Prothena Corporation plc (PRTA) Earnings Call Transcript & Summary

And welcome to the Bank of America Healthcare Conference. We are continuing our sessions with our next presenting company, Prothena. It's my pleasure to have sitting next to me Gene Kinney as well as [ Suzanne Lentzsch. ] I'll let you guys introduce yourselves and for your titles and then maybe Gene after that, you can give us a 2-minute overview of the company, and we can go into Q&A on some topics.

Sounds great. Thank you. And [ Suzanne, ] you want to start?

Yes, sure. Thank you, Tazeen for having us. [ Suzanne Lentzsch, ] Senior Vice President of Strategy and Operations.

Hi, everyone, I'm Gene Kinney, CEO of Prothena. And thank you very much for having us today. It's always great to be here.

Thank you for coming.

Looking forward to the discussion. So just a little bit about Prothena as you asked for those that don't know the company. What we focus on are diseases caused by dysregulated proteins or misfolded proteins. So obviously, our bodies make protein every day. The protein needs to fold in a certain pre-dimensional shape in order to do whatever it is that protein is meant to do. But in a number of diseases when that protein folding goes a rye and we start to see dysregulation of the protein, it can aggregate sometimes on itself, and we see deposition in critical organs of the body that they can either be in the brain or in the periphery, and that leads to disease. And so what we think a lot about at Prothena is how to understand how those proteins are causing dysregulation at the organ level, and ultimately have intervene in that process. It's something that we spend a lot of time thinking about. Our portfolio reflects that. We've got a number of therapeutics in the peripheral amyloid space. So these would be diseases caused by proteins that ultimately deposits in peripheral organs such as the heart, leading to cardiac dysfunction and cardiac failure that would describe our AL amyloidosis in ATTR amyloidosis programs. And then, of course, in the central nervous system, when we see deposition there and dysregulation there, we see relatively large disease spaces like Alzheimer's disease, Parkinson's disease, ALS and the like. We have a number of therapeutic approaches through a portfolio that address those areas as well.

Yes. So let's go into one of those programs, perhaps the one that's most mature birtamimab. So when I go way back, we were involved in taking Prothena public. And at the time, it was a story about AL amyloidosis. So for those who aren't familiar with the history of the molecule, can you just maybe walk us through what's happened and where you are now?

Absolutely. Yes. Thank you for that. So first AL amyloidosis is a disease caused by dysregulated protein. In this case, the protein that we're talking about is light chain. So light chain made by plasma cells, plasma cells normal day job is to make every chain light chain, which combine to make immunoglobulin. In this disease, what we see is an overproduction of light chain that does not combine with heavy chain appropriately, it starts to misfold, land on itself, and then we see deposition in kidney, heart and what have you. Cardiac involvement is definitely the most problematic for patients, it leads to -- this deposition leads to a progressive restrictive cardiomyopathy that can lead to death in patients. And depending on where those patients are diagnosed, the prognosis from a survival perspective can be quite different. So depending on how early we get it in the disease, we see these patients. So to your question, what we've done is something a little bit different. Most of the treatments in AL amyloidosis target the production of the protein. So they go after the plasma cells themselves. And it's obvious why that's the case is because we have drugs that target plasma cells in the multiple myeloma space. So they can be repurposed into AL amyloidosis to try and go after those plasma cells, reduce the amount of protein that's being produced. And then let's hope there's a benefit in terms of ultimately the patients can function better if we kind of stop the production. The limitation of that approach has always been that the resident protein that's already deposited is what's causing problems today and what's leading to early mortality, in particular for patients that are more advanced in the disease. So what we've done with birtamimab is developed a monoclonal antibody that targets specifically the light chain when it becomes dysregulated. So it binds to a very well-defined epitope that's only exposed on misfolding event. And it furthermore seize both [ Catha and land ] forms of light chain. So it's broad and specific with respect to how we interact with the protein. And what it's designed to do is actually remove the resident amyloid, which is very different than everything else out there. So we call it a depleter mechanism of action. That molecule first entered clinical trials in 2013, we ran a pretty robust Phase I/II program with that molecule, developed a lot of information with respect to cardiac response, renal response, and then actually initiated a Phase III study, which as a primary outcome measure had a component of survival and hospitalization and then initiated a Phase II study that used a biomarker readout as well. What happened, and this was in 2018, was when that biomarker study came in, the results weren't quite as we expected. So that caused us to kind of reevaluate the risk-benefit calculus from a patient perspective. We ask the DSMB to go do a futility analysis on the ongoing Phase III study. And although there was a benefit in favor of birtamimab, it was insufficient with respect to size of benefit that we would like to be successful. So we discontinued that trial. When we actually went and looked at the data, fortunately, we had actually pre stratified based on level of severity. So there's a classification system known as the Mayo Staging system. Mayo Stage I through Mayo Stage II, III, and IV. Mayo Stage IV being patients at highest risk for early mortality. So based on literature, you would expect those patients to survive somewhere around 6 months following diagnosis. And what we found in our data set was really the only meaningful amount of mortality, so the amount or number of events that we saw was in this Mayo Stage IV group that we had fortunately stratified for. And in that group, we had a hazard ratio on all-cause mortality at 0.413 across the first 9 months. That 9-month period was also where we looked at key secondary endpoints, including 6 Minute Walk Test, a measure function and Short Form-36, which is a measure of how patients are feeling from a quality-of-life perspective, nominally significant effects there as well. So in this group, a non-subtle survival benefit in really the only subgroup for which we had a meaningful mortality event number. And so with those data in hand, we obviously did a lot of analysis of that to make sure that we could convince ourselves that, that was a drug effect and unlikely to be explained by some baseline variable, went to the FDA, talked with them about both our safety database as well as this prior observed effect now and ended up with a special protocol assessment agreement through the FDA, which allows for our ongoing study, which is called the AFFIRM-AL study. This is a confirmatory Phase III study where importantly, the alpha for success is 0.10. So typically, folks will be familiar with the 0.05 alpha in terms of its powering and designing studies here. We were able to agree with the FDA on a 0.10 for success. So that allowed us to do a number of, I think, innovative things in the clinical design of AFFIRM-AL. But for the most part, at the highest level is really just confirmatory. It's focused on the Mayo Stage IV patients, and we are using all-cause mortality as the primary outcome measure.

So how many patients would that be that would qualify to be in the AFFIRM-AL study just overall estimate? And I guess, how has enrollments been proceeding taking into account everything from Omicron to just -- it's a rare disease. Yes.

So. Yes. So in the population, the Mayo Stage IV patients are about 30% of the AL amyloidosis in population. Globally, that puts you around 60,000 to 120,000 patients in total from a prevalence perspective. So that's the patient population that we're looking at today. Interestingly, in our Phase III study, it was about 30% of our total population flow. We're looking to enroll 150 patients in this study. It's a 2:1 randomization. So we lower the probability that any given patient would be randomized to placebo. And we have introduced an interim analysis as well with some de minimis alpha spend that we can talk about. So to answer your question, what we're currently guiding to is top line data in 2024. You don't typically give enrollment guides along the way. But I think what we'd say is, as long as we're holding to that timeline, we expect that enrollment is going as expected.

No, I would bring it up because some of my other companies, not necessarily in rare disease, some of them are, but I would say uniformly have said that Omicron has had an effect. And so just in general, inquiring whether you would agree with that?

Certainly, across the board. If we look at clinical studies, writ-large. I think that the COVID pandemic has impacted operations at the clinical level in terms of enrollment timelines and what have you. The -- I guess I hate to use the word, fortunate, but the fortunate thing about this disease in the context of clinical trial execution is that particularly when you're talking about Mayo Stage IV patients, when they come to diagnosis, it's basically a medical emergency. And those patients need to be treated relatively quickly if you have any chance of actually prolonging life span of those patients. So for this patient population, in particular, it's a relatively severe disease. It is a medical emergency. The patients once diagnosed, need to be treated. So that part, I think we feel very good about. I think the other part of the question is awareness of patients putting off care, they're putting off diagnosis. We know that this is a disease that's probably under diagnosed. If you look at the amyloidosis research consortium work. We know that patients go through a number of specialty physicians before they get an accurate diagnosis. So to the extent that they may be delaying that, it'd be very hard to quantify, but it is possible, certainly.

So what would prevent the patient from entering the study? I mean, what's available to treat them now?

Yes, it's a great question. So everything available to treat these patients today works on the plasma cell discretion. So if you think about what's happening here versus multiple myeloma, multiple myeloma is about a clonal expansion of a clonal set of plasma cells. Whereas here, the clone size doesn't matter as much. What matters is that, that clonal plasma cells is producing this protein. But nonetheless, you can envision treatment in very much the same way. So the hematologists would look at this and they say, well, we have approaches to deal with plasma cell production of protein. These would be things like prednisone inhibitors, IMiDs, anti-CD38 like daratumumab. Daratumumab is the first treatment that was approved on an accelerated basis to treat AL amyloidosis, specifically on label. Now I will say it's not indicated for more advanced patients. In fact, I think in the label, not recommended for more advanced patients, but it is relatively broadly used out there. From our perspective, if you look at the data around survival in these most advanced patients that we're talking about, we don't really see an impact of those types of treatments on these advanced patients. Said another way, you really need to survive long enough to get the benefits of turning off the spigot, if you will, when the bathtub is already full of water. And so we look at -- I think that the daratumumab data has been updated now a number of times. We have 6 months data out there. I think they're up to 2 years now. And at least to my knowledge, they haven't yet reported a survival benefit, which would infer that you may need to treat even longer than that and survive longer than that to get that kind of benefit. So I think really where we're headed, we believe that there's a unique profile of this depleter mechanism of action for patients that are relatively advanced have a high level of resident protein. As we start moving in the future into more mild patient populations, I think combination approaches turn off the faucet and drain the bathtub probably makes a ton of sense.

Okay. So if we would have fast forward to 2024, what would be good data?

Good data would be statistically significant effect on all-cause mortality. When we look at the Special Protocol Assessment agreement, that's defined as a p-value of 0.10. What's -- we are also, from a secondary perspective, looking at 6-Minute Walk Test in Short Form-36, which is a quality-of-life outcome. Those latter secondary endpoints are not considered in this pot agreements. So what we need to hit on is the all-cause mortality endpoint.

So 6-Minute Walk is being scrutinized pretty heavily now for ATTR. So Gene, you do have a separate molecule that you've partnered off for ATTR, but as you think about comparisons and contrast for AL amyloidosis versus, let's say, ATTR cardiomyopathy. What are some of the bigger differences which make it either harder or easier for a trial to be successful?

Yes. It's a great question. And the diseases are very similar from a biology and concept perspective. So they're both diseases caused by deposition of protein. The proteins are different, and they're made in different places and where they deposit and how they cause dysfunction, may be slightly different. In the case of -- so we talked about AL amyloidosis. In the case of ATTR amyloidosis, the offending protein, if you will, here is transthyretin, which -- transthyretin is a relatively abundant protein for healthy individuals. It's not toxic per se. But when it goes arise, then we start to see the problem. So when transthyretin is made, it's made in several places in the body, the liver gets the most attention, but also made the core reflexes and other places. It quickly assembles into a tetrameric structure, the 4 units of transthyretin. And that's its normal form and it underlines it's a normal function, which is to transport thyroxine retinal binding protein and other proteins. When that tetramer falls apart, it should be relatively rapidly cleared. That's kind of the normal process. If that doesn't happen efficiently, those constituent proteins can then become dysregulated and start to recombine in a structure that can be toxic. And so again, you can get that deposition in heart. You can get that deposition in other organ peripheral nerve as a focus of ATTR, and you can see dysfunction there. I think as we think about that space, we think about it very similarly. Everything today and the 2 big classes of treatments are the silencers, so these are your siRNA-based approaches, and the stabilizes, which basically try to stabilize that normal tetrameric structure, but the goal, at least at the highest level is relatively similar. Let's prevent new protein from coming into this pathological pathway. So in many ways, there's a lot of similarity between the way you treat ATTR and the way you think about treating AL. What's left in that space is very similar to what we just talked about with AL, which is a depleter approach. If your more advanced patients, actually, the problem that they have is the resident protein, then how do we go after the resident protein? In PRX004, the molecule you referenced is a molecule that has that depleter mechanism action designed to go after the resident protein and remove it. That's what the intent of PRX004 is. So why should we care about that? And I think one just needs to look at the available data. So tafamidis, of course, has a ratio of around 0.7 on mortality, overall survival. But if you actually look at the data that they've published, the majority of that effect is really driven by New York Heart Association Class 1 and 2 patients. If you look at New York Heart Association Class III patients, the p-value is 0.78. And if you go look at the survival curves there, there's not much separation until about 18 months, and that's when you start to see the separation. So we would pause at a very similar philosophical approach that if you are at risk of early mortality or morbidity, than getting in and actually dealing with the resident protein is going to be very important, and that's how we see PRX004. You had asked specifically about 6-Minute Walk Test and the 6-Minute Walk Test is obviously an important measure of function, particularly with cardiac involvement. In some diseases, there are a lot of other systems that are going to play into that peripheral neuropathy being one. And if I can't feel my toes, it may be harder to walk in a very purposeful way. Edema is another one. So with cardiac involvement, sometimes you have meaningful edema, that can be problematic for patients. So one needs to take those things into account, be very careful about how one actually runs the 6-Minute Walk Test. Ourselves, we tend to use specific groups that specialize in this to help us with the 6-Minute Walk Test. So to set those up at sites as well as to kind of go back and take a look at how things are going on a regular basis. But it can be challenging as we saw with some recent readouts where maybe the placebo effects weren't what we thought they were. And I think the question that remains at least in my mind, is what that was owing to, was that something to do with the change in the demographics of the patient population? Was that to do with the way that the 6-Minute Walk Test was conducted? And I don't know that we have a strong answer on that yet. So for us, at least with the AFFIRM-AL trial, an AL amyloidosis, it's important that 6-Minute Walk Test will be conducted very similar to how we did it in our prior study where we did see a normally significant effect. It's also important that, that's not the primary outcome, nor is it required for success on the pot agreement.

What would it give you if you were to have positive data?

Well, I think it's always good to see the function kind of track with survival benefit. We want to know that these patients are doing well that remain alive. We don't -- we want to make sure that we can have some assessment of quality of life and of function. And it's important to kind of assess that as you go. And I think it's just important to understand how these things are tracking. So if you think it -- kind of the 3 simple criteria from a regulatory perspective, it's survival function and feel. And so it's always important to have at least some component of each of those in your various clinical trials.

So how confident are you that the impact on mortality, the 2024 readout, you will have had enough time to show that benefit?

Yes. So we've looked pretty extensively in AL amyloidosis at survival, particularly with Mayo Stage IV patients. We know from a literature evaluation of median survival somewhere around 6 months. In our study, on best standard of care, we had a median survival with [indiscernible] on the control arm, we had a median survival, 8.3 months. So it did a little better coming to physicians on a monthly basis and actually seeing expert physicians on the regular. That said, we don't see that there's really much that's changed from a treatment perspective, that's really changed that all that much. And if you go do some more recent reviews of the data, it would suggest that these patients still have a pretty poor prognosis. So from our perspective, we've powered this study, and in fact, one could argue, overpowered at the p-value of 0.10 with 150 patients, we expect that we could actually be powered to see a p-value that's actually a little bit more stringent than...

Yes. Okay. So if that data set is positive, then what would be the next step?

Next step with that program would be commercialization. So we -- that's a wholly owned program within Prothena. Part of the reason it's wholly owned is because we think it's a very addressable market from the commercial perspective. We see about 75% of these patients show up around, somewhere around 500 specialty centers worldwide. Clearly, these are, in many cases, specialty centers where we're conducting clinical trials. So we know these sites. We're one of the few companies that have done randomized-controlled trials in this patient population. As far as I know, we're the only company that has shown a survival benefit in Mayo Stage IV patients in a randomized control setting. And so I think from our perspective, we feel that the most expeditious way to get this therapy to patients is for us to actually commercialize it ourselves. So that's our plan with top line data, we expect in 2024, that gives us an ability to start thinking about the transition from R&D to RDNC on a fully integrated basis in that time frame.

So would it be reasonable to assume if your data is positive in '24, you could be commercial with that product in '25?

I think we would move as quickly as possible. Obviously, the timing from -- in terms of when we see that data will matter, but we would expect to be moving very quickly.

Yes, yes. Okay. So maybe let's move on to some of your other areas of focus. A very easy drug area is, of course, Alzheimer's, just kidding. So you've taken on a challenge. You're trying a few different approaches there. So can you just talk us to why Alzheimer's?

Yes. Well, I mean, first, let's start with the need. I mean you've got the 6 leading cause of death. It's a fatal disease. You've got 6.2 million people in the U.S. with this disease up to 50 million people worldwide. And the only approved drug for slowing progression of disease right now is [indiscernible] on an accelerated basis. And obviously, as we know, that uptake has been, let's call it, slow. So I think from our perspective, the need here is enormous. And what we've seen having been in this space for some time. So folks may not know, we spun out of AlainPharmaceuticals back in 2012, which was really the genesis of targeting proteins in the context of Alzheimer's disease. So the AN-1792 vaccine trials, bapineuzumab, the initial collaboration with Eli Lilly. So I think there was a lot of really early work that was done there that really helped to define the biology. What I've actually seen over the last, let's call it, 10 to 15 years has been an evolution on the clinical side. How do we actually select patients that are most likely to respond to these types of treatments? How do we think about clinical end points that are a little bit more sensitive to that patient population? How do we think about the use of biomarkers? How do we think about the side effect profile, ARIA, something that people talk about a lot with anti-amyloid treatments? And what does that mean? What is the biology of that? Can we dose through that? Can we get to higher concentrations. Those learnings have actually iterated to a place where we are today, which I would argue is a pretty exciting time. So there's a recent review written by [ Eric Karen and Barcus Stroer, ] which actually showed that if you use PET imaging, PET amyloid imaging and you actually look at meaningful reductions and they define that on a scale called synuclein, they find as a 20-synuclein achievement of 20-synuclein signal on PET imaging in the last 5 trials that have actually seen meaningful PET reduction of amyloid. 4 of those have been positive on prespecified clinical outcomes. So these are functional clinical outcome measures, different clinical outcome measures through different programs. But as you kind of start to look at the totality of data, the group level correlations between reduction of amyloid and positive clinical outcomes, we actually think the science is at a place that may be a little disconnected from some of the sentiment. And obviously, we think there's enormous opportunity when those 2 things are a little bit disconnected. So what we've done in our portfolio is the following. We've got kind of a 3-pronged approach at this. The first is to think about anti-amyloid approaches. First-generation anti-amyloid approaches, we believe are going to be somewhat restricted to IV administration, will be important in terms of transitioning from symptomatic to disease-modifying approaches but may not be ideal from a patient experience perspective. So our PRX012 molecule, which is wholly owned and began clinical trials this year, was designed with that in mind. So it's at least tenfold more potent than aducanumab. It's designed for subcutaneous administration. In fact, we started our clinical trial subcutaneously and on a convenience schedule. So no special pumps or techniques or other approaches. This is straight subcu formulation once a month and off you go. So we're looking at that profile now. We think the combination of higher potency and more convenient administration, which one enables the other could also lead to a place where we can explore what that means from an efficacy and safety perspective.

So what about amyloid though? It just feels like so many different people have tried looking at it in so many different ways, and what's not been properly done?

Yes. So I think amyloid is a big term, right? So it's -- we have to be specific there. And I think when you talk about amyloid and what the current therapies that are coming up the Phase III trials are trying to do is really in my mind in testing the clinical hypothesis of what happens when you address both the soluble oligomeric species and the insoluble plaque. And the current set of treatments and so specifically donanemab, lecanemab, aducanumab and of course, PRX012, what affords then the ability to test that thesis is the targeting of [indiscernible], which remains relatively unstructured through the different misfolded confirmation. So if you target a different part of a beta or amyloid, you're going to get a different biological profile, which leads to different clinical hypothesis that you're testing. So first, let's be specific with respect to what we mean by amyloid because it ultimately informs what it is that we're testing. And I think what's nice about the current sets of treatments. When you talk about aducanumab, lecanemab, donanemab is not only we're seeing that if you reduce amyloid meaningfully, you're seeing positive impact as per this paper I just referenced. But also, the effect is very consistent across trials. So as we look at the effect over 18 months on CDR Sum of Boxes, which is an outcome measure, you're seeing effects that range from 22% to 26% slowing over 18 months across those molecules. If we switch to maybe an endpoint that has better dynamic sensitivity, something called the iADRS endpoint. You're talking about 32% to 34% change over 18 months across those molecules. I think that level of consistency is interesting as well when you're targeting amyloid in the same way. But it's not just that. If you go all the way back in the history here, when we first got these PET imaging agents and you can start deploying them in the clinical space, both Eli Lilly and the time Janssen and Pfizer, who are developing molecules in this space, ran substudies and in both cases, somewhere around 30% of patients in those substudies were not positive with respect to amyloid in their brain. So I think now we're pre-screening patients with amyloid, we have better use of biomarkers. And then the final point I'd make is in terms of dosing back in the early days when bapineuzumab started clinical trial. We didn't really know what ARIA was. I think we have much better biological understanding, much better clinical understanding of this presentation. And you're seeing now companies figuring out how to dose through or around ARIA in order to maintain exposure of these anti-amyloid agents. And now you're seeing more consistent results. And then the final point, which I know I said the last one as the final point. But the final point is the outcome measures, which back in the early days, I think we were using outcome measures that have been primarily developed to evaluate symptomatic drugs. And now we're deploying single primary outcome measures like CDR Sum of Boxes and iADRS, which are a little bit more sensitive to these patient populations.

Okay. So a similar question to what I asked about AL amyloidosis. What would be good data here because there's really nothing that works very well. And are you surprised that [indiscernible] has not had a good pick up?

I think what we're going to see in terms of data is we're going to get the lecanemab readout from Eisai here this year -- in the latter part of this year. We'll see again [indiscernible] readout for Roche and the next year, donanemab readout as well. These are all well powered Phase III studies. In fact, Eisai just yesterday announced that they have completed their submission for accelerated approval and have agreement with FDA that their upcoming lecanemab Phase III readout could be sufficient as positive to be confirmatory and allow for full approval submission effect this year. So I think we're going to see a lot more in this space, and we're going to learn a lot more around endpoints, around how these trials were conducted, the relative powering effect size and those sorts of things. So a lot of data to come from that perspective. What we have seen is an increased level of evidence and body of evidence that suggests that reduction of amyloid, we can measure that by PET in the brain of individuals with Alzheimer's disease correlates with positive clinical outcomes. And we certainly see that at a group level, at least Eisai in their press release yesterday suggested that they're seeing that at the individual level as well. So that's actually very encouraging. And I think what that allows us to think about as we start moving past the multiple dose study is looking at early time points at level of amyloid reduction by PET. And with this additional evidence coming and the data that we already know to be true, I think that gives us a good sense of where we are on that scale. We expect we could start seeing that sort of data as early as next year.

So would this be all [indiscernible] lifted situation, if you have one good data readout from one company would that increase? Do you think the confidence people have overall in Alzheimer's?

Well, I think the way -- again, I have to go back to science and sentiment, right? It's harder for me to predict sentiment. But I think from a science perspective, I think the field -- my perspective, at least, has already started to develop a data set that's becoming much more consistent in this space. There's still a few things to work out, right? So we're seeing CDR Sum of Boxes using 1 or 2 of the upcoming studies and iADRS is a primary and the other. So we're still trying to understand how to deploy those endpoints. The beauty for Prothena is that's all in front of us, right? So we've got Phase 1 risk right now, but we get to learn from all of that as we move into our Phase II or even Phase II/III studies depending on how that future unfolds.

Yes. And I think we're going to run out of time. [indiscernible] but what really would most excite you from these upcoming data readouts and how would that influence what you would look at, let's say, in the Phase II?

Yes. So there's 2 questions in there. One is what would excite me and somebody has been studying Alzheimer's disease for over 25 years, what would excite me is that a solid shift from symptomatic treatments to disease modifying treatments. I think we are very, very close from a science perspective, from a clinical science perspective, we're putting the bedding edges on. And I think we're going to learn a lot from these upcoming readouts. And I think for the up to 50 million patients for Alzheimer's disease that's an incredible thing to say that paradigm shift from a treatment perspective, it's absolutely enormous. I think from what is Prothena going to learn, we're going to learn everything. I mean, so if we had said CDR Sum of Boxes was the endpoint, the only end point, those are relatively large studies, you can see with the [indiscernible] study with almost 1,800 patients in them for a company with a wholly owned asset like ours, that's a meaningful layout of capital. On the other hand, when you look at iADRS the Phase II Eli Lilly study with better dynamic sensitivity with a better control of error, we saw a study with around 300 patients that got to statistical significance. So we think that changes the game significantly from an endpoint perspective. We're very eager to see how those endpoints actually perform in these upcoming clinical trials. And obviously, we're going to learn from that. I'll take this -- I know we're out of time, but patient selection is the other, slightly different approaches in terms of patient selection, how they're enriching their patient populations. We're going to be looking to that as well and making sure we incorporate best practices as we go forward.

Okay. Well, we'll be following all those steps, and we'll be in touch to get your thoughts on what's been happening. So Gene and Suzanne, thank you very much for coming to this conference. And thanks for participating. Thanks, everyone.

Thanks, everybody.