GSK plc (GSK) Earnings Call Transcript & Summary

Good afternoon, and welcome to this session featuring GlaxoSmithKline at Cowen's 41st Annual Healthcare Conference. We're very, very pleased to have with us Dr. Hal Barron, who is the Chief Scientific Officer and President of R&D at GlaxoSmithKline. Hal, thanks for taking away from your journey to transform GSK at this time to be with us. We're very, very appreciative of that.

Maybe we can start out by asking a high-level question on your thoughts on the progress being made within GSK R&D since you came on board. What would you say has gone very well? And what would you say has gone less well?

Well, thanks, Steve, for inviting me. It's a pleasure to be here. Yes. Let me start -- that's a high-level question. Let me just recount a few things that I think we've been focused on and I think are going relatively well. In 2020 alone, I'll start there, although I think we've had some significant progress before that. But just in '20, last year alone, we had 9 approvals, actually, and 4 new molecular entities approved in a given year, which is pretty impressive for any company and certainly in my career. We also had 9 pivotal study starts, which will hopefully ultimately end up unblinding and being positive and leading to new medicines for patients. 2021, it's obviously early, but we've already had some success as Cabenuva was approved in the U.S., and we're now submitting that for every 2 month dosing. We started the Phase III in the RSV older adults trial, and we'll get to that in a little later, if you want to talk about the science behind that and the opportunity that might have for patients and the company. We've shared some positive data on dostarlimab at ASCO GI. We're hoping that is an approval this year. We have, I think, a lot of progress on our COVID-related fronts from the treatment perspective. We just announced the Phase II data from otilimab, and although it just missed, actually, its primary endpoint in the overall population, there was a pretty significant effect in the elderly, where we had actually been cautiously optimistic that data would be more positive, given some preclinical data as well as some data -- and from patients, biomarker data in patients, looking at where GM-CSF is elevated. And a bunch of other interesting things will be evolving as it relates to otilimab through the year. We also have some of the VIR data coming out this quarter with a monoclonal antibody to COVID, which we're looking forward to. We have a little later in the year, we're going to be reading out some information from our ICOS, our agonist antibody to ICOS receptor that induces a T-cell proliferation and differentiation for immuno-oncology, the INDUCE-3 interim analysis and the descriptive Phase II study in ENTREE-LUNG, the randomized Phase II study. We also are enrolling the BLENREP gamma secretase, which hopefully we'll see positive data for this year. We're cautiously optimistic that, that combination should work. We can go through that, if you like. We should have some pivotal data on daprodustat, which I think is becoming increasingly important for us. We should have some COVID vaccine data and some other studies, including potentially some cell therapy with [ emoiso ] in lung, maybe even some data in lung cancer with cobolimab or TIM-3 antagonist, and as well as a number of Phase III starts, including our long-acting IL-5 program in asthma that's -- should have a pivotal study start; and linerixibat, our IBAT inhibitor for cholestatic pruritis is starting with a pivotal study. So a lot going on in terms of pipeline progress. Other things going -- I think the governance model that we put in place, a very clear accountable decision-making framework that I think will add a lot of rigor to how we make decisions and who makes them. And ultimately, that will, I think, be helpful for ensuring that we're allocating capital appropriately to optimally invest in those projects that are most likely to help patients and benefit the company and deprioritizing those that weren't. We've moved, I think, over the last year into a nice integrated development model for both pharma and vaccines. That's new, and I think that's coming very well. And although early and not tending to be something that investors focus on a lot, but I'm very excited about is our research strategy which we put in place, focusing on human genetics, functional genomics and machine learning, the real -- a triad of these, what I think will ultimately be very disruptive technologies, has gone exceedingly well. I think we now have the largest genetic databases with 23andMe and UK Biobank, FinnGen, Open Targets. A number of data sets gives us, I think, more information genetically on patients as well as really with the laboratory for genomics research that we started with Jennifer Doudna and Jonathan Weissman as well as a recently announced collaboration with the [ Broad A ]. And an internal functional genomics group that I think is second to none is giving us a lot of enthusiasm for seeing targets that no one else can see. We've already now, with 23andMe collaboration, had over 30 targets that have been advanced in research, and we're forming other collaborations like with Adrestia in the U.K. looking against -- at synthetic rescue and synthetic viability, as well as having built up the synthetic lethal research unit from scratch. Basically, we didn't have a focus on synthetic lethality in 2018. We in-licensed Zejula and have been doing a lot of great work on life cycle management there, but are trying to complement that with other synthetic lethal targets. And that's all going very well. So I think that's -- those are the things going well. We're in a high-risk field. Only 10% of things we put into the clinic end up translating into medicine. So of course, we're going to have disappointments, and we've had a few Phase III -- sorry, Phase II failures recently with interfused alpha not working in lung cancer, which was a very, very high bar, trying to be better than pembro. We thought this was a smart bet to make the EUR 300 that we invested to get access to what this program and have -- which had 400 patients worth of safety data and recently compelling preclinical data as well as some intriguing clinical data in Phase I, but that didn't pan out as well as 2 inflammatory bowel studies that were also in Phase II that didn't read out positively. But that's balanced with our Phase II readouts now. We track this pretty rigorously and have been underperforming industry average, as is in the past. And when we look at our -- the industry average of about 25% to somewhere between 20% and 30% of Phase IIs are positive, we've been tracking below that in the 2010 to 2020 era, at some points, significantly below that. And most recently, our numbers are above 40%. So we're still tracking to have, I think, made good decisions on what to put into Phase II, but again, the majority don't end up coming out to Phase III. But as long as you are rigorous in your design and when you kill things, you know you're killing it because they didn't work and you have clear evidence for that, we think that's actually something to celebrate, is making clear go/no-go and don't. So while the outcome isn't what we hoped for, the process of ensuring that the Phase IIs are robust as what we pride ourselves on doing and making the best choices for targets and capital allocation. So long winded, I hope that at a high level covers what we're doing.

It's a long answer because there's a lot going on and a lot of good stuff, too. So we would like to drill a little bit deeper on some of these assets. [Operator Instructions] So maybe we could start out with one of the more exciting products in your pipeline, and that is BLENREP. There are 2 important readouts in 2021, the DREAMM-5 study and the DREAMM-4 study. Maybe you could help us set expectations for these readouts.

Yes. Well, let me start with the highest level, just to make sure everybody realizes this is a pretty special medicine, I think. It binds BCMA, that's pretty straightforward, which I think now people are realizing is really the fourth pillar, if you will, in myeloma therapy, moduling the BCMA receptor. This therapy is special in 3 other ways, though. Not only does it bind the receptor, but it carries within a payload a toxin that is delivered to the cell and very, very potent at killing the cell, the plasma cell, the cell responsible for myeloma. And in doing so, actually, uniquely, because not all antibody-drug conjugates do this, but we've seen preclinically and I think we'll see this clinically, induces what's called immunogenic cell death. That's a death that exposes some of the foreign proteins to the immune system and activate the immune system, which may or may not translate clinically, and we're exploring that in DREAMM-5, so we'll talk about that in a second. It also -- and this is also very important to people, the Fc portion of the antibody is afucosylated, which means that by altering the fucos and glucose, the glycosylation pattern of an antibody, particularly in the Fc portion, you can actually very tightly or very weakly, depending on how you do it, engage the NK cells that create this immune synapse to kill the cell. And so this is afucosylated, which gives it a significantly greater killing potential. And this has been seen with other antibodies as well, both naked and antibody drug conjugates. So it's got multiple ways in which it induces its self-killing. We've seen in DREAMM-2 in the most refractory patients, a very significant, about 32%, 33%, 31% response rate in this heavily pretreated patient population. And the responses are very deep and durable. We've seen that. Now we're moving more proximately. And I should point out that one of the side effects we do see is this ocular toxicity keratopathy. And I think that the 12-0 ODAC vote in favor of Belamaf reflects the fact that this incredible efficacy is -- outweighs the potential concerns that patients have when they develop this keratopathy. But as we move more proximally, it's our view that will lead to show either similar profound efficacy and/or reduce the ocular tox, particularly in the front line. We see the ocular tox as being important to minimize, to the extent possible in all patients, but particularly in the frontline where we have a number of strategies I'll go over. So the DREAMM-7 and DREAMM-8 studies are pivotal trials moving into the second- and third-line setting. We are going head-to-head, if you will, with VELCADE and Darzalex, so very high bar, but we, again, believe that, that's warranted given the profound efficacy that we've seen. And the recent Trudel data from the ASH meetings recently, I think gave us confidence that the response rates were quite high. Even in the most heavily pretreated patients, we're seeing somewhere between depending on the population and the breakpoint, something like almost 90% response to almost 100% depending on the subgroup, but some of these patients were having very deep responses, complete responses [ MRD ]. We are evaluating lower doses. We're evaluating longer duration between doses. We're evaluating different regimens for stopping due to the keratopathy, all of which we hope can maintain this profound efficacy and potentially reduce the ocular tox that we've seen with this keratopathy. We have seen from the Trudel data that a small reduction from 2.5 mg to 1.9 mg per kg resulted in a reduction of ocular toxicity from about 70% down to 25%. So we're encouraged that dose reductions that, frankly, in terms of drug development, I would call, relatively small reductions, can have a pretty big effect. That all is important because one of the most exciting biologic studies done preclinically was the addition of a gamma secretase inhibitor, which is the -- something being studied with [ Belamaf in ]. And the gamma secretase inhibitor basically prevents the clippage -- the cleavage of the BCMA receptor off the plasma cell. And that's important for 2 reasons. First of all, the target no longer would be present on the plasma itself, it got clipped off, making it less effective. But more importantly, for an ADC, the soluble receptor now is able to bind the ADC. And instead of the tansinoid getting in the cell and causing cell toxicity, it's now available systemically. And that's, we think, an important problem we're trying to solve. So by inhibiting the cleavage of BCMA off the plasma cell, we think we can potentially reduce the dose and maintain efficacy. So that's the data that we'll be seeing from the DREAMM-4, DREAMM-5 looking at combinations. We're also looking at whether potentially pembrolizumab, which has -- all the PD-1s actually have been universally unsuccessful in myeloma. We're wondering, it's a bit speculative, but it's possible that inhibiting PD-1 in the setting of a drug that induces immunogenic cell death will bring back the potential for dostarlimab or other PD-1s to actually induce a response that has never been seen in myeloma. So we're excited about those 2 combinations, particularly the GSI, to be able to have greater efficacy or equivalent efficacy at a lower dose.

Great. Let's move on to another exciting agent on the market and lots of work in development, and that is Zejula. What key indications are there beyond ovarian cancer? And what is the time frame to get visibility on them?

Well, I think, Zejula is a very special drug. We've seen from PRIMA study, which is where Zejula was given in the frontline setting maintenance, which is a very big opportunity for us to help women with ovarian cancer because of very many people, even still today, which has somewhat sad, still are what's called watch and wait. They don't receive a PARP. They are in the maintenance setting despite what is clearly a very significant benefit achieved. And we think that this benefit in the ovarian cancer maintenance setting is driven by a number of things. First of all, the study design for PRIMA was to not just focus on BRCA patients, which represent about 15% of ovarian cancer, but to focus on what's called HRD patients, those who have a homologous recombination defect, a genetic defect that's analogous to BRCA. What we saw in this study was that the Zejula benefited the patients with HRD to the same degree, to be honest, in both in relative terms and absolute terms as was benefiting the BRCA women with BRCA mutations in their tumor. And interestingly, and I'll get to this in a second, we saw benefit in even the wild types. The all-comers population was pretty significant as well. And we believe that, that's in part driven by the fact that Zejula is unique. We don't completely understand why this is the case, but we observed it preclinically, and then we actually saw clinically where it accumulates in the tumor. And then actually, if you think about it, it could have a pretty significant effect. So when we look at the plasma concentration relative to its tumor concentrations, preclinically and clinically, in 2 different populations so far, relative to other PARP inhibitors, we see an exaggerated accumulation, whether this represents -- well, I won't get into why. But for whatever reasons, accumulating, and we think this could be why it's working so well in the HRD negative. We're not sure. It's hard to know. But it's the only PARP inhibitor with the broad label for all comers. We were waiting for the results of that to see if this tissue accumulation would result in the all-comers label in effect. And once we saw that, we decided to move to lung cancer in all comers. One could argue to tailor the therapy to just patients with HR defects. We decided to do all comers, and we're doing both squamous and non-squamous in a trial called [ ZIA-1 ] which I think should be sometime in 2024, ideally the first half, but we'll see how enrollment goes in the event rates. So that will be an important trial. We think that if it does work there, whether that's a class effect or specific to Zejula, tissue accumulation will have to be seen. There's other PARP inhibitors studying these -- those patients. What we do think is also unique and it doesn't actually manifest itself as being important in ovarian cancer, but it does seem to cross the [ blood-bred ] barrier preclinically, and we think that will translate clinically. And unlike ovarian cancer patients, lung cancer patients do have brain metastases. And so we are cautiously optimistic that, that will be another differentiating feature should it work in lung cancer relative to other parts. So we're very excited about that. And we have also enthusiasm about the potential move into breast cancer in a very novel study that we're in the midst of discussions with the regulators. So we see this as a terrific, terrific opportunity for being best-in-class in ovarian, and hopefully, the [ ZIA-1 ] data in lung will materialize in a way that will help a lot of lung cancer patients as well.

Great. That's helpful. Maybe we can move on to another exciting program within the GSK pipeline, and that is the RSV program. So maybe, again, you can update us on the program. And perhaps you can tell us what role you think vaccines will have versus antibodies in the future?

Well, the RSV OA program, the older adults is very important. I think it's -- people have spent a lot of time, pediatricians in particular, thinking about RSV in the maternal and pediatric setting, where it's a very important cause of neonatal and early life pneumonitis and pneumonia. Both passive and immunotherapy have been tried, and active immunotherapy have been tried. What's probably less appreciated, and we think there's a huge opportunity to help patients, is for vaccinating the older adults. And patients -- just in the United States alone, in patients over 65, there's upwards of 170,000 hospitalizations for infection. And so this could be an enormous opportunity to help them. We think that our vaccine is differentiated in a lot of ways. The first one is that unlike the vaccines that have been attempted to be developed over the last, I don't know, probably 50 years actually, is that we've finally figured out, we, the world, that the RSV antigen needs to be stabilized in what's called the prefusion state. And that -- in that state, it induces the right neutralizing antibodies. So we're pretty confident from the Phase II data that, that was true and that the induced T cell and B cell responses are appropriate for minimizing the effect. We also think that because this is an older adult population where vaccinations typically are less immunogenic in the elderly, that our adjuvant will be very, very important in differentiating this with the treatment effect that we hope is quite substantial. And we're using the same adjuvant as we use for RSV, where we've obviously seen the benefit of that adjuvant in terms of inducing a more robust effect, particularly in the elderly. So we're very excited about that program. Obviously, for paternal as well as pediatric, but this opportunity to prevent hospitalization and potentially even death, which is not uncommon in these patients who are hospitalized, could be a huge opportunity for GSK and certainly for patients.

Okay. That's great. Maybe we can move on to cabotegravir in PrEP. So what do you think the key points will be that regulators focus on as they go through this application?

Well, we don't tend to talk too much about the regulatory interactions, but I think the -- obviously, on the highest level, safety and efficacy are the main concerns of any regulator when they're reviewing data, and I think we have an enormous amount of data on the safety of cabotegravir. In the pre-exposure prophylaxis study, as you know, HIV patients, men, in the 083 study, it was stopped early for overwhelming superiority. And you're seeing treatment effects that, frankly, are pretty incredible. Particularly when you think about a trial going up against an active control, to have the 65%, 75% reduction that we're seeing is pretty impressive. Also, I think the unmet need, I mean there's more than 1.7 million new infections per year, and I think more and more people are appreciating the importance of PrEP as an effective option to help these patients. So I think it's the safety efficacy, and we're quite excited and confident that the significant benefit to patients will be appreciated by not only regulators but patients, in particular, and clinicians as well.

Great. Maybe we can move on to another unique dosing paradigm that GSK is exploring, and that's the IL-5 long-acting. Maybe you can similarly update us on this program. What is the target profile? What are the technical hurdles that have overcome -- been overcome? And what are those that you're still working on?

Yes. Well, this is a good example of -- if I step back of something that I think we are attempting to do a lot better than we have in the past at GSK, which is to do a robust life cycle management or what we call life cycle innovation. And when you think about Nucala moving from asthma to nasal polyps to eosinophilic granulomatosis to an auto injector, the COPD trial, we're quite confident that blockbusters, as the street calls them, are not always discovered, they're developed. To help the most number of patients, you really have to think about the life cycle. And the long-acting is a way of ensuring compliance really to ensure that we have the most convenient IL-5 inhibitor on the market. And we expect, as does the Street, that the number of patients using IL-5 inhibitors increases. Nucala is growing quite nicely, given all of those. And by playing with the antibody, increasing specificity, avidity, affinity and half-life, mostly, we think that we can administer this probably q6 months and obtain a treatment effect equivalent to or maybe even better depending on the results of Nucala. So we think this will be a best-in-class IL-5 inhibitor. And we've decided to take what we think is a very smart risk and that is skip Phase 2 and go directly from Phase I to the pivotal trials. So a bit of an aggressive call, but one that we think is warranted given the biology and the deep understanding that we have of the PK/PD relationship.

Okay. Let's move on to another thing you mentioned at the outset, and that is COVID. So maybe you could just kind of walk through the various ways that GSK is trying to provide solutions to COVID and where we stand now and how you see the future playing out?

Okay. Thank you. So yes, very challenging year and one where we have decided to sort of focus on a few different attempts at solutions for COVID. First, we really believe that one of the unique features of our vaccine business is that we have these adjuvants that make every vaccine more potent in different ways, whether it's allowing bigger treatment effect, as I mentioned, with RSV, and hopefully with -- sorry, with Shingrix, and hopefully, with RSV, or whether we use what we call more of a pandemic-like adjuvant that reduces the dose needed, which effectively turns out to treat twice as many people, if you can have the dose, 10x as many if you can 1/10th the dose. We think this is enormously important in pandemic situations where we're trying to vaccinate billions of people across the globe. And we had established multiple collaborations where we were providing what we think is a very differentiated and important adjuvant. We -- in addition -- and so those trials are ongoing with Sanofi and other partners. We also have a recent opportunity with CureVac that we -- to develop mRNA. We've seen clearly a big advance in mRNA and vaccinology with Moderna and BioNTech vaccines. And we believe that that's probably here to stay in terms of the impact mRNA is going to have in vaccines. And our internal SAM, our self-amplifying MRNA, our own internal mRNA efforts as well as those with CureVac, we think are going to allow us to be leaders in mRNA technology as it relates to vaccines. And we're working with CureVac on the multivalent vaccine that might be needed as these mutations evolve. On the treatment side, we took an interesting bet, and I'll take a few minutes to talk about it. We'll be unblinding a lot of this data over the next few weeks. So I don't want to get ahead of myself. I haven't seen anything yet, but we -- the company we're working with is VIR and they have a monoclonal antibody. We thought that monoclonal antibodies would be very effective. We weren't so convinced that the vaccines would be as 95% effective, particularly in the elderlies, what seems to be the case. And so we were developing monoclonals to both, be a bridge to the pre hospital setting for those patients who weren't melting a vaccination response or those who didn't get one. And we have a study called COMET-ICE that will unblind this quarter. Looking at administering monoclonals, this monoclonal 7831 to those patients as well as a combination study with the Lilly antibody. We chose VIR because we think the science here is the best of all the monoclonals. And that's because this antibody binds a very different epitope than all the other monoclonals, one that was identified from patients who had a robust monoclonal response to SARS-CoV-1, okay? And who is neutralizing antibodies to SARS-CoV-1 were neutralizing to COVID-19. And we thought that was a very important assay, if you will, because by having this assay that's [ orthotimal ], meaning it works really against SARS-CoV-1 and COVID-19, we believe that, that would be reflective of an epitope that doesn't change over time. And that's probably because there's some fitness disadvantage, some detriment to the virus for mutating at the spot. So far, that does appear to be the case when we look at all the mutations, the thousands and thousands of mutations that are being reported, they're not in the epitope that the very antibody binds. And so, so far, we're cautiously optimistic that we might have a unique profile against not only the dominant mutations that are emerging, but any ones that might subsequently occur. There may be a need to combine it, but we think that this is going to have unique activity. And in addition, it was modified to have a longer half life, to be more potent. So the dose is actually lower than most of the other monoclonals. We think this together could enable best-in-class monoclonal. There was one other feature we did to the antibody that we think could end up being very important, and that is we induce effector function. We talked about this earlier when I was talking about -- excuse me, Belamaf, and what we can do with effector function. We think it's -- from -- if you look at the endogenous antibodies that are made by humans against viruses, the ones that are most effective do have effector function. And most of the other monoclonals being developed don't have much effector function at all. And it was this scientific rationale that made the NIH decide to want to -- despite all the other monoclonals failing in the in-hospital setting, said it was at least a smart risk to study the VIR-7831 in the active program, the NIH active studies. So again, very high bar. The other ones didn't work. Who knows whether ours will, but that could be another differentiating feature should it pass the interim. So that's pretty exciting. And then we lastly decided that we should investigate anything that we could to help the in-hospital setting, which is where much of the morbidity and mortality occurs. Unfortunately, and we have been observing this field evolve and it's becoming very clear that patients admitted to the hospital sort of fall into a bunch of different categories that reflect some other biologic process, whether it be procoagulant where these patients end up succumbing sometimes to thrombotic complications, whether they be strokes or AMIs or PEs and things like that or more of a cytokine storm. So we initiated the otilimab, the anti GM-CSF, because we believe that the patients were suffering from an activated monocyte-like syndrome that really resemble what we've seen in the past from patients where that -- and this is one of the most potent inhibitors of monocyte activation. It inhibits many, many different cytokines rather than trying to block IL-1 or IL-6. So any given one, we thought this would be very potent at blocking one of them. What we saw in the study was, again, it missed its primary, but there was a strong trend. But we did see a pretty profound, actually, effect in the over 70. And if that's real, I think we should be able to see that in the extension study where we're repeating that cohort. So multiple different shots on goal to help these patients.

Great. We probably only have time for one more question, and let me put that question to daprodustat. What advantages do you see in this agent over its competitor from AstraZeneca?

Well, it's hard to know because I'm not sure, exactly. All the data has been a little bit hard to decipher from all the other trials. And we'll be learning about the label for roxa soon, so we can see what exactly the data is. And frankly, we don't have daprodustat data. So seeing how it's differentiated, it's going to have to wait to see data. I think what I can say is that our global clinical trial program has always been designed to be very, very robust. It's well powered for most treatment effects. It's a single sponsor, it's an active-controlled clinical program. It doesn't rely on [ med analysis ], et cetera. So it's very robust. Now that may or may not end up being particularly important. We think it will be, but we'll have to see the data. And those trials, we should see them in the second half of this year to see if it reduces cardiovascular events to the same degree as roxa and maybe more, maybe less, we'll have to see. But we're cautiously optimistic that this will have a very favorable risk-benefit profile relative to options patients have today and that payers will see this as an exciting alternative to EPO, and for the non-dialysis to be a very important thing.

Great. Well, sadly, we're out of time. So I'd like to thank you, Hal, for a great rundown on so many exciting assets for which much data is coming. So we'll look forward to the progress. And of course, an update at your major June meeting as well. So thank you for your time today, and thank you for everyone who participated in this session.

Thank you, Steve.

Bye-bye.