Forget about Herceptin. There’s a new HER2 agent that ought to be used in HER2-positive stomach cancer, and that’s Jazz Pharmaceuticals’ Ziihera.
Or at least that appears to be the consensus following the readout from Jazz’s Herizon-Gea-01 trial.
The phase 3 study showed that compared with Herceptin plus chemotherapy, both a two-drug regimen of Ziihera plus chemo and a triplet combo of Ziihera, chemo and BeOne Medicines’ PD-1 inhibitor Tevimbra can extend progression-free survival (PFS) time when used as first-line treatment for patients with advanced HER2-positive gastroesophageal adenocarcinoma (GEA), including cancers of the stomach, gastroesophageal junction and esophagus.
The triplet slashed the risk of progression or death by 37% compared with Herceptin-chemo, while the Ziihera-chemo combo mounted a 35% PFS improvement, according to data that were prematurely released ahead of the 2026 ASCO Gastrointestinal Cancers Symposium in San Francisco.
The median PFS time came in at 12.4 months for the two Ziihera arms, separately, versus 8.1 months for the control group.
The triplet also delivered a statistically significant overall survival (OS) showing, paring down the risk of death by 28% over Herceptin and chemo. The new combo extended patients’ median OS length by 7.2 months to 26.4 months, the longest reported in a phase 3 trial in GEA, according to Jazz. However, the 26.4-month median OS was a notable step down from the 36.5 months recorded in a phase 2 of Ziihera and chemo, without Tevimbra.
While the doublet has not yet met the OS goal at the first interim analysis, it showed a favorable trend marked by a preliminary 20% death-risk reduction, with a p-value of 0.0564. Although the number hasn’t yet reached statistical significance, the trial has power left to potentially meet the bar in the future, Herizon-Gea-01 presenter Elena Elimova, M.D., from the Princess Margaret Cancer Centre in Canada, said during a press briefing ahead of the official presentation at the conference.
In addition, while the objective response rates were roughly similar among the three trial arms, ranging between 65.7% and 70.7%, the median duration of response (DOR) was longer for the Ziihera-based regimens. For the triplet, the median DOR reached an unprecedented 20.7 months, Elimova noted.
On the safety side, the rate of grade 3 or above treatment-related adverse events (TREAs) was 71.8% in the triplet arm, 59% for Ziihera-chemo and 59.6% for control. Discontinuation of either Ziihera or Herceptin due to TREAs happened in 11.9%, 8.5% and 2.3% of patients, respectively.
“I believe, definitely, this shows that Ziihera is superior to Herceptin, and Ziihera should be the HER2 treatment of choice for front-line GEA,” Jazz’s global head of R&D and chief medical officer, Rob Iannone, M.D., said in an interview with Fierce Pharma.
“I really do think that this will be something that impacts clinical practice pretty quickly,” Rachna Shroff, M.D., from the University of Arizona Cancer Center, an ASCO-invited GI cancer expert, commented during the press briefing.
“What is clear is that we have found a better way to target HER2,” Shroff later added, referring to Ziihera as better than Herceptin.
In the HER2 stomach cancer space, a combination of Merck & Co.’s Keytruda with Herceptin and chemo initially won an FDA accelerated approval in 2021 based on data from the phase 3 Keynote-811 trial. In both Merck’s and Jazz’s trials, the control arm of Herceptin-chemo posted a median PFS of 8.1 months at an interim analysis. While cross-trial comparisons should be made with caution, Iannone contended that the similar control arm PFS suggests that the two studies’ patient populations were highly comparable. One key difference is that Keynote-811 did not enroll patients with esophageal cancer, who make up about 9% of participants in Herizon-Gea-01.
In Keynote-811, the Keytruda group’s median PFS was 10 months at an interim analysis, whereas the Tevimbra group’s median PFS reached 12.4 months in Herizon-Gea-01.
At the final analysis, patients who received the Keytruda-Herceptin-chemo regimen lived a median 20 months in Merck's study, whereas those who took the Tevimbra-Ziihera-chemo combo in Jazz’s trial lived a median 26.4 months.
“Do I think there will be a comparison to look at different PD-1s? I, personally, don’t think there will be,” Elimova said. “But I think this study does answer that [Ziihera] is a better HER2-targeted agent.”
While the data position Ziihera as the new HER2 agent of choice in GEA, the question remains whether the addition of an immunotherapy, Tevimbra, is necessary. For now, only the Tevimbra-containing triplet boasts a definitive OS improvement.
“I think further overall survival analysis for the [Ziihera] and chemotherapy arm is important,” Shroff said.
Following the initial accelerated approval and an updated OS analysis from Keynote-811, Keytruda’s HER2 stomach cancer indication was later narrowed to tumors that express PD-L1, as the regimen didn’t show a clear benefit in PD-L1-negative patients.
Then, last year, the FDA restricted the use of PD-1 inhibitors in HER2-negative GEA to only cover PD-L1-positive disease.
These past data bring to question the role of Tevimbra in Herizon-Gea-01 in PD-L1-negative GEA patients, who, according to Jazz’s Iannone, make up about a third of HER2-positive GEA cases.
Iannone argued that the latest data support the use of Ziihera and Tevimbra irrespective of a patient’s PD-L1 status. Subgroup analyses showed favorable results for Ziihera in all subpopulations by PD-L1 expression, he said.
A subgroup analysis in the PD-L1-negative group linked the triplet to a strong 51% reduction in patients’ risk of death and a 53% improvement in PFS, according to BeOne. The subgroup data do not bear statistical significance.
From a mechanism perspective, Ziihera, a HER2 biparatopic antibody, may lead to increased immune activation compared with Herceptin, creating more synergy with a PD-1 inhibitor, Iannone explained.
The surprising and dramatic benefit seen for the triplet in PD-L1-negative patients suggests Ziihera “may help turn cold tumors hot and expand the potential use of PD-1 inhibitors in GEA into PD-L1-negative disease,” analysts at Leerink Partners said in a Jan. 7 note.
Iannone declined to comment on Jazz’s discussions with the FDA or whether the company will seek all-inclusive labels for the two Ziihera regimens.
Rather than the PD-L1-negative population, the Leerink analysts are instead worried about the Ziihera triplet in the larger PD-L1-positive population. In this subgroup, the triplet improved OS by 18% and PFS by 35%. These results, as the Leerink team put it, do “not appear as clearly differentiated from” the Keynote-811 regimen. Keytruda’s label shows an OS and an PFS improvement of 21% and 28%, respectively, in PD-L1-positive disease.
“This could have commercial implications in the PDL1+ segment, potentially impacting physician willingness to switch [Keynote-811] to HERIZON GEA regimen given physician familiarity with KN-811, its lower overall cost driven by Herceptin biosimilars, and higher rates of grade 3 diarrhea with [Ziihera],” the Leerink analysts wrote in their note.
However, the team acknowledged that the Ziihera regimen showed clear benefits in all subgroups, and doctors may therefore accept the drug as a superior agent to replace Herceptin in GEA.
The GEA trial win gives Jazz additional confidence in Ziihera in other tumor settings. In what could be Ziihera’s largest potential indication, Jazz is running a phase 3 trial pitting Ziihera against Herceptin in their separate combinations with chemotherapy to treat HER2-positive breast cancer following AstraZeneca and Daiichi Sankyo’s Enhertu.
Ziihera may have to compete with Enhertu in first-line stomach cancer in the future once the phase 3 Destiny-Gastric05 trial for the star antibody-drug conjugate reads out, which is expected after 2026.
Jazz may consider other indications, too, such as early-stage stomach cancer, Iannone added.
Editor's Note: The story was updated at 8:30 a.m. ET on Jan. 7 to include additional subgroup data and comments from a note by Leerink Partners.