Biotech, in brief

The FDA, the US drug regulator, approved GSK's drug Jemperli on 19 October for adults with stage 2 or 3 rectal cancer of one genetic kind. [1][2]

These patients usually get chemotherapy and radiation, then surgery. Those treatments can harm the bowel, urine control, sex life and fertility. [1]

The approval covers tumours called dMMR or MSI-H, whose cells cannot fix the mistakes they make when copying their DNA. [1]

The mistakes pile up and make odd proteins that the immune system can recognise. [1] Jemperli blocks PD-1, an off switch on immune cells that tumours press to hide. [1]

In the main trial, AZUR-1, 154 patients got nine doses of Jemperli, one every three weeks for six months. [2]

110 of them, or 71%, needed no surgery, chemotherapy or radiation, and their cancer did not grow, after a median of just over two years. [2]

Independent reviewers have so far counted 47% of 116 patients free of visible cancer for a full year. [2][3] The trial's own doctors counted 68% of 144. [3]

An earlier trial at Memorial Sloan Kettering, a New York cancer hospital, gave Jemperli to 50 patients. [1] Of the 27 followed for three years, the cancer came back in one. [1]

This is an accelerated approval, given on early results. GSK must keep following patients to show the benefit lasts. [1]

Serious side effects hit 20% of patients, and 3% stopped the drug because of side effects. [3] Immune drugs like this can attack healthy organs. [1]

The FDA decided four months before its own deadline, under a fast-pass scheme for national priorities. It is the 11th approval under that scheme. [1][2]

Regulators in Canada, Israel, Switzerland and the UK are reviewing the same application at the same time. [2]

Why it matters. The next test is whether the cancer stays away as patients are followed for longer, with the full AZUR-1 results due at ESMO this month. [1][3]

Disc Medicine, a US biotech, said on Monday its pill bitopertin met all its goals in a final-stage trial. [4][5] It treats EPP, a rare inherited disease in which a substance called PPIX builds up in the blood and makes sunlight painful. [4][5] On the drug, PPIX fell about 46%, against about 4% on a dummy pill. [4] Patients spent 12.2 more hours a month in light without pain. [5] The FDA refused the drug in February, because earlier trials had not shown that lower PPIX meant more time in the sun. [4]

Why it matters. Disc plans to apply again by the end of the year and expects an answer by mid-2027. [4][5] Its shares rose more than 35% on the news. [4]

Merck KGaA, a German drugmaker, agreed on Monday to buy Silence Therapeutics, a British biotech, for about $1.3 billion in cash. [6][8] It pays $20 a share, 130% above Friday's closing price, or 82% above the average of the past 30 days. [7][8] Silence's main drug, divesiran, is for polycythemia vera, a rare blood cancer in which the bone marrow makes too many red blood cells. [7] The drug switches off a liver protein so less iron reaches the marrow, which then makes fewer red cells. [7] In a mid-stage trial, 88% of patients responded. [7]

Why it matters. Divesiran would compete with Mimrylo, a weekly injection from Takeda and Protagonist approved in August, and would be given every three months. [7][8] Merck must still pay for a final trial; Silence had $72 million in cash in June. [6]

Roche, the Swiss drugmaker, will pay Hansoh, a Chinese drug company, $400 million up front and $110 million after a year, the two said on Monday. [9] Up to $15 billion more depends on the drugs passing tests and selling. [9] The deal covers antibody-drug conjugates, medicines that use an antibody to carry a poison to cancer cells. [9] Hansoh will take two early drugs, and others aimed at targets Roche picks, through first human tests. [9] Roche then develops them everywhere except Greater China. [9]

Why it matters. It is Roche's sixth deal with a Chinese drugmaker since late August, more than any rival this year, BioPharma Dive counts. [10] Neither company has said which diseases the drugs are for. [10]

Gilead Sciences, a large US drugmaker, said its daily pill emvistegrast helped people with ulcerative colitis, a long-lasting inflammation of the large bowel. [11] In a trial of 238 adults, 31% on the top dose were in remission after 12 weeks, against 8% on a dummy pill. [11] The pill blocks a protein that guides certain immune cells into the gut. [11] Entyvio, a best-selling drug from Takeda, blocks the same protein but is given by drip or injection. [11] Gilead plans a final-stage trial in 2027. [11]

Why it matters. A pill would be easier to take than a drip, but the two drugs have never been tested against each other. [11] 78% of the patients had not failed an advanced treatment before, so harder cases are still untested. [11]

AbbVie, a US drugmaker, tested a mix of its drug Skyrizi and a second antibody, trosunilimab, in 300 people with Crohn's disease, a long-lasting gut inflammation. [12] After 12 weeks, 30.2% on the mix showed little or no damage when the gut was checked with a camera. [12] That compared with 21.3% on Skyrizi alone and 18% on trosunilimab alone. [12] Skyrizi blocks one inflammation signal, IL-23, while trosunilimab blocks the gut-guiding protein that Gilead's pill also targets. [11][12] Three in four patients had already failed another advanced treatment. [12]

Why it matters. Skyrizi brought AbbVie $10 billion in the first half of 2026. [12] A 2,000-patient trial that includes the mix began in July. [12]

OSE Immunotherapeutics, a French biotech, said on Monday its cancer vaccine Tedopi helped people whose lung cancer had grown after first treatment. [21] In a mid-stage trial of 106 patients in France, Italy and Spain, 47% given Tedopi with chemotherapy were alive after a year. [21] With Tedopi and the immune drug nivolumab, 46% were. [21] On chemotherapy alone, 29% were. [21] Only patients with a common immune type, HLA-A2, could join. [21]

Why it matters. A final trial of Tedopi against chemotherapy is recruiting, with results expected in early 2028. [21] Full results will be shown at ESMO, Europe's big cancer meeting, in Madrid on 25 October. [21]

AstraZeneca, the British drugmaker, said its drug Sone-Ve helped people with advanced stomach cancer live longer in a final-stage trial. [22] It was tested in patients whose cancer had grown after earlier treatment. [22] The drug carries a poison to cells that show CLDN18.2, a protein found on many stomach cancers. [22] About 60% of these patients carry enough of the protein to qualify. [22] AstraZeneca says it is the first final-stage trial of such a drug to show a survival gain in this group. [22]

Why it matters. The survival numbers are not public yet; they come at ESMO, which runs from 23 to 27 October. [22] Until then, nobody outside the company can judge how much longer patients lived.

IDEAYA Biosciences, a US biotech, and Servier, a French drugmaker, released trial results on Monday for darovasertib, a drug given with an older one, crizotinib. [23] It treats uveal melanoma that has spread, a rare cancer that starts in the eye. [23] In a mid-stage trial, tumours shrank in 22.1% of patients overall. [23] Older treatments shrink them in about 5%, the companies say. [23] Half the patients lived at least 13.4 months. [23]

Why it matters. The trial had no comparison group, so its numbers are set against past studies, which put survival at about 12 months for patients treated first. [23] Diarrhoea hit 82% of patients and nausea 76%. [23]

US President Donald Trump signed an order on Monday to create a women's health division at the NIH, the US government's medical research agency, within 90 days. [13] Melania Trump, the first lady, announced it. [13] A White House task force led by Susie Wiles, the chief of staff, was due to meet for the first time the same day. [13] Details of the new division were not disclosed. [13] Since 2025, NIH grants that used the word women have faced special reviews. [13]

Why it matters. Women have more bad reactions to drugs than men, partly because research has studied them less, a 2025 review found. [13] The new division sits beside the NIH's existing Office of Research on Women's Health rather than replacing it. [13]

Researchers at Mass General Brigham, a Boston hospital group, trained a computer on 40 million face photos and then taught it to predict how long cancer patients live. [14] In 3,820 cancer patients, 64 of the 122 with the highest risk scores died within 90 days of their photo. [14] The study came out on Monday in Lancet Digital Health, a medical journal. [14] The model learned mostly from white patients in the Boston area. [14] It is not yet ready for use in clinics. [14]

Why it matters. A bioethics professor at New York University warns that insurers or employers could use a tool like this to judge people. [14] A wrong score could also mean a sick person is not offered care. [14]

Veradermics, a US biotech, said on Monday its pill for male-pattern baldness passed a second final-stage trial. [10] The pill is a slow-release form of minoxidil, a cheap hair-growth drug that is widely sold as a copy. [10] In the first trial it regrew more hair than a dummy pill over six months. [10] An analyst at Cantor Fitzgerald, an investment bank, called the new results consistent and slightly better. [10]

Why it matters. Some investors doubt people will pay for it when cheap minoxidil already exists. [10] The same analyst argues that a version approved by the FDA could still sell fast. [10]

Nouscom, a Swiss start-up, raised $135 million from investors to keep testing NOUS-209, a vaccine that teaches the immune system to find and kill abnormal cells. [10] It is aimed at people with Lynch syndrome, an inherited condition that raises the risk of cancer. [10] The vaccine is in an early trial. [10] Data at this week's ESMO meeting suggest it might help prevent some tumours. [10] M Ventures and Revelation Partners led the round. [10]

Why it matters. Lynch syndrome comes from faulty genes for repairing DNA, the same fault that marks the rectal tumours Jemperli now treats. A vaccine would aim to act before any cancer has formed.

Ultragenyx, a US rare-disease drugmaker, said last Wednesday it sold a priority review voucher for $203 million. [10] The voucher is an FDA fast pass, which its holder can use to get a quicker review. [10] Ultragenyx earned it when its gene therapy Fayuvi was approved. [10] Less than two weeks earlier it sold another voucher for $210 million. [10] The company has been weighing cuts since its drug for Angelman syndrome, a rare brain disorder, failed. [10]

Why it matters. The two sales brought Ultragenyx $413 million without selling a medicine. [10] The buyer gets a quicker FDA decision on one of its own drugs, and the FDA's test for that drug stays the same.

Enlivex, an Israeli biotech, is laying off 60% of its staff and stopping its early laboratory research. [15] It began the year with 34 full-time employees. [15] Managers take a 20% pay cut and the chief executive 40%. [15] The money goes to one mid-stage trial of allocetra, a cell treatment for knee osteoarthritis, the wear-and-tear arthritis of the knee. [15] Results are due in mid-2027. [15]

Why it matters. Enlivex also keeps part of its money in prediction markets, websites where people bet on future events, which is unusual for a drug company. [15] Its future now rests on one trial in one disease.

UK Biobank, a British study holding health records and blood from half a million volunteers, will add detailed immune data. [16] IMU Biosciences, a London company, will profile blood from an initial 80,000 volunteers. [16] It aims to read more than 100 million immune data points from each sample. [16] Machines from Waters Corporation and Ultima Genomics, two instrument makers, will do the measuring. [16] The work starts this year. [16]

Why it matters. UK Biobank's chief scientist says it will show which immune changes come before cancer, diabetes and dementia. [16] The data will be open to approved researchers. [16]

PacBio, which makes machines that read long stretches of DNA, signed a deal lasting several years with Verily Health, a health-data company. [17] Verily will combine PacBio's DNA readings with patients' medical records on its computer platform. [17] Long readings can catch large changes in DNA that shorter readings miss. [17] The data will be used for research and to train AI programs. [17] The deal was announced at ASHG, the yearly meeting of the American Society of Human Genetics. [17]

Why it matters. Neither company gave the price, the customers or a timetable. [17] Verily names rare disease, cancer and brain disease as the areas it hopes to help. [17]

Scientists at Chiba University in Japan found how stress can restart joint inflammation in mice whose arthritis had calmed down. [18] Stress switched on brown fat, a kind of fat that makes heat. [18] The fat released IL-6, an inflammation signal. [18] IL-6 made cells in the joint lining call in white blood cells, and the joint swelled again. [18] Blocking one receptor for stress hormones reduced the flare. [18] The study is in the journal Science Advances. [18]

Why it matters. People with rheumatoid arthritis often report that stress makes it flare. [18] This was shown in mice, and most results in mice do not carry over to people.

Researchers reported a virus that stays switched off until it reaches a tumour. [19] They fitted a cap onto VSV, a virus that infects livestock, so it cannot enter cells. [19] Enzymes common around tumours cut the cap off, and only then can the virus infect cancer cells. [19] Injected into the blood of mice, a version carrying an immune booster wiped out tumours without harming the rest of the body. [19] The paper is in Signal Transduction and Targeted Therapy, a Nature journal. [19]

Why it matters. Viruses that kill cancer are already approved, but each must be injected straight into a tumour. [19] One that travels in the blood could reach tumours that have spread, though so far that has only been shown in mice.

Scientific American looked at rutaecarpine, a plant extract sold online as a way to drink coffee late and still sleep. [20] It comes from a fruit used in traditional Chinese medicine. [20] In rats, it raises CYP1A2, a liver enzyme that breaks down caffeine. [20] It has never been through a human trial. [20] A 2024 study in mice and human cells suggests it could be dangerous taken with paracetamol, also called acetaminophen. [20]

Why it matters. The same liver enzyme turns paracetamol into a substance that can damage the liver. [20] Sold as a supplement, it needs no FDA test for safety or effect in the US. [20]