Gilead Sciences, Inc. (GILD) Earnings Call Transcript & Summary

[indiscernible] represented by their Executive Vice President of Kite, Cindy Perettie, who is in charge of running their CAR-T franchise. So Cindy, thanks so much for joining us.

Thanks for inviting me.

So I'm going to start with a multipart question of something I think is on a lot of people's radars and minds lately, with anito-cel, the BCMA CAR-T for myeloma. We just saw some updated data at EHA. And I'm curious how this frames your views on some of its potential advantages and differentiation from other BCMA CARs. First off, on efficacy, just some of the potential predictors of response there, extramedullary disease, number of prior lines, et cetera, what do you think is most important that we should be looking for to compare your data set to other therapies? And then when we think about safety, what do you think is the most likely reason why we're not seeing as many neurotox events, we're not seeing any Parkinsonism, the degree to which this relates to better patient selection and prophylaxis and how -- and your expectations for how that's going to hold up? So I know that's a lot, but really it's about efficacy, safety and how this compares.

Yes. I am happy to talk about anito-cel. I think maybe we'll start with the efficacy. So if you look at the efficacy and you're may be able to see the abstracts that we're going to be sharing more data at EHA in about 2.5 weeks. The efficacy that we're seeing now, we've got about 12.6 months' worth of average follow-up for those patients. So you're seeing a more mature data set. And we continue to see a nice high response rate at 97% for the overall response rate. We're also seeing our CR rate mature from what you saw at ASH, and so we're at 68% on the complete response rate. And we expect that to continue maturing. And then on MRD, we saw 93% minimal residual disease. So we feel really great about this data. You asked about how do you think about the various high-risk subgroups. And I would say, take a look at our subgroups. We'll be showing that data with extramedullary disease, cytogenetics. And what we're seeing in those high-risk patients is we're also getting very consistent responses. And so we feel like the efficacy for this product is on par or better with what we see in the existing constructs in the marketplace today. And of course, that data will keep maturing. The second piece was around safety. And so for our safety profile, for grade 3 or above CRS or ICANS, we're at 1% or less. And what excites me about that is that can become an outpatient therapy. So if you think about patients who end up being admitted to hospital, it's usually from the grade 3 events that you see. And so how we're looking at this is that we have an opportunity to reach more patients. We have not seen any of the delayed neurotoxicities that you see with some of the existing constructs. So no Parkinsonism, no Guillain-Barre. We haven't seen any of the cranial nerve palsies. In addition, we haven't seen any enterocolitis either, which is something that has been reported with some of the other constructs in this space. So we're feeling great about the safety. We're going to keep looking. It's early days. We have about 150 patients that have crossed over the 4-month mark where we would expect to have seen some of these things happening. And we haven't. But again, we'll continue monitoring and iMMagine-1 as well as in the iMMagine-3 study that we've started. But that safety profile coupled with the efficacy is something, again, I'm going to stress could make a very nice outpatient therapy. And as we want to reach more patients where they are, that becomes really important. If you couple that with our manufacturing today, so we have 3 manufacturing facilities across our network, and we make our own viral vector. So we're moving the viral vector as well into our vector facility. But today, we're producing anito-cel for our clinical trials out of our Maryland production facility, and that's also where we'll launch out of. And within the clinical trials, we now have more than a handful of patients on iMMagine-3, and we're seeing those turnaround times in line with our commercial products, in line with Yescarta and Tecartus time lines today. And our goal is to really meet that is to, at launch, have that 96% reliability that we see with Yescarta and Tecartus to be able to get our turnaround times down. So that's really meaningful for patients. So we're excited about that. You asked kind of how do we think about the safety. And there's certainly a number of hypotheses floating around. We spend a lot of time with our translational teams. We've been able to make the other 2 constructs within our research organization and then looking at the iMMagine-1 translational data. And there's a couple of things that we notice that are different. The first is, for those of you who've seen pictures of the binding domain of the D-domain, the D-domain on anito-cel is a small compact binder. And when I say small compact, it's helical in structure. And if you look at the other constructs that are available on the market today, their binding domains are quite large proteins that have integral folding. And so what we found in mimicking cellular, intracellular uptake of these, we are not seeing the D-domain change based on PH or temperature, any of the things that would change physiologically in the body. But what we did see on some of the other constructs that we reproduced is we did see changes in the fold. And so when we look at binding in addition to that, we see a very fast on- and off-rate with the D-domain. And we think that, that again mimics more physiological conditions. That fast on- and off-rate allows for really rapid killing and effective killing of the myeloma cells. But it also doesn't hang on and create the immunotoxicity that you see when something has really strong binding. And so our -- today, our hypothesis that we're testing is really around the binding domain and our ability to not create those inflammatory and immunotoxic events that we're seeing with others.

Great. And what should we be looking for in this next set of data? And what's your sense of what would be required from a regulatory standpoint for approval in terms of just the magnitude of benefit, amount of follow-up numbers of patients?

Yes. So we're not commenting on the regulatory strategy per se, but I will say we continue to have engaging conversations with the agency around the iMMagine-1 filing. And we have all 117 patients now that we'll be sharing in the EHA data set. But obviously, those patients would be maturing over time and would be part of the full regulatory submission in addition to the Phase I study that was conducted by our colleagues at Arcellx. So we would be filing both of those data sets with the agency for the iMMagine-1 fourth line approval.

Got it. And you talked about manufacturing and it sounds like you've been able to really leverage a lot of your tight manufacturing facilities and expertise in this. What was the tech transfer like? Is that still ongoing at all? And I guess where would you -- it sounds like turnaround times are comparable to Yescarta. Where are you with regards to inspect rates? Is that something you would expect to be comparable? Or just because this is a different product and process, we should be kind of thinking about this differently?

Yes. Our goal is obviously to continue to be leading in manufacturing and get to those places where we have the 96% reliability we have with the Yescarta and Tecartus. And to do that, you obviously need high volumes, right? So we've treated over 29,000 patients today with the 2 constructs that we have today. So there's also learnings that come from it. From the manufacturing standpoint, our tech transfer went really well. So part of a successful tech transfer we found is we have our research facility located near our Los Angeles production facility, so that continuity between your research and process development folks really matters. Good news for us is Arcellx has their research facility located in Gaithersburg, Maryland, which is literally 2 exits from our Maryland production facility. So we had a really seamless, not just through the people, but also from the technology -- tech transfer, and that went really well. We had -- we literally had no hiccups, and I was waiting for something to come up and we had none. And so the tech transfer is complete. Again, we're producing today out of Maryland for the clinical trials. And that's great that we're doing that because it gets us used to producing. It allows us to put the automation that we wanted to put into that. So we're excited for launch.

Excellent. And as you think about launch, where do you see the highest priority market role for anito-cel? Is there -- I mean you sort of alluded to the outpatient opportunity. And even kind of beyond that, is there a certain type of center or a certain type of patient whose needs are not being met by the existing therapy? Where do you see this ultimately fitting in, I guess, at launch and then down the line for earlier line patients?

Yes. So at launch, so we'll be launching into that fourth line plus population. We get the question a lot, are the fourth-line patients today? And so I want to remind folks that today, for myeloma, only 1 in 10 out of patients -- 1 out of 10 patients who are eligible are getting access to CAR-T. So there's still ample room for us to grow that, and we will. In lymphoma, it's 2 out of 10. So we know that -- we now know some of the pieces that are going to help move that market, which is how we get closer to patients in the community. 90% of myeloma patients sit in the community, only 10% in academic centers because it's a more indolent disease, they don't get referred to later. And so reaching them where they are is important. So that goes back to my comment about having an outpatient therapy. So I just have spent about 3 weeks meeting with various authorized treatment centers and community practices across the U.S. And I know all of you meet with KOLs frequently. So I encourage you to ask the same questions that I'm asking. When they look at the paradigm they have today for CAR-T and their institutions, many of them are treating with CARVYKTI, but treating inpatients. So in a certain number of days, the patient may be dosed, but they come back into the hospital. What I found on average was 4 to 5 days they're admitted. The conversations we're having is they're not sure -- obviously, time will tell if they even will need to admit an anito-cel patient. And that's something that they're thinking through and certainly our clinical trial data will determine that. But bed space can become a capacity issue. And so how we think about coming forward with an outpatient therapy and what that paradigm could look like is really important. We tested it in about 10% of our iMMagine-1 patients. We're looking at about 20% of our iMMagine-3 patients being treated in an outpatient setting. And so we'll continue to collect data around that. But again, there is still a large unmet need. The great thing about myeloma is, there's been a lot of cool approvals in the last 5 to 10 years. So there's a lot of therapies that physicians can work with. And what we're finding is you're seeing the quads move on the front line, some in the second line and then the 3-drug regimens into third line. So there are still patients that have not received CAR-T by fourth line, and we really think we have an opportunity to treat those patients. So a combination of the academic centers and our existing authorized treatment centers as well as really pushing this paradigm for outpatient. Today, we have 555 authorized treatment centers around the globe. So we'll obviously be able to use that footprint, and that footprint grows every year. So by the time we launch anito-cel in 2026, we'll presumably be in more than just 29 markets and have more than the 555 authorized treatment centers we have today.

Great. Well, maybe speaking a little bit more about that existing footprint and the lymphoma space. Yescarta has been a miracle drug so far for so many patients, but there are also some challenges growing its use to the next level. You mentioned only 2 of 10 patients. Can you talk about CAR-T franchise commercial performance as of late, particularly coming out of first quarter? What have been the key positives that you've seen and some of the key challenges and any kind of additional data you'd be looking at to help potentially reaccelerate growth?

Yes. No, thanks for that question. So as a reminder, one of the reasons I came to Kite is that we're actually providing curative potential for patients. So if I think about our second-line data we have 4-year overall survival in 55% of patients, that's unheard of. That's second line. And third line, it's at 43%. We've shared 5-year overall survival data. And even in ALL, we're going to be sharing 5-year overall survival data at EHA, but our 4-year overall survival data in ALL is 40%. So we know these drugs are active, and we know these drugs can have a profound impact on patients. So like you were saying, it keeps me up at night that we're at 2 out of 10 patients. And so we have taken on a number of things to say, how can we make that look different? So I'll talk about quarter 1, and then I'll go through sort of the activities we have. So in quarter, we continue to face headwinds, both in-class competitors as well as out-of-class competitors with the T cell engagers. In-class, in the second half of last year, we saw a number of new approvals in the follicular lymphoma space, the CLL space as well as ALL. And so in those more -- those smaller tumor types is where we saw a lot of the headwinds in quarter 1. So our Tecartus franchise went down 22%. So we took a big hit there in-class and out-of-class. In lymphoma, we grew about 2%. So lymphoma, we're seeing a little bit more steady. But with the new entrants of both bispecifics in the ALL, MCL space and in-class competition, that's where we've seen a lot of headwinds. We're going to continue to face those headwinds because we can imagine that you'll see additional approvals. But some of the things that we're doing to unlock that is then how do we reach more centers, how do we reach patients where they are. I've talked about this previously. My hope when I joined Kite was that we'd be able to move those community practices in 6 to 12 months. And what we've discovered is it takes a little bit more time than that. They have to work through contracting with hospitals and a number of other things. But we're starting to see some of that work go into motion where we're seeing more community practices now taking on CAR-T treatment and doing it either in collaboration with a hospital or one of our existing authorized treatment centers. The second piece is for our commercial payers, commercial payment is often tied to FACT accreditation. And FACT is an accreditation that goes along with transplant and cell therapy was added to it. What's required to deliver a cell therapy is not the same as what's required for transplant. So holding that equal is really tough for a community practice who's never going to be a transplant center. So FACT has created a modular accreditation, and they will be piloting across 3 of the large integrated community practices, the second half of this year. We hope those will start probably in the late July-August time frame, and we'll have data from that. And I think it's that type of work that will show that you can accredit a community practice that could apply more broadly. And I think having those large integrated practices showing it can happen is going to be hugely helpful. We have a number of things going on the policy side as well, but those are some examples. And we need to move lymphoma because we need to make room for anito-cel as well.

Good. And then on lymphoma, on Yescarta and Tecartus manufacturing. Just maybe touch on the latest improvements that you're seeing there in manufacturing and turnaround time and where this may go in the future? And maybe also touch on the idea of kind of low-dose 3-day turnaround Yescarta.

Yes. So for the existing products we have today, the improvements in manufacturing, a lot of it comes from automation, so becoming fully automated. Our Maryland production facilities are state-of-the-art, and they operate primarily and fully automated. And we're bringing full automation versus semi-automation into some of the other production facilities. So that's a way we can make improvements. And we can make improvements on the back end on sterility testing. We learned so much during COVID. If you remember, we took sterility testing from 7-day plating to like 1-day turnaround on sterility testing. So we're applying a lot of the COVID technologies to the back end of our analytics and testing side. The second piece you brought up, we're super excited about. And so we have developed 2 products that look at a 3-day manufacturing. So today, manufacturing is anywhere from 5 to 7 days of the product. And when you look at 3-day manufacturing, you end up with more juvenile cells, so more naive cells, and these are more potent cells. And so one of the components of our next-gen products in lymphoma is that we have 2 constructs, 1 that looks -- that's just like Yescarta that has a 3-day manufacturing process. And then we have bicistronic which is a CD19, CD20. So it's 2 targets and 2 co-stimulatory domains, CD28 and 4-1BB, and that's in 3-day -- we have one that's 6-day and one that's 3-day manufacturing. We're going to show the 6-day data at ASCO in just a couple of weeks and then again at EHA and ICML. But the 3-day manufacturing, it allows you to go in with anywhere from 1/10 to 1/20 of the dose. So really low doses, but showing very similar potent efficacy and much better safety. And so if you think about going from a 6-day manufacturing to a 3-day manufacturing, right there is the savings in the manufacturing time. So we're looking forward to sharing data in the future on those products.

Excellent. And maybe speaking of the bicistronic, can you talk about maybe some -- I know it's been tried before, but you guys have some unique elements with these specific co-stim domains that are, I guess, independent coupling with the CD19 and CD20. Can you talk a little bit about the properties of the construct and then kind of what you're seeing profile-wise in terms of cell expansion and persistence relative to the sort of the first-gen CD19?

Look at you, you need to come to ASCO and we are -- press for us. So on the co-stim, the dual targets that we have, the bicistronics, we have done these -- we call it a bicistronic construct. You really have 2 CARs that are pulled together by a linker. You'll also see in this space tandem constructs, so they're 1 construct with 2 targets. And what we found in our own hands and others may find different is, when you do tandem, you only have so much. Our research had always said, you only have so much -- sorry, you only have essentially so much that you can pack into a single CAR, so much real estate on that CAR. And so by doing a single CAR in tandem, it's harder to get 2 targets in there most effectively and even harder to get great transduction efficiencies. So we went with bicistronic. So CD19 and CD20. CD19 has the CD28. So it gives you the sort of punch that you get with Yescarta, which is the rapid cell proliferation. But it's actually -- when I say rapid, it's muted just slightly. You get to the same peak, but you get to it over time because of the 4-1BB co-stimulatory domain. So you end up with the same number of cells, but the growth comes over a 3-day period versus a 24-hour period. And why is that important? It's super important as it comes to safety. So you want to get to the efficacy level where you have the same number of cells being produced, but doing that persistent component of having it over time and having it hang around longer is what we're seeing with the 4-1BB. So we get a little bit of the best of both worlds. If you have a chance at ASCO, we're showing our bicistronic construct with our 6-day manufacturing. It's called 363. And the goal of the bicistronic for us have been, we need to improve on efficacy versus Yescarta and existing products, and we need to improve on safety. So we're looking at both -- improvements on both, and we're looking forward to sharing all of that data with you.

Excellent. You've talked a little bit more about potentially going into I&I with CAR-Ts. And I know it's been an area where it's been tried. There was some initial -- initially a lot of excitement and then maybe the next rounds of data hadn't necessarily lived up to the initial enthusiasm and there's been challenges in rolling patients and finding the right patients and indications. So can you talk about your interest in expanding the bicistronic to I&I indications? What your current plans are? When we might see the readouts there? And how you maybe overcome some of the challenges? Do you have some data that you're seeing on B-cell depletion inhibition that gives you confidence you're going to be at a level that's more successful? And/or are there certain indications that you will be going after that you think maybe others have not pursued that? Where you think there's the most opportunity?

Yes. And similar to sort of hematologic landscape, if you think about rituximab, it's used both in autoimmune, the CD20 target, I think, about OCREVUS in multiple sclerosis. We know that CD20 is active in autoimmune diseases. So that's one of the main reasons we decided to look at the CD19, CD20 construct because we think having the opportunity to hit both, particularly if you have patients that end up having one or the other muted, you have a second chance. And that applies to oncology as well. But -- and then as I look at the construct, and I described, we're putting our 6-day manufacturing, the 363, the data we'll share in hematology at ASCO. We're putting that into -- we filed the IND, and that's going into the rheumatologic malignancies -- rheumatology indications now. And our belief really is to test that hypothesis is what is CD20 adding. The second piece is I described the 2 co-stimulatory domains about reaching the same peak number of cells, but reaching it over a slower period. That matters in autoimmune disease. I mean, these are patients who are living with a chronic disease. We don't want to see safety signals in the line of what you would see in an oncology patient. It just won't be tolerated. And so that was another component of why we wanted to look at our bicistronic compounds. And so we started we filed an IND. We're starting a basket study where we're looking at lupus, both SLE and lupus nephritis. We're looking at myositis, and we're looking at scleroderma. And we are in the process of filing a second IND where we would have -- we will look in neurological conditions, both myasthenia gravis and MS. And both of those we know have a CD20 component to it, and that's why it made sense to look at the bicistronics there.

Great. And when might we see any initial hints of those data, do you think? Is that sort of a 2026 event?

Yes. We're just getting started. So we'll keep you posted. As you said, it's about -- it's become a crowded space. So we're looking at running our trials in a global setting in the U.S., Europe and Asia, just to make sure that we can get a diverse patient population and understand not see differences -- not end up having not seen differences early in the Phase I.

Good. Just in the last 30 seconds, anything that we didn't touch on, on the cell therapies that we should be keeping an eye on or that's maybe under investors' radars but we should...

Yes. I would rather put in a plug for solid tumors. So we have an oral coming up at ASCO in a week in the glioblastoma space. So we have put together a construct with our colleagues at Penn that is looking at a dual CAR, which is an EGFR and an IL-13 RA that we are testing in glioblastoma. I think we'll be sharing somewhere between 15 and 18 patients worth of data there. These patients are heavily treated glioblastoma patients, probably coming up on second and third line end of life, unfortunately. But the patients -- the therapy is delivered through an Ommaya into the brain. You don't have to lymphodeplete the patient. It stays within the CNS. And so hopefully, you'll have a chance to come see that data. We are working on a version 2 of that construct because we recognize from the data we see in Phase I that we need a little more persistence. We want it to hang around in the CNS and still not have to lymphodeplete the patients. So we're putting armoring on that, TGF-beta armoring, and we'll be testing that second construct in the coming year. But solid tumors are possible now.

We'll look out for that data in the coming weeks for sure. So thanks again. Fantastic to have you here.

Great. Thank you so much. Thanks.

Thanks, everyone.