Biotech & Longevity · Sunday, 25 October 2026

01 Briefing what happened

Amgen's Imdelltra added to Imfinzi cuts the risk of death by 58% in small cell lung cancer, in a final trial shown at ESMO

In a final trial of 563 people, 79% given Amgen's Imdelltra with AstraZeneca's Imfinzi were alive after a year, against 53% on Imfinzi alone, as Europe's big cancer meeting also heard prostate, bladder and ovarian results.

21.1 vs 12.7

months that half the patients lived, on the pair against Imfinzi alone

given later in the illness, in an earlier trial, it added about five months [1]

79% vs 53%

of patients alive after one year, on the pair against Imfinzi alone

that is 26 more people in every 100 [2]

58%

lower risk of death with Imdelltra added

measured after about 16 months, at a planned early look at the data [1][2]

49.6%

of patients on Imdelltra had an immune overreaction

only one case was severe, and none was life-threatening [1]

The lead story — what happened

Amgen, a US drugmaker, said on Saturday that adding its immune drug Imdelltra to AstraZeneca's immune drug Imfinzi cut the risk of death by 58% in extensive-stage small cell lung cancer. [1][4]
Small cell lung cancer grows fast, and extensive-stage means it has already spread beyond the lung where it began. [4]
The DeLLphi-305 trial had 563 patients whose cancer had not grown after a first round of Imfinzi with chemotherapy. [1]
Half then got Imdelltra with Imfinzi, and half got Imfinzi alone, as maintenance: treatment kept up after the first heavy round. [1][4]
Half the patients on the pair were still alive at 21.1 months, against 12.7 months on Imfinzi alone. [1][2]
After one year, 79% of patients on the pair were alive, against 53% on Imfinzi alone. [2]
Imdelltra grabs a protein called DLL3 on cancer cells and a protein on T cells, the immune cells that kill, and pulls the two together. [4]
Almost half the patients on Imdelltra had cytokine release syndrome, an immune overreaction, though nearly every case was mild or moderate. [1][3]
Serious side effects linked to treatment hit 20.9% of patients on the pair and 10.8% on Imfinzi alone. [1]
Imdelltra is now approved only for patients whose cancer grew during or after platinum chemotherapy, and it sold $627 million in 2025. [4]
The trial compared the pair only with Imfinzi alone, not with other ways of starting treatment. [4]
The results were shown at ESMO, Europe's biggest cancer meeting, in Madrid, and published the same day in the New England Journal of Medicine. [1][3]

Imdelltra with Imfinzi21.1 months

Imfinzi alone12.7 months

How long half the patients in each group lived. The gap is about eight months.

Who is involved

Amgen

a large US drugmaker; it makes Imdelltra and ran the trial
AstraZeneca

a British-Swedish drugmaker; it makes Imfinzi and paid for part of the trial
Jacob Sands

a lung cancer doctor at Dana-Farber Cancer Institute in Boston; he led the trial and presented it
Stephen Liu

a cancer doctor at Georgetown University who reviewed the trial at the meeting; he said he would offer it to every eligible group

How it unfolded

Before Imdelltra is approved only after chemotherapy has failed [4]
DeLLphi-304 given later in the illness, half lived 13.6 months on it against 8.4 on chemotherapy [1]
Sat 24 Oct DeLLphi-305 shown at ESMO in Madrid and published in a journal [1][2]
Next Amgen shares the results with drug regulators around the world [1]

Where this points

The next step is the drug regulators: Amgen says it is sharing the results with them worldwide, and the 21-month figure will change as patients are followed for longer. [1][4]

Signals on this beat

The bar and the word are our reading of how hard each one is pushing, and the arrow is where it is heading.

Cancer drugs moved earlier Building↑

Amgen tested Imdelltra straight after patients' first treatment, not after it had failed. [4] Johnson & Johnson gave its pill Erleada before and after prostate surgery. [7] Bayer added its pill Nubeqa to radiation in men whose cancer had not spread. [6]

Judged against a rival Building→

AstraZeneca's bladder cancer pair beat surgery alone but fell short of Merck's Keytruda with Padcev. [3] Novartis's multiple sclerosis pill cut relapses less than a Roche drug did in its own trials. [9] Three rival drugmakers are racing Eli Lilly's new ovarian cancer drug to market. [12]

Germs spreading High↑

US measles cases rose by 178 in a week to 4,588. [17] A baby died of mpox caught at a clinic in Karachi, Pakistan's biggest city. [19] Nigeria's only Ebola case appears linked to a strain found in Kenya. [18]

Medicines running short Building→

Dr Reddy's, an Indian drugmaker, stopped supplying its first copy of Ozempic in July after finding an impurity. [14] In Mbarara, a city in Uganda, a twice-yearly HIV jab is often finished when people come for it. [20]

The rest of the day

17 more stories on this beat.

Each with its own sources. None of these is a link to the story above.

02
AstraZeneca's bladder pair trails a rival

AstraZeneca showed final-trial results at ESMO for its immune drug Imfinzi, given with Pfizer's drug Padcev before bladder surgery and alone after it. [3] The VOLGA trial was in muscle-invasive bladder cancer, which has grown into the bladder wall. [3] Against surgery alone, the pair cut the risk of death by 42%. [3] After two years, 78% of patients on it were alive, against 69% after surgery alone. [3]

Why it matters — Merck's Keytruda with Padcev, already the standard before surgery, cut the risk of death by 50% in its own trial, so STAT reports that AstraZeneca's pair falls short. [3] The two figures come from separate trials with different patients.
03
Bayer's Nubeqa holds back prostate cancer

Bayer said adding its pill Nubeqa to hormone treatment and radiation cut the risk of prostate cancer spreading, or death, by 58%. [6] The DASL-HiCaP trial, run by researchers rather than the company, had 1,106 men whose cancer was at high risk of coming back. [6] After four years, 95% on Nubeqa were free of spread, against 88% on a dummy pill. [6] Nubeqa blocks the male hormones that prostate cancer feeds on. [6]

Nubeqa added95%
Dummy pill added88%
Men free of spread after four years in each group.

Why it matters — After four years, 97% of men on Nubeqa and 95% on the dummy pill are alive, and the doctors are still following them. [6] The trial's 58% matches the lung trial's figure, but here it stands for 7 more men in 100 free of spread. [6]
04
Erleada before and after prostate surgery

Johnson & Johnson said its pill Erleada, given with hormone treatment before and after surgery, cut the risk of prostate cancer returning, or death, by 62% against surgery alone. [7] The trial had 408 men with high-risk cancer that had not spread, followed for a median 37.8 months. [7] After three years, 63% on the drugs had no sign of the cancer returning, against 41.4% after surgery alone. [7]

Erleada and surgery63%
Surgery alone41.4%
Men with no sign of the cancer returning, three years after surgery.

Why it matters — Serious side effects were more common on the drugs, 31% against 22%, and 64% of men had hot flushes. [7] Erleada is not approved for this use, and nobody yet knows whether the men will live longer. [7]
05
The melanoma vaccine is shown in Madrid

Georgina Long of the Melanoma Institute Australia presented the trial of intismeran, a vaccine made for each patient by Moderna and Merck, at ESMO on Saturday. [3] Its results had leaked two days earlier: added to Merck's Keytruda, it cut the risk of melanoma, a skin cancer, returning or death by 42%. [3][8] The cut in the risk of the cancer spreading to distant organs, 38%, was smaller than an earlier trial had led people to expect. [8]

Why it matters — Inge Marie Svane of the University of Copenhagen, who reviewed the trial at the meeting, said it is not yet known whether the vaccine helps people live longer. [3] Merck hopes for US approval next year, and the full results appear in a journal next month. [8][3]
06
Novartis's MS pill falls short of hopes

Novartis said its pill remibrutinib cut the yearly rate of multiple sclerosis relapses by 35% and 31% in two final trials, against Sanofi's older drug Aubagio. [9] In multiple sclerosis the immune system damages nerves, and a relapse is a flare of symptoms. Analysts at BNP Paribas had expected about 50%, and Reuters puts analysts' hopes at 45% to 60%. [9][11] Roche's rival drug fenebrutinib cut relapses by 51% and 59% in its own trials. [9]

Novartis, trial 135%
Novartis, trial 231%
Roche, trial 151%
Roche, trial 259%
Cut in yearly relapses against the same older drug, Aubagio. The two companies ran separate trials.

Why it matters — Heinz Wiendl, the neurologist who presented the data, said the rate on remibrutinib works out at about one relapse every 13 years. [10] STAT reports that its safety record is cleaner, which could still help it compete. [10]
07
Lilly's ovarian cancer drug shrinks most tumours

Eli Lilly showed early results at ESMO for sofetabart mipitecan, a drug that carries a cell-killing poison to cancer cells bearing a protein called FRa. [12] In 33 women whose ovarian cancer came back at least six months after platinum chemotherapy, given it with the drug bevacizumab, tumours shrank in 79.3%. [12] In 20.7%, scans found no cancer left. [12] Serious drops in white blood cells hit 42.4% of the women. [12]

Why it matters — AbbVie, Genmab and AstraZeneca have rival drugs aimed at the same protein, and Lilly began its final trial a year ago. [12] Patients may grow resistant to the poison most of the new drugs share, so Fierce Biotech says being first to market could matter more. [12]
08
A drug pair without chemotherapy for pancreatic cancer

An experimental drug from Tango Therapeutics, given with Revolution Medicines' approved pill Rasonque, shrank tumours in 82% of patients with one kind of advanced pancreatic cancer, results shown on Friday at ESMO. [13] After nine months, 88% were alive with no tumour growth. [13] The pair uses no chemotherapy, long a standard treatment for this cancer. [13]

Why it matters — STAT says the result backs the idea that pancreatic cancer patients gain more from pairs of targeted drugs than from one. [13] Tango's drug is still experimental. [13]
09
Dr Reddy's swaps in new Ozempic copies

Dr Reddy's Laboratories, an Indian maker of cheaper copies of drugs, has started selling two new brands of semaglutide, the active ingredient in Novo Nordisk's Ozempic and Wegovy, Reuters reported. [14] Supplies of its first brand, Obeda, stopped in July after it found an impurity in the ingredient. [14] The new brands, Obeda Pro and Obeda Plus, are made by another company, Hetero Labs. [14] A recall in Canada cost it 710 million rupees of sales in its latest quarter. [15]

Why it matters — Dr Reddy's launched Obeda in India in March, after semaglutide lost its patent there, and a race among Indian drugmakers followed. [14] It says Obeda returns next month in India and this quarter in Canada. [14]
10
UK approves a daily pill for psoriasis

The MHRA, Britain's medicines regulator, approved Johnson & Johnson's daily pill Icotyde on Friday for moderate to severe plaque psoriasis. [16] Plaque psoriasis is a skin condition that causes red, scaly and often itchy or painful patches. [16] The approval covers adults and children aged 12 and over who weigh at least 40kg. [16] It rests on four final trials in 2,500 people, in which the pill beat a dummy pill after 16 weeks. [16]

Why it matters — The US approved Icotyde earlier this year, and J&J, which makes it with Protagonist Therapeutics, expects $5 billion a year in peak sales. [16] It is also being tested in ulcerative colitis, psoriatic arthritis and Crohn's disease. [16]
11
US measles cases reach 4,588

The CDC, the US national public health agency, reported 178 more measles cases this week, taking the 2026 total to 4,588 as of Friday. [17] That is more than double last year's total, across at least 43 outbreaks. [17] Pennsylvania alone has 1,338 cases, and South Carolina and Utah each have more than 500. [17] The CDC counts three measles deaths nationally, while Pennsylvania reports five and is looking into a sixth. [17]

Why it matters — The measles vaccine prevents 97% of cases, and one in five unvaccinated people who catch measles ends up in hospital. [17] The US health department also proposed this week to let doctors bill for reporting suspected vaccine injuries. [17]
12
Nigeria's Ebola case linked to Kenya's

Ali Pate, Nigeria's health minister, said the country's only confirmed Ebola case appears genetically related to a strain found earlier in Kenya. [18] He said there is no evidence yet that the Kenyan patient went to Nigeria, and more tests are under way. [18] Both patients are believed to have passed through Entebbe airport in Uganda this month. [18] Uganda said on Friday that people who have been to DR Congo can no longer change planes there. [18]

Why it matters — The outbreak began in DR Congo in May and has more than 9,500 cases there and over 4,000 deaths across Africa. [18] It is caused by the Bundibugyo kind of Ebola, which has no approved vaccine or treatment. [18]
13
A baby dies of mpox in Karachi

A 52-day-old boy died of mpox early on Saturday at a Karachi hospital, the first baby to die of it in the Pakistani city, Dawn reported. [19] Dawn's sources say he caught it during a circumcision at a hospital in Malir that officials have since sealed. [19] Mpox is a virus that causes a rash and spreads through close contact and shared objects. [19] His death takes Sindh province to 12 mpox deaths and 98 cases this year. [19]

12 of 98
mpox deaths among cases in Sindh this year
Each dot is one mpox case in Sindh province this year. The filled dots are the 12 deaths.

Why it matters — Nearly all who died were babies infected in clinics without proper infection control, Dawn's sources said. [19] Another baby treated at the same Malir hospital is in critical condition. [19]
14
Uganda runs short of a twice-yearly HIV jab

Lenacapavir, a jab given every six months that nearly stopped new HIV infections in two big trials, is often unavailable in Mbarara, a city in western Uganda, africanews reported. [20] Uganda has rationed a few thousand doses across the country, so many people get a two-monthly jab, cabotegravir, instead. [20] Mbarara's HIV rate is 10%, double the national average. [20] The founder of a local HIV group says funding cuts ended outreach that used to test people. [20]

Why it matters — The Global Fund, which pays for HIV drugs in poorer countries, says 430,000 doses had been delivered by mid-September. [20] Gilead, which makes lenacapavir, says it supplies it at no profit while licensed makers of cheaper copies scale up. [20]
15
AstraZeneca puts $50m into Abu Dhabi trials

AstraZeneca will spend about $50 million over the next few years on roughly 30 clinical trials in Abu Dhabi, with Cleveland Clinic Abu Dhabi, a hospital there. [21] The trials will study heart, kidney and metabolic diseases, and the first should start before the end of this year. [21] Heart disease causes 34% of deaths in the UAE, its health ministry says. [21] AstraZeneca also announced a project with Oracle to use AI and health records to find high blood pressure earlier. [21]

Why it matters — Iskra Reic of AstraZeneca said the UAE suits such work because its medical records are well connected and it holds Emirati genetic data. [21] She said more patients there could join global trials. [21]
16
An ear device for stubborn blood pressure

Researchers at Queen Mary University of London tested a small device that sends gentle electrical pulses into nerves in the ear, in 63 people whose blood pressure stayed high despite medicine. [22] Half used the real device and half a fake one, at home, for 12 weeks. [22] Across everyone, blood pressure fell no more with the real device than with the fake. [22] In people with more severe high blood pressure, a group chosen before the trial began, it fell more. [22]

Why it matters — The device was safe and most people used it as told, but the trial was small. [22] The team says larger trials must show whether it works, and for whom. [22]
17
Boehringer's rival drug tried in rarer cancers

Boehringer Ingelheim, a German drugmaker, showed early results at ESMO for obrixtamig, which, like Imdelltra, links immune cells to the DLL3 protein on cancer cells. [5] In neuroendocrine cancers outside the lung, given with platinum chemotherapy as a first treatment, tumours were held back for a median 7.7 months. [5] This year Boehringer began two final trials: one in people starting treatment for small cell lung cancer, and one in these rarer cancers. [5]

Why it matters — Boehringer's lung trial pairs obrixtamig with Roche's immune drug Tecentriq at the start of treatment, the stage where Amgen has now shown a survival gain. [5][1]
18
China nearly matches India on malaria drugs for Africa

China's exports of malaria drugs to 12 of the African countries worst hit by the disease reached $89.3 million last year, a new study found. [23] India's reached $89.6 million, so the two are now almost level, according to Sol Gravitas, an Indian market research firm. [23] In 2021, India sold $121 million of these drugs to the same countries and China $62.6 million. [23] Malaria is a disease spread by mosquito bites, and most of its cases are in Africa.

Why it matters — Indian drugmakers have long been the main suppliers of malaria medicines in these countries, and Chinese firms closed almost all of the gap in four years. [23] The figures cover only these 12 countries and come from one research firm. [23]

02 Lesson why it matters

Why the same 58% cut in risk can mean very different gains for patients

A "58% lower risk" compares two groups, so it cannot say alone how many people it helps: that depends on how many would have died anyway.

The twist

A 58% lower risk does not mean 58 people in 100 are helped. In the lung trial it meant 26 more people in 100 alive after a year, and in the prostate trial 7 more men in 100 free of spread after four years.

The picture

Lung: Imdelltra added79%

Lung: Imfinzi alone53%

Prostate: Nubeqa added95%

Prostate: dummy pill88%

Both trials reported a 58% lower risk. Compare each pair of bars: 26 people in 100 in the lung trial, 7 in 100 in the prostate trial.

How it works

A trial splits patients into two groups and counts deaths in each
The headline compares the two groups' risks as a ratio, such as 0.42
A ratio of 0.42 is reported as a 58% lower risk
Where many in the comparison group die, 58% of their risk is many lives
Where few would die anyway, the same 58% is a handful
So the real gain shows only in each group's own numbers: months, or people per 100

The same force, elsewhere today

Where this chain is also running, in today's other stories.

Erleada before and after prostate surgery

A 62% lower risk of the cancer returning became 63 against 41 men in 100 free of it after three years, about 22 more.
AstraZeneca's bladder pair trails a rival

A 42% lower risk of death became 78 against 69 people in 100 alive after two years, 9 more.
Novartis's MS pill falls short of hopes

A 35% cut in the yearly relapse rate was turned by the presenting doctor into about one relapse every 13 years.
Bayer's Nubeqa holds back prostate cancer

The same 58% as the lung trial meant 95 against 88 men in 100 free of spread, because few men in either group had the cancer spread.

Where you've seen this

Road safety

a new junction that halves crashes saves more people at a busy crossing than at a quiet one

Food scare headlines

a food that doubles the risk of a rare cancer adds very few cases

School exams

a scheme that halves failures changes more lives where many pupils fail than where few do

The catch

The ratio is still useful, because it often stays similar from one group of patients to another, while the gain in people per 100 changes with how sick the group is.

And the whole of it

A patient wants to know their own chance, a health service the cost of each extra year, and an investor the sales. All three start from the same 58%, and none of them can get an answer from it alone.

03 Truth what's really going on

What is really going on

At ESMO in Madrid, Amgen showed that adding its drug Imdelltra to AstraZeneca's Imfinzi kept more people with a fast lung cancer alive, so both companies stand to sell their drug to more patients for longer. [1][4] Most of the day's other cancer results were measured against surgery alone, a dummy pill or an older drug, and two of them fell short of a rival's own trial. [3][6][7][9]

Why it works on us — One large percentage such as 58% sounds like a share of patients saved, when it is a comparison of risk between two groups.

Who gains

Amgen — Imdelltra is now approved only after chemotherapy fails, and a survival gain at the start of treatment could widen its use beyond the $627 million it sold in 2025. [4]
AstraZeneca — Its Imfinzi is given in both groups of the trial, so every patient who gets the new pair still takes Imfinzi. [2]
Merck — Its Keytruda with Padcev, the standard before bladder surgery, cut the risk of death by 50% in its own trial, against 42% for AstraZeneca's pair in a separate one. [3]
Roche — Novartis's pill prevented fewer relapses than expected, which Bloomberg says may give Roche's rival fenebrutinib an advantage. [9]
Hetero Labs — It makes Dr Reddy's two new semaglutide brands, which filled the gap left when the first brand was halted. [14]

Who pays

Patients given the Imdelltra pair — Half had an immune overreaction, serious side effects doubled from 10.8% to 20.9%, and each must be watched in a clinic for hours after the first two doses. [1]
Men given Erleada around surgery — Serious side effects rose from 22% to 31%, and 64% had hot flushes. [7]
People seeking HIV protection in Mbarara — Lenacapavir is often finished when they ask for it, so many take a jab every two months instead of every six. [20]
Families of babies in Sindh — Nearly all of the province's 12 mpox deaths this year were babies infected in clinics without proper infection control. [19]
Dr Reddy's — The semaglutide recall in Canada cost it 710 million rupees of sales in its latest quarter. [15]

What nobody knows yet

Open questions from across today’s stories — ours included.

01

How long the Imdelltra pair really keeps people alive.

The 21.1 months comes from a planned early look at about 16 months of follow-up, and the trial's leader says the figure may change as patients are followed for longer. [1][4]
02

Whether the pair beats other ways of starting treatment.

The trial compared it only with Imfinzi alone, so no result exists against other first treatments. [4]
03

Whether the melanoma vaccine helps people live longer.

The trial measured the cancer returning, and the doctor who reviewed it at ESMO said a survival benefit is not yet known. [3]
04

Whether Nubeqa and Erleada lengthen men's lives.

After four years, 97% of men on Nubeqa and 95% on the dummy pill are alive, and both companies say follow-up for survival goes on. [6][7]
05

Whether remibrutinib's cleaner safety makes up for fewer relapses prevented.

Novartis and Roche tested their drugs in separate trials against Aubagio, and no trial has put the two side by side. [9][10]
06

How many people were in Tango's pancreatic cancer trial.

The 82% and 88% figures come from the opening of a report whose full text is for subscribers, and the patient count was not in it. [13]
07

How many people have died of measles in the US this year.

The CDC counts three deaths and will not say where the third was, while Pennsylvania alone reports five. [17]
08

How Nigeria's Ebola patient was infected.

Her virus appears related to a strain found in Kenya, but there is no evidence the Kenyan patient went to Nigeria and tests continue. [18]

04 Hope carry this

In a final trial of people with small cell lung cancer that had spread, 79 in every 100 given Amgen's Imdelltra with Imfinzi were alive a year later, against 53 on Imfinzi alone.

Also true today

Among 1,106 men with prostate cancer at high risk of coming back, 95 in 100 given Bayer's Nubeqa with radiation were free of spread four years later.
Britain approved Johnson & Johnson's daily pill Icotyde for plaque psoriasis, for adults and for children aged 12 and over.
By mid-September, national programmes reported more than 118,000 people receiving lenacapavir, a jab given twice a year that nearly stopped new HIV infections in two big trials.