Roche Holding AG (ROG) Earnings Call Transcript

[Audio Gap] [Operator Instructions] One last remark, if you would like to follow the presented slides on as well, please feel free to go to roche.com/investors to download the presentation. At this time, it's my pleasure to introduce you to Bruno Eschli, Head of Investor Relations. Bruno, the stage is yours.

Thanks a lot, [ Henrik. ] And could I have the first slide, please? So welcome to our fifth IR call here in 2023 and this time focusing on the Roche Neuroscience franchise. And especially, we want to focus today on the latest data presented during October at recent conferences, ECTRIMS and CTAD. So let me quickly take you through today's agenda. We have today 3 speakers with us, 2 internal ones and 1 external. The first one is Paulo Fontoura, our Global Head of Neuroscience, Immunology, Ophthalmology, Infectors and Rare Diseases for the Clinical Development. Paulo will provide us an update on the overall Neuroscience franchise strategy. And he also will especially cover the molecules, which we will discuss in more detail today. The second part then is by Dr. Stephen Hauser, Director of the UCSF Weill Institute for Neuroscience and the Professor of Neurology at the University of California in San Francisco. He will lead us through 2 important trial results in multiple sclerosis, which were recently presented at ECTRIMS beginning of October. His data comprise the randomized Phase III study data of OCARINA II for OCREVUS 6-month subcutaneous formulation. And secondly, the first Phase II data for Fenebrutinib, FENopta data, which -- on efficacy and safety. And the third part then will go to Luka Kulic, our Therapeutic Area Leader for Dementias and Medical Director of Neuroscience and Rare Diseases. Luka will provide an update on some early data, which we are -- in Alzheimer's disease, which were presented at CTAD. So the Phase I/II dose finding data, efficacy and safety for trontinemab, our novel Brainshuttle anti-amyloid beta antibody. And secondly, Phase I data for a highly differentiated and potentially first-in-class gamma-secretase modulator also in Alzheimer's disease. Overall, the event today will go scheduled for 75 minutes. We have about 50 minutes planned for the presentations and then 25 minutes left for the Q&A session. All speakers will be available to take your questions. Could I have the next slide, please? So before we start with the presentations, I just wanted to use this occasion to quickly provide an overview of how these molecules fit into the broader picture for our portfolio. This was a slide here, which you see here, which we have presented for the first time at Pharma Day. And what you can see here is that we have divided our portfolio in 3 buckets to outline a bit future growth opportunities. On the left side, you see here our current portfolio, 11 of the 16 blockbusters, which we currently have on the market. And these 11 molecules are molecules which belongs to what we define as the Young portfolio, which will deliver growth going forward for a long time and some of them having additional potential due to line extensions. Then here highlighted in the middle bucket, you see the midterm opportunities. So these are all the NMEs, or major line extensions, which could be filed until the end of 2026. Since we first presented this slide, we have added here a significant opportunity with an additional molecule. We have acquired the leading anti-TL1A antibody, which is which is to be developed in IBD, but might even have significant potential beyond that in many other autoimmune diseases. And from the profile, you can see that this molecule is a bit similar to the previous molecule which we acquired zilebesiran. It's a molecule which fulfills again the criteria we have been looking for. So highly differentiated on the science and the novel drug target. Excellent, strong early data, Phase III ready with the potential to redefine the standard of care eventually here even in several diseases. And on the right side, then you basically see the long-term opportunities with filings projected after 2026. Can I have the next slide, please? So on this slide, I just wanted to show where are the molecules plotted which we talk about today. And these are 4 molecules. You see here in the bucket on the left. We have the OCREVUS data, so an important life cycle opportunity coming with the 6-month subcutaneous formulation. The OCARINA data in the middle bucket, you see the first Phase II data, the fenebrutinib efficacy and safety data for our potentially best-in-class BTK inhibitor. And then we have 2 molecules to talk about, which are in the -- on the right side here. The first one is trontinemab, our Brainshuttle anti-amyloid antibody, where we just showed a couple of days ago, the first data at CTAD. And then also, I think, very exciting the first data of Phase I data for our gamma-secretase modulator. And can you please go to the next slide? And this is just a slide also picked here from Pharma Day just to remind people that they can look it up. How do these molecules work out in current analyst estimates. So you can find a few of these molecules on this slide. The gamma-secretase molecule and it is a Phase I as it is too early. It's not yet picked up, but you see here what currently is in the models with regards to the opportunities we discussed today. And with that, I would actually hand over to Paulo for the update on the neuroscience strategy.

Thank you, Bruno, and hello, everyone. Can I have the next slide, please. So we've been working on the neuroscience pipeline now for over a decade. And what you see here is really the result of that. It's a very strongly differentiated portfolio, both with different targets and different platform technologies, all the way from small molecules and large molecules to gene therapies across all stages of clinical development with 4 launched products. Next slide, please. And as you can see here, at Roche in terms of portfolio value that neuroscience represents, this has really been a story of success in the past 5 years, going from nonexistent on the market now to having a substantial proportion of our sales represented there and being the #1 company in terms of sales for this area. This is really due to the strong growth in products like Ocrevus and Evrysdi, but also due to the emerging products like Enspryng. And you see here on the right-hand side that there are beyond these assets and a number of other assets as well. They have very strong sales potential as well. Let me just point out a couple of them that we're going to be talking about today, fenebrutinib and MS; of course, trontinemab and gamma-secretase modulator, but also new molecules like GYM329 or anti-latent myostatin antibody. Next slide, please. So just thinking about the strategy for the neuroscience pipeline, we're focused on 3 core strategic areas. This is multiple sclerosis, Alzheimer's disease and neuromuscular diseases. And for each 1 of them, our vision is really to have a transformational impact and bring real value for patients and for society. As you can see here in all of these 3 areas, we have a strong pipeline with assets in all phases of development. For MS, our strategy has been for the longest time really to build on the success of Ocrevus to stop preventing a reverse MS. For Alzheimer's, we think this is a momentous time in disease modification for this condition. And our vision is to halt and prevent Alzheimer's disease. And for neuromuscular disorder diseases are mainly rare genetic disorders with a high degree of visibility. It's really to create the future of strength and independence for the neuromuscular disorder community. And I'll go a little bit deeper into each of these as we go along. Next slide, please. So starting with MS. Again, we're building on the Ocrevus' leadership story that started around 2017, the launch of Ocrevus in relapsing MS and that subsequently in PPMS. And Ocrevus is really a landmark medicine in a sense that it was the first drug approved in this very broad label that includes both the more common relapsing forms as well as progressive forms. And since then, the data that we've kept generating that you can see here, that includes a shorter infusion data, the subcutaneous data that Dr. Hauser is going to talk about and the high dose data that we're looking forward to, our vision is to keep building on that success. And we feel that really the big unmet need right now here is around better control of patients for progression to preserve the quality of life for these patients. Of course, Ocrevus is the most study anti-CD20 therapy with over 10 years of data that we recently presented. But at the same time, we're very excited about new molecules. It's just fenebrutinib that I'll briefly touch on as well as employing new platforms such as our Brainshuttle technology that allows us to shuttle antibodies into the CNS and hopefully, again, improve efficacy in terms of controlling disease per progression. Next slide, please. So briefly touching on the Phase III OCARINA subcu results, and Dr. Hauser will go more deeply into this. Our vision right from the beginning is that we wanted to have a medicine that was very convenient for patients that would fit the lifestyle of the typical MS patient, who is a young adult that has an active life. Of course, an IV administration every 6 months is already very convenient. But the time of administration, we realized very early on might be a limitation. So initially, we worked actually to reduce the infusion time from the regular IV 5.5 to 6 hours to a shorter IV infusion of only 3.5 to 4 hours. And now with the OCARINA data, we are going to reduce that much, much further into something that initially will be done in a doctor's office under supervision, but for which the ultimate goal is to bring to the home care setting, allowing really for a very short, very convenient administration for these patients. The data from these trials, again, Dr. Hauser will speak to those, show the comparable efficacy in safety to the Ocrevus IV formulation. And we feel already based on the market potential here that this will really fit into the care settings, whether those are centers that have constrained IV capacity or areas where subcu or even oral are preferred as a therapeutic option that this really is where the subcu formulation will have strong growth. So, we see this very much as expanding further the market and the population that will be capable of taking Ocrevus rather than switching from Ocrevus IV. Next slide, please. As I mentioned, we keep providing several data updates on ECTRIMS -- just a couple of weeks ago, we provided some more. The subcu data, as I mentioned, for the noninferiority of subcu versus IV both on efficacy as well as safety. The Ocrevus 10-year data, this is really landmark data because this is following 10 years after the Phase III OPERA and ORATORIO trials. And these data show that in these patients, 8 out of 10 relapsing MS patients and 3 out of 10 of PPMS patients who are completely progression free. This is again our data that I've never seen -- been really seen with any MS drug. At the same time, recognizing that MS is mainly still a disease of young women, the data that we're showing of Ocrevus in terms of safety for women with MS, they have received Ocrevus during pregnancy and postpartum showing the safety of that, we feel it's really relevant. And we have ongoing Phase III trials that will evaluate immune responses in infants born to and breastfed by women treated with Ocrevus. Again, further supporting the patient centricity of this program. And of course, we're looking forward to getting the Ocrevus high-dose data. This is based on an analysis of our existing Phase III data that shows a deeper levels of B cell depletion, which are obtained by higher exposure to Ocrevus lead to better control of progression. And so we're hoping to see that result, this high dose in the further control of what's already a great profile in disease in progression. Next slide. Switching now to fenebrutinib. This is our BTK inhibitor that we've just shown data for, which, again, Dr. Hauser will go much deeper into. We feel this is the most specific, its brain-penetrant. And at the same time, it's the only non-covalent reversible agent that's currently being developed in Phase III trials. Now we feel this is very relevant, not just from an efficacy standpoint. Obviously, the brain penetration counts there, but at the same time from a safety standpoint. And as you can see on the right-hand side, we have a very broad program looking both at relapsing MS patients versus teriflunomide and PPMS patients versus Ocrevus. And what we hope to show in this program is that this is -- would be really the best-in-class oral BTK inhibitor, given its high potency, high selectivity, reversibility and the head-to-head, the studies that we're doing here against Ocrevus in PPMS. At the same time, we have a very large safety database, over 2,500 patients that have been dosed with fenebrutinib, not just in MS, but in other conditions as well. So we really feel that if this medicine is successful, this could really be a new disruptor in the oral segment market, which currently comprises about 40% of the global MS market. Next slide, please. Now switching over to Alzheimer's disease. This has been a long-standing commitment from Roche. And despite recent setbacks with Gantenerumab we really feel we still have a very strong pipeline here with medicines that could potentially be best in disease invested in class. And of course, what we're trying to do here is, on one hand, maximizing the therapeutic potential of what's really a more validated pathway, which is targeting a beta and combining that with new platform technologies that allow us to do that much more effectively. As you can see here on the right-hand side, we also have a growing portfolio that encompasses not only pharmaceuticals, such as trontinemab, Dr. Luka Kulic, will speak about or gamma-secretase modulator, but also anti-tau antibodies that are currently in Phase II trials. But at the same time, our emerging diagnostics portfolio that support early diagnosis that will enable treatment prevention and access to therapy. Next slide, please. Now Luka will go more in detail to these data. I just wanted really to provide here the broader context. The Brainshuttle platform is a technology we've been working on for over a decade now. And the idea really is to be able to bridge the blood brain barrier and provide access to biological agents to the CNS. Of course, if this is successful, it opens up a tremendous therapeutic space for new modalities and really, our Brainshuttle technology applied to an anti-A-beta binding -- antibody is really the first proof of mechanism for this new technology. Now trontinemab is specifically designed to have efficient transport across the blood brain barrier. And on the right-hand side, you see here some of the details of what that looks like. On 1 hand, one of the things that the Brainshuttle platform does is that it increases CNS exposure, not just in terms of quantitative, but also qualitatively, very differently. So you can see here in this Immunofluorescence image standard IgG versus a Brainshuttle IgG molecule, we see what's really a more homogeneous brain distribution, which we feel should lead to a deeper and faster amyloid clearance. At the same time, this antibody was really engineered on one hand to prevent certain potential targeted immune responses that might result in the worst safety profile. But at the same time, maintaining full Fc effector function to maintain the efficacy at removing amyloid. And again, Luka will give us a little bit more detail on that. Next slide, please. At the same time, we've been working for a long time on the gamma-secretase modulator. Now this is really hard to do to find a very specific molecule. On one hand, that cleaves that -- modulator cleavage of the amyloid precursor protein without touching any other gamma secretase substrates such as Notch, which we know are associated with safety events. But at the same time, it really shows that modulation of the processing of APP, meaning you hope to see an increase in nonaggregating Aß forms, such as Aß37 and 38, but at the same time, a reduction in toxic more aggregating A beta forms, such as 40 and 42. This sweet spot has been very, very hard to reach and we're very, very happy to apparently have a molecule now, and Luka will speak more about this that really threads that needle. So we're very excited about the potential that this has. Next slide, please. And this is my last slide. I just wanted to briefly touch on our neuromuscular diseases efforts. Of course, we're very happy with the success that Evrysdi has fast becoming a global leader in spinal muscular atrophy. At the same time, we're having combination trials now. We have anti-myostatin antibody. That's in Phase II right now. But at the same time, we're the emerging data for the gene therapy. This is the first approved gene therapy for Duchenne muscular dystrophy. We should have the Phase III data any day now. At the same time, our emerging efforts with our anti-IL-6 receptor antibody, Enspryng, in generalized myasthenia gravis. Again, this is an area of tremendous medical need, and we're just happy to be able to keep providing what we feel are really transformational medicines. And with that, I'll turn it over to our next speaker, which I believe, is Dr. Hauser.

Thank you, Paulo. Let's -- there we go. Thank you. As a clinician, it's difficult for me to overstate the impact that Ocrevus has had on MS. A generation ago, the average person with MS would become nonambulatory within 15 to 16 years. And now for young people, whose disease is just beginning today, we can be reasonably confident that they have a superb long-term prognosis for a life free of significant disability. In the Phase III Ocrevus trials, the impact of treatment on focal brain inflammation, which is the pathologic substrate of relapses or attacks of multiple sclerosis, was over 99% with Ocrevus. This is an effect size that I think is unequaled in modern clinical trials. It's really quite remarkable. And that -- those data from those trials, as you've heard from Paulo has been amplified in the 10-year follow-on open-label extension studies and in increasing volumes of real-world evidence supporting the sustained safety and efficacy of Ocrevus in all people with MS. So it's really quite an amazing success story. However, MS is not cured. And the part of the disease that is incured, as you've heard from Paulo, is the progressive phase of the disease that we now know because of Ocrevus is operating in most patients throughout the landscape of their life history with MS. Early on, it's very difficult to measure. Patients know they're not quite as good but we can't see it on our neurologic exams typically. This is called silent progression. Later on and in patients who have had sustained secondary or primary progression prior to beginning treatment, it is much more evident in terms of year-on-year changes. And what we've learned from the Ocrevus and other anti-CD20 trials is that the substrate of that progression is not primarily B cells in the periphery that move into the nervous system, the mediator of attacks, but rather is due to locked inflammation within the central nervous system mediated in chronic MS by plasmablasts and microglial cells, a form of macrophage in the brain. Two cells that are not directly hit by CD20 because they don't have CD20 on their surface. So that is the unmet need. And I am very excited about the programs with Ocrevus and fenebrutinib that will address this unmet need as well as better address underserved groups and patients with inadequate access to IV capacity. Next slide, please. So let's begin with OCREVUS and the ECTRIMS data. As you've heard from Paulo, the trials of Ocrevus subcutaneous administration, were presented at ECTRIMS. And these trials basically have looked at pharmacokinetics as a primary endpoint and pharmacodynamics as secondary as well as exploratory endpoints. Looking at a noninferiority comparison between Ocrevus subcutaneous versus intravenous, 920 milligrams subcutaneously, 600 milligrams intravenous open-label study one-to-one selection for Group 1 to 2 with a 24-week observation period, followed by an extension, but a primary endpoint that is pharmacokinetic similar overall exposure for -- during the control period of the initial 12 weeks comparing subcutaneous with intravenous administration. And what you can see on the right side of this slide is that the results showed in terms of pharmacodynamics, rapid and sustained B-cell depletion in blood, similar in both treatment arms and the pharmacokinetics on the left side of the trial, the primary endpoint also shows a similar overall exposure during the control period of 12 weeks with the subcutaneous compared with intravenous administration. So let us move to the next slide, please. Not a primary endpoint, but obviously, a very important endpoint is the relative effect of subcutaneous administration on clinically valid endpoints, which, in this case, MRI changes, specifically evidence of new or growing lesions. And here, if we look -- I don't think you can see my -- can you see my arrow? I don't think you can. But if you look at week 24, there was essentially 0 evidence of focal brain inflammation with the subcutaneous or intravenous compared with intravenous doses showing comparable efficacy to the IV formulation for the new subcutaneous delivery mechanism. I will not show that no new safety concerns were identified. You can see this in a very rough way on the right side of the slide. There were more adverse effects in the subcutaneous dosing, and that was due entirely to local tissue reactions, which were mild to moderate in severity, usually mild around the injection site. So the safety profile of the subcutaneous was clinically non-inferior to that of the intravenous dosing. Next slide, please. So going back of the 300 -- more than 300,000 patients treated with Ocrevus, we have been presenting ongoing results from the open-label extension study, which is now out to 10 years of clinical experience. And in the data presented, almost 8 out of 10 patients, who were initially in the Ocrevus arm of the OPERA relapsing remitting OCREVUS trial, almost 8 out of 10 were progression free 10 years after onset. That is really quite remarkable. If we look at subgroups of patients, those with recent onset disease and no disability when Ocrevus has started, even do better at 10 years. In patients with primary progressive MS, they're older, they have more advanced disability on admission compared with the Ocrevus patients in the relapsing study. More than 1 out of 3 of these patients were progression-free at 10 years. Another quite remarkable result at a 10-year follow-up point. This is extraordinarily clinically significant. So very impressive long-term data with continued safety in both the OPERA and ORATORIO trials of relapsing and primary progressive MS. This is a dramatic change in the landscape of multiple sclerosis. There was also data presented from the international MS-based pregnancy neonates outcome -- neonatal outcome in women's health registry on the value of Ocrevus for family planning and specifically, its utility as a protective means of preventing MS disease activity throughout pregnancy and the postpartum period. Remember that more than 3 out of 4 people with new onset MS are female and the majority are in the childbearing age. This slide shows the annualized relapse rates here in green with Ocrevus compared with platform low efficacy therapies interferon and glatiramer, natalizumab and dimethyl fumarate. I would make the point that it is clear from the Phase II studies that the effects of Ocrevus, the benefits, the protection against relapses is sustained even when B-cells recover at about 4 to 8 months. And that sustained protection against relapses goes out beyond 12 months, at least to 18 months, we feel. Now it's Q6-month dosing is essential because the protection against progression requires 6 months dosing for this to be maximum. But protection against relapses, the main concern for over the short term from many women with multiple sclerosis in the childbearing age is conferred over a long period with just 1 dose of Ocrevus. So one could treat patients 3 months before conception and that benefit against progression will be sustained throughout the pregnancy and postpartum period. I'd also make the point that there is no rebound following B-cell repletion with Ocrevus unlike natalizumab and the S1P inhibitors like fingolimod. So there is a rebound in a proportion of patients with a dangerous increase in MS disease activity above the level for ceding treatment. Another risk for patients who are in pregnancy, unable to continue taking their other disease-modifying therapies for MS. And they are not only not protected, but they are at heightened risk because of discontinuation of other MS therapies. So the benefit of Ocrevus in this situation, in my opinion, is really quite unique. Next slide, please. Let me move from Ocrevus to fenebrutinib. Next slide. I had mentioned the pathologic substrate of progressive MS plasmablast and microglia. 2 cell types, not directly affected by Ocrevus but both the direct targets of Bruton's tyrosine kinase inhibitor. In addition to its effects on B-cells and plasmablast in the periphery, BTK inhibitors, if they can penetrate CNS effectively, should have active inhibitory effects on microglia. And we know well that pro-inflammatory, neurotoxic microglia are a key, key component of the MS pathology underlying progression. So in many respects, a BTK inhibitor is a perfect selective therapeutic at least in theory for patients with MS because in addition to inactivating disease-causing B-cells, plasmablasts and microglia are also affected. The problem with BTK and with kinase inhibitors, in general, as many in the audience knows, is a selectivity problem. And one of the -- and perhaps the most important advantage of fenebrutinib and you can see this on the right side of the slide is that it is more than 100-fold selectively acting on BTK compared with any other kinase inhibitor that can be studied here, close to 60 kinase inhibitors were studied, and this is different from the less selective kinase inhibitors here shown for tolebrutinib and evobrutinib. Also, the non-covalent mechanism of action of fenebrutinib, something that Paulo spoke about and also its slow off rate makes it a very appealing choice for chronic daily use in patients. Next slide. So we presented at ECTRIMS the fenebrutinib data from the FENopta clinical trial, a Phase II clinical trial and also, a CSF and microglial in vitro data as well as B-cell in vitro data at ECTRIMS. And just very briefly, enrollment in the FENopta trial was 2:1. FENopta -- fenebrutinib over placebo. The primary endpoint was the number of gadolinium-enhancing lesions. And a number of exploratory endpoints, as you can see here. And it was the MRI scans at weeks 4, 8 and 12 were summed for the primary endpoint in FENopta. And this is followed -- this will be followed by an open-label period. So let's go to the primary data now. And what you can see here in the FENopta study is that fenebrutinib reduced T1 gadolinium lesions new areas of focal brain inflammation by more than 90% at here at week 12, beginning of an effect in week 4 and a quite substantial effect in week 8. The lower number of lesions at week 12 in the placebo group is due to just the variation that we typically see with small Phase II studies. This greater than 90% effect is really quite striking and very significant. Next slide, please. Importantly, fenebrutinib was also able to be detected in a nested study out of FENopta at clinically significant concentrations at levels that can impact mechanisms underlying both microglial and B-cell progressive disease biology within the nervous system in people with MS. And here, you can see the mean fenebrutinib CSF concentration in nanograms per milliliter. And here is the IC90 the IC50 concentration. So very impressive maximum inhibitory concentrations achieved through twice daily oral dosing of fenebrutinib. Next slide, please. And finally, the safety studies were also quite promising. All adverse events were mild to moderate, Grade 1 or 2, except for 2 Grade 3 asymptomatic transaminase elevations. There may be questions about this later. I think that although we will be required to see how data continues to develop in the Phase III fenebrutinib studies. It appears that the hepatotoxicity seen with other BTKi inhibitor with other BTKis is quite modest with fenebrutinib compared with those other 2 compounds. So some degree of transaminase elevation may be a class effect of BTKis. But much of it may well be these off-target kinase effects that are not present with fenebrutinib at the dosing that is currently being used in the MS trials. Next slide, please. So in conclusions, fenebrutinib was present in the CSF at levels sufficient to reduce activation of both B cells and microglia and especially plasmablasts and microglia in vitro, suggesting that fenebrutinib may impact mechanisms underlying progressive MS and may go beyond even what Ocrevus is able to do. The primary endpoint was met. And I would point out that by -- that after week 4, reductions in new gadolinium enhancing focal inflammatory MRI lesions was greater than 90% and that no new safety concerns were identified. And unlike 3 other BTKis in -- currently in trials for MS for which a halt has been placed because of high liver function elevations. This has not been the case for fenebrutinib and this may well be due to the selectivity of fenebrutinib for BTKis. I believe that's my last slide.

Yes. Thanks very much, Dr. Hauser. Good evening, everybody. Good evening from Basel. My name is Luka Kulic. I'm an expert medical Director and Therapeutic Area Leader for the Dementias at Roche Pharma Research and Early Development based in Basel. It's my -- great pleasure to walk you through some of our recent data that we presented at the CTAD Conference in Boston last week. We at Roche, as many of you know, and Paulo also highlighted it in his intro, have a long-standing footprint in Alzheimer's disease. We are and remain committed to this disease and to transforming every single step of the 80 patient journey. We do this by providing tailored solutions across the breadth of the Alzheimer's disease continuum. And this includes the development of advanced diagnostic solutions and biomarkers as well as our efforts to develop an advanced innovative therapeutic agents, Agents that target the core disease pathologies of AD like A beta and tau along with novel targets and mechanisms that we are currently exploring in our earlier pipeline. If you go to the next slide. Today, I'd like to give you an update on 2 of our clinical stage assets, Trontinemab and RG6289, our new gamma-secretase modulator. If you could go to the next one. So Paulo already introduced the molecule, Trontinemab is a novel molecular anti-TA so-called Brainshuttle A beta antibody. Actually, the first of its kind that is currently in clinical development. Trontinemab combines by recombinant fusion and anti-A-beta IgG-1 backbone with the so-called Brainshuttle module that specifically binds to human transferrin Receptor 1. This is a receptor that is highly enriched on the surface of brain and [ cerebral ] cells. And this whole Brainshuttle A beta antibody construct and crosses the blood brain barrier via an active transport mechanism by a so-called receptor mediated transcytosis at the capillary level to reach its target amyloid plaques and aggregated A beta in the brain. In comparison with standard anti-A beta and monoclonal antibodies, this is, we believe, a truly differentiated approach, which may not only enable a substantially higher brain exposure. But due to this capillary entry mechanism of the [ shape ] also a more widespread and more homogeneous distribution of the therapeutic molecule in the brain and also a superior target engagement. A substantially higher brain exposure is something we were able to demonstrate in various of our preclinical studies that we did with trontinemab, including so-called single-dose pharmacokinetic studies in Cynomolgus monkey, non-human primates. In these studies, this is the figure in the middle. Trontinemab led to a 6- to 17-fold higher brain exposure across different brain regions compared to gantenerumab. The first in human evidence that we didn't obtain for a shuttling effect to CNS compartments. This evidence came actually from our first-in-human single ascending dose study of trontinemab in healthy young volunteers. In this study, this is the figure that you see on the right-hand side. Trontinemab showed an approximately 8-fold increase in the CSF plasma ratio compared to historical gantenerumab data. To on with the slide. In 2021, then we initiated our first inpatient study called Brainshuttle AD. We presented interim results from this study at CTAD last week, and allow me please to walk you through some of the key data that we shared with the scientific community. So the Brainshuttle AD study. The study is a global double-blind, placebo-controlled Phase Ib/IIa multiple ascending dose study of trontinemab in people with MCI due to AD and people with mild-to-moderate Alzheimer's disease. The primary objective of the study is safety and tolerability. A key secondary objective is pharmacodynamics, as revealed by the change from baseline in amyloid plaque burden measured by PET. Go to the next slide. The study uses a staggered parallel group design with participants recruited in 4 sequential dose cohorts that you see here. In the initial dose escalation part of the study, a minimum of 10 study participants per dose cohort are randomized in a 4:1 ratio to receive either trontinemab or placebo IV every 4 weeks for a total of 7 doses. The study is adaptive in nature. That means that those levels that have been tested in the dose escalation part of the study, that these dose levels may be investigated further in the so-called expansion part in order to establish a robust PK/PD and safety profile of the molecule in a larger number of study participants. Today, we would like to share with you the interim results from the first 3 dose escalation cohorts in this study and the cut-off date for this interim analysis was actually recently on June 30, 2023. Next slide, please. For time reasons, we will be really focusing today just on the key results, beginning with the PD, the amyloid PET results. The interim analysis of amyloid PET data that you see here revealed the dose-dependent amyloid plaque lowering across all 3 active dose groups at 0.2 mg per kg, so the lowest dose level that we tested in the dose escalation part of the Brainshuttle AD study, trontinemab led to a meaningful reduction of minus 20 Centiloids versus baseline. The reduction in the second dose cohort at 0.6 mg per kg was minus 31 Centiloids versus baseline. And at 1.8 mg per kg, so our third dose level, we observed a very strong and rapid amyloid PET reduction of minus 62 Centiloids versus baseline at week 12 and minus 84 Centiloids versus baseline at week 28. In this third dose cohort, 36% of the study participants on active drug fell below the amyloid positivity threshold of 24 Centiloids after 12 weeks. 75% of the study participants became formally amyloid negative after 28 weeks. Next slide, please. Just to put these numbers in relation to what we have previously seen with standard anti-beta antibodies. So what you see here is the amyloid lowering trajectories of trontinemab at 1.8 mg per kg in blue and of several other anti-A beta monoclonal antibodies that have been tested in Phase III so far. Gantenerumab based in turquoise and purple, lecanemab based in brown and donanemab is depicted in green. Please note the different time scale on the axis on this slide compared to the image that you've just seen on the previous slide. The magnitude and the speed of amyloid lowering with trontinemab, as I said, trontinemab is depicted in blue is quite impressive and exceeds the effects of the standard antibodies even of those that read out positive in Phase III like lecanemab last year and donanemab earlier this year. In comparison with gantenerumab, there is an approximately 9-fold faster amyloid plaque clearance if we take the 3-month PET time point in the 1.8 mg per kg group as a reference. Next slide, please. Let's have a brief look at the blinded safety profile. So please note, this is an ongoing study. This study remains blinded to individual treatment assignment. So that means that the data from participants receiving trontinemab and placebo that these data will be presented here together as cohort level data in order to protect the blind at an individual participant level. In brief, there were no deaths and no serious adverse events that were related to the study drug by the investigators. Among the treatment-emergent adverse events, mild to moderate infusion-related reactions, or IRRs, and a mild transient anemia showed a dose-dependency and were more common or common adverse events in the highest dose group at 1.8 mg per kg trontinemab or placebo. What we can say so far is that both IRRs and the mild transient anemia are monitorable and manageable adverse events. The vast majority of IRRs that we observed, for example, occurred after administration of the first dose of the study drug in the absence of premedication. We have, meanwhile, in our study protocol included premedication to effectively mitigate IRRs at those levels of 1.8 mg per kg and higher from the outset. Next slide, please. A common -- and occasionally sometimes serious adverse event associated with anti- A beta antibody therapies are the so-called amyloid-related imaging abnormalities, or ARIA. Our current data set is, of course, still relatively small. So this is preliminary data. In total, we have 44 participants included in the interim analysis. What we can say though is that the ARIA incidence with this molecule has been fairly low so far, especially if we put this in relation to the magnitude and the speed of the amyloid lowering that you've just seen at 1.8 mg per kg for example. In Cohort 1 and 2, we didn't see any ARIA events at all. And in Cohort 3 at 1.8 mg per kg trontinemab or placebo, we had in total only 1 radiographically mild ARIA-E case. This ARIA-E case resolved on a follow-up MRI scan 4 weeks later. It was associated with my clinical symptoms, a transiently impaired attention, and these symptoms resolved after approximately 1 week. Another participant in this third cohort developed 2 ARIA-H events. So both events were asymptomatic and stable on follow-up MRI scans. Next slide please. So to conclude on this first part on trontinemab. Trontinemab Is a novel molecular entity, a Brainshuttle A beta antibody that crosses the blood brain barrier via active receptor mediated transcytosis at the capillary level. In our current study, the Brainshuttle AD study, trontinemab demonstrated rapid and robust amyloid plaque reduction in people with AD and at a relatively low doses, so at 1.8 milligram per kilogram. So dose levels at which standard anti-A beta antibodies usually have little or no effect on amyloid plaque pathology in patients. Collectively, our preliminary PD and safety results, including a so far low ARIA incidents, these results support further investigation of trontinemab in the ongoing Brainshuttle AD study. And our encouraging amyloid PET results moreover provide the first pharmacodynamic proof-of-concept for Roche's Brainshuttle platform approach. This may, of course, have important implications not only for AD and the treatment of AD but also for the treatment of other neurological disorders. Next slide, please. Moving on from trontinemab to our second exciting A beta targeting asset, namely RG6289, our new gamma-secretase modulator that we investigated recently in a Phase I study in healthy volunteers. So as Paulo already mentioned in his intro, RG6289 is an oral, a small molecule with a complementary mechanism of action to anti-A beta monoclonal antibodies. The gamma-secretase modulator acts upstream in the amyloid cascade. It shifts the processing of the amyloid precursor protein, APP, to produce less of the longer aggregation prone and toxic A beta fragments like A beta 42 and more of the shorter nontoxic A beta isoforms like A beta 38 and A beta 37. This unique mechanism of action is expected to significantly slow down or even hold the formation of toxic A beta aggregates and of amyloid plaques in the brain. Importantly, there is recent evidence from 2 independent natural history cohorts suggesting that these shorter A beta isoforms like A beta 38 are protective. A beta 38 has recently been associated with better clinical outcomes in people who are at risk for AD-related cognitive decline. On the right-hand side, you see the results of 2 in vitro pharmacology studies highlighting the unique mechanism of action of the gamma-secretase modulator. The longer A beta species like A beta 42 go down. The shortest species, like 37 and 38, go up. There is no change in the overall net production of A beta. What is important to note here and Paulo also alluded to it in his intro is that this mechanism of action is fundamentally different from the former secretase inhibitors, including the gamma secretase inhibitors. Gamma secretase inhibitors, as the name already tells you, inhibit -- that means they block the cleavage and also the signaling of a number of important cellular substrates of the gamma secretase complex. For example, the cleavage of the Notch receptor, which really plays a pivotal role in the central nervous system. In previous studies, this inhibition of Notch cleavage was associated with toxicity and various safety problems in the clinic. Gamma-secretase modulators in contrast are completely different from the secretase inhibitors, gamma secretase inhibitors in particular. As you may appreciate from the figure on the right-hand side, they do not affect Notch cleavage at all. This is really important to keep in mind and crucial also for the understanding of the differentiated safety profile of gamma-secretase modulators. Next slide, please. At CTAD, we presented data from the first-in-human Phase I study of RG6289 healthy volunteers. Next slide. So we shared the results from the first 3 parts in this study that are highlighted here in blue. A single ascending dose study in healthy young volunteers, a multiple ascending dose study in healthy young and healthy middle-aged participants and a multi-dose study in healthy elderly individuals. The total number of participants enrolled in the whole Phase I study was 127. Next slide, please. Let's have a brief look at the data, beginning with the primary objective that is safety and tolerability. RG6289 in this study exhibited a favorable safety and tolerability profile after single and multiple oral administrations for 14 days. Almost commonly reported adverse events were of mild intensity, and there was no dose-dependent increase in adverse events in any of the study parts. Moreover, we did not observe any noticeable treatment or dose-related changes in vital signs, ECG or lab parameters nor in any of the clinical and neurological assessments that we performed. Next slide, please. For time reasons, we left out the PK results in this presentation. What is relevant to mention here is that the PK profile of RG6289 was well behaved, dose proportional and supportive of daily dosing. What you see on this slide are the pharmacodynamic results after multiple oral administrations of RG6289. And as you may appreciate from the figure on the left-hand side, we observed a very nice dose-dependent target engagement and lowering of A beta 42 levels in plasma. At the highest and second highest dose level tested and A beta 42 reduction of 60% to 70% was achieved. Next slide, please. These plasma results were further confirmed in CSF, cerebrospinal fluid. And the figure in the left upper corner, you see a nice dose-dependent lowering of A beta 42 levels in CSF, similar to what you've just seen in plasma. In agreement with the gamma-secretase modulating activity of RG6289, there is also a decrease in CSF A beta 40 levels and a parallel increase in the shorter A beta forms, A beta 37 and A beta 38. Lastly, when we plotted our plasma A beta 42 results versus the CSF results, this is the figure on the right-hand side. When we did this, we observed a very good correlation actually between the plasma and CSF levels, suggesting that Plasma A beta 42 levels may, in principle, be used as a proxy for CSF A beta 42 levels in the clinic. Next slide, please. With this I'm at the end of this short overview of RG6289 results in the recently completed Phase I study, the new gamma-secretase modulator showed a favorable safety profile across all single and multiple doses in young as well as healthy elderly study participants. It exhibited a favorable PK profile supporting daily dosing. Importantly, we were able to confirm successful gamma-secretase modulation in human and establish a proof of RG6289 mechanism in this study. The longer toxic A beta species, 42 and 40, decreased as expected and the shorter protective forms, A beta 37 and A beta 38, increased in a dose-dependent manner in CSF. The good correlation between plasma and CSF A beta 42 levels, moreover, suggested that plasma A beta 42 levels may serve as a proxy for CSF A beta 42 levels in all future clinical studies of RG6289. And the overall positive results from this first-in-human study support further clinical development of this new gamma-secretase modulator for the treatment of Alzheimer's disease. With this, thank you very much. And I'll pass it over to you, Bruno.

Thanks a lot, Luka. And then with that, we will jump into the Q&A session. The first questions come from Harry Gillis from Berenberg.

Yes. So I know you touched on the fact that there are the elevated liver enzymes in FENopta, but these may be less severe than the other BTKis. So I was just wondering how confident you are that you could receive a differentiated label versus competitors if they were to have warnings? And is there a risk of a sort of class-wide safety warning here? And then secondly, if I may. Just on the gamma-secretase modulator, just wondering if there's any rationale for combining even with the Brainshuttle or other anti-amyloid antibodies?

Who's going to take the questions? Paulo, maybe?

Yes, I'll take a shot at it. Thank you for the questions. I mean the first one is obviously very difficult to answer because at the end of the day, it's a result of conversations with the regulators when we have the full Phase III data. What I think is fair to say now is that we are taking a very thorough approach to monitoring liver safety. And as Dr. Hauser mentioned, a few signals we have seen so far are mild to moderate. They're easily monitorable and they're reversible, and there's no high loss cases. That gives us some confidence that in approaching. The regulators will be able to have a good conversation around that. Also taking into account that the mode of action of the drug the fact that it's a non-covalent reversible binder might give us credence to why those data, in fact, show that it's a differentiated mode of action. At the end of the day, of course, this is a conversation with the regulators. So I think it's very premature. First of all, to say, there will be a class label. Secondly, I think even if that is applied, I think we would have good data to substantiate our own position. Regarding the potential of combinations, maybe Luka, you want to comment as well. As I mentioned in the slides, it is actually one of key part of our strategy moving forward is that we want to build on the backbone of new -- of existing and new anti-A beta therapies, including new antibodies. And obviously, gamma secretase modulator would be a perfect pairing partner if you want one of those immunomedicines. But our combinatorial approach is not limited to A beta mechanisms only. So we have other mechanisms in the pipeline that we're studying, and we'd obviously be interested in looking at the combination potential for those. Maybe Luka, you want to add something?

Yes, I can only echo what you just said, Paulo. So I mean as I already mentioned also in the presentation, it's a complementary mechanism of action. I think the idea of a combination with an anti-amyloid monoclonal perfectly makes sense. So the gamma secretase modulator acts upstream in this amyloid cascade, while the antibodies essentially clear pre-existing aggregate. So this is also what fenebrutinib is doing, obviously, very efficiently and rapidly. So yes, there is a strong rationale for a combination.

Very good. With that, we would move on. The next questions would come from Stephen Scala from Cowen.

I have 2 questions for Dr. Hauser. First, this is following up on the liver tox. You suggested that other BTKs have liver enzyme elevations that meet the definition of Hy's law. Can you say which agents do so? Roche, Sanofi and Novartis all have said they have not seen Hy's, and Merck hasn't said. So which one does have Hy's law cases of liver tox? And the second question is regarding the local reactions to subcutaneous Ocrevus. What did medical personnel do when they appeared? And would they ultimately inhibit patient administration in the absence of medical personnel?

Yes. So for the first question, I -- if I said that I knew that -- if I intimated that I knew the details of the BTKis that resulted in the hold, I would have misspoken. So I don't know the details. But the drugs that have been put on FDA hold are tolebrutinib, orelabrutinib and evobrutinib. So that is all that I know about those trials. Plus for evo and tolebrutinib, we know the Phase II data. We have not seen anything thus far in the Phase II or in the ongoing Phase III trials of fenebrutinib that bring us to have concerns about that point, that the drugs are -- that we're seeing cases of Hy's law or of less significant LFT, liver function and particularly transaminase elevations that put us at risk for a similar FDA action. The -- so I guess that's all that I would say about that. And with respect to the second point, Paulo, I believe that the -- most of these areas of erythema, warmth, some itching, some pain can be managed with analgesics. Interestingly, they are present more with the first dose than with later doses and do not seem to affect the provider's impression that this is an attractive therapy, and that's anecdotal. I would say that this could be an outstanding means of delivery should this subcutaneous formulation reach approval for the reasons that I stated. In addition, for insurers' perspective, the cost differential, which is huge in a country like the United States where, as you may know, the cost of infusion can be 12 to 15x the cost of the drug, I think it could be enormously valuable for patients.

And maybe just a brief comment. Thank you, Dr. Hauser, and I agree with you. Most of these local adverse events are pretty easily managed. They're not severe in any way. But actually, we continue to work on this technology. And the first administrations now and in the trial are administered in the office under physician supervision, but their goal is very much to keep developing that, to develop a patch pump, to do the studies that prove that comparability to be able to move it to the home setting in which, obviously, we're looking for a very clean, safe and tolerability profile.

Thank you. Did we answer your questions?

Yes.

Let's move on. Next one would Emmanuel Papadakis from Deutsche Bank.

Maybe a couple on trontinemab. Just curious on the rationale and indeed perhaps even the ethical basis for using placebo even in a Phase I study. Why not use gante or one of the approved alternatives? Perhaps more importantly, just some clarity on time lines and next steps. It looks like you're continuing to run the Phase I/II Brainshuttle study. So are you going to do another separate Phase II dose escalation? Or is this -- you now go into Phase III? And then what does the Phase III look like? Would that be a head-to-head study? Given the speed with which you're reaching amyloid negativity, would that be a relatively short, fixed duration study, so perhaps even less than 6 months? Any comments there would be helpful.

Yes, thanks very much for the question. So the first question was related to the use of placebo. So first of all, I mean, this is the first inpatient study and the first multiple ascending dose study, sort of multi-dose study of the molecule, so to really establish -- and the primary objective of the study is safety and tolerability. So placebo controlled in such an early trial is really important. And by the way, the study was initiated, I mentioned this at the beginning, in '21. So I think at the time, we also didn't have these readouts that were presented last year and then earlier this year. So yes, in any case, in such an early study, a placebo-controlled is important, especially also to get a good understanding of the placebo. The randomization ratio is 4:1. So we really have obviously many more patients on active drug. The second question relates to the next step. So I mentioned, so we are -- so we presented today data from an interim analysis from the first 3 dose escalation cohorts from the Brainshuttle AD study. We have now fully enrolled the highest dose within this dose escalation cohort, 3.6 milligram per kilogram. And then the idea is, as I mentioned, to then expand cohorts with the most promising risk/benefit profile essentially. So to add additional numbers of participants to really get a robust understanding of the PK/PD and safety profile of the molecule in the so-called expansion part. And this is something that, yes, is currently underway and will be also done the next year. So this is where we are currently in terms of clinical development. And from there on, of course, we will take it based on what we see, especially in the expansion part, once we have a robust understanding of the PK/PD and safety profile of the molecule. Yes, there are different considerations currently in the team with regards also to the Phase III development. But thanks for the question.

Emmanuel, all answered?

Yes.

And we move on. Next one would be Richard Parkes from Exane BNP.

So two questions, both on trontinemab. I wonder -- I didn't see anything about neutralizing antibodies in the slides. So could you talk about whether you've looked at incidents and neutralizing antibodies and how much comfort you've got about the immunogenicity given the profile of the drug? And then secondly, in terms of ARIA incidence, I wonder if you could help us to understand. What do we know about rates of ARIA relative to amyloid or rate of amyloid reduction for the current drugs? I wonder if there's any correlation now that might raise a concern given the rate of amyloid reduction with trontinemab.

Thanks very much for this question. So today, indeed, we didn't have time to touch on the pharmacokinetics and immunogenicity, but we presented the data at CTAD last week. So what we observed after administration of the first dose of trontinemab in this multiple ascending dose study was a well-behaved PK profile. This was not the case after administration of the seventh dose. So at the end of this treatment period, which was due to the incidence of, you alluded to it, anti -- drug antibodies -- PK-relevant antidrug antibodies or ADAs. What is important to note is, so yes, ADAs have been observed across the dose levels that we explored in this study. What is really important to note is that the impact on PK appears to be dose dependent. So -- and significantly less pronounced with higher dosing in the study. So at 1.8 milligram per kilogram. So this is the dose level with these impressive amyloid lowering results, what we observed was actually a lower incidence of ADAs, first of all, in -- at this dose level, lower titers of ADAs and also less of an impact on the AUC, the median AUC brain -- the median AUC systemic exposure, apologies, which was reduced by approximately 25% after administration of the seventh dose at 1.8 milligram per kilogram. So there was a moderate effect of ADAs on exposure -- systemic exposure at this third dose level. What is also important to note is that ADAs were not associated, what we can say so far based on the available data, with any safety signals. And obviously, they did not prevent the rapid and robust amyloid plaque lowering that we observed so far with this molecule. And the second question, I think, was related to the incidence of ARIA. I think this is really an interesting question, and we've been thinking also a lot about this. Obviously, amyloid lowering monoclonal -- standard monoclonal antibodies are associated. They are all associated, including gantenerumab, as you know, within -- with the incidence of amyloid-related imaging abnormalities or ARIA. What seems to be the case is that, yes, I mean, the more amyloid you removed, the more likely it is that you get ARIA. However, there are also differences across the molecules. So we know for molecules like lecanemab from Eisai that the ARIA rate is lower than, for example, with the donanemab or lower than with aducanumab. So it's not only about amyloid lowering obviously, and there are certain molecule-specific differences that appear to play a role. In our case, we believe that -- I mean, of course, it's still early days, right? We have only 44 participants. And it looks like a relatively low and maybe lower-than-anticipated ARIA rate so far if we put this in relation to the amyloid lowering effect. Currently, we are discussing different hypothesis why, if this really can be confirmed in the expansion part of the study, what could be an underlying mechanism. So one of the theories is actually that this access to the brain via the capillaries might be a major difference that could explain this lower ARIA incidence. Other aspects are, for example, that the dose that we administer systemically is significantly lower. We achieved at a significantly lower dose more amyloid and fast amyloid removal than standard antibodies with higher doses. So that might also play a role. And the third potential factor that we're currently discussing is actually the pharmacokinetic properties of trontinemab. Thanks very much for the question.

The next questions would come from Eric Le Berrigaud from Stifel.

Two questions on MS. First, on Ocrevus subcu. And referring to one of your initial slides, it looks like Ocrevus subcu goes with double bullets, meaning over 2 billion peak sales potential. If we think about where Ocrevus is now, i.e., around 6.5 billion at the end of this year and still growing double digits, probably we might think of Ocrevus reaching 7 billion by the time the subcu formulation is coming to the market. So if we had 2 billion on top, would bring the brand to 9 billion or around this. Is it what you're suggesting? Or should we think about any cannibalization or any patient switching at some point from IV into subcu because of being in trouble finding the IV centers? Or -- and if there is any shift to model, could you help us maybe understand how much would that represent out of the existing Ocrevus patient base? The second question is on fenebrutinib and the timing for the data. It looks like when we look at the agenda, which was on one of the slides, the arrow suggests the 2 reporting Phase III data would be in the second part of 2025. Is it your best guess as we speak? And are the 2 Phase III data set to report about the same time or one ahead of the other?

Maybe, Eric, I will take the first question on the peak sales. So in general, we don't guide on peak sales for an individual molecule. What we have outlined here with the 2 billion plus opportunity is really the totality of the outstanding Ocrevus trials, and this is the Ocrevus subcutaneous opportunity as well as the Ocrevus high-dose opportunity. And you're right, it's both. It's -- we believe there is an additional opportunity here. So there is an element of market expansion. But there probably also will be a certain element of switching in the hospitals -- in the large hospitals where Ocrevus is primarily established today. The other question, then I -- who's taking this one?