New stomach cancer combination keeps nearly 40% alive at 3 years

Close to four in ten people with advanced HER2-positive stomach or oesophageal cancer were still alive three years after starting a combination built around zanidatamab, according to updated results from the phase 3 HERIZON-GEA-01 trial. The figure for the long-standing standard, trastuzumab plus chemotherapy, was about one in five.

The data were presented on 23 October 2026 as a late-breaking talk (LBA79) at the ESMO Congress in Madrid. This was the second interim, and final formal, survival analysis. It follows the US approval of two zanidatamab regimens for this cancer on 25 August 2026.

The cancer in question

Gastroesophageal adenocarcinoma (GEA) covers cancers of the stomach, the junction where the oesophagus meets the stomach, and the oesophagus itself. Jazz Pharmaceuticals, which markets zanidatamab as Ziihera, cites estimates that about 20% of these cancers are HER2-positive. That means the cancer cells carry too much of a growth receptor called HER2 on their surface.

For people whose cancer has spread, the outlook is poor. Jazz quotes a global five-year survival rate of less than 10% for metastatic disease.

For years the standard first treatment for HER2-positive advanced GEA has been trastuzumab, an antibody that latches onto HER2, combined with chemotherapy. HERIZON-GEA-01 set out to test whether a newer HER2 drug could do better.

What zanidatamab is

Zanidatamab is a bispecific antibody. Instead of binding one spot on HER2, as trastuzumab does, it grabs two separate sites on the receptor at once. According to the prescribing information summarised by Jazz, this pulls HER2 into the cell and lowers the amount on the cell surface. It also flags the cancer cell for attack by parts of the immune system.

In the trial, some patients also received tislelizumab, a PD-1 checkpoint inhibitor that releases a brake on immune cells. Zanidatamab was developed by Zymeworks and is now developed by Jazz and BeOne Medicines.

How the trial was set up

HERIZON-GEA-01 (NCT05152147) randomised 914 adults from about 225 sites in more than 30 countries. All had HER2-positive GEA that could not be removed by surgery or had spread, and none had received treatment for advanced disease. HER2 status was confirmed centrally.

Patients were split into three groups:

• zanidatamab plus chemotherapy plus tislelizumab
• zanidatamab plus chemotherapy
• trastuzumab plus chemotherapy, the control

The trial had two primary endpoints: progression-free survival, judged by independent reviewers, and overall survival. The trial was open-label, so patients and doctors knew which treatment was being given.

The three-year results

At the data cutoff, median follow-up was 35.3 months, 9.4 months longer than at the first survival analysis.

Three findings stand out.

The triple combination held its lead. Adding tislelizumab to zanidatamab and chemotherapy cut the risk of death by 35% against the control (hazard ratio 0.65). That hazard ratio was 0.72 at the first analysis, so the gap widened with time.

Zanidatamab without immunotherapy now also beats trastuzumab on survival. The zanidatamab plus chemotherapy group lowered the risk of death by 24% (hazard ratio 0.76, P=0.0088). At the earlier analysis this comparison had not yet reached statistical significance for survival.

Tislelizumab's own contribution is less certain. Comparing the two zanidatamab groups directly, median survival was about four months longer with tislelizumab added (27.0 against 22.9 months; hazard ratio 0.81). The P value was 0.0502, just above the usual 0.05 cut-off, so the difference was not statistically significant.

To put the three-year figures in plain numbers: for every 100 people starting trastuzumab and chemotherapy, about 23 were alive at three years. For every 100 starting the zanidatamab triple, about 40 were. That is roughly 17 more people alive per 100 treated, based on estimated survival curves.

Side effects

Jazz reports that the safety profiles of both zanidatamab regimens were in line with the first analysis. Patients on the triple combination stayed on treatment longest, with a mean of 64.1 weeks against 42.3 weeks for the control, which partly reflects that they lived longer without their cancer growing.

The US prescribing information carries a boxed warning for diarrhoea and for harm to an unborn baby. In clinical studies, diarrhoea affected 85% of people on zanidatamab with chemotherapy and tislelizumab, with grade 3 diarrhoea in 24% and grade 4 in 2.1%. Fatal diarrhoea occurred in 1.5%. Doctors are told to give the antidiarrhoeal loperamide as prevention during the first cycle.

Other warnings include a fall in heart pumping strength, seen in 9% of people on the triple combination, and infusion reactions in 22%. The risk of severe diarrhoea and of fatal side effects was higher in people aged 65 and over on the triple regimen.

These are not small numbers. Anyone weighing this treatment will want a clear talk with their oncologist about how diarrhoea is prevented and managed, and how often heart scans will be done.

Who can get it

In the US, the FDA approved on 25 August 2026:

• zanidatamab with tislelizumab and chemotherapy for HER2-positive disease scored IHC 3+ or IHC 2+ with a positive ISH test
• zanidatamab with chemotherapy for disease scored IHC 3+

Jazz says the US NCCN guidelines now list both as preferred options. Zanidatamab is already approved in the EU for a different cancer, HER2-positive biliary tract cancer, but we found no EU or UK decision for stomach or oesophageal cancer at the time of writing. Jazz says it is pursuing further approvals.

What to keep in mind

• Company source. These updated numbers come from Jazz's press release and an ESMO presentation. A full peer-reviewed paper of the three-year data was not available when we wrote this. The first analysis was published in the New England Journal of Medicine.
• Open-label design. Overall survival is an objective measure, so open-label bias matters less for this endpoint than for others.
• HER2 testing matters. The benefit applies only to HER2-positive tumours, and the approved US uses depend on the exact HER2 score.
• Tislelizumab's role. The data support zanidatamab as the bigger driver of the survival gain. Whether immunotherapy adds a meaningful extra benefit for everyone remains unsettled.

This article explains research findings and is not medical advice. Talk to your cancer team about what fits your situation.