Kymera Therapeutics, Inc. (KYMR) Earnings Call Transcript & Summary

Great. Good morning. Welcome, everyone. Let us go ahead and get started. My name is Vikram Purohit. I am one of the biotech analyst at the Morgan Stanley Research. This is the fireside chat with Kymera Therapeutics, and I am very happy to have with me on the stage, CEO, Nello Mainolfi. Nello, thanks for joining us.

Thanks, Vikram. Great to be here.

And before we get started, let me read a brief disclosure. For important disclosures, please see the Morgan Stanley research disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative. With that, let us get started. Nello, before we dive into the pipeline, maybe you could just level set for us, provide some brief opening remarks and just kind of cover some of the key milestones you believe Kymera has hit throughout the course of the year.

Great. So if you do not mind, before that, just a brief description of Kymera hoping that there will be new people listening to this or viewing this. Kymera was funded about 6 years ago in 2016 on the concept of taking a new drug modality called targeted degradation into a drug development and commercialization company. So targeted degradation is, I would say, the only really new small molecule-based modality of the past 50 years that allows us to drug proteins, targets, and address diseases in a completely different way. It allows us to use small molecules to remove disease causing protein. So the excitement that we had 6 years ago and is unchanged, it is not more today, it is really to build a company that can go after a new generation of therapeutic hypotheses. So we have been really focused on building the company, focusing on a pipeline, obviously, building a pipeline of programs, where we believe protein degradation is the best or the only way to tackle that particular biology and hopefully develop game-changing therapies. And we have also invested heavily on our technology platform, discovery engine; however, we want to define it to obviously being efficient, productive, but also to being able to evolve not only how we understand the technology works, but more importantly, to allow the technology, the modality, the protein degradation approach to being able to go after new things, new hypothesis, new targets, new ways to go after on drug or poorly drug protein. So that is the back story of the company. So where we are today, we have 3 programs in the clinic. They are all actually at different stages of Phase I. We have KT-474, which is an IRAK4 degrader that is really in the final stages of our Phase I study. We have run, I would say, comprehensive Phase I, not probably the most compliant vaccine at least recently. And so we can talk about where we are and what we expect. I would say that, as we have said in the past, we are on track to share data from the patient cohort before the end of the year in Q4; I think we have said most recently. So everything is on track for that. Then we have 2 oncology programs that are in Phase I dose escalation. As you know, obviously active volunteer and patients for 474 outside of oncology, those trials are run differently than oncology. In oncology, we went directly into patients were in dose escalation for KT-413, which is an IRAKIMiD degrader; again, this is a dual degrader mechanism. They use this hydro bi-functional degradation and molecular glue to go after MYD88-mutant lymphoma, so we are dosing patients, and we expect to share what we call proof of mechanism data also in Q4, we can talk about where those are, hopefully, in the next few minutes. So we are still on track for that. And then we have another program, another first-in-class program, KT-333 going after STAT3, STAT3 is a key transcription factor of the JAK/STAT pathway. JAK/STAT pathway continues to deliver pretty impressive drugs. We just saw recently the TYK2 inhibitor that was approved just Friday last week. So again, another key important note that has not been dragged before. So also there, we are in dose escalation in patients with data, proof of mechanism data before the end of the year or Q4; however, we want to define that time frame. I would also say we have a commitment as a company to deliver at least one new IND per year, one new clinical program per year. And so this year, as everybody knows, we have our MDM2 degrader that we expect to have a clear IND before the end of the year, and we will share when that happens.

Great. Great. Thanks for the overview. Maybe let us start with KT-474. It is currently your most advanced program in your pipeline that is your IRAK4 degrader. You generated, as you mentioned, some healthy volunteer data throughout the course of the year. Do you want to just touch on some of the highlights both from a degradation, efficacy, and safety standpoint on what you have seen so far?

Yes. So just very briefly, though, why did we decide to go after IRAK4? I would be very brief 2 minutes. There was really only a few pathways. So we always think about target selection and going from pathway and critical node that has not been dragged on drug well because that we feel is the de-risking mechanism. So there is only a few pathways that have been validated in immunity with drugs on the market. One is the JAK-STAT pathway, which we have already touched on, and the other one is the IL-1 pathway, the IL-1/TLR pathway, with several drugs heating nodes in those pathways, mostly upstream that have shown positive proof of concept. Some of them have been approved or to be approved soon. The challenge with that pathway with the L1 pathway is that all the L1 family cytokines and TLR agonist have to signal through the Midazolam complex and blocking only one singular cytokines, whilst this shown clinical activity, I would say is insufficient to block the broad inflammatory signals and this pathway is responsible for. So the opportunity with an IRAK4 degrader not with an inhibitor, and we can get into that is to block the pathway fully and basically deliver hopefully, a clinical profile. But for sure, we have shown already a biological profile that is superior to all other agents in the pathway. So it is expected that an IRAK4 degrader is going to be active in patients that I will say. So what have we shown so far? So the question that we have answered already is KT-474, is that the right molecule to be asking the clinical question. So we like to say that we will probably talk about that also in the oncology program. And I think the answer is yes. We have shown that we have an IRAK4 degrader that is able to degrade the target almost what I would say completely in blood. We have shown it in blood. We have shown it in skin at doses above 50 mg already, we are able to show complete degradation of the target at steady state, which, to be honest, you do not see even complete inhibition with any small molecule that you dose in the clinic, given that it is a different modality. So we have shown that degradation of IRAK4 has an impact on downstream biomarkers. We had to use an Ex-Vivo assay healthy volunteer, because healthy volunteers do not have inflammatory conditions. So we took the blood of active volunteer. We activated a human blood and show that the subjects that have been dosed with IRAK4 degrader and where IRAK4 was a really low level, it is almost absent. We protected them from inflammatory signaling. And in fact, we were able to block up to 9 different cytokines in the bladder patients. So we have shown the degradation is potent, is functional, and we had a good safety profile. We had really no serious adverse events and all adverse events were quite mild, and headaches and nausea and palpitations, things that are somewhat subjective, I would say. So we feel good about going into patients with that data set to start to understand more the PK/PD in the patient population.

Got it. And then on safety, there has been a fair amount of investor focus an observation of palpitations and QTC prolongation that you've reported through the healthy volunteer data. Could you just remind us exactly what you saw and what the scope of that signal was?

Yes. I think I would like to separate the 2. I think we have observed population in 2 or 3 subjects out of, I think, more than 40 that we dosed in the MAD study or at least they were recruited to the trial. And these were shortly that feelings of palpitations that were not when there was an opportunity to measure functionally did not give rise to any measurable changes in heart rate or even Q2 that was different from others. So we see them or at least we saw them also according to what our consultants tell us as an observation that was restricted to some patients that had this filing. So that is one. We like to separate because we have no data connecting QT to capitation. Then what we observed, actually, this was well after we completed the healthy volunteer study. I do not want to rehash the whole history, but we decided to go and look because he was not actually apparent from just observing the QT traces from a safety standpoint because we had no safety signal, but we went there and did an analysis, and we saw this subclinical we have called it QT prolongation, where we see basically a change of mean from what is normally seen in placebo or in healthy volunteer, which can range sometimes even up to zero the way to 10 milliseconds, and we have seen an increase that went between 10 and 20 milliseconds. So we did not see any adverse events. We obviously did not see any QT changes that went into high-risk range, which usually is above the QT delta is about 60 milliseconds or the actual interval above 500 millisecond, ours was always within $450 million. So in the normal range and the debt, again, was very minor. But the more important thing, I mean, those are important facts. The most important aspect of what makes us and others feel quite confident that this profile is not going to impact our development and our potential commercialization of the drug is the fact that this was non-dose responsive, self-limiting. There seems to be a saturation of this mechanism that happens very early and that does not allow this QT to have excursions that go into areas that would be concerning. And so that is why it is something that obviously we have shared for transparency, but it is not something that we are overly concerned about.

Got it. Let me ask you one last question on this, and then we can move on to discussing the readouts for 4Q '22. Since you have had these observations, what kind of work have you been able to do to it or figure out what exactly might be causing these observations? And what have you learned so far from like a mechanistic and scientific national perspective?

Yes. I mean, first, what I would like to say is that we have conducted plenty of studies pre-clinically, including Forman chronic tox study, where we were measuring QT never were able to find a signal pre-clinically. So obviously, there could be arguments that either the signal is so small, I will say, quite small that the instrumentations in preclinical species or Preclinical patients in themselves were not as sensitive. I guess that could be the scenario. But obviously, with this observation in hand, we did do a lot of work to try and figure out what was driving it. I think what we can say with a high degree of confidence is that this has nothing to do with IRAK4. And I cannot say that enough because I keep sometimes hearing that there are questions about that. We know that there are subjects humans that have no IRAK4 that obviously do not have any cardiac phenotype. We know that there are molecules that inhibit the function of IRAK4 understood is different, but they have been dosed to hundreds of patients without any real QT signal. But I would say we have shown also in-house that we have done some human cardiomyocyte work, and we have shown that KT-474, that has a very, very low affinity in terms of binding to the mind channel, basically we can replicate that kind of the QT signal that we see in the clinic in cardiomyocytes, concentrations, there are overlapping what we see in terms of and channel binding, then we take another IRAK4 degrader we made, obviously, 1000s of them that does not have a facility for design channel, and we do not show that there is a change of current in cardiomyocytes. So we believe that it is driven by [indiscernible] channel. Our affinity is very mild at relatively high concentration. And we believe that is the reason why the signal is very modest and nonadverse.

Got it. Okay. That is helpful. Let us turn to the Part C data from the Phase I study that you are expecting to read out by the end of the year. The first basic question, what can we expect to learn. What parameters of data do you think we will be able to see there?

Yes. So before I do that, I just want to close on the QT. I just want to say, so non-adverse modest baseline change, no excursion, we believe, will not require monitoring post approval. That is the reason why we do not believe this is going to have an impact on how we develop and commercialize this drug. Why did we run our patient cohort? Some would ask, why not just go directly to Phase II study, which, to be honest, was also on our plan. And the reason for the patient cohort, this was designed years ago was new modality, degradation outside of oncology, how does degradation profile translate from healthy volunteer to patients. Hopefully, people appreciate that when you work outside of targeted oncology, your target population is a variety of cell types; B cells, T cells, NK cells, microphages, neutrophils. And we wanted to make sure that in a higher inflammatory state; where there might be or we know, for example, in [indiscernible] there is a higher target expression. Can we maintain this amazing, I would say, degradation profile. So that was the reason for running the study. So what we want to learn, and I think what we want to see from this study is really good degradation, degradation that is similar to what we have seen in SC Volunteer. We also would like to see now that we have inflammatory conditions, and we do not have to do this artificial Ex-Vivo assay. We would like to see that degradation of IRAK4 has an impact of IRAK4 dependent cytokines and chemokines. And we will look in the skin, most likely, where we should be able to see changes. We also look systemically, but it depends on how much systemic inflammations patients would have. And so for us, it is important to continue to draw that line between PK/PD and impact on cytokines. And then we also are keen on demonstrating that our safety profile is in line with what we have seen with the healthy volunteer. And I think this is the data set that we would like to see to then transition the program into Phase II. We are measuring, I do not want to now completely dismiss. We are measuring clinical endpoints. And I know that this is a topic of discussion, I am sure we will discuss today in all our one-on-ones. So we are measuring clinical endpoints because we can because it has been shown that in 20 studies, you can measure those. But we cannot have expectations on clinical endpoints for this type of study, right? It is a 28-day study. There is no placebo. It is small numbers, right? HS and ADS will be 10-inch if we get the numbers exactly right. And more importantly, I also would like to remind everybody that in order for us to get steady state degradation in the skin, it takes us more than 2 weeks. So it is asking a lot of the compound that in 2 weeks of maximal pharmacology at best will have a clinical signal that can be measured that gives us confidence that can be measured objectively outside of the placebo arm. What we would like to do, though, is to measure it, see if there are any trends that we are seeing and whether those trends can be correlated to BioMarkers data that we want to see.

Okay. Got it. And what kind of patients are you enrolling into the study? What is their overall disease profile from HS and from an AD perspective?

Yes. So we have not discussed obviously in detail the type of patients that we are recruiting. But I would say there are going to be patients that have measurable disease. So these would not be mild patients.

Okay. Got it. You alluded to the difference between IRAK4 degrader and an inhibitor in your opening remarks, but just to put a fine point on it. Pfizer recently discontinued their IRAK4 inhibitor in HS, and then they posted some data at EADV that there was a fair amount of focus on. So given all of that, could you just tease out what you think are some of the key differences between a degrader and an inhibitor and why you think a degrader could be significantly more potent for this patient population?

Yes. So for people that have followed the company for 6 years now, 6 plus at this point, I have been saying and other, I get the help of Jared now in the past 3 and half or so, been saying that in order to block this pathway, you need to degrade the target and the kinase inhibition is going to be insufficient, especially in high inflammatory conditions. So, I think actually the data from Pfizer is validating what we have been saying for the past few years. Now yes, I appreciate that with a molecule that is active, let us not forget, IRAK4 inhibitor from fighter, which is actually a good molecule for an IRAK4 inhibitor, go positive proof of concept in array. So we are clearly seeing, first of all, that a suboptimal pathway inhibition, IRAK4 inhibition, TA2 inhibition, IL-1 antibody, suboptimal pathway inhibition, IL-33, R36. These are all one-fourth or one-fifth or maybe 50% of the pathway that they can inhibit. All of those have shown proof of concept. So we should appreciate that if you have an agent that is going to block the pathway fully, you have to expect clinical activity. Now with the IRAK4, so yes, you have seen proof consistent in RA. I mean, I think if you want to be positive, you have seen some hints of activity in HS. I think maybe if you look at the time to flare, plot, maybe you can squeeze and say, it looks different from placebo. I am not sure about the stats in the graph. But it is clearly different. And we know that. We know RA and HS are very different diseases. So I think what we have learned from this study is, first of all, the molecular is good safety profile. Molecule well behaved in the study, clearly did not show meaningful differences from placebo. So the question is why? We have actually a few or anybody listening goes on our corporate deck of this month. We are probably supposed to do it for a while. But we decided to compile all the data we put together on differentiation between inhibitor and degrader in a wide variety of biological context. There is data from Kymera and there is also data from academics outside of Kymera that we do not even collaborated it. These are just people that have been interested in this differentiation for a while. So you will see there that there is a biochemical difference. So IRAK4 is central to midazolam assembly and signaling. If you don't have IRAK4, you will have a big impact on how the midazolam signals. If you actually have IRAK4, even if you inhibit it, you continue to allow the signal to propagate. So biochemically is very different. And then we have shown function in so many ways that especially in diseases where there is both IL-1 signaling and TLR signaling, which we believe HS, AD, probably IPD, and many diseases that have both local and systemic inflammation, IRAK4 inhibition is insufficient to drive pathway inhibition. We have shown that biologically in many contexts. So in some context, the IRAK4 inhibition is sufficient, like IL-1 only activation. Sometimes you can see good effects. And so it is not surprising that in some disease, that drug should be or could be approved if the sizes decided to continue. I think they are continuing in RA. But in many, especially the more diseases where there is higher inflammatory state it is just insufficient. So now can we guarantee that our IRAK4 degrader is going to be active NHS, we cannot. But what we can say is that we believe an IRAK4 degrader is the best way to block IL-1 TLR pathway. There are many diseases where this pathway is competent and is relevant, and we have seen them already by other agents in this pathway. So it is realistic to expect that an IRAK4 degrader will be effective in many conditions, and hopefully, we will have many patients.

Got it. All right. That is helpful. Maybe one question on the design of the Phase I study. So there was a bit of investor focus on some changes that were posted a couple of weeks ago to that program. Do you want to just unpack for us what those changes were and why you made them?

Yes. So I think we learned that clinicaltrials.gov is followed very closely. And to be honest, these changes were operational changes that the clinical team, Jared and his team made to allow us to either learn more or have more flexibility. So I know there are a few, so I am going to address them one by one. And if I did not include them all, please remind me of what I missed. So there was a change to increase the number of recruitable patients from 20 to 30. So this was maybe something that we should have done a lot earlier is just that once we got close to the study, we realized, okay, let us increase the number. The reason was we wanted to include up to 20 patients, and we wanted to have 20 patients on study. We also know that in order to recruit 20 patients, you probably need to screen more than 20 patients. We did not want to be in the position where maybe we had 21 or just to make sure we had the right mix between HS and AD and we could not have the 21 because it was not in the protocol. So we have no plan to have 30 patients. You will never see 30 patients worth of data. I can guarantee you that already. But they might be 20 or 21, and so that is the flexibility that we wanted to build. So that is very, very much of a really process change. There was another change; I believe it was the infrequent dosing. So that is different than the number of patients. This was for people that have followed the story closely in the SAD study; we have seen some really profound degradation that lasted for multiple days. Actually, to us, that was a welcome surprise. I think seeing a drug that can degrade the target after a single dose for up to 6 days, above 80%, 85% for some doses was quite impressive. So we always had the question, can you dose this drug less frequently. What does the PK/PD look like? So this is both a drug question and a platform question. And because in the assumption and the program will go into Sanofi's and from the end of Phase I, and we will not be able to run any more studies with the molecule. This is the window that we had to run the study.

Got it. Okay. And that is a good segue, I think, into one final question, and then I would like to move quickly on to the STAT3 and IRAKIMiD programs. Just remind us, what are the time lines for data delivery to Sanofi? And then generally, how much time do they have to get back to you with their decision?

Yes. So we will share the data with Sanofi as soon as we have the data complete. I will add that we have a really close relationship. Hopefully, they will feel the same way. And so we share information along the way. So obviously, there will request as is contractually defined the whole data set. So when we will have it, we will share it and then have X number of days contractually to make that decision. We never share how many of those days are so we are not going to do it today. But it is a reasonable amount of time for company specialties to have to analyze and then make decisions.

Okay. Understood. Great. Let us move on to your other 2 data sets expected by the end of the year, STAT3 IRAKIMiD. You have mentioned it is going to be proof of mechanism data, so 2 quick questions. What can we learn and what do you think is a win here?

Yes. So I think what I have said more recently, we like to think about when you develop a drug, especially in product degradation, which is this beautiful opportunity to measure what the drug is doing in real time. There are few other technologies that allow you to do that. So the first question, a company should ask or any scientific endeavor should ask is the drug doing what it is supposed to be doing. So what for us, both 413 and 3D3 need to do is to degrade the target at levels that we have seen pre-clinically are therapeutically relevant and is that done with an advance safety profile. So for us, the win is, for example, KT-413, we have seen preclinical. So this is a combo in a single molecule, so [indiscernible] and IRAK4. And we have shown that between, let us say, 60% and 80% degradation of all leads to really profound anticancer activity. So can we show that in our dose escalation at a particular dose that hopefully, we reach this year? And is that coming with a good safety profile. We know that in that program, especially we expect that for the IMiD biology there is an expectation that there will be some neutropenia. We have done a huge amount of work to actually develop the drug where that will not be limiting our exploration of efficacy. So we want to be able to share that. For 333, the question is going to degrade at 90% because now this is not a combo. So it is in a way, it is in a way to single agent. So you need to go deeper in degradation to see maximal effect. So it is about 90%. Can we degrade a 90% for about 48 hours? And does that come with a good safety profile. So those for us are close to that. Those for us are wins because so that as we go into expansion and focus on our key patient population, we go there with expectations of seeing clinical activity and not a question of we [indiscernible] dose or with the right TD. So we can say the drugs are doing what they are supposed to do and now let us start look for efficacy. And if not, we have a clear answer. So the drug was good, the mechanism was best. What you do not want to have is you have no idea if the drug is good, then you are going to ask the mechanistic question for the clinical question.

Great. So we are just right about at time. Now maybe one final housekeeping question for just to round out the discussion. Your current cash balance associated runway? Do you want to just mention that?

Yes. So as you probably know, we had a closed the pipe investment financing in August or September. I do not remember actually 2 weeks ago. And so before then, we had about $480 million. Now we closed August with over $600 million. Our previous cash position will allow us to get us into 25. So you should expect that this injection of cash will extend that. We just have not come out with a new expectation just because we have a lot going on between now and at the end of the year. And I think probably early next year, we will update that guidance.

Okay. Great. With that, we will close it out. Nello, thanks so much for this. Appreciate it. Thanks, everyone, for joining.

Thank you. Have a good day.