So delighted to introduce our next speaker, Dr. Marjorie Green at Merck, who heads oncology development. Marjorie came five months ago from Seagen. We're also joined by representatives of Merck's IR team. I'm gonna leave opportunities for those in the audience to ask questions throughout the day. So just raise your hand if you would like me to put a question to Marjorie, and we can kind of go from there. Maybe we start with TIGIT, or TIGIT, as Roger used to like referring it. I like TIGIT because it rhyme with the title of the first report we wrote about it, "TIGIT: We Totally Dig It!" And on that note, Roche inadvertently disclosed their phase II data, which is a whole another story.
Yes.
But it's quite informative, and we obviously, you know, did bad statistical analysis, trying to assess the probability of it hitting, depending on what you assume for the alpha spend, was for the PFS read, and what do you think the hazard ratio is gonna be for the remaining patients? It's not complicated stuff. One thing from Merck perspective, one might imagine, is why not drop PFS as an endpoint, given that the treatment effect seems to be all driven by a, y ou know, if there's a real treatment effect there, it's not like you saw in CITYSCAPE with curves at six weeks going like this.
Mm.
It's a late response. So given that, why not either repower, but then it's gonna cost you time, or just drop the PFS, and in that way, at least you're gonna get something. So how's Merck thinking about KEYVIBE-003? There we go.
No, no. Thanks, thanks for the, the question, and, you know, we, we tend not to talk about the details of our statistical plans, which is frustrating and annoying to everybody in the investor community. We do think that overall survival is the key endpoint for, for, for TIGIT, and particularly for these diseases, where, where modest efficacy improvements in PFS may not always translate out into overall survival. But we have not publicly disclosed any changes in our, our studies at this point. We do keep an eye on the emerging landscape. You can look at historical studies that Merck has done and how we have assigned alpha when we have had different kind of endpoints for both PFS and OS, and we usually protect both. So we're able to, to adequately test for, for both endpoints.
OS, we think, is needed, and I think that you, you bring up a great point.
So in your KEYVIBE-003, because it's very different from SKYSCRAPER-01, in that it's testing across the whole range of PD-L1 expression, from lows to highs, and I can't remember if it's 1,000-something patients. But when you do the math and say they're split a 1/3, 1/3 , 1/3 , you know, your sample size isn't any different from, you know, what Roche is. In fact, it may even be smaller. So in that sense, when you say that you can protect both, I'm not sure how you could protect both unless you run it. Isn't it just-- I mean, let's put it like this: If I said to you, it would be a reasonable suggestion to drop PFS if one works on a Bayesian basis on the input from other trials, would you agree?
I'd say thank you. So yeah, you know, I just can't talk about-
Yeah, okay
the statistical designs for the study. You know, I, for those who don't know, the KEYVIBE-003 study is building on, you know, the KEYNOTE-042 and KEYNOTE-024, which is looking at KEYTRUDA with monotherapy versus chemotherapy, and so this is looking at our anti-TIGIT co-formulation compared to KEYTRUDA. And you know, we are one of the few drugs, as you mentioned, Andrew, that has efficacy across levels of PD-L1 expression. And one of the things that's great about our development program is that, in this study, is that we did increase the sample size of the study to specifically test the higher PD-L1 population to the greater than 50s. And so we did increase the sample size previously-
It's not a 1/3 . I shouldn't assume a 1/3, a 1/3, a 1/3.
We did increase the sample size to address the greater than 50s, and so because you did see, when you looked at the published data for 42 long-term, there was efficacy across the entire population, but numerically, the hazard ratios improved the higher that you went-
Mm-hmm
with the PD-L1 expression. And there are many open questions for, I'm gonna call it TIGIT to be difficult. But with TIGIT, you know, is the efficacy gonna be better where the immune system is more active? Is it gonna be better where there is maybe a little bit less activity? We're well positioned to answer both of those questions.
So, maybe segueing to your chemo combination trial,
Which is KEYVIBE-007.
Which is KEYVIBE-007. Now, previously, I've been somewhat circumspect about the outcome of that trial because of the inclusion of the patients with PD-L1 low expression, given the CITYSCAPE data, for what it's worth. And on that basis, and also, another observation is one of your colleagues, Emmett Schmidt, will make a very good case that immunogenic cell death with chemotherapy, at least from human studies, suggest it's of limited or zero utility, and there's no evidence of synergy at all. So it's a nice thing in mice, but it doesn't exist in humans, and therefore, what's the point? So if we take his observation as real, and if we say that-
I would argue with that observation.
Okay, so-
I think being from Seagen, I would argue, immunogenic cell death, I think, is real. But okay.
I-
So we'll believe that Emmett's correct on that aspect.
Okay, so just, you know, hang with me for a second.
All right.
Just to get everyone on the same page in the audience, what we're talking about is, if you combine a chemotherapy in a tumor which is necessarily hot, with a PD-1, can you somehow, by causing the cell to die in a particular way, ignite a flame that synergizes with a PD-1, and you have tumor shrinkage? The alternative thesis, which Marjorie's colleague puts forth, is that it's just a question of not synergy, but addition. Chemo shrinks tumors, you're getting some PD-1 activity in patients who have PD-1 high.
You put two together, and you see a benefit, and the debate is whether we're seeing this immunogenic cell death, and that's very important in this particular trial that you're running, combining of the vibostolimab + PD-1 + chemo. Because unless you believe in immunogenic cell death, it's not entirely obvious why I should feel confident about the outcome of this trial, given the SKYSCRAPER-01 data. And if I believe in immunogenic cell death, then I can embrace it, so interested in your thoughts.
Yeah, no, no, thanks, thanks for the question. So I think that you can still have, with chemotherapy, increased antigen presentation without immunogenic cell death, which causes specific immune-related responses, that cause this preclinical synergy. And not all chemotherapies cause it. Some do, some don't. And so I think you still can get that kind of benefit, and I think if you look at KEYNOTE-189 and look at the efficacy across a broad expression of PD-L1, the hazard ratio for survival, with five-year follow-up in KEYNOTE-189 for patients whose tumors PD-L1 expression is less than 1% is still in the 0.6-something range. It's suggesting that you don't necessarily need to have immunogenic cell death.
It is antigen presentation, potentially some immunogenic cell death, that you're getting more than what you would expect from, from KEYTRUDA by itself, where you need the-- to, to take the brakes off the immune system. So I think that with 189's data showing that activity across the range of expression, it is reasonable to say that you could see, by taking the brakes off even further with an anti-TIGIT, you could have activity across that same spectrum. It's an open question that the study will answer.
If there is that activity, then seemingly it's not manifested by high response rates. It'll be manifested by more durable response rates in those patients who do experience tumor shrinkage.
Potentially. I think that in KEYNOTE-189, where KEYTRUDA was added to chemotherapy, you did see increased response rates. Whether you would get a marked increase with the anti-TIGIT, I think is to be determined. But I do think overall survival is going to be the key endpoint for the TIGIT drugs.
Mm-hmm. And on KEYVIBE-007, so the chemo trial that we're referencing, the timing, I think, is, I wanna say 2026 on ClinicalTrials.gov, but I'm assuming there's interim before then. Can you talk to when we should anticipate data from that trial? And the extent to which I think that Dean has referenced, that you've already had a look at the data where it's passed a futility analysis. So if you could just comment on that as well.
No, no, thanks, thanks for the question. Yeah, we tend to guide to the primary completion date because that's the study's really powered and designed for that. We do interim analyses. The number vary from studies and have different assumptions to it, so I can't be more specific than that.
Within the KEYVIBE-003, so this is your PD-L1 high trial.
Oh, it's with all levels, but it is the company-
Yeah, yeah, sorry.
- therapy.
Yes. Yeah, sorry, with that chemotherapy. I, I misunderstood.
It's KEYTRUDA vs. KEYTRUDA.
Yeah. So within that, you obviously have a cohort of PD-L1 high. Can you share anything about the statistical plan? Because that has a final, a primary completion in 2025, from memory. I would imagine that within the high is where you'd imagine the magnitude of the dominant treatment, and the fact you powered, as you indicated earlier, you know, up in that. So you added patients in that segment. Is there any carve-out that would enable you to selectively unblind? And I'm assuming there's... Or not. What can you share about the hierarchy? How does it, w hat can you tell me about it?
I can't tell you anything about it-
Yeah, I thought you might say that.
- sorry.
Um, okay.
No, exactly. Nice try, nice try.
One does what one can. Okay, so, ADCs, which is very pertinent, given your background from TIGIT, seem to be in the hands of Pfizer if the FTC doesn't get in the way. So there is a debate proposed about whether the lack of biomarker-dependent efficacy with ADCs is instead indicative of the true mechanism of these drugs is just chemo hangs around for longer, just as a PK-driven effect. And I know my friends at AZ would violently disagree with that and just say it's because we haven't got a decent biomarker. How do you think about that and the wisdom of biomarker selective enrollment? And, you know, this is a question that Dean also touched on the last analyst call, but obviously, you know, Merck has benefited from the embrace of biomarkers-
Definitely
at the expense of others thinking about Bristol. So, so how do you think about what we're really seeing here?
Yeah, I agree with the colleagues at AZ in that our biomarkers may not be truly where they need to be because of the heterogeneity of tumors. And you're looking, when you get a slice of a tumor, you're not getting the totality of what the actual tumor expression is for your antigen. And so there's that difficulty in trying to understand, for the totality of someone's cancer burden, what is the expression level of a specific antigen. So we don't have a precise way of measuring that.
Mm-hmm.
I think that with ADCs, because of the instability or stability, depending on how it's designed, of the linker, and recycling, you do get bystander effect.
Mm-hmm.
So I do believe that actually happens. It can be shown pre-clinically, and that also likely explains the efficacy that you're seeing, where you have more heterogeneity or lower levels of expression. I think that there's public data, for example, with ADCETRIS, which is a treatment of Hodgkin disease, that showed efficacy even when you couldn't even measure the CD30. And so there is likely a biomarker issue combined with bystander effect, as opposed to just having the chemo hang around longer. Chemo hanging around longer is likely part of this as well. If you think about the half-life of most chemotherapies, they're quite short, and where ADCs are measured in days. And so the ADCs are, you know, the activity goes through multiple different kinds of routes. It's the antibody, does it have an effector function?
You know, do you have bystander effect? You know, is it something that it can generate ICD? Despite what Emmett says, I think that actually does happen, and all of these things play in together, and that's also why you likely see the difference, efficacy, between the different ADCs, too, is 'cause each of these factors can be tweaked to some degree.
So staying on the theme of ADCs, can we go to SKB264, which is your TROP2 that you licensed from Kelun, one of several. I think your Dean has guided that you're gonna be opening an expansive phase 3 program this year, I assume in lung and other indications. I'm sure you'll have several folks at ESMO, presuming that's where AZ shares the data, but they're being quite tight-lipped on that, but I'm sure you'd be interested in that. In terms of as it informs your trials, because I know you will only want to talk about things that impact Merck and not competitors, where would you see the natural home in a PD-1 experienced patient group in lung? Because there's this sort of guessing game of in which cohort have they seen benefit.
They reference a majority of patients, and it could be histology, because TROP2 is expressed on squamous. I wondered about prior taxane exposure, because that's common to all other TROP2 positive trials, but maybe not in second and third-line lung, because they could have got gemcitabine. So perhaps for some reason we don't understand, taxane inhibition upregulates TROP2 or internalization of TROP2. What's Merck's working thesis about patient selection? Because it can't be TROP2 biomarkers, because you've got the data showing that it's not predictive, at least in the earlier data. So how are you thinking about it? Because you must surely be thinking about this as you think about your own clinical trial program.
No, thanks. There's a whole lot to unpack there. No, no, thank you. So the ability of a biomarker to be predictive, there are different ways to slice and dice data to support the hypothesis that you want. And so generating ROC curves that find a cutoff to benefit a majority of patients, there are ways to look and see. But you know, the thing that I think is, becomes challenging with ADCs is that when you are first to market and you have activity in patients who've got low levels of expression, why would you want to limit your drug? And so if you're first to market and you see.
Like, you can't select necessarily that a patient who's got IHC 1+, however you want to measure that, that antigen, and that person has a benefit. You wouldn't want to omit them from getting a therapy. So an all-comer approach is definitely, if you're first, a smart way to go.
Sure.
But that doesn't mean you can't actually enrich. I would say that you probably could-
You're talking enrichment based not on a, you know, phenotypic, but you're talking about biomarker-
About the actual biomarker-selected group.
Even with a suboptimal biomarker, suboptimal IHC biomarker, as we have with TROP2.
So, I think, you know, there are different companies that develop biomarkers, and they're not all the same. So you can look at the PD-1 and PD-L1-
Sure
and see that they're not all the same. And so depending upon how you develop your biomarker, you might have better ability to enrich. So that's one aspect of the question. I think that the aspect about taxanes is interesting because we test our drugs often in late-line disease, and the patients have been exposed to a variety of different therapies. And so there has been a comment on the Kelun data about patients who've had prior—who have EGFR mutations-
Mm-hmm
in their tumors, and the efficacy looks better there. They've also had less chemotherapy.
Mm-hmm.
And so, is the taxane question, is it really that they had to be exposed to taxane to get TROP2 expression? I think when you look at archival samples of lung cancer tissue across different settings, you see TROP2 expression. So, it may not be expression but is it the resistance to. Like, if there was a microtubule inhibitor on this as opposed to a topotecan, you know, to a Topoisomerase I inhibitor, would you have different kinds of responses? So, I think it's more not the taxane specific, as opposed to the taxanes are just commonly given to lung cancer, where if you have mechanisms of action that are a little bit more similar to a Topo 1 inhibitor, you might actually have resistance. So, I think it's the-
Yeah, interesting point.
Mechanisms of payload. That's a guess.
Topo 1s are not used in non-small cell management, right?
No. No, no, no.
We can't test it. There's no-
No. And it's hard, so you look at breast cancer potentially where you do get some, but they're just not used as widely. But it is. The thought is having to do with just more chemo resistance as opposed to sensitivity.
Okay. That's interesting. So, one thing that struck me, because obviously AZ is prosecuting, or trying to get approval for, HER2 High, tissue-agnostic indication, right? And then separately, obviously, they've got the HER2 Low approval. We know that HER2-Low is found in up to 50% of solid tumors, right, outside breast. So, if you had a HER2 Low, and you may do from Kelun, we don't know, but would you be looking at HER2 Low outside breast as a potential biomarker to use and run trials, given you've got an active drug? And if not, why not?
No, thanks for the question. You know, it's. I'm probably—my IR team's gonna, like, yell at me, I think. No, they won't yell at me. I guess they—you're getting into the theoretical stuff where I'm a breast oncologist by training, so I love talking about HER2, and so I think it's really interesting. You know, we have not publicly disclosed any activity in the HER2 space. So in a theoretical world, you know, it's—some of it to me is still ADCs are still chemotherapy-
Mm-hmm
at their core. And so the agnostic, you need the chemotherapy to work. And so not all tumors are gonna be sensitive to the payload, and so that's the interesting thing is, like, how many, when you look at data from these, are they picking tumors where there's some evidence? And I've not dug all into competitor data to where I can comment, and I shouldn't comment on their data anyway. But are these tumors that there's some evidence that something like a irinotecan or topotecan or some other tecan has had a little bit of activity, or are they showing activity, irrespective of ability of that class of chemotherapeutic to work?
I think that is one of the challenges of something like an agnostic ADC approval compared to something like, you know, we've got approvals in TMB-High and MSI-High because the mechanisms are different, and so the agnostic labels are easier there because the immune system applies broadly, even though they're in those tumor types compared to chemotherapy, which will be more tumor-dependent about where there's response and where there is not.
Going back to SKB, when you look at differentiation, you're gonna be late, right? Because obviously you're yet to go into phase III. But-
This year.
This year, soon. But you're still gonna be late, but one year is not much, and stuff can happen, and clinical trial design can dictate ultimate successes we've seen with Bristol and Merck and KEYTRUDA and OPDIVO. But one advantage you seem to have is ILD, which may take on additional importance. We'll see when combined with PD-1. Could you talk to how strong your conviction is in the data that's emerged from the Chinese trials, that this is real mechanism driven? And there's a paper on macrophages, and the alveoli is underpinning this. But you know, it's a pretty early paper, and I'm not sure how well-founded it is.
Or do you, is your mind still open that actually this, we may well see similar rates of ILD as we run Western trials, you know, four, five, six mg/ kg in the phase III? So, what's your conviction that this drug is differentiated from an ILD perspective?
We are still in early days with the Kelun TROP2 ADC with MK-2870, and what we can say, there have been probably 200-300 patients treated so far at a variety of doses, and they recently have completed and reported out a phase III trial in triple-negative breast cancer, showing benefit over standard of care chemotherapy in a press report. What we can say is that no serious rates of ILD have been reported so far, and the question would be: Is there something inherently different about clinical care in these studies in China versus what you would see in Western countries? And I think, you know, when you're looking at these academic centers, they, the standard, the treatment is very comparable to what you would get in Western populations.
In addition, Kelun has phase I sites open in the United States, where patients are being treated, so that does add into the safety database. So I think that where we're really gonna find out is when we get into the phase IIIs, is that when you get the large, randomized data set. So I am cautiously optimistic that we will continue to see this as we go into our phase III program, but I can't tell you that ILD won't be seen because it's seen often with many chemotherapeutics. And also, KEYTRUDA has ILD too, and so it may be that we will see it in combinations. And so it's too early to tell. I don't want to differentiate based solely on toxicity.
I want to differentiate in different ways, but that is something that, you know, if you can avoid a fatal toxicity, that's always a benefit to patients.
Has Merck confirmed whether or not you've got rights to their Claudin-18.2 or not?
So we have publicly disclosed that we have, you know, the TROP-2 ADC MK-2870, two additional clinical ADCs, and then we have preclinical ADCs as well. So we have not—Merck has not publicly disclosed the other two clinical ADCs.
But you would. Okay, so would you find Claudin-18.2 as an interesting target?
Definitely.
There we-
So definitely. So I think that, you know, the data with as Astellas' monotherapy antibody is interesting, and it lends itself to be a good target for an ADC.
Maybe we could segue to the Moderna PCV. I'm not allowed to call it a PCV, am I?
INT.
INT.
Individualized Neoantigen Therapy.
It's because PCV could get confused with the pneumococcal vaccine, so there's been a terminology shift. That was my understanding. The data, I'm assuming that you're not gonna be able to file it, but you're gonna have to run a registration trial, and I think that's been Merck's commentary till now. I just want to make sure that's still the case. Perhaps you could, I was very struck. I mean, melanoma is great, but I didn't think that's surprising. It's a hot tumor, and it's an adjuvant setting. You know, look, until you've got the data in hand, it needs to be replicated, but I wasn't sort of blown away by it, but it's obviously great and an important new frontier.
I thought the BioNTech data in pancreatic cancer was eye-opening, given the nature of that indication and the prognosis, albeit a tiny patient cohort, and it was a translational trial. But I thought that dataset was actually much more encouraging for the potential breadth of indications that you could pursue. So I'd just be interested, where are you gonna take this? So I think you've indicated you're gonna go broad. You've said you're gonna start seeing lung. How long, y ou know, how can you accelerate the development program as the regulator becomes familiar with the profile within the melanoma setting? So, I mean, is there a flywheel here from a regulatory perspective that you can move faster and quicker and go broader if the data pans out as we anticipate?
No, no, no, thanks for those. So maybe starting first with the melanoma data, and what are we doing with the melanoma data? You know, so it's been presented twice. It's the melanoma data is a fairly small phase II open-label study conducted by Moderna, and they did a lovely job with the trial, comparing you know, INT or V940 plus KEYTRUDA vs. KEYTRUDA for patients who had Stage III and Stage IV NED, melanoma. And showed that you know, patients relapse-free survival was markedly improved, distant metastasis-free survival. The hazard ratio is in the 0.3 range, and so you say not blown away. It's like 0.3, that hazard ratio is awesome. That's what people die from, with their distant metastasis. But, okay.
So, I think, you know, that data is very compelling, and we have started a phase III study in this setting. So from the regulator standpoint, you can tell we've been talking to regulators and getting buy-in to be able to conduct these studies. We started, you know, in Australia. We've already started enrolling into the phase III studies, and so we've got good relationships with regulators. I think that the mRNA technology is well-established, well known by regulators, and so thankfully, you know, with the COVID vaccines recently, health authorities have it in the top of their mind about mRNA.
Mm-hmm.
And so it gives us a good lead-in to being able to conduct phase III studies. So I think that there's a lot of data historically, even prior to the COVID vaccine studies about mRNA, and so our ability to go into studies there, I think, is well established. The ability to register that, you know, I think that we have Breakthrough Therapy Designation, which means that we ask the FDA if they think this data is compelling because there are advantages you get with a Breakthrough Therapy Designation. It doesn't mean that you could file or not file, but it does give you a different relationship. We'd be able to ask a lot of questions if you're developing the drug, and so the FDA felt it was compelling enough to give Breakthrough Therapy Designation.
We have PRIME designation. You know, it's the reality is it is a fairly small phase II trial, and we did want to get the phase III up and going right away. So I think that we have been fairly cautious about saying that it's something that we know we can get approved. I think Moderna might be saying something a little bit differently. We both will say the data is compelling. So we think that the data is compelling, and we have started our phase III programs there. How broad and how large we're gonna go, we've talked about lung cancer publicly.
I think Dean said stuff about this, and, and so you will see a lung cancer study, looking at, INT, building upon, you know, KEYNOTE-091, our adjuvant study, and then we also have data now with KEYNOTE-671, readout due date in October. And, and so we will continue to move forward there.
Where do you, with regard to, chemo and, PD-1, in terms of the regimen-
Mm-hmm
when do you administer the vaccine, at the beginning or at the end? I'm presuming not during.
Yeah, there, there are a lot of open questions, just like we were talking about TIGIT, all the open questions. That's here, as well. And Moderna is conducting phase I and phase II studies, and we are as well to help try and answer some of these questions about the optimal timing.
Because you never had that, obviously, in the melanoma setting.
There is definitely an open question. I think that when you look at how KEYTRUDA has worked and other vaccines, there is the ability to develop an immune response during chemotherapy.
Mm-hmm.
So, it does exist. It might be blunted depending upon what the vaccine type is. So is it out of the realm that you could develop-
Mm
an immune response? The way that the studies are set up for melanoma, that one is, you know, it's not a chemotherapy question. We've not disclosed what we're doing in our lung study.
And then the extent to which, you know, beyond lung, right? Pancreatic is a, you know, significant unmet medical need, terrible prognosis. 20% of patients can get a Whipple's. I don't know what you thought of—you look at 16 patients' worth of data, the Balachandran data, but I thought it was interesting and compelling, given the backdrop and the setting, and the fact it seemed to be—the hazard ratio seemed to be aligned quite nicely with neoantigen-reactive CD8 response. So, how soon before you go into pancreatic or, you know, just more work to be done focusing on non-small cell?
No, no, thanks for that. I think, you know, it was a very interesting paper to read. We keep up with all the external data-
Sure
and where we're going. When you look at KEYTRUDA and what we've done there, is we've really invested heavily into earlier stages of cancer. We have seven approvals, we have more studies reading out, and we think the immune system is most robust there in our ability to be able to cure patients. That's, that's sort of where our goal is, and so a lot of our INT focus is in that setting. So for, for metastatic disease, you need to have a disease where you can have an indolent enough course to be able to get the T- cells to be generated-
For metastatic. I was-
For metastatic. I was referring to metastatic.
Okay, fine.
Going for metastatic disease. No, thank you. You do need to be able to get the T-cell response to happen, which takes some time. You know, with the INT for melanoma, it's nine doses of INT given concurrently with KEYTRUDA, after people have had a couple of doses of KEYTRUDA, while INT is being manufactured. And so depending upon the regimen, it would y ou know, it's less clear about the efficacy in the metastatic setting, so it's an open question, I think. You know, what that data showed is, it definitely showed the immunogenicity of the approach, but I don't know if it told us a lot about the efficacy.
But that was in the adjuvant setting.
Oh.
The Balachandran was in the adjuvant setting, not the metastatic.
So I apologize. I thought it was in the metastatic setting.
No, it was adjuvant.
Okay.
Okay. Should we talk about CLL and the ArQule transaction, and nemtabrutinib? Am I saying it correctly? Okay.
Thank you.
So, the IRA is. You know, it was unforeseen, I suspect, at the time of the ArQule transaction, and, you know, these are relatively long trials to run. There is an advantage in patients who fail earlier lines of BTK therapy, and that's not going away, and you'll be competing with pirtobrutinib. You've got a couple of large phase III trials running with nemtabrutinib. How much is the IRA changing your commitment to that molecule, compared to what it was previously? And then I've got a follow-on question.
Yeah, the IRA is definitely having an impact on drug development. You've heard this from multiple,
Yeah
pharmaceutical companies, and Merck's lawsuit, I think brings the concerns we have about innovation. I think when you look historically, 60% of approvals have happened in a timeframe where they might have been impacted by the IRA from drugs, because of lifecycle management. And so, you know, for nemtabrutinib, we are strongly invested in developing this molecule and trying to bring an optimal regimen to patients. And there are different ways of developing a small molecule in a way that you can still meet within the IRA timeframe. Some of them might have increased risk, in how you develop that drug.
And so we haven't publicly disclosed our full development plans for nemtabrutinib at this point, but we are fully committed to ensuring that we can provide benefit to patients, where we think this drug's going to have activity. And we like that it can potentially prevent the development of resistance to BTK inhibitors. That's something that we think is the hypothesis of these new molecules, because it works where resistance occurs. Can you also further delay the development of resistance, too?
Mm-hmm. I had a question, and I warned you I was gonna ask you a mean question at the beginning. So when I go through the Merck oncology pipeline and portfolio, so belzutifan was from Peloton, KEYTRUDA from Binovium, ILT4 from Agenus, LYNPARZA from Astra, SKB264 from Kelun, zilovertamab vedotin from VelosBio, the ODM-208 from Orion, bomedemstat from Imago. I mean, you get the idea. The point I'm making is that I could only find two molecules, vibostolimab and your LAG-3, which are in your oncology phase II, phase III, or on the market, that came from Merck Discovery, compared to, I think, 18, which were all either acquired, licensed or partnered. So the question is, is it that there isn't been really much focus within Merck Discovery on oncology, or is it just, it's really not very good at it?
Okay, that part was mean. So the latter part. No, I think it's a great question, and there are different ways that companies can enrich their portfolio. You know, I'm newer to Merck. I've been here for 4.5 months, and when I first joined, and when I was interviewing, people I met with were in discovery. And there has been significant growth, I think, over the past two-three years of the discovery group, with some key hires from other companies, as well as from academia. And with that, I've been very impressed with the science and some of the work that may not be as visible to industry, about the preclinical work and the molecule development we're doing internally.
But that being said, I think we're still opportunistic, so I think you're gonna see probably a continued mix of acquisitions and licensing, as well as internal drugs. Whether that shifts to where it's a higher percentage from internal, external, I think it depends on what our groups come up with. But there's been significant investment in discovery, and so when I talk to the group at Merck, I don't feel that they are any less proficient or scientific or innovative compared to early development groups I've worked with at other companies.
Can we talk about afucosylated CTLA-4 and Xilio? So Agenus showed this very provocative data in MSS colorectal cancer, which is, I know, 90% plus of colorectal cancer, and seemingly showing, albeit with lots of talks, some sort of profound improvements in outcome, albeit in patients who didn't have liver mets, which is not unimportant.
Mm-hmm.
You have a collaboration with Xilio, who has this masked afucosylated CTLA-4. I'm just interested in whether, given this is an unmet medical need, how impressed or not you were by that data, which is being shopped around, and how that impacts your commitment to, you know, pursuing, given you've got an internal CTLA-4 candidate that has the advantage of likely superior efficacy.
Yeah, it's. Yeah, I think there's a lot of debate about fucosylation and its ability. You know, do you increase ADCC for things that are not fucosylated, and can you get the immune response to being more robust? And there have been multiple different molecules that have been touted as having more efficacy with not being fucosylated and have not necessarily panned out. So I think it's early intriguing data. And so when we look at our internal molecule with the CTLA-4 that we have, we really try to develop a drug that we think that could have a better therapeutic index than drugs that are on the market because of the toxicity profile, and focused on that aspect even more than potentially the immunogenicity and the ability to spark ADCC.
and so we do have phase III studies ongoing with our CTLA-4 combination with KEYTRUDA, looking to see, you know, in renal cell, are we able to do better than what's already out there? And so whether or not that particular aspect of it plays a role, I'd look at it as an upside, but not the key driver of its activity.
You have a KRAS compound, which is sitting in phase I. KRAS G12C inhibitor.
Mm-hmm.
-that I think came from Merck Labs. Obviously, it's-
In discovery.
In discovery. I know. I give you credit. So my question is, you know, this is an increasingly crowded field, and we have lost track of the number of G12Cs there are. Is there anything unique about this model? For example, the AstraZeneca molecule, which is pre-IND, I think it's about to go into clinic, is brain penetrant because of its size, which would be one way of differentiating. So I'm curious about your molecule. Is it brain penetrant? Is there anything that is particularly unique about it?
Yeah, we're gonna have data that is being presented at ESMO, looking at some of the phase I data, as well as in combination, you know, as monotherapy in combination with KEYTRUDA, and so I think you'll get a better sense of what the drug has to offer there. Brain penetrance is sort of interesting because people who have metastatic disease, you know, in lung cancer patients, a significant number of people will develop metastasis in their brain. The CNS and the blood-brain barrier get damaged and leaky very early, and there's data that chemotherapy can actually have activity sometimes. So I don't know how much of competitive advantage that is. I think these drugs do get into the brain.
Eventually, in an intact model, they may not as well, but whether that's gonna have a therapeutic advantage is not clear.
Have you got a pan-KRAS or a pan-RAS?
Not in the, in my group.
Not in your group. It's not in development yet.
No.
But are you aware of anything that's-- I would imagine, given the-
We-
The magnitude of the commercial opportunity, it would be something that every pharma organization has to be pursuing.
Yeah, I, I think when people talk, another group I talked with today talked about the disappointment in the first KRAS, you know, G12Cs, and I like the oncologist is the optimist going: "Hey, they work. What do you mean? I'm excited about this because it means it's targetable." And so then, if you can extrapolate that, then it's like, how can we use that information to go to Dean and then to pan-KRAS? And so I think Dean has publicly talked about these are areas of interest for us, and for our discovery group.
VelosBio, the ROR1 ADC.
Mm-hmm.
So it looked like the solid tumor work has been terminated. Is that right? When I look at clinical trials, it looks like all the focus really is on heme, that you're no longer looking at solid tumors.
We still have ongoing cohorts in solid tumor. So, I think that that's where we're still waiting for data to make a data-informed decision about what to do there. Expression is definitely highest in hematologic malignancies, and so, you know, I think we have Breakthrough Therapy Designation in mantle cell lymphoma. We've got phase II, phase III in diffuse large B-cell lymphoma, and a lot of work is ongoing in combinations in hematologic space. But, you know, I think it's premature to say that we're not interested in solid tumors.
Okay, but it's, it's more heme-focused as of now.
At the moment, most of our investment is in heme.
Got it. We've only got a couple of minutes left, so I'm gonna look for one last chance for an arm in the air. And whilst I'm looking, with my other eye, I was going to say, we've covered a lot of ground today, but I'm sure there's stuff I haven't asked. Is there anything that you wanted to highlight that I haven't raised in my questions that you think I've missed or I should have highlighted?
Yeah, I think that definitely our focus on early stages of cancer is something that I think is really important, and we're very committed to trying and improve cure rates. And so that's-- We talked about that and touched on that a bit, when we were talking about INT and sort of our development plans there. I really think that, you know, WELIREG doesn't always get the focus that potentially it should. We had the LITESPARK-005 study readout showing improvement in progression-free survival in patients with renal cell carcinoma who had received prior IO therapy and TKI therapy.
And so again, being able to go. We talked a little bit about this, but being able to go in the areas where angiogenesis is a driver beyond patients who have VHL, or other predisposing conditions, I think is really exciting there. And then I'm really also excited about bomedemstat, so the LSD1 inhibitor.
This is the Imago-
It is the Imago drug.
Gotcha.
Those patients have a lot of burden. People who've got essential thrombocythemia, myeloproliferative disorders, these are diseases that really have a negative impact on their quality of life and can lead to horrible malignancies, and so to have something potentially disease-modifying is very exciting to me.
Very good. Well, on that note, I'm gonna end the conversation here. Many thanks, Marjorie, Merck team, for joining us today, and thank you for the audience for attending. Thanks again.
Thank you.