Good morning and welcome to the zanidatamab HERIZON-GEA-01 Investor Call and Webcast. All participants are in a listen-only mode. After the speaker's remarks, we will have a question-and-answer session. To ask a question at this time, you'll need to press star followed by the number one on your telephone keypad. As a reminder, this conference call is being recorded. I would now like to turn the call over to Renee Gala, President and Chief Executive Officer of Jazz Pharmaceuticals. Thank you. Please go ahead.
Thank you, Operator. Good morning, everyone. My name is Renee Gala, President and CEO of Jazz Pharmaceuticals. Thank you for joining us. Before I begin, please note that today's slide presentation accompanying this webcast and call is available on the investor section of our website. Yesterday, we presented groundbreaking data from the Phase III HERIZON-GEA-01 Clinical Trial assessing zanidatamab in combination with chemotherapy with or without tislelizumab in first-line HER2-positive GEA patients at the American Society of Clinical Oncology Gastrointestinal Cancer Symposium. We have the privilege of being joined today by Dr. Geoff Ku, Associate Attending Physician on the Gastrointestinal Oncology Service in the Department of Medicine at Memorial Sloan Kettering Cancer Center, who will discuss the HERIZON data.
While today's call will be focused on GEA, it is important to note that these data also support our efforts to maximize the full value of zanidatamab across multiple potential indications. The data we will discuss today reflect our strong belief in zanidatamab's novel mechanism of action and solidify our belief in the promise of zanidatamab to transform the treatment landscape for multiple HER2-expressing cancers. As outlined on slide two, I'd like to remind you that today's webcast includes forward-looking statements which involve risks and uncertainties that could cause actual events, performance, and results to differ materially. We encourage you to review the statements contained in our latest SEC filings and reports, which identify certain factors that may cause the company's actual results to differ materially from those projected. We undertake no duty or obligation to update our forward-looking statements.
Our agenda for today is outlined on slide three. We're pleased to start with Dr. Ku, who will walk us through the trial results. Next, Dr. Rob Iannone, our Global Head of Research and Development and Chief Medical Officer, will provide perspectives on the shifting treatment paradigm and clinical development plans for zanidatamab. Finally, Sam Pearce, our Chief Commercial Officer, will provide a commercial perspective and then will open the call for Q&A. Turning to slide five, results from the pivotal Phase III HERIZON-GEA-01 trial represent the first global randomized data set for zanidatamab, which increases our confidence in zanidatamab's mechanism of action and adds to the proof of its clinical efficacy. In addition, this is the first Phase III trial to demonstrate the benefit of a novel HER2-targeted therapy when directly compared to Herceptin as part of a combination regimen in HER2-positive first-line GEA.
We believe the combination of a truly unique MOA, along with the outstanding clinical efficacy and safety data we've generated for this molecule to date, are supportive of our efforts broadly to look at zanidatamab as a potential new standard of care across HER2-positive indications. I'm delighted to introduce Dr. Geoffrey Ku, a medical oncologist who specializes in the treatment of malignancies of the gastrointestinal tract at Memorial Sloan Kettering Cancer Center in New York. He is an Associate Attending Physician and Head of the Esophagogastric Section of the Gastrointestinal Oncology Service and a member of the Cellular Therapy Service, both in the Department of Medicine. His research focuses on the evaluation of novel therapies, including cellular therapies, and combined modality treatments for esophagogastric cancer.
In addition to being a member of the Esophagogastric Task Force of the National Cancer Institute and of the Gastrointestinal Non-Colorectal Cancer Committee of the NRG Cooperative Group, Dr. Ku was a contributing author to the HERIZON-GEA-01 oral presentation given at ASCO GI yesterday. Dr. Ku?
Thank you, Renee. It's my pleasure today to present the results of the HERIZON-GEA-01. And this is the same slide deck that my friend and colleague, Dr. Yelena Janjigian, presented yesterday at ASCO GI. So I'll endeavor to do as good a job as she did. Turning to slide eight, the key takeaway points are as such. HERIZON-GEA-01 supports zanidatamab as a new anti-HER2 therapy replacing trastuzumab, as well as the use of tislelizumab in the first-line treatment of HER2-positive esophagogastric cancer. There is a 35% reduction in the risk of disease progression or death, both zani and chemo, as well as zani, tisleli, and chemo versus trastuzumab chemo, which translates into a more than four-month improvement in median progression-free survival. There was a strong trend towards a statistically significant improvement in overall survival, favoring zani + chemo versus trastuzumab + chemo.
There was a numerical five-month improvement in median overall survival. I would also note that this is the first of several planned interim analyses. However, there was clearly a statistically significant and clinically very meaningful 28% reduction in the risk of death for zanidatamab + tislelizumab + chemo versus trastuzumab, which translates into a more than seven-month improvement in median overall survival. And both of these overall survival benefits were generally observed across all key specified subgroups, including patients with PD-L1 TAP scores of less than 1% and 1% or more. Finally, the safety profile was also consistent with the known profile of each individual treatment, but we will absolutely delve into this in more detail. Turning to the next slide, this is the background.
As we all know, the outcomes for HER2-positive gastric cancer remain modest despite improvements in the last several years, with a median progression-free survival of less than one year and median overall survival of less than two years. zanidatamab is a bispecific or specifically biparatopic antibody that targets two different extracellular domains of HER2. This unique molecular structure leads to also a unique mechanism of action that differentiates it from other anti-HER2 therapies. Separately, tislelizumab is a high-affinity immune checkpoint inhibitor targeting PD-1. There was promising efficacy as well as tolerability observed for zani + chemo as well as zani tislelizumab + chemotherapy in two independent Phase II studies, which led to this Phase III study. Turning now to the study design. The eligibility criteria are as expected for a standard first-line HER2-positive global study.
Stratification factors were geographic region, HER2 status, as well as ECOG performance status. Patients were randomized in a one-to-one-to-one fashion to arm A, which was a standard arm of trastuzumab and chemotherapy, arm B was zanidatamab + chemotherapy, and arm C was zanidatamab + tislelizumab + chemotherapy. The study had dual primary endpoints of progression-free survival for BICR, so blinded independent central review, as well as overall survival, and an important secondary endpoint was the objective response rate. I would note that arm A excludes pembrolizumab, which is the current standard of care. It was also, at the time, FDA-approved only in the U.S. And as a result, this study did not open in the United States but accrued in the rest of the world. Turning to the next slide, we focus on statistical design. Again, the dual primary endpoints are PFS and overall survival.
The primary PFS analysis would be triggered after a target event count was reached, and patients had at least seven months of follow-up. The first interim overall survival analysis would be performed at the time of data cutoff for the primary PFS analysis. You see below here also the sequence of testing. We start by testing PFS in arm C versus arm A. If that is positive, we then test for PFS in arm B versus arm A. If that is positive, we look at overall survival in arm C versus arm A, followed then by overall survival in arm B versus arm A. Only if this is positive do we finally compare both experimental arms, C versus B. I would note that the study was not designed to directly test both experimental arms. We now turn to patient disposition.
Overall, 914 patients were randomized, and there were approximately 300 patients in each arm. The one thing I would note is that a higher proportion of patients in arm A have discontinued treatment versus the zanidatamab continuing arms. Specifically, 86% of patients in arm A have discontinued treatment versus 76% in arm B versus 70% in arm C, and all three arms, the median follow-up is a little bit more than two years. Turning now to baseline demographics and disease characteristics, we see that these are essentially well-balanced across all three arms. Approximately 50% of patients were enrolled from Asia, and then the rest were enrolled from Europe, North America, and rest of the world. The study did enroll patients with esophageal tumors, which will comprise approximately 10% of the study population. 80%-85% of tumors were HER2 IHC 3+, which is what we expect from other contemporary first-line studies.
Somewhat surprisingly, in terms of PD-L1 status, 40% of patients had a TAP score of less than 1%, while approximately 60% had a TAP score of 1% or more. You'll note that these numbers do not add up to 100% because there were some missing data points. This study allowed either CAPOX or 5-FU cisplatin every three weeks, and overwhelmingly, 90% of patients received CAPOX.
Turning now to the primary endpoint of PFS, and this is the PFS in arm B versus arm A. And here we see that there is a clear superiority. In arm B, PFS is 12.4 versus 8.1 months in arm A, has a ratio of 0.65, and the p-value is highly statistically significant. And this translates into a two-year progression-free survival rate of 31.5%. We next look at PFS per BICR in the arm C versus arm A comparison. And again, this is also highly statistically significant and clinically meaningful.
The median PFS is 12.4 versus 8.1 months, has a ratio of 0.63, p-value also highly statistically significant. This translates into a two-year progression-free survival rate of 38.2%, which is a little bit higher than in the arm B versus arm A comparison. We now look at the PFS in key specified subgroups. Essentially, I think the take-home point is that all subgroups seem to benefit. One exception is the IHC 2+ FISH-positive population. Again, this comprised only 15% of patients, and therefore, it could be confounded by small numbers as well as other factors. What's notable here is that all patients seem to benefit irrespective of PD-L1 status. Both the TAP 1% or more patients as well as the less than 1% patients benefit. We now turn to a primary endpoint of overall survival, and this is the overall survival comparison in arm B versus arm A.
Here we do see a numerical improvement to 24.4 months from 19.2 months. This is a hazard ratio of 0.80, but the p-value is 0.0564, which just missed statistical significance. Again, I would emphasize this is the first of several planned interim analyses. Now, turning to overall survival in the arm C to arm A comparison, this is statistically significant. The median overall survival here is 26.4 months versus 19.2 months, and the control group has a ratio of 0.72. p-value now is highly statistically significant. And that translates into a two-year overall survival rate of 54.3% or a 30-month overall survival rate of 43.8%. Focusing now on the overall survival by key specified subgroups. Again, we essentially see that most subgroups seem to benefit from treatment. Again, the patients who seem to benefit irrespective of PD-L1 score.
Here we do see that the IHC 2+ FISH-positive patients, the point estimate at least is less than one, suggesting that they also derive benefit from zanidatamab. Again, these are small numbers which could confound the analysis. We turn now to anti-tumor activity. On the left, we see the response rate. The response rate was 66% in the control group, which is only slightly improved to 70% in both zanidatamab-containing arms. However, the complete response rate was nearly doubled in arm C, 19.6% versus 11% in arm A. I think what's particularly impressive here is the median duration of response. In particular, if we focus on the DOR in arm C, it is 20.7 months compared to 8.3 months in arm A and compared to 14.3 months in arm B. Such a long durable period of response is truly unprecedented in this disease. We turn now to safety summary.
Again, very busy slide, but I would focus on the fact that there was increase in grade three or more toxicities in the zanidatamab-containing arms. Specifically, it was 71.8% in Arm C versus 59.6% in Arm A. It actually was comparable in Arm B versus Arm A, 59%. These toxicities did lead to an increased rate of discontinuation of zanidatamab versus trastuzumab. In Arm C, there was an 11.9% discontinuation rate versus 8.5% in arm B versus only 2.3% in A. In terms of adverse events of special interest, infusion-related reactions occurred in 25% of patients in both arm B and C receiving zanidatamab. It was only 13.2% in patients receiving trastuzumab. The incidence of non-infectious pulmonary toxicities was also increased in Arm C, consistent with the known toxicities in this case of tislelizumab, but left ventricular dysfunction seemed to be comparably low in all three arms.
We focus now on graphical representation of common toxicities seen in at least 20% of patients. We'll see that there was a significant increase in diarrhea in the zanidatamab-containing arms, potentially some slight increase in nausea and vomiting as well. There was also an increase in decreased weight and hypokalemia. We'll drill down on that in more detail in this slide on treatment-emergent diarrhea. Now, diarrhea had emerged as one of the toxicities associated with this combination in the Phase II studies. As such, moving into Phase III, there was mandatory diarrhea prophylaxis for patients in the zanidatamab-containing arms consisting of loperamide twice daily for the first seven days. Subsequently, it could be discontinued at patient and physician discretion if they did not develop diarrhea or they developed constipation.
Possibly as a result of this, we actually see that the time to onset of diarrhea is at about the one-week mark, and the time to onset of grade three diarrhea is at approximately two weeks. The incidence of grade three diarrhea was increased in the zanidatamab-containing arms. It was 20% in arm B, 24.5% in arm C, and only 12.9% in arm A. But irrespective of the severity of the diarrhea, the time to resolution was approximately three weeks.
So finally, in the last slide, the discussion summarizes much of the information that I've already discussed. It is also a more lengthy explanation of the key points at the beginning of this presentation. I will leave it as is for those on the call to review, but it really is a summary of all of the salient points that I've discussed. Thank you all very much for your attention. I'm happy to take any questions following this, and I'll turn things back to Rob.
Thank you, Dr. Ku, for reviewing these unprecedented data, which we view as practice-changing. The strength of the results reflects the rigor and precision with which this study was executed. I'm thankful to the patients who participated in the clinical trial and all who collaborated with us to make this achievement possible. Being here at ASCO GI the last couple of days has increased the excitement around zanidatamab. The momentum we continue to build around this program is invigorating as we move quickly to engage the FDA and plan to submit a supplemental biologics license application in the first half of this year. In addition, we plan to expeditiously seek NCCN listing based on these published data.
Turning to slide 27, I'll cover zanidatamab's unique mechanism of action, which differentiates it from other HER2-targeted antibody therapies. zanidatamab is a dual HER2-targeted bispecific antibody that binds to two extracellular domains on HER2, which is distinct from the traditional HER2-targeted antibodies that bind a single epitope. The antibody is engineered to bind in trans to two non-overlapping epitopes of the HER2 receptor, which is known as biparatopic binding. This enables zanidatamab to cross-link neighboring HER2 proteins, leading to receptor clustering. Treatment with zanidatamab has been shown to block HER2 receptor growth signaling in HER2-positive cancer cells, and in doing so, can slow or stop the growth of HER2-positive cancers. In addition to zanidatamab's direct effect on tumor cells, it has been shown to exert several immune-mediated effects.
In preclinical studies, zanidatamab has been shown to activate the immune system via complement-dependent cytotoxicity, antibody-dependent cellular cytotoxicity, and antibody-dependent cellular phagocytosis. We believe that activation of the complement pathway is unique to zanidatamab within the HER2-targeted antibody class. We believe zanidatamab's mechanism of action drives the differentiated and compelling clinical profile Dr. Ku spoke of today and contributes to both the strong activity when zanidatamab is combined with standard of care chemotherapy, as well as when it is combined with tislelizumab, where strong benefit has been observed irrespective of tumor PD-L1 status. On slide 28, I will provide context around the magnitude of clinical benefit these results demonstrated compared to the current standards of care.
In the case of first-line HER2-positive GEA patients, that includes trastuzumab + chemotherapy based on results from the ToGA trial and trastuzumab + chemotherapy with pembrolizumab based on results from the Keynote-811 trial. Both zanidatamab investigational treatment arms demonstrated a statistically significant improvement in PFS, corresponding to a median PFS exceeding 12 months, reducing the risk of disease progression or death by 35% compared to the trastuzumab control arm. Likewise, both zanidatamab combinations exceeded two years of median OS, representing the longest survival outcomes ever reported in this setting. The zanidatamab + tislelizumab and chemotherapy arm demonstrated a clinically meaningful and statistically significant improvement in OS, with more than seven months' improvement and a 28% reduction in the risk of death versus the trastuzumab control arm.
zanidatamab + chemotherapy showed a clinically meaningful survival benefit with a median OS of over two years compared to the control arm, with a strong trend towards statistical significance at the time of this first interim analysis for OS. An additional planned interim analysis for this comparison is currently expected in mid-2026. Slides 29 and 30 illustrate the PFS and OS Kaplan-Meier plots, respectively, for the HERIZON-GEA-01 trial compared to the Keynote-811 trial, which was the basis for approval of Keytruda in PD-L1-positive patients. As you can see from these graphs, the zanidatamab treatment arm shows a clinically meaningful improvement when compared to the Keynote-811 regimen. Further differentiating the zanidatamab combination, the benefit is seen in combination with tislelizumab irrespective of PD-L1 status, which we view as practice-changing.
On the PFS graph, you can see how similarly the trastuzumab + chemo control arms performed by the overlap of red and maroon lines. On the OS graph, the HERIZON-GEA-01 trastuzumab + chemo control arm may have benefited from more frequent crossover to subsequent therapy than arm B. However, despite the better-than-anticipated performance of the control arm, both zanidatamab combinations outperformed the control arm with more than two years of median OS, representing the longest survival outcomes ever reported in this setting. Importantly, benefits were observed for both PFS and OS for the zanidatamab + tislelizumab + chemotherapy arm, regardless of PD-L1 tumor status. The benefit, regardless of status, may be driven by zanidatamab's unique mechanism of action, as I described earlier. Enhanced immune activity of zanidatamab compared to trastuzumab may in part explain the efficacy of zani and tisleli irrespective of PD-L1 status.
Turning to slide 31, the secondary endpoints of confirmed ORR and DOR were supportive of the primary efficacy outcomes. For the zanidatamab + tislelizumab and chemotherapy arm, the median DOR nearly doubled the current standard of care to 20.7 months. These secondary endpoints demonstrated deeper and more durable responses in the zanidatamab-containing arms, leading to clinically meaningful improvements for patients with HER2-positive metastatic GEA. In summary, we view the overall data as practice-changing compared to the current standard of care and supportive of approvals in this clinical setting for zanidatamab and tislelizumab, regardless of PD-L1 status. This is very exciting news for the field and, most importantly, a significant advance for patients who are facing a devastating diagnosis with limited options in first-line HER2-positive, locally advanced, or metastatic GEA.
Slide 32 shows the breadth of the current development program we have ongoing for zanidatamab, including multiple registration-enabling studies focused on areas where we believe zanidatamab has the potential to emerge as the preferred HER2-targeted therapy. Results from HERIZON-GEA-01, the first Phase III global randomized clinical trial of zanidatamab, not only underscore zanidatamab's ability to deliver meaningful outcomes for patients with GEA, but also reinforce its promise across a number of HER2-expressing tumors. This gives us strong conviction in the clinical utility of zanidatamab more broadly.
We currently have a robust development program exploring zanidatamab in various HER2-positive solid tumors, including a Phase III trial in metastatic breast cancer patients who are intolerant to or have progressed on previous anti-HER2 treatment, a Phase III trial in metastatic frontline biliary tract cancer, a pan-tumor trial that may be registrational, and a Phase II perioperative breast cancer trial. The compelling data and durable responses seen from previous trials of zanidatamab are already resonating with oncologists, and there is excitement around potential to combine zanidatamab with novel therapies. With these positive GEA results in hand, we are focused on unlocking the full potential of zanidatamab and are excited to share more on these plans in the near future. Now, I will turn the call over to Sam to discuss our commercial perspective. Sam?
Thanks, Rob. As Dr. Ku and Rob have outlined. These results are truly impressive and represent the kind of scientific progress that can redefine outcomes for patients. It's advances like these that really drive our purpose to innovate to transform the lives of patients and their families. Turning to slide 34, I'll begin by highlighting the significant unmet need in GEA, which is the fifth most common cancer globally, and approximately 20% of patients have HER2-positive disease. This represents about 63,000 cases each year across the U.S., Europe, and Japan, including approximately 8,000 patients in the U.S. alone. Unfortunately, HER2-positive GEA is associated with substantial morbidity and mortality. For metastatic gastric cancer, five-year survival remains below 10%, underscoring the seriousness of this disease. And despite ongoing advances in treatment, there has not been a new HER2-targeted option in the first-line setting since the ToGA regimen was introduced in 2010.
Importantly, because of the strong historical precedent of HER2-directed therapies, HER2 testing is routine, with adoption rates above 80%, ensuring patients can be readily identified for treatment. Turning to slide 35, I'll outline how our commercial strategy, combined with strong data and commercial foundation, positions us to accelerate the launch of Ziihera in GEA if approved. The Phase III HERIZON GEA results reinforce that HER2 is the key oncogenic driver in this tumor type. These data clearly demonstrate that zanidatamab delivers superior efficacy versus trastuzumab, independent of PD-L1 expression. We believe these practice-changing outcomes position Ziihera as the preferred HER2-targeted agent of choice in first-line GEA, replacing trastuzumab as the new standard of care, offering unprecedented durability and survival benefits. Our commercial momentum in BTC provides a strong springboard for rapid uptake in GEA.
Over the last year, we have expanded our solid tumor infrastructure, drawing on deep expertise within our oncology organization. Awareness and experience with Ziihera is already high across major academic centers and large community practice networks. Importantly, the compelling BTC data are translating into real-world results that oncologists are seeing in their patients, building confidence that will support adoption in HER2-positive first-line GEA. There is also substantial customer overlap across our solid tumor footprint. Approximately 40% of our targets are shared across Zepzelca, BTC, and GEA, and over 90% overlap exists between BTC and GEA prescribers. Because of this, and assuming regulatory approval, we anticipate leveraging our existing field force without significant expansion. Any incremental hiring will be highly targeted and fit for purpose. On the payer access front, Ziihera already benefits from an established permanent J code through its FDA approval in second-line HER2-positive BTC.
This simplifies reimbursement and reduces administrative burden for providers. In addition, our comprehensive Jazz Care support services, along with flexible ordering and fulfillment options, ensure that providers can access Ziihera efficiently and without delays. While we will share more detail on launch execution as we approach approval, our existing footprint, capabilities, and experienced teams position us to move quickly and efficiently to realize the GEA opportunity. Turning to slide 36, I'll speak to our regulatory path and the broader opportunity ahead. Following the positive HERIZON-GEA-01 results, we are preparing to submit a supplemental biologics license application in the first half of this year. We believe that the HERIZON-GEA-01 results will support near-term inclusion of zanidatamab in NCCN guidelines. More broadly, zanidatamab represents a global multi-indication opportunity for Jazz.
We remain excited about its potential to redefine outcomes across HER2-expressing tumors and to build a meaningful, durable franchise that expands our presence in solid tumors. Before I turn the call over for Q&A, I want to express my sincere thanks to all who made this progress possible: to the patients and their families who participated in the study, to all of the investigators and clinical teams across numerous sites, and to our colleagues and partners at Jazz, Zymeworks, and BeiGene. The dedication and collaboration have been essential in bringing zanidatamab to this point. And with that, I'll conclude the slide presentation and turn it over to the operator to begin Q&A.
As a reminder to ask a question, please press star, followed by the number one on your telephone keypad. In the interest of time, we ask that you please limit yourselves to one question. Thank you. Our first question today comes from Marc Goodman from Leerink Partners. Please go ahead. Your line is open
Yes, Dr. Ku, could you talk about the surprise with the PD-L1 negative subgroup having better data and your thoughts on that? And maybe you could just give us a sense of, okay, we've got this data. What % of your HER2-positive patients now are going to be using the HERIZON versus the Keynote-811 regimen?
I mean, I think it's an interesting question. I mean, I wouldn't say that my surprise. I mean, I think you're referring to the fact that the point estimates suggest that the TPS less than 1% patients do minimally better than the 1% or more. I mean, that's not really a take-home point for me. If you look at the confidence intervals, they clearly overlap.
I think, to me, the kind of slightly bigger picture is that I think the very pleasant surprise is that both the PD-L1 essentially negative and positive populations benefit from the addition of tislelizumab. And I think that's in contrast to the Keynote data set for two reasons. First, in Keynote-811, only 15% of tumors were PD-L1 negative. In this study, it was up to 40%. And it's unclear why, but I think one potential explanation is that, again, the Keynote-811 regimen was initially FDA-approved only in the U.S. and then subsequently became approved much later on in the rest of the world, including Europe. So potentially, by the end of the study, investigators may have been referring only patients with PD-L1 negative tumors. But I think that's actually ultimately very helpful because 40% of PD-L1 negative tumors really allows for a more robust subgroup analysis.
Really, my take-home point is that all patients benefit irrespective of PD-L1 status. I think, actually, there's a lot of interest in investigation into why that is. It may really refer or relate to the unique biology of zanidatamab versus trastuzumab. I think turning to the other part of your question, that honestly is not a patient that I would consider for pembrolizumab, trastuzumab, and chemotherapy at this point. I mean, I think that the cleanest comparison is arm B versus arm A. Arm B versus arm A clearly establishes that zanidatamab is superior to trastuzumab. Again, I mean, arm C, while the median progression-free survival is similar, two-year PFS is improved with the addition of tislelizumab. The hazard ratio moves a little bit down. If we look at median overall survival of C versus B, I mean, clearly, there's immaturity there.
But there is a suggestion, especially in the median duration of response, that the immunotherapy is providing an additional boost, as we would expect, and as really we saw in Keynote-811, where there was an incremental benefit for the addition of pembrolizumab to trastuzumab and chemotherapy. So maybe coming back to your point, I mean, I think, I mean, maybe to answer the question directly, I guess the somewhat contrived situation that arises is if someone had, maybe I think it's a good example, ongoing inflammatory bowel disease and actually was not eligible to receive any immunotherapy. And really, because of their baseline bowel issues, the additional diarrheal toxicity of zanidatamab might make me offer trastuzumab and chemotherapy.
But certainly, I think pembrolizumab and chemo, I think in the next, well, certainly, I mean, once it becomes, once zani, chemo with or without tislelizumab becomes an NCCN compendium listing, I think the Keynote-811 regimen will become a historical regimen.
Our next question comes from Mohit Bansal from Wells Fargo. Please go ahead. Your line is open.
Great. Thank you very much for taking my question. So just following up on that, the previous question, so Dr. Ku, do you think you would, if you're given a choice, would you still use tislelizumab with zani in your PD-L1 high patient setting or even PD-L1 low ? Or could you use a Keytruda in combination with zani, even though the data are not there?
Yeah, that's a good question. I mean, I think it's probably a broader question of whether all PD-L1 inhibitors are comparable. And again, I mean, tislelizumab was designed rationally to be a slightly better PD-L1 inhibitor. But whether it's superior to pembrolizumab or nivolumab, I think, is unclear. I think my honest appraisal is that all the PD-L1 inhibitors are largely similar.
So I think that—so I think clearly, I think the story here is zanidatamab. I think tislelizumab is an important contribution. And I think ultimately, it will depend on the specific wording of the NCCN compendium listing and ultimately the FDA approval. If it's specifically in combination with tislelizumab, happy to give that. I mean, if it's with any IO drug, then potentially. In fact, if anything, I would prefer nivolumab. I think pembrolizumab clinically is not a great drug because it's given every three weeks or every six weeks. And we in the U.S. typically prefer FOLFOX every two weeks. So the pembrolizumab has sequencing issues.
I mean, I would say that, I mean, I can maybe speaking to the Memorial experience, up to this point, tislelizumab has not been on our formulary, even though it is FDA-approved as first-line treatment for HER2-negative esophagogastric cancer. But based on these data, we are planning to add it to our formulary in anticipation of the fact that it will be given in combination with zani and chemo. And while kind of pricing issues are completely opaque to us as medical oncologists, my understanding is that tislelizumab is priced 10% cheaper than pembrolizumab and Nivo. And certainly, any additional healthcare savings to what will ultimately be an expensive regimen would be appreciated.
Our next question comes from Akash Tewari from Jefferies. Please go ahead. Your line is open.
Hey, this is Manuel Joan for Akash. Just a few from our end. Do you see the overall survival curve separation dynamics within the PD-L1 high population consistent with the ITT population, both in doublet and triplet arms? We have seen the PD-L1 component adding a late tail effect for the survival benefits, like in the Keynote-811 final analysis. Is there a possibility that zani's effects could be more front-loaded and zani doublet showing less of a further overall curve separation going forward? Lastly, in general, what are your expectations on the doublet hitting the overall survival endpoint in the next interim? Thanks.
Yeah, I'm not sure I completely got all the questions. But I think maybe one of the questions, if I understand correctly, is whether the, well, first of all, I've not seen any of the survival curves based on PD-L1 positive and PD-L1 negative. I think, honestly, in my discussions with colleagues and other KOLs, I think—I mean, again, I think one of the things you have to be cautious about with subgroup analyses is that if you have 15 variables and if you delve deeply into them enough, you're liable to see an association or a result that you did not expect. I think, again, the broader picture is that I think the very pleasant surprise and I think good news for our patients is that all patients seem to benefit.
Certainly, I've not seen separate Kaplan-Meier curves based on PD-L1 cutoffs. I think certainly in the future, that would be—I think that would be, I think, an interesting and welcome analysis. I think maybe the main point of your question really was whether immunotherapy offers a tail at the end of the curve, as we have seen with other studies, and yes, I mean, that actually would be my expectation. Again, I think the median duration of follow-up is right around two years, which is really right around the median PFS and overall survival numbers. So I'm hoping that with time, I mean, especially when you have a median duration of response of nearly 21 months, that we will begin to see a tail at the end of the curve for arm C favoring arm B, but we'll have to see.
And I think finally, I think with regards to your question about whether I expect that arm B will eventually hit statistical significance, it is on the verge. There is a numerical superiority at this point. I certainly don't have a crystal ball. But I think, yes, I mean, the expectation is that hopefully with the next interim analyses, arm B will have a statistically significant improvement in overall survival versus arm A.
Thanks, Dr. Ku. This is Rob Iannone. I would love to add my perspective on some of the questions that have come so far. The place that I'd like to start is just to remind everyone that fundamentally, this study was designed to compare zanidatamab to Herceptin in this highly HER2-driven cancer. And the results we believe show that zani is definitively better than Herceptin really on all efficacy measures and therefore should be the new HER2 agent of choice. That's really the first decision that any practitioner needs to make. Secondarily, we explored whether tislelizumab would improve on those outcomes. And the results we have certainly show that irrespective of PD-L1 status.
And that last finding, as Dr. Ku mentioned, is really quite unique and differentiated and we think could be related to zani's unique immune mechanism of action. And we also find the results that tislelizumab adds to the benefit to be highly consistent with all of the other data in the literature showing that this disease is sensitive and responsive to PD-L1 inhibitors. When you look at the PFS curves by PD-L1 status, the upper bound of the 95% confidence interval is below 1 for both positive and negative subgroups. And then when you look at the overall survival curves, these results are also consistent regardless of PD-L1 positive status, with the upper bound of the confidence interval slightly crossing 1 in the PD-L1 positive subgroup.
But I do want to call out some limitations and to ensure that people are not getting confused when they're looking just at hazard ratios. Remember, this was not a stratified analysis for PD-L1. And therefore, there may be confounding factors due to imbalances in other prognostic factors that are contributing to those small differences that otherwise show that the results are highly consistent between these two subgroups and favor the zanidatamab-containing arms. I also want you to be aware that for the PD-L1 analysis, it's likely to be further confounded by subsequent IO therapy on the control arm, which would not have been possible when Keynote-811 was conducted, noting that the zanidatamab-tislelizumab-chemo arm showed a median overall survival of more than 26 months regardless of the PD-L1 status.
So again, for me, the key message here is that zanidatamab went head-to-head with Herceptin. It definitively beats Herceptin across all efficacy measures and truly should be the HER2 agent of choice for all patients, as Dr. Ku mentioned. And then, just coming to that very last question of, I agree with Dr. Ku. I don't have a crystal ball either. But I do want to just be sure that people are clear around the statistics here. This was sequential testing. Both PFS measures hit. That will not be retested. arm C beat arm A on overall survival. That will not be retested.
So the small amount of alpha that was allocated to the first interim analysis, sorry, the remaining large amount of alpha will be carried forward to the next interim OS analysis, which will focus on arm B versus arm A. And then, of course, there's a final analysis as well. And given the strong statistical trend, we're optimistic about that.
Thanks, Rob. I agree with all those points. And I actually think that. I think one of the very happy situations where we're slowly kind of finding ourselves in is just like in breast cancer studies, progression-free survival is a validated endpoint because there are so many subsequent lines of treatment. And I think while that would be an overstatement in HER2-positive gastric cancer, I mean, we now have trastuzumab, deruxtecan. We have ADCs. Certainly, patients in arm A who did not receive immunotherapy at some point could have received that post-approval of Keynote-811 in their countries. So I think that the overall survival results are great, but they are likely to be confounded by post-progression therapy. So I think the purest distillation of the results is if you focus on the progression-free survival. And certainly, arm B versus arm A shows that zanidatamab is superior to trastuzumab.
And certainly, I mean, arm C versus arm A suggests that all the hazard ratios and all the median numbers and all the percentages of long-term survival are all in favor a little bit versus B, although again, that's not a statistically valid comparison at this time. So I think certainly zani and chemo, I think likely with tislelizumab for many, many patients is, again, I think will be the new standard of care, hopefully very shortly.
Our next question comes from Ami Fadia from Needham. Please go ahead. Your line is open.
Hi. Thanks for taking my question. I wanted to just ask about the difference in the PFS and OS data that we saw between the pivotal trial and the prior Phase II studies. How do you think that the baseline HER2 status or the choice of chemo may have contributed to the difference that we saw between the two studies? And then maybe just a follow-up for Dr. Ku. If you sort of think about the various settings where GEA patients are treated, do you think that there might be any difference in the rate of adoption based on either PD-L1 status or HER2 status? Thank you.
Yeah, I think that it's a great question. I mean, I think that—so I mean, I was the last author on the ZW25-201, which was the Phase II study that was performed with zani and chemo. And as you know, it evaluated several chemotherapy regimens, specifically FOLFOX, CAPOX, and 5-FU cisplatin. And certainly, I think virtually all U.S. investigators preferred FOLFOX.
That certainly will be a preferred chemotherapy backbone to combine with this regimen moving forward. Again, I mean, I think as everyone on the call is aware, I mean, there can be differences. I mean, typically, the efficacy drops from Phase II to Phase III for numerous reasons. I think that the results from Phase II to Phase III actually are broadly consistent. Probably the one kind of outcome that I was a little bit surprised about was that there was not a more significant improvement in response rates for the zanidatamab arms versus trastuzumab chemo. In the Phase II studies, that response rate was in the 80%-85% range. But I think, I mean, and one of the analyses I'm certainly will be interested to see is kind of sites of disease and maybe delving a little bit into some of the RECIST measurements.
But other than that, I actually think that the Phase II results translated pretty reliably into the Phase III. I think maybe I can't remember whether it was a question about toxicity or not. I think it's difficult to comment on the toxicities of FOLFOX versus CAPOX, for example, because there are also kind of ethnic and geographic differences in patients who receive FOLFOX versus CAPOX. I guess finally, in terms of the last question, in terms of adoption, again, I think my personal interpretation of the data is that in a patient who's otherwise fit, I think the triplet combination of zani and tislelizumab and chemo would be something I would strongly consider. But I think at the same time, if there are specific reasons or contraindications against an immune checkpoint inhibitor, zani + chemo itself is highly, highly active.
Certainly, not to speak for community oncologists, but I think in the community, and I think certainly with patients as well, I mean, there typically is a strong preference for an immunotherapy-based treatment. So I actually do think that there will be pretty brisk uptake of the triplet combination. HER2 testing and PD-L1 testing are extremely robust. I don't think there's going to be any issue in terms of biomarker testing with rolling this out. I mean, I think a more granular response would be the study uses TAP. TAP is rarely used either through commercial panels or at academic institutions. But we also know that TAP and CPS are highly concordant.
So I think that, again, I mean, to me, it seems like there is benefit irrespective of PD-L1 status. But PD-L1 and HER2 assessment are the two biomarkers that are almost always performed. And we have that data available when we make treatment decisions about first-line therapy.
Our next question comes from Annabel Samimy from Stifel. Please go ahead. Your line is open.
Hi. Thanks for taking my question. So I just wanted to look at some of the subgroup analyses. I'm curious if you think there would be any limitations based on IHC status for IHC 3 versus IHC 2 and for GEJ patients given what we've seen in some of the subgroup analyses.
Yeah. No, that's a great question. So sorry. Go ahead, Rob.
No, I didn't mean to speak over you, Dr. Ku. I was just going to maybe start the answer if that's okay. We've given this actually a fair amount of thought. It's very important to note that as you do multiple subgroup analyses, there's going to be some variability. And it's particularly prone to variability when you have very small subgroups. As was expected, this trial recruited largely IHC 3 + patients more than 80%. So you wind up with a very, very small subgroup in the IHC 2 + patients. And it's clear that that small group could be potentially confounded by other prognostic factors that are driving different results in the control arm. So we don't think that's a meaningful result and don't expect that will be a limitation either on approval or in practice.
Yeah. I mean, sorry. Go ahead. Well, I mean, let's just say I would agree with that response. And maybe in the interest of time, we can certainly move on.
Our next question comes from Joon Lee from Truist Securities. Please go ahead. Your line is open.
Hey, congrats on the strong data. And thanks for taking our questions. How will you be addressing FDA comfort around U.S. generalizability given the absence of U.S. patients in the HERIZON GEA trial? And will you be seeking label expansion for both the doublet and the triplet, or prioritize triplet? And will you be seeking priority review? Thank you.
Yes. So we certainly will be seeking broad label approval of zanidatamab as well as the triplet with tislelizumab as a broad label PD-L1 agnostic. And we do think the data are compelling enough for this to be considered under priority review. But of course, that is an FDA decision. With regard to questions about regional results, we're actually quite pleased in this trial with what we're observing in terms of results across the globe. I just want to remind folks that the regulatory standard here is that the patient population is representative of the U.S. population.
It's not necessary to recruit patients directly in the U.S. And as Dr. Ku mentioned earlier, there were very good reasons not to include U.S.-specific patients in this trial. But we did include other patients from North America and Europe. We do believe the enrolled population is very representative of the U.S. population. This is in part based on the eligibility criteria, but also in terms of the supportive care standards that were implemented as part of the study itself. And there's strong evidence that this turned out to be the case. When you look at the PFS curves in the control arm for this trial, HERIZON 01, and compare it to the most recent trial in the similar population, Keynote-811, they essentially are identical in a pretty remarkable way.
And so that demonstrates that there's high comparability in the patients who were recruited when judged against how they performed in the control arm. And of course, Keynote-811 did recruit patients from the U.S. And then I would just highlight that Dr. Ku I think made a really critical point around focusing on PFS as being the best distillation of the effect of the experimental treatments given that it's not confounded by subsequent therapies. If you look at PFS, highly consistent and clinically meaningful results across all regions. And the OS results are consistent with that as well. So overall, yeah, we'll be seeking broad label. And we do feel this trial was enrolled patients who are representative of the U.S. population and that the results are consistent across regions. And we do not think that this in any way would be an approvability issue.
Our next question comes from Andrea Newkirk from Goldman Sachs. Please go ahead. Your line is open.
Good morning. Thanks for taking the question. I apologize if I missed this. But Rob, just as a follow-up there, as you think about the profile for the zani doublet and the triplet, many of those measures that you're speaking about, including PFS, do appear to be somewhat comparable across the doublet versus the triplet. Would just be curious to hear your thoughts on why you think that might be. And then second, does that pose any risk to the regulatory agencies as they assess or view the value of the incremental benefit of a triplet versus a doublet here?
Yeah. Thanks for the question. So as Dr. Ku pointed out, when you look at PFS at the later time points, you do begin to see separation with the triplet relative to the doublet. Of course, the trial was not designed to compare the triplet to the doublet, and that's not the regulatory standard, so it's not necessary, but as you look at that sort of underpowered comparison, you do begin to see separation, but also, we've known for a long time that the benefit of immunotherapy PD-1s is often better appreciated on survival, and that shows up in this trial as well where the survival curves tend to separate even more. So for me, my interpretation is that this trial was fundamentally a comparison of zani versus Herceptin, and zani definitively beats Herceptin across all efficacy measures and all measures of benefit-risk.
And then when you look at the addition of tislelizumab, it also contributes to efficacy really on all measures. The important thing is this is consistent with the broader, very large body of literature showing that PD-1 antagonists are effective in gastric cancer. Whether you look at HER2 positive or HER2 negative, whether you look at the metastatic setting or more recently in the neoadjuvant setting, PD-1s are effective in this setting. What's unique about this trial, and we do think it could relate to the differentiated mechanism of action for zanidatamab, is the benefit of the triplet seems to be across PD-L1 subgroups, which is really differentiated.
On balance, I do think that with the rare excepti on that Dr. Ku highlights of the patient having a contraindication of sorts, that the best therapy for this still poor prognosis disease is to get the best HER2 agent, zanidatamab, and a PD-1 inhibitor upfront.
Our last question today will come from Jason Gerbery from Bank of America. Please go ahead. Your line is open.
Hey, guys. This is Chiung for Jason. Thanks for squeezing us. And I have one question for Dr. Ku and one question for Jazz. And so Dr. Ku, now that you've seen the Phase III data and while we are waiting for the FDA approval, would you or would you expect some of your peers might start to incorporate zani in frontline GEA either right away or when NCCN updates its recommendation guideline?
In that case, is there a certain patient population you would think could be an early adopter for either zani doublet or triplet regimen? My question for Jazz is, what's your expectation for the duration of exposure or treatment exposure for the zani component in frontline GEA given the disconnect between the 12-month median PFS and the seven- to 10-month duration of therapy described on the safety slides? Thanks so much.
To answer your question, I mean, actually, the simple answer is yes, actually. A colleague of mine yesterday surprised me by saying that he's actually, since the press release, managed to get private insurance to support this regimen in the first-line setting for one of our HER2-positive gastric cancer patients. So I think that certainly, I mean, I think really just about everyone I've talked to, I think, really feels that this is practice-changing.
I personally cannot wait for it to be included in NCCN guidelines, which will then make it an option for patients on Medicare and Medicaid. I think based on my colleague's example, the next patient I see with private insurance, I think now, especially since the data set has been presented, I certainly would be willing to appeal to the insurance to see whether they will approve zani even at this point. I think that once it's included in listings, once it's FDA approved, everyone I've talked to, I think, will move quickly to adopt it.
Based on the results published now at ASCO GI, we have submitted for NCCN consideration. Of course, we'll update that if need be with a full manuscript publication when that comes, which we're working quickly to get published.
We are out of time for questions today. I would like to turn the call back over to Renee Gala for closing remarks.
Great. Thank you, operator. And just in closing, we are thrilled to deliver the results we've discussed today and to deliver a scientific advancement of this magnitude that can redefine outcomes for patients with first-line HER2 positive GEA. We do believe zani has the potential to transform treatment for multiple HER2-expressing cancers, and we are focusing on its broad development plan. zani exemplifies our approach to corporate development as well as we seek to identify and acquire differentiated therapies that improve patient care, are strategically aligned with our development and commercial capabilities, and can create substantial shareholder value. We're excited about the potential for zani to become the preferred HER2 targeted agent of choice and the opportunity that zani data mab represents going forward. So thank you all for joining us today.
This concludes today's conference call. Thank you for your participation. You may now disconnect.