hard lesson in why the body's defenses resist a simple off switch.

The week medicine kept losing its footrace with resistance →

A phage-therapy result for stubborn urinary infections, two studies of how bacteria dodge antibiotics, and a cancer finding that shows the same escape logic - resistance is the theme running through this week's science.

The weaker drug passed. The stronger one failed. Same FDA week. →

In one week, an FDA advisory panel endorsed Replimune's twice-rejected melanoma drug, which was tested without a control group, and voted down Capricor's Duchenne heart therapy, which ran a full randomised trial. Plus GSK's cost cuts, a J&J cell-therapy bet, another Novo Nordisk stumble, and an Ebola vaccine only a charity would fund.

A blood test for colon cancer, and the rush to catch disease sooner →

The FDA cleared Freenome's blood test to screen for colorectal cancer, part of a wider push to spot disease years earlier. Plus a melanoma drug backed over its own reviewers' doubts, a Duchenne therapy rejected, and a weight-loss drug that cut heavy drinking.

A drug passed its trial. Its stock still fell 60%. →

MapLight's schizophrenia drug hit its main goal and lost two-thirds of its value the same morning, Lilly posted huge weight-loss numbers with murkier heart data, and an FDA panel rejected a Duchenne therapy - three trials this week judged less on whether an effect was real than on how big it was.

Fibromyalgia turns out to be written in the genes, and the brain →

A study across 2.5 million people found 26 genetic signals for fibromyalgia, planting a long-doubted condition firmly in the brain. Plus a weight-loss drug that missed the endpoint it also measured, an FDA fight over a Duchenne trial, and a gene-editing death kept quiet.

An FDA panel backed unproven peptides over its own scientists →

The week's biggest biotech story was a regulatory one - a US panel voting to widen access to barely-tested peptides. Alongside it: two large "linked to" health studies, a weight-loss drug that fell short on the heart, and a record measles year.