All right. good afternoon, everyone. Tyler Van Buren here, Senior Biotech Analyst at TD Cowen. Thank you very much for joining us here at TD Cowen's 46th Annual Healthcare Conference. For our next session, it's a privilege to have a fireside chat with Summit Therapeutics, and it's my pleasure to introduce Dave Gancarz, the Chief Business and Strategy Officer, and Allen Yang, the Head of R&D Strategy. Dave and Allen, it's a privilege to have you guys here. Thank you very much for joining me.
Thanks, Tyler. Really appreciate the invite.
Maybe we'll start. We've got a lot of questions, but maybe we'll start with just one high level question to set a foundation. Again, I'm sure most are aware at this point, but maybe you could just briefly recap the Akeso partnership for ivonescimab, how Summit's role has evolved over time as the program has progressed into late-stage development and global studies and now regulatory review.
Sure. No, I appreciate the question, Tyler. We entered into the deal with Akeso Bio, who are our partners based in China, for ivonescimab, a PD-1/VEGF bispecific antibody. That deal was signed in December of 2022, after, you know, traditional customary review, that it went effective in January 2023. Akeso at that point was running two phase III studies in non-small cell lung cancer. One post-TKI, EGFR mutant, non-small cell lung cancer, one in frontline PD-L1 positive non-small cell lung cancer, so effectively HARMONi-A and HARMONi-2. As we entered into the transaction with Akeso, our immediate intent was to also move into late-stage development immediately.
We expanded HARMONi-A into a global study, that became HARMONi, as well as looking to start immediately into frontline non-small cell lung cancer, and that was the chemo combination, squamous cohort of HARMONi-3. We part of the benefit in working with such great partners at Akeso, a significant amount of data had been generated at the time and also very quickly, and then they've continued to invest heavily in ivonescimab, given the plethora of products that they have in progress and now approved at different stages in China. We work very closely with Akeso on an ongoing and regular basis.
Now as we have introduced our third and now fourth phase III clinical study globally, HARMONi in post-TKI EGFR mutant lung cancer, HARMONi-3, which was then expanded into two cohorts, squamous and non-squamous chemo combination frontline non-small cell lung cancer, HARMONi-7 in PD-L1 high non-small cell lung cancer, that's the monotherapy setting. Finally, recently last quarter, we introduced the first-line microsatellite stable colorectal cancer study as well. That's Ivo in combination with chemotherapy compared to standard of care bevacizumab plus chemotherapy. We've announced that we intend to continue to expand that clinical development pipeline as you alluded to, Tyler. We also announced last week on our earnings call the ILLUMINE study, which will be a cooperative group study based in...i n multi-regional study in Europe and China.
We'll consider expanding that into the United States as well, and that's in frontline PD-L1 positive head and neck squamous cell carcinoma. That's the first study that we've announced in 2026, and we've mentioned that we plan to continue to announce additional studies as those are ready to launch as well, so.
Great. We'll get into each of those studies and indications in just a moment, but wanted to spend a little bit on the molecule's design and the clinical implications, given the data that you all and Akeso have reported. A lot of PD-1/VEGFs out there now, following in your footsteps. You know, as we spoke about the other day following earnings, it's very possible that ivonescimab may have kind of just hit the Goldilocks level in terms of its design and as you think about cooperative binding and daisy chaining. Just maybe elaborate on the molecule and how that's been, in your opinion, supported by the data that you've seen so far.
Yeah. Tyler, thanks for the question. You know, I think it's now a validated class of agents to PD-1/VEGF, right? There's now multiple phase III studies that show a benefit of PD-1/VEGF bispecifics over PD-1. I think the class is de-risked. The one thing I'd like to communicate is that ivonescimab is unique, and unlike the history of PD-1s, which was the first generation of checkpoint inhibitors, where the PD-1s seem to have similar data and almost interchangeable, it was a question of PD-1 versus PD-L1. With the bispecifics, because of this cooperativity, the format will matter. Okay? You know, ivonescimab is designed around, you know, VEGF-A binding sites and PD-1 binding sites. Some of the other molecules will try to bind PD-L1 and not PD-1. Some of the other molecules won't have the daisy chaining ability, right?
We know that there's what we call intramolecular cooperativity. The binding of PD-1 increases the binding of VEGF in vitro, and VEGF increases the binding of PD-1. There's this also unique feature of daisy chaining, and some of those molecules don't daisy chain, and I think that will impact the efficacy. We're beginning to see differences arise. The last part I'd say about the format being important is where the binding is. Some molecules have an X shape where there's a VEGF and then the PD-1, and then there's some that have the VEGF PD-L1.
There's some that have like a super Y structure where they have a PD-1 and a VEGF, and we think that will be important as well, and they may daisy chain a little bit differently. I think the daisy chaining is important in terms of increasing the cooperativity for binding of PD-1, but we now know that it has to do with sort of how it interacts with the PD-1 on the surface of cells, and that may cause sort of downregulation and degradation of the receptor. We believe that's gonna be important in the future as well. Given that complexity, it's not just a molecule blocking the receptor, right? It seems to be that there's some dynamic structures to this, and that will be important and the format will matter.
I think there's enough differences in the different companies' formats that will make clinical impacts as well.
Great. Thank you for that. Let's start with second-line lung cancer or the EGFR mutant second line setting. You know, obviously, positive studies with HARMONi-A, and then HARMONi, the global study that you all ran. The FDA accepted the BLA. We have a PDUFA date now. Can you walk through the interactions that you had with the FDA as part of that filing, the acceptance and how things have progressed to the extent that you can share those details with us?
Yeah. Similar to what we mentioned last week on the earnings call. We have continued to have very fruitful conversations with the agency. We don't get into kind of the meeting by meeting discussions at this point, but the review timeline was provided, and that includes the PDUFA date of November of 2026. We continue to look forward to, you know, the continuation of the review by the agency. We have the utmost respect for the FDA, as I think we've, you know, been very clear on. With that, we don't necessarily go back and forth, you know, in terms of discussing, you know, meeting by meeting with publicly. What it does is it informs, you know, the continuation.
Obviously, the acceptance, was, announced also recently, and then there's, you know, a traditional cycle that comes with that, mid-cycle review meetings, late cycle review, so on and so forth, after the Application Orientation Meeting, the, earlier on in the cycle. There, there's, you know, that normal process set up and we look forward to continuing to work with the agency through that process.
Great. I'm convinced you guys should get approval, you know, based upon the product that's out there and its safety profile and your data. Maybe for the audience you can elaborate about the HARMONi data response, PFS, OS, and what, where your conviction lies following the data that you guys should get approval?
Yeah, it's a great question, Tyler. If we step back, right, there have been four phase III clinical studies that have read out at this point with ivonescimab, three of which were in China in single region, randomized phase IIIs, and then the global HARMONi study, a multi-regional phase III. Four phase III clinical studies that have read out. Four phase III clinical studies that have read out with positive data. That's a pretty good batting average with which to start. When we dive specifically into HARMONi, we had a statistically significant progression-free survival benefit.
Obviously our, you know, as we continued to review, the overall survival benefit was not statistically significant at the time of the primary analysis, with additional follow-up of Western patients. Part of that dealt with the delay in enrollment in the West. It was a phase III study that had never been tested in the West. That delayed enrollment about, you know, four months or so in terms of physicians getting comfortable with the data, getting comfortable with the bispecific molecule, risk of metastases to the brain and the ability to control that, bleeding, so on and so forth. Once we kind of accounted for that long-term follow-up, we had the nominal p-value of 0.03, which, you know, implies the benefit for overall survival as well.
That threshold below the, you know, the magical clinical meaningful 0.80 was seen in both the primary analysis and the long-term follow-up. When we look at what's in the, in the field right now, right? There's one fully approved regimen, that's the amivantamab plus chemotherapy regimen. That was approved on a progression-free survival benefit, a trend, but not statistically significant overall survival benefit. Very much akin in terms of efficacy data. We look at the safety profile. You know, there's been some well-established tolerability concerns with respect to amivantamab. We believe that the data that's been shown in the HARMONi study as well as HARMONi-A, HARMONi-2, HARMONi-6, you know, have shown a very manageable safety profile.
We think when we look at the totality of the evidence, you know, that's meaningful with respect to the need out there in terms of options for patients and physicians to choose from. We also, you know, overall response rate was numerically higher in the ivonescimab arm. Duration of response did not have overlapping confidence intervals in terms of the ivonescimab versus the chemo only arm. As we look at the data, we see, you know, very clear risk benefit profiles that favor ivonescimab as compared to standard of care chemotherapy in the study, as well as what we see and then just in the marketplace, what's available for patients and the need for additional options.
We think this, the study show that this regimen has a nice place there. I think that's, you know, been generally consistent with the feedback we've gotten from KOLs in that space as well with respect to the desire for optionality, and that this regimen has, you know, shown what they're looking for in terms of that optionality.
What about the enrollment in U.S. versus ex-U.S. in that splits as well as the consistency of the data between the Western and the Eastern populations? Do you believe that based on everything you know that the FDA is kind of comfortable with that enrollment split and that the data are consistent enough for approval?
I mean, with respect to the enrollment split, we saw about, you know, 38% in the West, about 62% in the East. That's reasonably consistent in terms of many of the EGFR mutant trials. EGFR mutation-positive lung cancer is, has a bit of a higher prevalence in patients of Asian descent as opposed to, you know, many Western patients. We typically do see a higher enrollment in Asia in non-small cell lung cancer in general. That 50%-65% is a reasonable, you know, range in terms of what's seen from an enrollment percentage there. I don't think there's been a lot of question with respect to enrollment splits by any stretch.
I mean, I think when you bring up consistency, you know, we talk about gold standard of overall survival in terms of what that looks like. It's really difficult to make an argument that the data is not consistent. It's actually more consistent than is often seen in randomized phase III in general from a regional breakdown, especially when those individual regions are not powered. Median overall survival in the West, 17 months for the ivonescimab plus chemo arm compared to 14 months for the placebo. Median overall survival in the East, 16.8 months versus 14 months. Incredibly consistent from a numerical median perspective. Hazard ratios in Kaplan-Meier curves, I think if you overlay them, you know, when you look at earlier maturity versus including later maturity with Western patients, remarkably consistent.
I think we've shown that in a couple of presentations as well. I think the consistency, you know, has been established, pretty solidly. Now it's really, you know, making that case with respect to the benefit risk profile for the application.
Yeah. Just adding to what Dave said, you know, the number of patients, right? That as he said, the EGFR-positive non-small cell lung cancer is more prevalent in Asia, but those numbers of 150 patients from the West were pre sort of negotiated with the agency. We think that we've met that obligation. You know, there was never that number designed to show independent sort of statistical significance in a region, but just to show consistency of the results globally.
Okay. That's very helpful. What's your latest perspective on the market opportunity in this setting for Summit in the Western population?
Sure. I think, you know, we've spoken earlier. When you look at the vast amount of settings by which we believe ivonescimab has the opportunity, so when you look across not only non-small cell lung cancer, but when you go beyond lung cancer into those tumors that are PD-1 sensitive or checkpoint inhibitor sensitive, those that are VEGF sensitive, as well as places like EGFR mutant non-small cell lung cancer, which doesn't have an approval in either checkpoint inhibitor or anti-VEGF therapy, you know, that total available market, this is about 3% of that. From that perspective, it's a small portion of the overall opportunity for ivonescimab. However, it's an opportunity to lay the groundwork, right? It's a thoracic oncology indication, so you're able to build experience with thoracic oncologists.
We look to our next readout that's in frontline non-small cell lung cancer, in the squamous histology cohort. The opportunity to really work with thoracic oncologists, have that availability of the drug to be able to use and administer in patients and understand, is always helpful in terms of setting the ground for the next step. Obviously, there's commercial underpinnings as well that come with that interactions with payers, you know, health systems, so on and so forth as well.
Yeah, as a clinician, it is a rare population, but it is the same physicians that treat the EGFR positive in the frontline. As Dave said, it's an opportunity to enter the market early. The HARMONi-3 readout is gonna be critical for us.
That's great. We'll go ahead and move to the frontline then. Was there in the audience at the plenary session at ESMO when you guys gave that presentation, again, remarkably consistent, a beautiful PFS separation across the subgroups as you guys have shown in prior trials. You mentioned on the recent earnings call that you are now doing a PFS interim analysis in Q2 with HARMONi-3 in squamous population, so as you split it around ESMO. Maybe you could talk about what changed there and what is giving you confidence to take that earlier interim analysis in Q2 based upon the historical data in HARMONi-6, et cetera.
No, it's a great question. When we look at the historical readouts that have taken place to date, there's been two randomized phase III studies that have gone head-to-head against a PD-1. HARMONi-2, ivo monotherapy, pembro monotherapy. That's in the PD-L1 positive frontline lung cancer. HARMONi-6 that you're alluding to, Tyler, read out at last fall at ESMO, ivo plus chemo versus PD-1 plus chemo frontline in squamous, right? HARMONi-6 is pretty much a direct read-through in terms of HARMONi-3. It's very. There isn't much better of a read-through in terms of a phase III study than a randomized phase III done with effectively, you know, incredibly similar combinations in an incredibly similar setting, right? When we look at HARMONi-2 and HARMONi-6, both were highly statistically significant in PFS.
Both achieved that at an interim PFS look. Right. One, when we look at the opportunity to accelerate the timing, it's one following the data that's been generated by our partners at Akeso. We said we'll continuously learn from the data generation from our partners. We've now seen two randomized phase IIIs at the interim PFS, that begs the question in terms of almost why not, as opposed to why here. Secondly, and very importantly, what it also does is it accelerates the conversation with the health authorities, right? You can have discussions with the health authorities, but it's a different conversation when you go in with, "This is a statistically significant positive primary endpoint in a phase III study." That's a different conversation, right?
With an interim PFS, that analysis to be conducted, in the second quarter as we said, that accelerates the opportunity to potentially have a conversation with the health authorities, based on, you know, potentially, you know, positive readout in that scenario.
Great. As we think about that interim analysis, is it fair to say that the duration of follow-up and the number of events is gonna be designed to be very similar to what you saw with the positive interims with HARMONi-6 and HARMONi-2, essentially a little bit?
Yeah, I don't know that we've commented specifically in terms of expectations on, like, median follow-ups or whatnot. We started enrolling the squamous cohort of HARMONi-3 a little bit earlier than the non-squamous, as I mentioned in the opening there. Nonetheless, I think, you know, we're confident with respect to what we've seen in HARMONi-2 and HARMONi-6 and then, you know, what that should read through, based on the similarities of the study.
Okay.
Yeah. if I could
Mm-hmm.
30 seconds just tie it together. You know, you asked, like, what is the partnership with Akeso? You know, they've been a great partner. You know, they've provided a lot of data. They continue to provide great phase II data, phase III data. If you look at what we've done with HARMONi-3, I mean, that study was designed before any of these studies read out. The Akeso HARMONi-2 study read out. This is when we had the confidence to throw in the non-squamous cohort there, right? Because we saw in HARMONi-2 the efficacy benefit existed in both histologies, right? Now the HARMONi-6 data has read out very strongly in combination with chemo, and that's a very similar population to HARMONi-3. I think that informs that decision.
As Dave alluded to, you know, the goal is to bring this to patients earlier, right? As fast as possible, right? As Dave said, you can have that discussion. You know, PFS and OS are dual primary endpoints. When you have one of those endpoints hit, this becomes a positive study, and you can have that Pre-BLA meeting. Prior to having that data in hand, you can't have a BLA meeting, right?
Yep. For HARMONi-3 in squamous, which is reading out first, what do you think is the bar for PFS and OS in order to file and receive approval? Remind us when OS is coming?
Sure. The last question is easiest. Interim OS consistent with the previous disclosure and what we said about the final PFS analysis in the second half of 2026, right? In terms of bars, I think, as opposed to, you know, I never wanna speculate in terms of the agency, you know, and what specific thresholds, and part of that is it's a total package consideration. Safety data becomes very important, and then the endpoints reading out niche is important. When we talk to KOLs and physicians, you know, what they've said pretty consistently in terms of what they're looking for, you know, about a 0.75 for PFS, about a 0.8 for OS.
Those are When you have, you know, dozens and dozens and ultimately hundreds of conversations with, you know, KOLs as well as, you know, significant, you know, high treating physicians, they tend to coalesce around those two points and obviously the study, you know, takes that into consideration in terms of this design as well.
Okay. Would you say that that KOL feedback on PFS and OS in terms of expectations is pretty consistent with the non-squamous population? Maybe you could remind the audience when we're gonna get that data as well.
For clinicians, yeah. I mean, I think that seems reasonable. I mean, if you go back in history, like the 25% improvement in PFS and 20% in ruins, you know, this dates back to prior to these targeted therapies and immunotherapies. Back in the old days when you had chemotherapy, what's the bar that would change your practice? That seemed to be a reasonable bar. I mean, what we're seeing with ivonescimab has been better than that, you know, we're hoping that we can beat that minimum. You know, it's more of a what you say, and it seems to go across products. It's not specific to ivonescimab. The other question was-
Timing of the PFS OS for non-squamous.
Yeah.
Yeah. We've said the plan is to complete enrollment in the second half of 2026, and that would lead to a planned PFS analysis in the first half of 2027. We've been a little bit less specific in terms of overall survival, not because there's not one plan. You know, it's still in the first quarter of 2026, as we look at the first half of 2027, you know, we have the plan, but we wanna make sure we get to the number of events before we, you know, give any formal guidance with respect to that timing. In general, final PFS would be in the first half of 2027.
Yeah. The enrollment started after the squamous, and it's been very brisk, but it is a larger sample size, so...
Mm-hmm.
You know, it's hard to give specifics, but it seems to be enrolling very well.
All right. That's good to hear. HARMONi-2, ivonescimab's obviously already approved on the basis of that in China, Akeso doesn't necessarily have to update that OS data. HARMONi-6,
You're gonna ask anyway.
Yeah. Well, there's no expectations for HARMONi-2, but what about HARMONi-6, right? Akeso might update that, and again, it's not in your hands, but is it fair to say that we might get HARMONi-6 or we're probably not getting HARMONi-2 or?
I'll remind you that this is Akeso's study. They haven't, you know, issued formal guidance with respect to either study in terms of an OS update. What I would say is the protocol's been published with respect to HARMONi-6 and that's a larger sample size as well, and that's an important distinction between HARMONi-2 and HARMONi-6. The HARMONi-6 study is a bit of a larger sample size, you can kind of infer from looking at the protocol there's likely a 26 event somewhere in there. HARMONi-2, because that was, as I mentioned in the beginning, that study was in flight when we did the deal, right? That's specifically designed single region exclusively in China, PFS, not really thinking about, you know, global implications.
If that study reads out positively and everybody goes, "Great, but we're changing the goalpost after the study. We wanna talk about OS." Like yourself. It's a fair question, but I think if we look back in terms of history, what that goalpost did effectively move after. In fairness to our partners at Akeso, they really haven't, you know, provided guidance, and we won't leapfrog above them. HARMONi-6, that was kind of upsized in order to take into account some of these conversations.
Okay, fair enough. We have four and a half minutes to talk about all the other indications.
Yeah.
Frontline lung cancer is huge. I think everyone knows that.
Yeah.
Let's move to colorectal.
Sure.
Why did you guys choose that as the first expansion indication beyond lung? How's that program progressing, and when could we get data?
I was about to say we just began enrollment in that study, so I appreciate the when will we get data? We're not quite there yet in terms of guiding on that point. Look, I think a couple of important points, right? Microsatellite stable colorectal cancer, not a PD-1 indication, right? That's a historical VEGF indication. We had very encouraging phase II data there. We tested that, as we mentioned, with multiple chemotherapies. We provided the FOLFOXIRI data in conjunction with our partners at Akeso and ESMO 2024, continued to enroll that both in China and the U.S. with additional cohorts for different chemotherapies. Settled on the FOLFOX6 regimen based on discussions with KOLs review of the data. That's matured a bit.
We're highly encouraged that led to the go forward there. It's also a cornerstone opportunity. That's a significant number of patients. That's an unmet need. There hasn't been a lot of progress in the last 20 years in terms of MSS CRC, a couple of EGFR inhibitors, right? But not the broader market for all patients there. That's a significant opportunity for patients, you know, with a, with a need, and it's another cornerstone opportunity for ivonescimab.
Tyler, maybe I can whet your appetite for when we'll look at data. A couple things. Remember, we're very excited about this CRC because it indicates that ivonescimab is unique, right? At the simplest way you can think of it is the VEGF component is competing with bevacizumab. We think that there may be something unique about ivonescimab and sort of cracking that immunotherapy going into these colder tumors. As Dave mentioned, there was FOLFOXIRI data published by Akeso in ESMO 2024, which was the excitement clinically that led to the HARMONi-GI3 study. We decided to move forward with FOLFOX, and during that, we did a global phase II to expand Akeso's ongoing phase II. Now there will be global data for phase II looking at the FOLFOX regimen with ivonescimab, and that data you might see sooner.
Can we say it's like maybe this year?
Yeah, probably this year. Yeah.
That will be very important data because Pfizer has released their FOLFOX combination data as well, at the end of last year at SITC. That'll be a sort of a data point that can be compared. You should never compare single arm studies against each other, but everyone does.
Thanks for that, Allen.
No problem.
You guys, you earlier, Dave, you mentioned GORTEC.
Yes.
ILLUMINE trial, head, neck.
Mm-hmm.
Basically, same questions for head, neck. You know, why do head and neck?
Yeah.
How is that program gonna progress? When can we see data?
That, that will, we said early Q2 with respect to FPI there. Look, there's a head and neck is another setting where you would, you know, biologically makes a lot of sense for ivonescimab. It probably wasn't one of our top three or four settings in terms of where to go next. However, we were approached by the GORTEC cooperative group. Anytime you're approached by an external party who has significant interest in running a study of unmet need, you know, with the compound, that becomes really important to take that seriously. That's, you know, that led to, you know, some pretty quick movement with respect to, you know, moving forward with the study. That's a multi-regional study. Again, we'll take a look at bringing that into the United States, as well.
We'll consider that over the course of the next few quarters. What that really does is it continues to broaden the scope. In addition, you know, as I said, if it's not one of our top three or four places where we intended to go, that means there are additional settings that we are looking at. Those. You know, looking at sponsored indications as well in those settings that we plan to continue to, you know, move forward later this year.
What are your top three or four settings in terms of where to go next?
I love the question. We have them. You know, I think as an anecdote to that, when we announced colorectal cancer in October of 2025, that we had planned to go first line colorectal in phase III by November of 2025, there were two other competitors, shall I say, were gonna go into colorectal cancer in the frontline setting. We're gonna wait until, you know, we're basically ready to go forward with those indications as opposed to providing too much groundwork. We're, we're in the background, you know, laying the foundations for those settings right now, and part of that is, you know, We'll announce that when we're basically ready to start enrolling.
Fair enough. Will do my best to exercise some patience here. In closing, since we're up on time, I'd like to ask you both what aspect of the Summit story do you believe is most underappreciated by investors?
I think I alluded this before. I mean, everyone's asking us about HARMONi-3 and what indication we're going to next. You know, I think people think of this as operational execution. Who can throw the more studies out there the fastest? Again, I think the format of the molecules is very underappreciated. I think ivonescimab is at that Goldilocks or sweet spot in terms of design, binding affinity, daisy chaining, that will make it competitively much more advantageous to use across some of these other formats. The other formats may work, but they still have to prove themselves, right? There's a lot of things to look at.
I think on that point, not all PD-1/VEGFs are interchangeable, and we have now shown data in four phase III positive clinical studies. That's... If you're gonna make an adjustment, there's a lot more floor, room to the floor than there is to improve upon that, one, and two, there aren't a lot of compounds that have, you know, started with that track record, you know, that people are not excited about going forward and ultimately get proved out.
Wonderful. With that, Dave, Allen, thank you very much.
Thanks.
Really appreciate it.