Summit Therapeutics Inc. ($SMMT)

All right. We'll go ahead and get started. Good afternoon, everyone. Thanks for being with us. This is the second day of the Citizens Life Sciences Conference here in Miami. It's my pleasure to welcome Summit Therapeutics joining us here. Joining us is Dave Gancarz, CBO and Strategy Officer; as well as Allen Yang, Chief Medical Officer. So thank you guys both for being here. I never know exactly who's in the audience who knows the Summit story, who may not. I don't know who's listening in on the webcast as well. So I always like to start these discussions off with maybe a 2- to 5-minute overview of Summit as a whole.

Sure. And really appreciate, Reni. Thank you for the invitation, and happy to be here. So Summit Therapeutics focuses primarily on our main pipeline asset, which is ivonescimab. So we entered into a strategic partnership with Akeso Bio in December of 2022. That deal went effective in January 2023 for ivonescimab. So Summit Therapeutics has the rights to ivonescimab in North America, South America, Europe, Africa, the Middle East and Japan. So major markets effectively outside of China and Korea. And so as we look through the strategic process, what that involved was getting going in late-stage clinical studies as soon as possible. So part of the impetus behind that deal was really a significant amount of data that was highly encouraging in Phase II in order to allow for that immediate progression into the pivotal studies. And so immediately upon in-licensing ivonescimab, we began by expanding the HARMONi-A study into the global HARMONi study. And that was performed over the course of '23 and '24. And so we've read that data out, and that was positive with results displayed at World Lung last year. And so that was statistically significant in progression-free survival, although not statistically significant in overall survival in the primary analysis and part of that due to the follow-up period with respect to Western patients based on kind of expanding that study from a China-specific study into a global study. That became with additional follow-up time of Western patients, showing a nominal p-value of less than 0.04. And so as we look through the strategic steps for ivonescimab, frontline solid tumors was the core of the priority. And so we went -- shortly after initiating the HARMONi study, we launched HARMONi-3 in frontline non-small cell lung cancer in combination with chemotherapy. While that started as a squamous specific study, that ultimately became squamous and non-squamous. And so we had mature Phase II data in squamous -- frontline squamous non-small cell lung cancer in the AK112-201 study. That prompted us to go immediately into the squamous cohort once we saw the data from the Phase III HARMONi-2 study in China as well as maturing data from the 201, that same Phase II study in the non-squamous cohort that went into both histologies. So that brings us to a full frontline chemo combination study. We also have the HARMONi-7 study, the monotherapy non-small cell lung cancer study in PD-L1 high-expressing populations. And then recently, we initiated the HARMONi-GI3 study, which is a global study in microsatellite stable colorectal cancer. So 4 Phase III studies are sponsored by Summit at this point, 3 in non-small cell lung cancer and one in MSS colorectal cancer. And so where do we sit then today based on some of the work that we've done as well as our partners at Akeso. So between the 2 companies, 4 Phase III clinical studies have read out at this point, 4 Phase IIIs have read out with positive data. So it's been a good start from that perspective. And so all 4 were positive with respect to progression-free survival. And in addition, so HARMONi-A was statistically significant in overall survival in China post-TKI EGFR mutant lung cancer. 3 of the 4 have had some level of readout in overall survival. All 3 of those have had a hazard ratio less than 0.8, if you will, that generally accepted clinically meaningful threshold for overall survival when we speak to physicians, KOLs, investigators and so on and so forth. So as we look forward, we have quite a bit coming now in 2026. And so that includes the squamous cohort of the HARMONi-3 study has now completed enrollment. So we'll look to an interim PFS analysis in the second quarter and then a final PFS and an interim OS analysis in the second half of 2026. And then the non-squamous cohort completes enrollment in the second half of this year and then final PFS, interim OS -- or final PFS in the first half of 2027 and then there is an interim OS planned. And so we haven't definitively guided in terms of when that is. And so that puts us in a pretty good position with respect to upcoming events. We have the PDUFA for the BLA with respect to the HARMONi study in November of 2026 as well. So several catalyst events in '26. And then I think finally, the larger point, if you look at the totality of ivonescimab, 10 Phase IIIs currently being run by our partners at Akeso. We have the 4. There's a fifth that's a head and neck frontline study that's multiregional. And so that is being performed by a cooperative group in CorTec, called the Illuminate study. So 15 Phase IIIs throughout. Over 4,000 patients have now been dosed in clinical studies sponsored either by Summit or Akeso. And Akeso has achieved approval in China for ivonescimab in 2 settings, and that has led to over 60,000 patients who have now been treated in a commercial setting in China. So good uptake there. I think the final point I would highlight is the enthusiasm across solid tumors from investigators. So over 140 clinical trials are registered on clinicaltrials.gov. And so a little over 40 of those are sponsored by either Summit or Akeso. So that means there's a significant amount of interest globally, both in our regions as well as in China with respect to investigators initiating -- investigator-initiated trials in order to evaluate ivonescimab in several settings beyond those of which we've discussed.

I think -- thank you very much, right, for that overview. We'll dig in a little bit more on maybe some specific HARMONi studies because I always get the nomenclature a little bit mixed up 2, 3a, right, all this stuff. But I think we unanimously hear that this could be one of the game-changing molecules in IO, right? So there was KEYTRUDA, Opdivo, right, Yervoy. And people have always been trying to improve upon that, right? Those are phenomenal molecules. I don't want to say it was. They are phenomenal molecules. We've been trying to improve on it forever, whether it's other checkpoints, right, combinations and the like. And it seems like ivonescimab has really potentially cracked the nut. So can you just tell us the uniqueness of this molecule? I know everyone knows it's a bispecific, but the uniqueness of the molecule. And if you had to pick like out of the many studies you highlighted, had to pick one study to showcase the ORR, the PFS and the OS, what would it be?

Okay. I shall take that. So in terms of the molecule itself, it has 2 validated targets, PD-1 and VEGF. And I think what was underappreciated at the time when that molecule came out is like, "Well, why can't you just give pembro and bevacizumab together? And there were studies with atezo/bev that went around and showed maybe some activity, no clear better activity than pembro alone. But I think what people underappreciated was that the way the molecule was designed. It wasn't just targeting the molecules. It's the way that it targeted. So we talk about cooperativity. And when I talk about cooperativity, there's 2 types of cooperativity with one we call sort of allosteric cooperativity where the PD-1 binding increases the binding of VEGF and the VEGF binding increases the binding of PD-1. So you get the tightest binding where you have both ligands present, which is the tumor microenvironment. The other thing that's underappreciated is that it's a tetramer, so it has 2 VEGF binding sites, 2 PD-1 binding sites. VEGF-A is actually a dimer. And what's been well shown is you can actually cross-link one IVO to VEGF to another IVO to another VEGF to another IVO. This is what we call associative cooperativity. And this cross-linking does 2 things. It presents these high-affinity PD-1 binding sites. So you can imagine multiple PD-1 binding sites binding to a T cell. But we now have evidence that it actually causes clustering of the PD-1 receptor on the cell membrane, and that leads to actually internalization and degradation of the PD-1 receptor. And we believe that's going to be important as well. And I think the take-home message is that unlike the PD-1/PD-L1 story from the previous checkpoint inhibitor era, where there's about a dozen approved and they seem to be interchangeable, the format is going to be very important. We think ivonescimab is not only the first-in-class, but probably the best-in-class. However, there's not enough data for the other molecules to sort of compare it to. But I think as more data evolves and more data is compared, you'll see that the format matters. The one study that I think sort of demonstrates this the best is the HARMONi-2 study run by Akeso. That was a pure experiment. We always knew that, that was going to be important. That study was ongoing at the time of the deal, where Summit licensed this from Akeso, Monotherapy pembrolizumab versus monotherapy ivonescimab. Clear in non-small cell lung cancer, sort of a showdown between the 2 drugs. What we were very interested is looking at the data, not only was it -- would it win, we assumed that we would have an advantage and those patients that had low PD-L1 expression because of the VEGF. But what was interesting was that IVO beat pembro not only in the low PD-L1 expressing, but also the high PD-L1 expressing. So we know that IVO was making the PD-1 aspect of the molecule better. And now there's good evidence that it's not only the sort of mechanical cooperativity, but there's probably some biological cooperativity because VEGF probably plays a role in immune sort of exhaustion and so forth. So we're very excited about the molecule. The molecule has performed very well and several studies have been positive now.

And can you just remind us of kind of the delta? Like that's a great example, right, head-to-head. Everyone asks in the [ IVO ] space, like do you have monotherapy activity? And this like concretely showed you, right? So now when you're going into chemo combinations or ADC combinations, no one can sit back and debate, "Oh, does it have monotherapy activity." So can you just remind us of that benefit that you saw over KEYTRUDA?

Yes. So the -- you know the number...

Yes, hazard ratio in the monotherapy study, PFS was 0.51, so about a 49% improvement over monotherapy pembrolizumab. I think equally as important, following on to what Allen said was then that was tested in the chemo combination setting as well, right? And so the HARMONi-6 study read out in frontline chemo combination non-small cell lung cancer, head-to-headed PD-1 plus chemotherapy. And so that also showed a significant improvement that was clinically meaningful hazard ratio of 0.60. And so now we're seeing activity clearly in monotherapy head-to-head to Allen's point with respect to that monotherapy activity. But now in the -- what is the overwhelming backbone therapy in solid tumors, PD-1 plus chemo, a clear separation of those curves within the HARMONi-6 study performed by our partner, Akeso in China that showed clear benefit in that chemo combination study as well.

So everyone kind of asks us about the HARMONi study, right? So this is what you just -- you filed on. You have a BLA in November -- sorry, PDUFA date in November. It was a very anticipated data set, right? And it was bittersweet, like you hit in PFS, but the first look at OS didn't meet. But then as you mentioned, the second look did meet. Normally, one would require some encouragement from the FDA, from investigators, something to kind of give you the gumption to go ahead and still file, right? And you did. Can you just give us maybe a little bit of the background or why did you decide to just go ahead and file for this indication?

Yes. So a couple of key points, right, in terms of the setting. So this is the post-TKI setting for EGFR-mutated non-small cell lung cancer. And so osimertinib, either as a monotherapy through the FLAURA trial or in combination with chemotherapy through the FLAURA-2 study. So that's an EGFR TKI, is standard of care as well as the MARIPOSA regimen of amivantamab and lazertinib. So that's your kind of frontline standard of care. Second or post the TKI regimens are really a dearth of opportunity for those patients, right? And so that's amivantamab plus chemotherapy is the only post-TKI fully approved regimen at this point. And so part of what becomes important, one, is taking a look at what's out there from a landscape perspective or from what patients and physicians have the opportunity to in one regimen. I think there's general understanding in terms of some of the tolerability concerns within that regimen. And so when we look at the results of the HARMONi study, we see very reasonably consistent, if you will, results from an efficacy perspective, statistically significant PFS and then not statistically significant, but a good trend in OS, both with hazard ratios in the 0.7 in that, but neither statistically significant from the statistical analysis perspective. And so very comparable on efficacy. And then we look at the tolerability, and we do see a difference in terms of a tolerability profile for the ivonescimab plus chemotherapy arm that's -- we get feedback from physicians that there's a high level of tolerability there. And so anecdotally, we hear feedback with respect to treating physicians on the HARMONi studies, it was IVO plus chemo compared to placebo plus chemo. And oftentimes, those physicians wouldn't know which of the 2 arms the patients were on initially because they weren't seeing tolerability concerns that would traditionally come from add-on therapy. So a good kind of read-through in terms of that. So when we looked at the opportunity, we saw a lot of enthusiasm from physicians. We saw the package that we're able to have in totality, and we thought that, that was highly relevant to patients and physicians, and that drove the decision.

Okay. I don't want to speculate on the regulatory interactions and what might be an ODAC panel or not an ODAC panel. So we'll leave it at that. That's going to come one way or another in November. We do hear from clinicians and KOLs, especially when they're on scientific boards of other companies, when they're giving advice, they're already thinking about IVO being there and being one of the frontline molecules. So that's already a good step of approval, I think, for it. Let's talk about HARMONi-3 just in the 8 minutes or so we have left. Some people kind of challenge us, right? They're like, well, in HARMONi, they should have just waited longer with OS. They shouldn't have unblinded it together. How do we know that something like that isn't going to happen in HARMONi-3. HARMONi-3 now is the global study you guys have run, right, following HARMONi-A, right, which had both patients from China as well as global. This is all you. How are you feeling about the statistical analysis? You just -- I think at the most recent earnings call said that we're going to have an interim analysis in the second quarter. It was a surprise to me, maybe not to other people. Kind of just take us through how confident you feel about the stat plan and your pacing of the unveiling of data?

Yes. So importantly, the HARMONi study was sequential enrollment to China and then expanding into the U.S. and Europe, right? And so this is a global study from the start with HARMONi-3. So that involves multi regions around the world, Asia, the Americas, Europe, but it's a traditional global study from that perspective. When we look at the recent announcement, as you referenced with respect to the interim PFS, important to note that there are 2 Phase III studies that have replaced PD-1 and have been successful, right? And so those are the HARMONi-2 and HARMONi-6 studies. The only replacing of PD-1 to be successful is an ivonescimab-based regimen or base study, right? And so both of those were successful in PFS at the interim PFS analysis. And so when we talk about learning from the data, we've said this a lot in terms of the importance of learning from the data being generated by our partners at Akeso and learning from data that we're generating here, always applying that information. This is an example we read out the HARMONi-6 study at ESMO last fall and taking a look, seeing another study hit at the interim PFS that became important in terms of leveraging that information with respect to our design. The other piece in terms of why do that is the easiest way to have a conversation with the health authorities, in particular, the FDA is to go with a statistically significant primary endpoint, right? And so that -- what that effectively does is pulls in that conversation with the health authorities in terms of a statistically significant primary endpoint should that come true within the interim analysis. And then that allows for accelerating that conversation and what does it look like in terms of bringing this opportunity forward to patients as well. And so we learned from the data that has been generated previously, not once but twice against -- head-to-head against PD-1 and then said, how do we make this regimen available to patients as soon as possible.

Have you provided any more detail in terms of what the alpha spend is for this interim, if it hits, obviously, everyone knows what happens there. But if it doesn't hit, does it preclude you from other further analysis? Like how should we be thinking about that?

Yes. No. So this is an interim analysis. We haven't given details with respect to statistics other than to say the interim analysis is a higher bar, if you will, than the final analysis, right? So no, there's nothing that would preclude moving forward with the preplanned final analysis that we had talked about previously in the second half of this year. So nothing changes that. It's a higher bar at the interim. So certainly nothing that would say not meeting the interim then leads to the inability to hit the final analysis. One thing we've also said is in effect, we did not move the bar very much at all in the final analysis to kind of maintain where we were to not really change probability of success throughout 2026, but bring in an opportunity at a higher bar based on what we've seen head-to-head with IVO against PD-1s historically. And that gave us the confidence to add in that step, if you will. In effect, if you think about it, there's no better read-through to a randomized Phase III study than a randomized Phase III study using your own compound in a very similar regimen with an extremely similar control arm. And so we saw HARMONi-6, and that was highly positive, and that gives us the confidence to pull in an interim PFS analysis and then take that opportunity to go forward in order to accelerate. I think some will say, "Hey, are you hedging?" Not at all. In reality, we're not moving anything with respect to the final in terms of the bigger picture. But what we are doing is bringing in an opportunity a little bit sooner.

I think it's important for investors to also realize that enrollment for these large studies, this is a fairly large study, right? And...

Yes, there's 600 patients in this cohort.

Enrollment went pretty rapidly, right? And you've been able to maintain these. Normally, we see companies that are pushing out for one enrollment issue or another. You guys went very quickly. You're bringing in this interim into the second quarter. So now just help me understand or just correct me if I'm wrong, interim in the second quarter, final in the second half, interim OS in the second half and then final OS in the first half of next year, possibly?

Haven't talked about final OS yet. So we were always very clear when we guide, we want to be reasonably certain about something and so that we're not constantly moving things around, to your point, in terms of changing timing. And so we haven't given timing with respect to final OS, but there would be an interim OS look in the second half of this year as well as the final PFS, if necessary, in the second half of this year. But then what we've said is the non-squamous portion also, that completes enrollment in the second half of this year to be able to look at final PFS and then shortly around that period an interim OS.

So both squamous and non-squamous, right, coming out, which is great. Everyone is going to be looking out at that. We only have about -- I could talk to you guys for a long time about this. So we're really fascinated by the molecule. We love the data that we see. It seems like it's going to be a game changer. But I guess I'd love to maybe take a step back and think about strategy a little bit. We know from Merck how they were able to get KEYTRUDA to be the best-selling checkpoint inhibitor, right, by a significant amount, right? It's not like there weren't fast followers, right? But clearly, there was a game plan there. As you think about IVO, is it this BLA that's coming up? Is it the squamous, non-squamous data from HARMONi-3? What is it that kind of triggers you to really kind of go out and get the broadest footprint possible for IVO? Is that even like part of the plan? Or do you want to kind of slowly grow this and just have maybe a partner come in and take over? Give us an idea as to how you're thinking about this.

Yes, I'll name a couple of bullet points, and I want to make sure Allen has a chance as well. So a couple of important things. Cornerstone indications for the PD-1 VEGF class are non-small cell lung cancer as it was for the PD-1 class. And then a place where PD-1s were not successful in microsatellite stable colorectal. First 2 tumor settings that we took a look at is others, if you will, in class are looking at starting Phase III clinical studies in first-line non-small cell lung cancer. We're now talking about readouts in that path forward. So that's a pretty significant difference with respect to placement there. And then obviously, we are enrolling now in the colorectal setting as well. What differentiates here? So 4 Phase IIIs have read out for ivonescimab. There are 4 that have read out for the class, right? And so only ivonescimab has had a readout, 4 for 4. So Allen made a really important point. These are not monoclonal antibodies interchangeable. And so if you make a difference to the molecule construct, that -- when you're 4 for 4 at ivonescimab and you're going to make a change to that, there's more room to go down than there is to go up. That's one. Two, the 10 Phase IIIs being run by Akeso right now. That speaks to the broad applicability of ivonescimab because there's obviously data supporting moving ahead there. So that's another differentiator. We do plan to go much more broadly, but what we are going to do is kind of give you that information as we're ready to enroll as opposed to showing everybody else where to go and where to start planning, right? So that becomes important. Anything else that?

Yes. No, I would just say that we're going after all of those, right? Like I think Dave mentioned that non-small cell lung cancer is our anchor or cornerstone indication, right? With the studies we have on hand, we will probably sort of corner the market on metastatic non-small cell lung cancer, right? And that's probably about 80% of the market. The one indication that could actually be bigger than non-small cell lung cancer is microsatellite stable colorectal cancer. And so we have a Phase III, and we're ahead on that as well. So I think those are the important cornerstone or anchor indications, right? If you look at how Merck beat atezo and nivo in the old days, it was because they won in lung cancer, right, even though all 3 of them started about the same time. Now we're looking across all the indications, right? So with our Akeso partners, we're running 4 global Phase IIIs. They have over 10. So there's like 15 Phase IIIs running. You should probably ask yourself like why aren't we doing certain indications? Like Akeso is doing TNBC. They're doing small cell lung cancer, right? They're doing pancreatic cancer. It's not that we're not able to join those studies. We think that the standard of care in those diseases are going to change. And so maybe we're waiting for something to happen there. And then the other indications, you'll hear more. I think Maky, our CEO, sort of alluded to that you'll hear more Phase IIIs out there. So the idea is to go after all of these things in parallel.

And I think it's important for investors to also know that you guys are disciplined with your -- with the utilization of your cash. You've set up some collaborations with Revolution Medicines and GSK so that you're not necessarily growing this R&D engine to come up with new ADCs or new combination metrics. There are potential partners there to work with?

But that's an important point also because what we're not going to try to do is force the optimization of our own pipeline, right? So we're going to follow the science. We're going to go out. We're going to combine with several different ADCs. We've seen there isn't a single ADC platform that is optimized across solid tumors. So we can combine with platform A in this setting, platform B in this setting. And that, to your earlier point, what led KEYTRUDA to be as explosive as it ultimately ended up being. Merck was able to go after each of those different combination agents or different settings based on following the data as opposed to trying to optimize within its own pipeline that may not be the optimal combination in each setting.

Terrific. Dave, Allen, thank you very much for coming. I apologize for going over time, but really appreciate it.

Thank you.

Thank you.