Technically day two of our Jefferies Healthcare Conference, in New York. I know all of you guys are enjoying Times Square, the elevators, the Elmos that you can see downstairs. But, you know, for those who don't know me, my name is Akash Tewari. I cover pharma and biotech here at Jefferies. I have the pleasure of hosting the CEO of Zymeworks. Ken, I'll hand it off to you for some brief intro remarks, and then we'll get into it.
No, it's great. Thanks, Akash, and, oh, this is, it's working. Yeah, happy to be back here again, for a fireside chat. Not that I'm not looking forward to doing this right now, but a year from now, I'm really gonna look forward to that fireside chat.
Mm
... 'cause I can actually hopefully talk about actual data from the pending HERIZON-GEA-01 study, as opposed to trying to help you forecast and read the tea leaves, which I can't really do. But really excited to be here, really excited about where Zymeworks has gotten to at this place since I started in early 2022. We've really advanced zanidatamab with our partners, Jazz and BeiGene, to the standpoint of now being in front of regulators for our initial indication, biliary tract cancer, and looking at a substantial Phase III readout in first-line GEA targeted before the end of this year, and with Jazz obviously talking about investing in further indications beyond that. So really happy with our partner of choice in 2022, and where zanidatamab has gotten to and its potential.
Mm.
And beyond that, we've been working on the rest of our portfolio that we talked about. So we've got now five planned INDs in the next 24 months on innovative medicines, both antibody-drug conjugates and our own approach to trying to make T-cell engagers more effective. So we're really excited to move all five of those agents into clinical studies over the next 24 months, with the first two coming up pretty soon. So we're really excited about where we are positioned. I think we've been very scientifically disciplined around that portfolio and still maintaining our financial discipline to make sure that we can fund all of the great R&D ideas in the side of the company, and certainly looking forward to the rest of this year and moving into next year with the potential that I see in front of Zymeworks.
So, Ken, I'm glad you started with ZW25 because I can tell you, we're getting very built up on the story, both for Zym and actually Jazz from a risk-reward perspective. And I wanted to get some very simple blocking-and-tackling questions out of the way first-
Sure
... 'cause I think they're important. From a milestone perspective, if you get BTC and GEA approved, how much are you going to get paid just on milestones?
Good question that I can't answer. Obviously, the totality of the financial arrangement we have with Jazz and BeiGene are public. Individual amounts that you're asking about are not public. They're between Jazz and BeiGene and ourselves. Obviously, we have, in the Jazz deal, $525 million in regulatory approval milestones still to come, some of which will be on BTC and GEA. We have $900 million in commercial milestones based on annual sales targets that are hit by Jazz, and we have a royalty that scales from 10%-20%, based on annual sales levels. So I think from my perspective, I think we have some additional financial elements that are, you know, based on the GEA outcome.
and ultimately, we think, you know, zanidatamab, we don't disagree with Jazz. We think zanidatamab has a place in BTC, GEA, and other indications, and we think it's gonna be a meaningful medicine, and I know Jazz has guided last year to this being a $2 billion+ peak sales potential product-
Right
... for them. We don't disagree with that. BeiGene's not made the same forecast yet. So we believe zanidatamab is gonna be a meaningful medicine in a number of different clinical indications, and biliary tract cancer was the first place we had data in back in 2022 to show us that, and it's in front of regulators now. And with the GEA data being targeted for later this year, hopefully, that gives us additional data to confirm the belief that we think zanidatamab is a very different medicine, and in the HER2 space, has a key role to play for patients beyond what others have been able to accomplish.
Ken, so I'll push you on that.
Yeah, sure.
Jazz, I think, has stated that the majority of that $2 billion is BTC and GEA alone. Would it be logical to think then the lion's share of that $500 million plus in milestone payments would be tied to commercial approval in BTC and GEA?
Yeah, I can't... Yeah-
That's fair
... I can't go further than the comments that we already have. I mean-
Yeah
... you have the benefit of being the only analyst, I think, who covers both Jazz and Zymeworks. So
It's very rare.
You know, you're the only one, so I think with some of these questions you can put to Jazz. You know, it's fair to say that BTC and GEA are important indications, and in the deal construct, we would've wanted to make sure that we, you know, get our share of the benefit that's provided by having those indications approved. And the actual amounts, once we earn them and receive them, we'll announce them, and then that will answer the question.
Understood. Okay, so, you know, I was just at ASCO. We hosted our kind of takeaways webcast, and one of the things we talked about, you know, it's everyone's talking about this PD-L1 VEGF bispecific from Summit and then BioNTech. So kind of old is new. I mean, you have a mAb where you're having higher internalization, you're hitting the receptor and Perjeta epitopes. And I wanted to start off with just the BTC data because I think this is important. It's not a big indication. I don't think there's any argument on that. But I wanted to see how that translates to, let's say, other indications, where it's kind of similarly heterogeneous HER2 expression.
Talk to me about the data you showed at BTC, not just from a response rate perspective, but also in terms of durability of response and overall survival that you think stands out, and how do you contextualize that, A, with what Herceptin Perjeta separately have shown in that population, then B, say anything in HER2. Even if you don't even express HER2, you gotta think about it in HER2. And how do you think about some of the pan-tumor data they've shown in that setting, too? Why would ZW25 actually stack up competitively with both of those regimens?
... No, good series of questions, all in one statement. Let me try and parse those as much as I can. You know, I think we learned a lot about zanidatamab's potential back in 2022, and I was probably sitting here two years ago, having the same feeling. We had a pending pivotal study readout at the end of 2022, trying to understand if zanidatamab had a differentiated mechanism versus other agents. It's not Traz and Pert in a combination. It's much different than that, that the clinical data would show that. And I think two years ago, there was concern, probably the same concerns in the biliary tract cancer indication as to what you would see.
And you'll remember when the data came out, that data was very compelling, and it was very clear from the clinical data in biliary tract cancer that we were able to provide a patient benefit that had not been seen with Traz, Traz with Pert, T-DM1, or with other agents that were under development. And what we saw from that data was very clear. Those who get a response from ZW25 or zanidatamab get a very quick response. Median time of response is 1.8 months. And those who got a response were very consistent across the patient population, very deep response, and very durable, even though it was zanidatamab in monotherapy, in a very difficult-to-treat indication, where heterogeneity was really relevant to that population.
So you saw clinical benefits that were very different from what you would have seen with other agents that had been tried or that might still be trying. And since that data in the end of 2022, that data continues to mature and even become more compelling-
Mm.
For the patient benefit that you might see, including at ASCO last weekend, where the longer-term follow-up, including the first OS data, was disclosed. And that median OS is well beyond what patients would have seen with prior therapy that was available now or standard of care, and much different than you've seen with Traz plus Pert, and even much more than you've seen with T-DXd. So clearly there's a mechanism with zanidatamab, where in circumstances where heterogeneity is an issue, zanidatamab can provide a benefit that's far beyond what you might have seen with prior trials with other HER2 agents. So we learned a lot about that. That data was very compelling at first.
It got even better over time, and you can see the benefit on a long-term basis of having deep, durable responses in a substantial-
Right
... part of the population. We're two years later, waiting for a GEA readout, where our early data with Zani plus chemo and our Zani plus chemo plus a PD-1 was equally as compelling as our phase I data in biliary tract cancer, and just points to the fact that zanidatamab will provide a clinical benefit well beyond what you've seen in previous studies with Traz. And I think in those tumor types, you know, beyond what you might be able to see with an HER2, because we have the ability to combine with chemo or immunotherapy in GEA.
Right.
And we also have the ability to do that in biliary tract cancer, which is the subject of the confirmatory study that's ongoing now, looking at standard of care, which is PD-1/gemcitabine in most cases, added to zanidatamab. And in that first-line setting, we might be able to go well beyond what we've seen so far. So clearly, with BTC in our pivotal study and with GEA in our earlier studies, which is now the subject of the Phase III study, you can see there's a clear difference in the mechanism that Zani gets you that's different than anything else. And in some tumor types, that benefit might be the way to optimize the potential benefit for patients more than others, either in monotherapies we started or in combination.
Mm.
Our clinical data is targeted to be available, you know, later this year. And we hope that once again, as it did with biliary tract cancer, just confirms the hope and promise that we have for zanidatamab for a pretty large, underserved, you know, patient population where, you know, zanidatamab could play a really key role, if the data supports that.
Understood. Maybe just to put a finer point on that-
Mm-hmm.
heterogeneous HER2 expression, you see that in BTC. Is it fair to see to say you see a very similar dynamic in GEA, or are there some nuances here that we should be paying attention to?
You know, different tumor types, obviously different patient populations. But I, you know, I think the very nature of how you think about the mechanism of Zani working in areas where the expression has heterogeneity in it seems to lend itself well to Zani's mechanism.
Mm.
That's been in all the datasets that we've looked at. In addition, zanidatamab's tolerability profile allows it to be combined with other agents. So in GEA, we're combining with chemo, we're combining with potentially PD-1 as well. In our metastatic breast cancer studies that we did before our partnership with Jazz, we obviously looked at a whole range of combination approaches as well as early lines with Zani as monotherapy. So different tumor types, different clinical studies, we need to see the results. But I think if zanidatamab, you know, has a particular mechanism to support its use, it would be in these tumor types where there's heterogeneity, which has been difficult to treat with other agents and also with T-DXd.
Hopefully, our clinical data supports that zanidatamab is the workhorse to deal with those really difficult-to-treat tumors, where in GEA, you know, there are patient populations that were not as well served maybe by the regimen in KEYNOTE-811-
Right
... which is why it maybe wasn't as successful. There are some hard-to-treat patient populations in there where, you know, Traz in that regimen was just not able to deal with that heterogeneity. And that is where we think Zani could have a place, and hopefully, the data confirms that we have a very consistent response across those difficult-to-treat tumor types and subpopulations.
Okay, so I want to hit on that. What do you mean difficult to treat? I think everyone is thinking, "Okay, is there, you know, low HER2-low expression? Is it heterogeneous?" Or, actually, when I've talked to you about this, it's not that, right? So talk to me about what you've seen with KEYNOTE-811, and the subpopulations where you think you have an ability to really differentiate.
... Yeah, I think if you look, you know, if you look through the ePAR for KEYNOTE-811, that was the subject of the EMA approval, and again, we like to look to try to learn from others who are ahead of us. And you can clearly see that in some of the subpopulations, not just high PD-L1 scores, but if you look at some of the older patients, male patients, Asian ethnicities, you saw the performance of that regimen from KEYNOTE-811 was not as successful as the median, and I think that's what held back the overall trial result and maybe where it wasn't as successful as the sponsor hoped it would be or KOLs-
Right
... would hope it would be. You know, what we've seen with our clinical data in BTC, in our early GEA studies, is zanidatamab tends to be able to get a consistent response, despite some of those characteristics that have definitely been considered harder to find a response in.
Understood.
So if we can repeat that consistency in our phase III study, then hopefully we'll find that the consistency of result lends us to an overall result and a median result, which is higher. So we're definitely focused on not just the headline of what median PFS might be out of a study, reported later this year, but then in looking at the populations, just understanding the consistency response across different characteristics of patients. That's really important from a KOL perspective and dealing with their patients and their practice.
Right.
Obviously very important for the commercial opportunity that will be the result for Jazz and BeiGene.
Now, you know, I think several months ago, I don't know, I'm starting to forget exactly when, but you talked about enriching your study for OS, right? You added more patients for OS. I think one of the things that, you know, you're right, like, you have to look at the EMA review for KEYNOTE. It's not like KEYNOTE 811, that Merck regimen had no effect in PD-1 high- in PD-1 low patients, right? It's actually very, very borderline. It was like a month or two, a month and a half differences. The FDA only gave an approval in one subtype, and then, of course, you saw, or EMA so far, but you also saw that they hit on OS ITT. So the question I have for you, let's say that you have the exact same drug as the Merck regimen.
With your increased powering, would it be fair to say you would actually be able to show potentially OS in a PD-1 high and PD-1 low population? Forget differentiation on ZW25, simply from a statistical design perspective.
I don't know how to answer that question absent data. But, you know-
But is that the way you thought about enriching?
Yeah, no, I think I can tell you the way we think. I think in developing zanidatamab in this indication, GEA, you know, we found responses with zani plus chemo, regardless of PD-L1 status. And to us, that makes sense because I think in that case, zanidatamab is the workhorse in that regimen. I think when you do a regimen including a PD-1, which includes trastuzumab, you know, our view is that zanidatamab provides you know, a different clinical mechanism, and in those tumor types, maybe a better potential chance for response than trastuzumab. That's our basic belief. And PD-L1 status did not really come into that discussion.
You know, we added a third arm to the GEA study, the HERIZON-GEA-01 study, because we wanted to understand for physicians, if you added a PD-1 to the patient population, which potential patients might benefit from having a PD-1 as a part of the regimen, in addition to zanidatamab and chemo. And that data will come out of the study with a pinch of how efficacy might be enhanced and which patients it might be enhanced with, and also to understand, you know, what the additional tolerability profile is of adding a PD-1, in this case, tisle with zanidatamab and chemo. So I, you know, I don't think we went into this thinking that zani would not work in, you know, high PD-L1 patients.
You know, we found a very consistent response across, regardless of PD-L1 status, but it was important to inform ourselves and inform physicians eventually and regulators. If you added a PD-1 to that combination of zani plus chemo, where, where would the benefit be seen? You know, we haven't seen that data out of KEYNOTE-811 yet, because other than-
Yep
... dividing between PD-L1 negative and positive, we haven't seen where the most benefit came from that patient population. You know, if that's presented at some future date, then I think it will, it will inform KOLs and physicians about, about where the, the most benefit may come from. You know, at least with our data, it's all pre-specified as subpopulations, so that data should eventually become available from that study to understand exactly where a PD-1 may be relevant beyond what benefit you might get with zanidatamab and chemo. So I think we just came at it from a different place than-
Understood
... than the sponsor for KEYNOTE-811.
I know there's limited you can say about your stat design and how you're kind of assigning alpha. Is it hierarchical? You know, I don't know. But, tell me this, if we had a scenario where the triplet did have a benefit in both PD-1 high and PD-1 low, and the doublet, let's say, had a, you know, an outsized benefit in that kind of PD-1-low population, fair to say both of those regimens technically would be fileable to the FDA? Or no, you would have to run separate studies to kind of explore that signal.
Yeah, again, I don't know how to answer that question, but again, you know, we're really focused on, you know, getting the study fully recruited, including the extra 100 patients-
Yeah
... for the OS, the OS endpoint. I think, you know, we clearly understand when we would be able to look at the PFS endpoint at the end of this year. There is an interim OS analysis because that would be necessary to share with regulators, to understand the trend in OS if it's not mature enough to be statistically significant. I think, you know, the study design is essentially there's no comparison between the triplet and the doublet because of the way it was structured.
Right.
So it is, you know, zani plus chemo compared to Trast plus chemo, and zani plus chemo plus tisle compared against Trast plus chemo. So there are two independent calculations which are done. You know, I think the intention of Jazz and BeiGene would be to get that data and discuss it with regulators around the world to understand what their view of that is. And again, I still think, you know, from our standpoint, we think zani plus chemo is a great regimen regardless of PD-L1 status, and we'll have to wait and see if adding a PD-1 inhibitor to that combination is beneficial for some patients, and in that case, you know, which patients it may be beneficial for, and if that's clear in the data, that would obviously be helpful for regulators and physicians to understand.
So can you know, when you think about stat design, you always, you know, prepare for the worst, you kinda hope for the best. So generally speaking, would it be fair to assume with your current stat design on PFS, you would be able to have a statistically significant signal, with a PFS benefit as low as two months? Is that two or three months, is that generally ballpark how we should be thinking about this, or can you not make those comments right now?
Yeah, I mean, I can't really make that commentary. I mean, obviously, you know, there obviously is a delta that's required to be statistically significant, and that's pre-specified. You know, I think coming out of having the benefit of looking at the KEYNOTE-811 data, you know, I think we felt very comfortable with the statistical design for the PFS endpoint out of HERIZON-GEA-01. Obviously, Jazz decided when looking at the OS endpoint to add an additional 200 patients in the study to give us a higher probability of being statistically significant on OS basis. And I think that was a smart decision-
Right
... after having the ability to look at the KEYNOTE-811 data in a very similar patient population. So I think that gives us the ability to have PFS be a fileable end, you know, a fileable endpoint at the earliest possibility, while giving us a higher probability and a better chance of getting to OS later if it takes longer to mature, which did happen in KEYNOTE-811.
Yep.
So I think that was a very good strategy to get this medicine to patients as quickly as we can prove a PFS benefit, that's clinically meaningful and statistically significant, while still continuing to follow the study, for OS and hoping that matures also to that, to that standpoint. So I think that's- it's a well-constructed study, it's well-executed. I think it's designed extremely well, and we just need to wait, to see the data at the end of this, you know, targeted before the end of this year, to, to understand if our belief in zanidatamab in this patient population, can be confirmed and, and supported to regulators.
So, Ken, maybe on that point, it's funny 'cause like I, I have asked you this question in the past, and, you know, you're like, "Yeah," you've never given an exact answer, understandably. But I also sense that your belief in terms of a clinically meaningful benefit for ZW25 is maybe a PFS benefit of 4-5 months or, you know, something larger than that, what I just kind of outlined. Would you agree with that? That, you know, there's maybe data you've seen so far in humans you've published, that would make you feel confident that a 4-5-month PFS benefit could be delivered with ZW25?
Well, obviously, you know, our phase 2 studies with Zani plus chemo and also Zani plus chemo plus tisle are very compelling. But again, they're smaller studies. But those have continued to mature. So, you know, we've presented the phase 2 data in first-line GEA with Zani chemo twice already. We do continue to follow that study, so it continues to mature. So I think, you know, from everything we've seen from Zani, both inside GEA and outside of GEA, you know, it gives us hope and belief that we really could change the standard of care, if data supports that, for this patient population, which is large, unmet need, and hopefully in a very, you know, substantial way. I think you do try to be conservative.
I think for biliary tract cancer, back 2 years ago, it's a similar situation. Our phase 1 data was very compelling. The question was, could we replicate it in a larger study that's, you know, a phase 2B for pivotal?
Right.
What other benefits would we see longer term? We didn't know that. I think if you look inside the company in 2022, I was probably the most conservative forecaster of what that study would, what the outcome would be. And I got a really big surprise to the upside, that it was much better and compelling than even I had anticipated. And that's okay. CEOs don't mind a good upside surprise as well on clinical data. So I think we try to be conservative, run the best study we can, make sure the data is clear and interpretable, and take a look.
But clearly, in biliary tract cancer, that data was very compelling in terms of speed of response, the deep and durable responses, and now we know from the ASCO data that that's turned into a longer-term benefit in OS, which is pretty compelling compared to the other opportunities available to that patient population and certainly to what's been studied before. And certainly, you know, although you can't do cross-trial comparisons, what you're seeing with T-DXd in that patient population.
Ken, are you seeing the same speed and durability and depth of response in your GEA phase 2 data that you saw in BTC?
Yeah, I think if you—I mean, the data's been published, you know, twice, you know, twice for each the doublet and the triplet, so I think everything you need to know is in the data. You should look at that. You know, even interesting things like, I think if you look at the GEA data, you know, there's IHC 2+ patients in there and IHC 3+ patients in there. You know, if you look at the data, it doesn't suggest any difference in potential response for the patients we studied. So you don't see that. Obviously, in BTC, we did see that in our phase 2B study, and that was reported at ASCO last weekend. So response does correlate with expression.
Right.
You know, in GEA, in our earlier studies, we didn't see that. So there will be some differences between those-
Mm-hmm
Tumor types and the indications, but I think there's no reason for us not to be confident at this point. I think we've designed a good study, we've executed it very well. The data is hopefully be very clear on PFS as we come up to this year, at the end of this year. And then we're really excited to take that data with Jazz in BeiGene to discuss that with regulators, and understand, you know, what the regulatory pathway would be to bring that to patients based on that data.
And-
If I'm more conservative inside the company about the results, and I get an upside surprise, that's not a bad thing either.
Yeah, like, exactly. Okay. On ADCs, and, this will be unfortunately just one question, but-
Sure
You know, there are a lot of people, DAR, TOPO, GPC3, right? There's a bunch of Chinese ADC companies that have done this, pharmas in-licensed them, whether it's CDH6, whether it's Nectin-4, everyone seems to have copied, admittedly, the Daiichi approach. In some ways, so are you.
Mm-hmm
And that's okay, but, talk to me about like, let's say, in folate receptor alpha, the Genmab ProfoundBio deal, right? Where it's TOPO, it's—I think it's DAR. You're going to get potentially expression in PD-1 high, PD-1 low. I think everyone gets that type of pitch with these types of ADCs. But when you think about either internalization, differences in terms of tolerability profile, how could you maybe differentiate versus some of these other DAR, TOPO ADCs with your platform?
Yeah, I mean, we tried to emulate what we've seen work really well in the DXd portfolio and tried to make it even better on the antibody side, but through optimization. So that's important for us. We've got three really interesting ADCs coming to the clinic this year and next year. Our folate receptor alpha ADC, which is a DAR-8, our NaPi2b ADC, which is a DAR-4, with a muted Fc, and our GPC3 ADC, which is also a DAR-4. So there is some diversity in that portfolio, but that's the same payload, same linker strategy. All three of those have optimized antibodies, which we think would not just binding, but internalization and tumor penetration characteristics, which will enhance payload delivery and hopefully get you more activity, but not at the expense of tolerability.
All those programs are designed to treat broad tumor types, different expression levels, and be used in combination, provided the tolerability data from phase 1 supports that. We're really excited about the portfolio of ADCs that we have-
Mm-hmm
That we're gonna put in the clinic this year and next year. And if we're right, I think we can take this idea of the camptothecin analogs and having our proprietary payload, and optimizing the antibody side of an ADC to improve efficacy. And that'll be really interesting to study that clinical data from all three of those ADCs, which are all gonna be in the clinic, this year and next year.
Understood. We are out of time, but again, always a pleasure to talk to you, and, really interesting conversation as always.
That's great. Thanks for having us. Appreciate it.