Biotech, in brief

Caribou Biosciences, a California biotech company, said on 6 October that it is stopping all work on its medicines and cutting most of its staff. [2]

Caribou was one of the first companies set up to use CRISPR, a tool that edits genes, in medicine. Jennifer Doudna, who won a Nobel Prize for CRISPR, co-founded it in 2011. [1]

Its main treatment, vispa-cel, is a CAR-T therapy for lymphoma, a blood cancer. CAR-T means a patient's immune cells are given new instructions to find and kill cancer cells. [3][1]

Approved CAR-T treatments are made from each patient's own cells, which takes weeks. Vispa-cel is made from healthy donors' cells, so it could be ready on the shelf. [2]

In an early trial reported this summer, vispa-cel held lymphoma back about as long as two approved CAR-T treatments, Breyanzi and Yescarta. [2]

The FDA, the US drug regulator, had already agreed the design of the large final trial that approval would need. [2][1]

Caribou could not raise the money to run that trial. Its chief executive, Rachel Haurwitz, blamed a funding climate that made the money impossible to find. [2]

An analyst at Leerink Partners, an investment bank, wrote that investors were unwilling to fund the trial. They doubted Caribou's method of matching donor cells to patients, and the strength of its data. [1]

Caribou held $113.8 million at the end of June and had 97 full-time staff in February. It has not said how many will lose their jobs. [2]

Investors and big drugmakers have moved to a newer idea: changing a patient's immune cells inside the body, with no cells taken out. Eli Lilly, Johnson & Johnson, AstraZeneca and Moderna have all entered it. [2]

Money is still reaching other young drug companies. In the same week KymaThera raised $80 million and RougeTx, a new company, raised $58 million. [4]

Caribou's board will now look at selling the company or its drugs. Haurwitz said she hopes another group will take the treatments forward. [2]

Why it matters. Watch whether a buyer takes on vispa-cel and runs the final trial the FDA already agreed. If nobody does, nobody will learn whether donor-cell CAR-T can match a patient's own cells. [2]

A gene therapy built on this week's Nobel-winning science improved light sensitivity in six of ten blind people, researchers reported in the New England Journal of Medicine. [6] All ten had retinitis pigmentosa, a group of inherited diseases that kill the eye's light-sensing cells. [6] One injection gives other eye cells a light-sensing protein, and goggles turn the world into light pulses those cells can read. [6] Some patients could find a door or walk along a line, and the effect lasted four to five years. [6] They cannot yet see faces. [6]

Why it matters. It works whatever gene caused the blindness, and the disease group affects more than 1.5 million people. [6] Nanoscope, a company using the same idea, is already waiting for an FDA decision on its own version. [7]

A Dutch court has blocked Merck from selling Keytruda Qlex, the under-the-skin form of its cancer drug Keytruda, in eight European countries. [4] Keytruda is a cancer immunotherapy, a drug that helps the immune system attack tumours. [4] The court sided with Halozyme, a company that owns patents on a way to push large drug doses under the skin quickly. [4] Halozyme already won a case in Germany and is suing Merck in the US. [4]

Why it matters. Bristol Myers Squibb and Roche paid to use Halozyme's method for their own under-the-skin cancer drugs, and Merck did not. [4] An analyst at TD Cowen said the ruling could improve Halozyme's path to a licence fee or another payment from Merck. [4]

Cerevance, a private Boston company, said its daily pill solengepras met the main goal of its final-stage trial in Parkinson's disease. [8] The trial had 341 patients who still had 'off' hours, when their usual medicine wears off and symptoms return. [8] At the higher dose, off time fell by 0.61 hours a day more than on a dummy pill after 12 weeks. [8] Patients started with about 5.65 off hours a day. [8]

Why it matters. Most Parkinson's drugs raise dopamine, a brain chemical the disease runs short of, and can cause jerking movements; this one works on a different protein. [8] Cerevance will now ask the FDA what it needs before it can apply for approval. [8]

Atsena Therapeutics said its gene therapy improved the sensitivity of the retina, the light-sensing layer at the back of the eye, in six of nine patients on its chosen dose. [10] None of three untreated patients improved. [10] The patients have X-linked retinoschisis, an inherited disease of the retina that affects about 30,000 men in North America and Europe. [10] It has no approved treatment. [10]

Why it matters. The trial is still small, and the data come from the company at an eye doctors' meeting this week. [10] Atsena expects final results in the first half of 2028 and to ask the FDA for approval by the end of that year. [10]

MimiVax's vaccine SurVaxM missed the main goal of a mid-stage trial in 233 people newly diagnosed with glioblastoma, the most common and aggressive brain cancer in adults. [9] The vaccine trains the immune system to attack cells that carry survivin, a protein found in nearly 95% of these tumours. [9] In patients aged 65 and under, 44% were alive without the cancer growing after a year, against 24% on a dummy. [9]

Why it matters. Most patients die within about 18 months of diagnosis, and no new drug has been added since 2005. [9] MimiVax says the results in younger patients give it the confidence to seek FDA approval anyway. [9]

The US government, Meta, Google DeepMind and Isomorphic Labs are joining Biohub, a charity founded by Mark Zuckerberg and Priscilla Chan, to build cell data for AI. [11] The three companies will put in $300 million together, and the US energy department more than $500 million over five years. [11] The goal is to record how cells react across many more conditions than science has measured. [11] A first dataset is due in about a year. [11]

Why it matters. The companies paying get early use of the data before it becomes public. [11] Rivercell, a Paris start-up building its own cell model, raised $25 million the same day. [12]

Cadrenal Therapeutics, a Florida company, is looking to sell or license its drugs, or the whole company. [5] In August it agreed with the FDA how to run a final trial of CAD-1005. [5] The drug is for heparin-induced thrombocytopenia, a rare reaction to the blood thinner heparin that makes the body form too many clots. [5] Cadrenal's shares fell from $7.22 in January to about $1. [5]

Why it matters. Like Caribou, it has a final trial already agreed with the FDA and is looking for someone else to pay for it. [5] Its chief executive says it has had strong interest from possible partners. [5]

Enliven Therapeutics has agreed with the FDA on a final-stage trial of relcobatinib, its pill for chronic myeloid leukaemia, a slow blood cancer. [4] The drug blocks the faulty gene that drives the disease. [4] The trial will enrol 450 people and test whether it beats another drug of the same kind. [4]

Why it matters. It would compete with Novartis' Scemblix and with a drug Merck bought for $6.7 billion. [4] An analyst at Jefferies wrote that recent Enliven data, and a slightly weaker outlook for the Merck drug, have improved how investors see it. [4]

Scientists at UC San Francisco changed immune cells into cancer-fighting CAR-T cells inside the bodies of mice, with no cells taken out. [3] Two tiny particles carried the CRISPR editing tools and the new genetic instructions straight to the cells. [3] One injection cleared leukaemia from nearly all the mice within two weeks, the team reported in Nature. [3] It also worked against a solid tumour. [3]

Why it matters. It has not been tested in people, and the team has started a company, Azalea Therapeutics, to try. [3] Jennifer Doudna, Caribou's co-founder, is among the authors, and this is the kind of treatment investors now prefer to Caribou's. [3][2]

Four large US-funded networks that test vaccines and treatments for HIV, tuberculosis and new infections face an uncertain future. [13] The Trump administration has delayed starting the process to renew their funding, which ends on 30 November 2027. [13] Last Friday the National Institutes of Health told HIV researchers to hold off on new studies that could not finish before then. [13] These networks also helped test the Covid vaccine. [13]

Why it matters. Scientists warn delays could put trial volunteers at risk and stall studies that depend on timing. [13] The US health department did not say whether or when renewal will start. [13]

Stanford Medicine scientists built a two-sided molecule that grabs BCL6, a protein that keeps many lymphoma cells alive. [14] It then joins BCL6 to other proteins that switch on the cell's own self-destruct genes. [14] In mice given it twice a day, lymphoma tumours disappeared within 11 days, the team reported in the journal Cell. [14]

Why it matters. It has only been tested in mice and needs more work before any human trial. [14] Two of the scientists founded a company, Shenandoah Therapeutics, that holds the licence. [14]

The World Health Organization advised that children under 10 should not be given obesity drugs such as Mounjaro and Wegovy, or weight-loss surgery. [15] Most of these drugs are licensed only for older children, but some doctors prescribe them younger. [15] Drug companies are testing them in children as young as six. [15] The WHO said care should focus on food and exercise. [15]

Why it matters. About 70 million children aged five to nine are obese worldwide, four times the share in 1990. [15] A WHO official asked when a child started on these drugs would ever stop, since there is no evidence they work as a short-term treatment. [15]

Surgeons in Edinburgh put frozen testicular tissue back into a 19-year-old who lost the ability to make sperm after chemotherapy. [16] The tissue was taken and frozen three years ago, before treatment for a serious blood disorder. [16] It holds stem cells that can make sperm after puberty. [16] It is the first such operation in the UK and the second reported anywhere. [16]

Why it matters. Boys treated before puberty cannot freeze sperm, so this could one day let them have children. [16] Doctors expect to know within months whether it worked; more than 3,100 people worldwide have stored such tissue. [16]

Kenya has quarantined ten people who were close to a man who died of Ebola in Nairobi, and is tracing 57 more. [17] He had been ill for about a month and travelled by road and air from Congo through Uganda without being stopped. [17][18] Congo's outbreak has killed more than 4,000 people since May. [18] Médecins Sans Frontières, a medical charity, says spread in North Kivu province is out of control. [19]

Why it matters. It is the first case outside Congo since Uganda was declared free of the disease. [18] The strain, Bundibugyo, has no approved vaccine or treatment. [17]

The WHO says it does not yet have the full picture of a lab worker's death at a plague research institute in Irkutsk, Siberia. [21] Russian officials say she died of pneumonia of unknown cause. [21] A WHO spokesperson said tests found no dangerous germs in any of her contacts. [20] Russian media reported nearly 200 people under medical watch. [22] The US State Department formally asked Russia for details. [21]

Why it matters. Plague in the lungs can spread between people and kills almost everyone left untreated, but antibiotics given early work. [20] President Trump said the US does not think it was a bioweapon. [22]

Henri Kagan of France and Kenso Soai of Japan won the 2026 Nobel Prize in Chemistry. [23] Many molecules come in two mirror-image forms, like left and right hands, and living things use only one. [24] Kagan found in 1986 how to make reactions favour one form. [23] Soai made a reaction in 2003 that produced only one. [23]

Why it matters. More than half of medicines in use are mirror-form molecules, and the wrong form can do nothing or cause harm. [23] Their methods help drugmakers make only the form that works. [24]

Scientists read the genes of Jonathan, an Aldabra giant tortoise on St Helena thought to be about 194 years old. [25][26] He has unusual versions of 287 genes, some involved in repairing DNA. [26] The chemical tags that switch his repair and energy genes on and off look like those of much younger tortoises. [25] The study appeared in Science Advances. [25]

Why it matters. Those tags usually drift with age, so their stability is one clue to his long life. [25] The study compares one animal with four others, so it offers leads, not proof. [25]

A study in newborn mice found that the first gut bacteria trigger the gut lining to make calprotectin, a protein that kills germs. [27] Pups that made little of it, or had few gut bacteria, were more likely to die from infection. [27] Giving them calprotectin rescued them. [27] The protein was also found in the guts of human babies under two days old. [27]

Why it matters. Infection is a leading cause of death in newborns, and this points to a way probiotics might one day help. [27] The main results are in mice. [27]

Researchers tracked 9,419 meals eaten by 173 cancer patients in hospital and checked their gut bacteria every day. [28] Each extra 100 grams of sugar in the two days before antibiotics went with a 21% bigger drop in the variety of gut bacteria. [28] Sugar also went with more Enterococcus faecium, a germ that causes hard-to-treat infections. [28] The study was published in Nature. [28]

Why it matters. It shows a link, not proof that sugar causes the damage, and the authors want trials of diet changes. [28] The patients studied, with blood cancers, are among those most harmed by gut damage. [28]

Danaher, a US life-sciences company, said it will open its first research lab run by AI and robots in early 2027. [29] The lab will design antibodies and other molecules that stick to a chosen target, then build and test each one. [29] Every result goes back into the AI to improve the next design. [29] Danaher expects it to find molecules eight times faster. [29]

Why it matters. It is part of a wider push, with Biohub and Rivercell, to let computers do more of the early work in finding drugs. [29][11] The speed figures are the company's own expectation, not a measured result. [29]