Merck KGaA (MRK) Earnings Call Transcript & Summary

Ladies and gentlemen, welcome to Merck's R&D Update Call 2020. [Operator Instructions] May I now hand you over to Constantin Fest, Head of Investor Relations, who will lead you through this conference. Please go ahead, sir.

Thank you, Abby. Dear, ladies and gentlemen, a very, very warm welcome to this R&D Update Call. My name is Constantin Fest. I'm Head of Investor Relations here at Merck. And it's great to have all of you here joining this call, which is also the third deep dive session of last week's Capital Markets Day. And this is why I'd like to directly hand over to Amelie Schrader, Director, Investor Relations and Principal Healthcare Lead, to kick off this deep dive.

Thank you, Constantin. Dear, ladies and gentlemen, a very warm welcome also from my side. I kindly ask you to please turn to Page 3 of the slide deck for a brief introduction of the 4 representatives of our R&D and commercial leadership joining us on this call today. I'm very pleased to welcome to team members of our Health Care Executive Committee whom you know very well; Luciano Rossetti, Global Head of Research and Development as well as Rehan Verjee, Head of the Global Innovative Medicine Franchises and President of EMD Serono. Furthermore, I'm delighted to also introduce you to 2 senior leaders from our R&D organization: Peter Halle, Global Head of Research as well as Klaus Edvardsen, Head of Oncology Development. Over the next half an hour, the team will be walking you through our key pipeline highlights, major achievements and upcoming milestones. And following this, we will then be opening the floor to your questions. And with that, let me hand it to Rehan.

Thank you, Amelie, and welcome all also from my side. So on Slide 4, we have the agenda for today, and we're going to start with an overview of our research and early innovation, followed by a focus on select really promising clinical candidates. And we're also going to include some updates on the launch momentum for Bavencio in the U.S. and also give you a sense of how the Mavenclad launch is continuing to progress. On Slide 5, I just want to link back to where the Executive Board left off at the Capital Markets Day. So Stefan introduced you to the 3 main drivers of growth for the group for 2022 and beyond, and you can see that more than 80% of the growth is expected to come from these Big 3. And of course, the Healthcare pipeline is a big part of this growth. And as you know, the EUR 2 billion by 2022 has been our ambition. We remain committed to that. We're very confident in that. And not only the EUR 2 billion, but we also believe that based on the progress in the pipeline, beyond 2022, we will see even more contribution from the Healthcare pipeline to the growth potential of group. Slide 6, please. Also just from the Capital Markets Day, it's just a reminder that from a Healthcare perspective, despite the pandemic, Healthcare has some really strong underlying momentum. As you know, in H1 -- in the first half of the year, we posted good organic growth numbers. And so that showed some real resilience. And also, we've had some really important developments. The approval of Bavencio, June 30, and now filings accepted by other major jurisdictions and agencies. We've seen tepotinib's approval in Japan as the first selective and highly targeted MET inhibitor, and we continue to see real progress with Mavenclad while we're continuing to drive the pipeline forward. So now let's get into the meat of the R&D update, and let me hand over to Luciano.

Terrific, Rehan. Thanks, and I want to also send from my own personally, welcome to this conference. Many of you have been following our journey for a long time. And I feel that Slide 7 will give you a little bit of a very quick summary of the journey before we get into the R&D review. It has been an interesting ambition for us and a journey to get to the point to create an R&D organization that could be a serial innovator within specialty medicine. That's the ambition. And we will give you a little bit some point of validation related to where we are today. And then I hope you're going to hear more from Peter in particular, but also from Klaus, about some of the evolving potential for sustainability of this success over the last few years. I'll start from the left side of the slide very quickly. These are 10 new molecular entity that are still active in the pipeline. So these are first in-humans since 2014 that are still active in the pipeline. We are also very proud of making rapid choice on other NMEs in the pipeline once they were tested in humans. The point here is the quality of these NMEs, the factor of testing first-in-class hypothesis and also the yield. The majority of the discovery output has really translated in the mechanisms that are currently tested in the clinic. And I think that's an important point. They're trying to reduce the cost of failure. I move very rapidly. I think additional evidence of our progress is particular emphasis on the scientific leadership. We can see now that not only the number of scientific publication and report to meetings but the quality has increased quite a bit. You see some of the examples of the various journals, peer-reviewed journals, where we publish some of our findings. I just want to remind everyone about the 2 New England Journal of Medicine reports on the VISION trial for tepotinib and on our JAVELIN 100 urothelial cancer very recently published in the New England, as I mentioned. Another way to look at our aspiration to go forward with, again, excellent quality of clinical development is the success in obtaining what FDA and other regulatory agency consider way to accelerate and to support the availability of real major, major innovation for serious diseases within oncology, in particular. I mentioned that we had, in recent year, 4 breakthrough designations with tepotinib and Bavencio, Sakigake designation and recently also the real-time oncology review mechanisms used by FDA for some of our programs. Finally, validation also on the fact that we now have 3 major programs that are going into approvals -- global approvals, 6 approvals within 3 different brands, Mavenclad, Bavencio and TEPMETKO. And you're going to hear more from Klaus about that. And finally, I certainly think that Rehan has already illustrated our ambition that this novelty will contribute significantly to our revenue in the future. And you see that we start to see that uptick of contribution already, and Rehan will talk about this at the end, particularly in view of 2022. What we now want to focus on is the sustainability. I know many of you have been asking for us to show you not only the success we had with these various programs in the clinic but also see -- give us a window in the future, a window in how the company is trying to create the research and clinical operation that can be sustained for the future. And to do that, I decided it's very important for you also to know our leadership talent. And we created a position of Head of Research that Peter Halle has accepted to take. And I just want to mention that Peter has been also for a long time, the Head of our Immuno-Oncology Discovery and Development Operation early on -- Early Development Operations as well as Head of External Innovation. So he has been already an engine for this research success. We cannot share with you details of many of the programs, but we have taken some samples to give you an idea of our thinking, how we want to innovate in the future. Thank you. Peter?

Yes. Thanks, Luciano, and also a warm welcome from my side. I'm really looking forward to our interactions and discussions in the future. So Luciano shared with you how our R&D organization has gained some momentum in recent years, and I would like to give you a better understanding about the drivers behind this success and why I'm so excited about the research team. If you look at Slide 9, you see everything we do starts with the understanding of the disease biology. We run all steps of the discovery process and identified then highly differentiated drug candidates. But our work doesn't stop there. Preclinical combination data, biomarker data, modeling simulation, all of that informs our clinical development colleagues. And they then close the cycle by providing us in research with data and insights from clinical trials and real-world evidence to help us better understand the disease biology. So let me explain to you which areas we have selected, and we are focusing on Slide 10. At the Capital Markets -- actually, at the Capital Markets Day in 2040, we shared with you our translational innovation platform concept, we call those TIPs. Ever since, the 3 TIPs are fully empowered to drive the innovation strategy, and we applied a thorough peer review and established scientific advisory boards. As you will see in a minute, this really paid off and boosted our productivity. And so we kept these 3 TIPs in oncology with targeted therapies, immuno-oncology and immunology. This continuity builds the deep expertise we need and help us to double down on our success. What evolved, however, since 2014 is the collaboration between the TIPs. I will share with you an example later. But we are not only evolving the organization, we are constantly reinventing ourselves, exploring new disease areas, biology areas and partnerships with academic institutions such as the ICR in U.K., NCI in U.S., Karolinska in Sweden and the Weizmann Institute in Israel, but also with biotech companies such as BioMed X and Heidelberg. This leads to new biology areas such as phagocytosis in TIP immuno-oncology, stress and plasticity in TIP oncology and others you see here on the slide. And we have mature areas such as oncogenic signaling and immune complex signaling that TIPs developed drug candidates, Luciano was already talking about. On Slide 11, you see a full list of internally discovered products and development candidates such as Bavencio, evobrutinib, bintrafusp, tepotinib; you're very familiar with them. As we continue on this successful path in research, so what we are doing, we really take conscious risk early on in hit discovery phase and then apply rigorous bars, high-quality bars for the quality of the molecules and the differentiation about the standard of care. And currently focus on 8 projects in exploratory development, some of them actually close to the nomination of a development candidate. All of this is supported by central discovery platforms for NCEs, NBEs and ADCs as well as development platforms such as safety pharmacology, DMPK, CMC, all part of the research organization. So despite this internal success, we don't have [indiscernible]. And allow me a side note, this shouldn't come as a surprise to you with my former roles in biotech and business development. And the external innovation team, that Luciano just mentioned, spends the whole chain from life cycle management to discovery and is still part of my team. So let me talk about this in more detail on Slide 12. On this slide, I want to give you a better understanding on how we make rigorous portfolio and partnering decisions to maximize the value of our pipeline and portfolio. As you can see from this internal bucket at the left side, we have projects that we execute from A to Z internally, such as tepotinib, which we discovered in-house and brought to approval in Japan recently. We complement this internal discovery approach with co-discovery deals, Domain, Sutro, Ablynx are examples. And also in-license in focus areas, such as the deal with Vertex in the area of DNA damage response. This approach leads to such a rich pipeline and portfolio of development candidates that we cannot develop everything internally. We maximize the potential of our assets in co-development deals, GSK, Pfizer as examples. All spin-off assets into start-up companies that have better internal expertise for the specific areas, iOnctura and Asceneuron are examples here. We also licensed back recently M9831 to Vertex for specific indications in gene editing. All of that to maximize the value of the portfolio according to the 3 lenses you see on the right-hand side. But let me switch gears. We promised you a closer look to the early portfolio. In the interest of time, let me focus on 4 assets. I'm particularly excited about them. But I also wanted to highlight that there is still significant risk as they are early development projects. So on Slide 13, let me start with M1231, our bispecific ADC. As you know, ADCs which the market next-generation approaches have created a lot of momentum and team-making activity recently. But current technologies still have limitations as listed here on the slide. Together with our partner, Sutro, we have generated a bispecific EGFR MUC1 ADC that you see depicted in the cartoon and this ADC is addressing many of these limitations. It incorporates several advancements to standard ADCs, but let me focus on one specific element, the bispecificity. As you can see from the red color on the pictures next to the cartoon, this antibody is much better internalized than standard ADCs. And this higher internalization leads to much -- significantly more preclinic activities, sorry. The green bars on the right depict this, relative to the standard monospecific ADCs in blue that you can see on that chart. In addition to this preclinical activity, we expect a better differentiation between cancer cells and normal tissue, as normal tissue sometimes express both targets, but on polarized cells and opposite side of those polarized cells. So the bispecific probably will not recognize these normal tissues. We published this data at this year's AACR and more ADCs are in our discovery pipeline. So stay tuned. We will report more about this. On the next slide, Slide 14, another example for oncology, but that actually originated in TIP immunology. So an example for this great collaboration I was talking about between the TIPs. It is M3258, a highly selective LMP -- LMP7, sorry, proteasome inhibitor. And in contrast to panproteasome inhibitors, this one specifically blocks the immunoproteasome. And because the initial idea was to develop it in immunology, it has actually a favorable safety profile. But in preclinical models, we saw the best activity in multiple myeloma models. Data we presented last year at the AACR in oral presentations, and we are now in Phase I clinical trials and are so far very pleased with what we have seen. On the next slide, an asset that we haven't yet published, Slide 15. It's our TIGIT antibody that Stefan Oschmann mentioned in the Q2 earnings call and that created some excitement since then. So let me share some internal, not yet published data. Like avelumab and bintrafusp, it comes from our team in immuno-oncology. We consciously selected an IgG1 effector function competent antibody and actually an antibody that has a rock-solid preclinical profile. We demonstrated that its profound antitumor activity in preclinical models is driven by multiple mechanisms, not only inhibit in TIGIT and allow stimulation of CD226 via 226, it also depletes high TIGIT expressing Tregs and other immune suppressive cells. And very important, it combines well with bintrafusp alfa. This makes a lot of sense. If you think about the mode of action, TGF-beta is not only an important inhibitor of T-cell activation, but it's also blocks NK cells and trapping TGF-beta via bintrafusp allows these NK cells in the tumor microenvironment to do their job via ADCC depleting these immune suppressive cells. So we started a Phase I trial with this antibody in monotherapy, but the protocol already foresees combination with bintrafusp. And last but not least, on Slide 16, let me come to M4344. You may wonder why I, as the Head of Research of Merck, present this ATR inhibitor, which we in-licensed from Vertex. The reason is that we don't stop research when we deliver development candidate, but continue to support our clinical teams. In this case, we run patient-derived xenograft models and could demonstrate that M4344 does not only have an interesting monotherapy activity but that via combinations, we could broaden the response rate of PARP inhibitors independent of BRCA mutation status in individual patient-derived tumors. With this example for forward transformation, I really hope that I could convince you that we have an exciting pipeline and a team in place. And I would like to hand over to Klaus, who will guide you through the oncology clinical development. Klaus, over to you.

Thank you very much, Peter, and also warm welcome from me. Turning to the development aspects of our pipeline, I will, on Slide 18, share a few highlights on Bavencio, bintrafusp alfa, tepotinib and our DNA Damage Response portfolio. Turn to Slide 19, talking about Bavencio and on the left side of the slide, showing the OS Kaplan-Meier curve of Bladder 100 JAVELIN with a hazard ratio of 0.69, meaning 31% reduction in risk of death by adding avelumab to best supportive care in a maintenance setting. This is so far the only IO data in bladder with a significant OS advantage. And therefore, certainly, this data has the potential to transforming the way we treat patients with advanced bladder cancer. Turning to updates given on Bavencio at this year's ESMO, and I will focus on updates given on the JAVELIN Bladder 100 trial. And as you see on the right side of the slide, there was presentations about subgroup analysis, demonstrating an OS benefits irrespectively of the induction chemotherapy leading to a complete response, a partial response or even stable disease. There were also data presented about associations between clinical outcome and exploratory biomarker, indicating that a potential predictor of OS is expression of some Fc gamma receptors variants that indicate that potentially with avelumab could be a direct cell killing mechanism mitigated by an ADCC mechanism. And then finally, on quality of life data showing that there were no negative effect on standard patient-reported outcome measures, adding avelumab on top of better supportive care. Please turn to Slide #20. Moving into bintrafusp alfa. This is a program that has been designed with multiple shots on goals beyond looking into a potential differentiation to checkpoint inhibitors in a PD-L1 high non-small cell lung cancer population as monotherapy, actually testing multiple independent hypothesis beyond that monotherapy setting. And as an example, we can turn to a combination with chemotherapy to initiate and enhance the immune reaction and target fibrosis via the TGF-beta pathway that would allow a bigger penetration, addressing some of the low resistant mechanisms to checkpoint inhibitors. We can also look further into the lung program moving bintrafusp alfa into earlier stages of disease, testing it against durvalumab in a stage 3 non-small cell lung cancer nonresectable patient population. There are also attempts to explore the more direct role of adding a TGF-beta sequestering mechanism into the PD-L1 PD-1 axis inhibition in HPV-driven disease, that is primarily being tested out in cervical cancer and very likely also in head and neck cancer in the future. And finally, let me then mention combination with bintrafusp alfa as the backbone. Peter did mention to you that we have initiated a dose escalation trial with our anti-TIGIT, and there's a very clear aim starting combinations with a rarity of solid tumor using bintrafusp alfa as the backbone. You're also very familiar with our biliary tract cancer program with monotherapy in second line and against chemotherapy in the first line. If you please turn to Slide #21. Let me give a few highlights of what was presented for bintrafusp alfa at ESMO. There has been some long-term follow-up data presented showing that the 3-year survival update confirms efficacy in second-line non-small cell lung cancer, especially in the PD-L1 high patient population with a medium overall survival that has not yet been reached, but showing that 67% of the patients are still alive in that trial after 60 -- sorry, 36 months. Also showing data in biliary tract cancer in an Asian patient population after 28 months following in second-line biliary tract cancer that continues to show durable long-lasting responses and long-term overall survival with a median of 28% OS rate at 2 years. Please turn to Slide #22, turning to tepotinib. We have been laying the foundation for that program with the VISION trial, showing an overall response rate or 46% with a median duration of response of 11.1 months in patients with exon 14 skipping mutations in non-small cell lung cancer. We have obtained an approval in Japan in March under the trade name of TEPMETKO in all lines of the disease driven by that specific mutation. There is an expectation of a U.S. approval in the first quarter of next year. We are not stopping there with tepotinib. We want to tap into a disease setting where disease is driven by MET amplification. We have initiated a trial that is called INSIGHT 2 that is testing MET amplification as a resistant mechanism for patients that are failing first-line treatment with Tagrisso in an EGFR-driven mutation in non-small cell lung cancer population. That is based on the notion that MET amplification is one of the predominant drivers of resistance to Tagrisso with numbers as high as 25% of the patients failing treatment based on that specific amplification. INSIGHT 2 is testing the resensitization by combining tepotinib with Tagrisso. And finally, we are also exploring what does MET amplification mean as a disease driver outside of non-small cell lung cancer by doing an exploration in colorectal cancer with a backbone of Erbitux, testing in a second line advanced colorectal cancer with a combination of tepotinib and Erbitux. And I want you to turn to Slide #23. And just to highlight our expectations of the INSIGHT 2 program because we have fundamentally shown that tepotinib in the MET-amplified setting is working. We have shown proof-of-concept by the INSIGHT 1 trial showing a significant benefit of rescuing patients that fail with a MET amplification using a combination of tepotinib and Iressa as indicated on the right the slide. I now want you to move to Slide #24, turning into our DNA Damage Response portfolio. We have previously discussed our strategy. So I'm not going to focus on the left part of the slide, but I want to draw your attention to the right side of the slide. First of all, showing the breadth of our program, starting with ATR. We have positive clinical proof-of-concept data for berzosertib in ovarian cancer as well as in small cell lung cancer, and I will come back to that in the subsequent slide. We have shown progress with our oral ATR program. With the front-runner molecule M4344, we have defined a dose and are in the process of testing that specific oral ATR compound in specific central arm expansion, testing out a number of mutational hypothesis, among that RITA1, ATR and ATM mutations and the plans of moving that program into PARP combinations to tackle moving the PARP field beyond BRCA mutated patients. On the DNA-PK program, M3814 is progressing well in dose-finding with chemoradiation in neoadjuvant rectal cancer. And when dose is established, the plan is to progress that into chemoradiation proof-of-concept study, clearly making us, as a company, very well placed to deliver the first proof-of-concept, industry wide, for DNA-PK inhibitor. Now I want you to turn to Slide 25. And as I indicated, I'll go into a little bit more detail on the positive proof-of-concept for berzosertib, first all on the right, in a randomized controlled trial in ovarian cancer in combination with gemcitabine against gemcitabine, we have shown significant PFS benefit in all patients, more stronger benefit in the most platinum-resistant patients. And finally, if you look on the left side of the slide, we have data in Phase II now in collaboration with NCI in small cell lung cancer that, unfortunately, has not yet been published, and therefore, I cannot share the actual data here, but I can clearly state that they are confirming the early signals that are shown on the left of this slide, very well positioning us to start registration-driven trials in small cell lung cancer, in combination with a chemotherapy agent, tepotinib. And by that, I would like to hand it back to Rehan.

Thank you, Klaus. So to close out this view in terms of the oncology section, we're going to touch a little bit on the performance of Bavencio, now that it's been approved and introduced into clinical practice in the U.S. So let me start by really saying that the significance of this data set in frontline urothelial cancer is really underscored by the reception and the recognition that it has received in the community. From the ASCO presidential plenary to the FDA approval with the broad label, the ESMO guideline inclusion, the NCCN evidence recognition and an update to the guideline, and of course, the encore of data at ESMO with orals. All of this has really led to an incredible halo of some very good coverage in terms of education, which is really important as we basically try to rapidly change the standard of care. I will say as well that the data dissemination has potentially been the one thing that has potentially benefited from the pandemic context that we're in because, as many of you know, all of these conferences have been virtual, all of them reduced their fees, all of them saw far greater participation than normal. And basically, there was no fighting over who is going to get the front row seats. So I think, in a way, this is potentially one of the net advantages of the context that we now find ourselves in. But now let's shift to how we're actually doing in terms of the actual introduction. So the first real statement to make is that our assets are working. We have leading share of voice amongst all of the players that today offer an immune therapy to patients with bladder cancer, whatever the setting that may be. And this means that basically, physicians are being more frequently contacted by us and have more frequent retention of the critical messages by us than they do for some of the other players. So this is -- this means that despite the COVID context that we're in, we're actually able to reach and actually have an impact in terms of starting to move people's mental model towards adoption of Bavencio. We've seen very broad reach also. And obviously, in the context that we're in, this was going to be a concern. Can we still get to people given that it's going to be virtual and people are overloaded with e-mails and request for virtual sort of interaction? And the answer to that question is we're doing really well. When we look at our priority targets, so these are really some of the top decile physicians that have, therefore, this most significant volume of patients, we've reached 70% of our targets, basically within the first 2.5 months. And this is very consistent with basically benchmarks, leading benchmarks that you might look at in a more standard context, not a pandemic context. Finally, also our virtual programming. We're getting really robust participation, really robust dialogue and very positive feedback. And so basically, there is a genuine thirst, I think, to understand the data and to understand how this needs to be integrated into their clinical practice regime. When we look at more concrete measures, I think, of performance, we share here basically that the unit growth of Bavencio has really demonstrated an inflection. So if you look at the 11 weeks sort of post approval versus the 11 weeks prior approval, you're looking at a 40% jump, and that continues to accelerate because if you just took a rolling 4 weeks view for the most recent 4 weeks versus basically the prior 4 weeks, you're at 50%. So the point here is that this is a real inflection. Now what's also important is that the units are going up and also that the base of prescribers is also increasing, right, so the breadth. And what we've seen basically is just a tremendous increase in the number of accounts that are starting to order Bavencio. And as you know, Bavencio has been approved. It's been approved in MCC. It's been approved in RCC. It's been approved in second-line bladder. And so to have a significant increase in the number of accounts starting to order is really, really positive. And this continues to accelerate, August better than July, September also looking really good. And finally, when we look at penetration into the indicated segment, so these are patients that have received platinum therapy, patients who then have stable disease or better. These are the candidates for Bavencio maintenance. And what we see here is that basically since -- from July to mid-August of the candidate patients, we estimate a penetration of about 30% into that segment. So what that means is that of all of those available patients that should be candidates for Bavencio coming through in July, coming through in the first half of August, we're getting about 30% of them. So it was about 27% in July, 34% in August in that first half, and this is really, really encouraging. Just a word on the data itself. It's difficult to track this particular launch because of the unique context of a switch maintenance type approach. So what we have is a very customized view with a specific provider where we actually do comprehensive chart reviews of a significant group of physicians on a monthly basis, and that's been accumulated on a quarterly basis. And this is going to be our primary means of really understanding treatment patterns and integration of Bavencio as basically a maintenance agent following stable disease or better induction by platinum therapy. So this is really positive. And to put the 30% also in context, we, of course, look at benchmarks. And 30% penetration at this point into the target segment is very positive. We looked in particular at one benchmark. We look at many, but one in particular, just to put on your mind, is the introduction of Imfinzi, basically as a maintenance therapy in stage 3 lung. That was obviously a very successful introduction. And if you put the 30% here that we have in terms of the maintenance penetration into bladder cancer frontline versus that, you actually see that we are overachieving, I think, the level of penetration that they did at the comparable time period. So this is very positive. So now we'll move on to the neurology and immunology section, and Luciano and I will cover evobrutinib, the M5049 and also give you a sense of how we're progressing with Mavenclad from a data characterization perspective and also from a launch recovery perspective. So let me start now with evobrutinib. So I want to say a few things. There's a lot on the slide. We're not going to cover everything. But I think that there is a reason why we have always been excited about this mechanism and there is a reason why we were the first to clinic and there is a reason why we were the first to push the button on significant Phase III trials. And a big part of that is that we have always maintained that there is a significant unmet need today in the field of MS despite the tremendous progress that has been made over the last few decades. What's also important is that there is a significant continued momentum towards the use of higher efficacy agents in the treatment of MS. But when you think about higher efficacy agents, yes, there were quite a few that are now approved. But to be honest, there are only 2 mechanisms, 2 mechanisms that offer higher efficacy, disease control and that can be taken orally with the convenience of oral administration, only 2. And therefore, there remains a need still for better disease treatments, more high efficacy treatments, but also there is significant value in having oral higher efficacy treatments. It's also important to recognize that in the field over the last decade, there has been a growing understanding of the role of B-cells in the immunopathology of MS. Mavenclad has a profound impact on the B-cell compartment. We believe that, that could be one of the reasons why it's having such a profound efficacy. But of course, BTK is another way to get at that problem. And when you look then at the total market environment, considering high efficacy agents, considering agents that are impacting the B-cell compartment and considering the growth of the oral segment, it's just a no-brainer to consider something like evobrutinib. So with that, let me hand over to Luciano.

Terrific, Rehan. I will be going to Slide #29 and very briefly. I know many of you have been following the evobrutinib story with us for a while. So I don't want to reiterate too many points. But I want to make one point. This large Phase IIb trial that we have completed in MS, with now very significant and conclusive information on annualized response rate, it's really the real proof-of-concept for the Bruton's tyrosine kinase inhibition hypothesis in MS. This is really the most comprehensive data set. And even in our 108-week follow-up, we can confirm that we have efficacy on this primary endpoint for Phase III, that is very comparable to CD20 and other very high activity. The other one is the safety database. We have now more than 1,300, actually, patients in various trials that have been exposed for up to 2 years and plus, actually. And basically, systemic adverse events are really minor, and there is no signal for infection at the dose that we selected to go forward. Another very important point that I'm not going to reiterate in detail is that this compound had CNS penetration, but most important also by being an irreversible inhibitor is able to label and to block BTK in cells, the traffic, in and out of the CNS. That sets a huge upside potential in having an impact on many other phenomenons that are BTK affected in the CNS, including microglia, inflammation and, as I mentioned, trafficking within the CNS of cells. The final and critical point is that it's absolutely critical for this particular indication to have up to 95% BTK occupancy. And because of that, we felt that we had to spend a lot of time to really create moderate simulation to really push forward with our analysis of the ideal doses, and that's why we selected the BID approach to really maximize the efficacy. So in terms of efficacy and safety database, this is very, very robust data. The most important update is that despite COVID-19, the trial is really doing very well. It's on track. And we still can get to the conclusion we projected now by the end of 2023 for this trial. Let me move now very briefly to an exciting clinical development compound in the Slide 31st and make very quickly point that the Toll-like receptor 7 and 8 inhibitor, it's a very unique program that really is trying to block the sensor for RNA in many immune cells. That's what TLR7 and TLR8 is. And in that sense, there is a very strong similarity between our COVID-19 pathology driven by SARS-CoV-2 that is an RNA virus. And the way this is sensed and trigger very robust immunopathology and cytokine responses that are probably a major, major driver of the COVID-19 pneumonia and other inflammatory component of these very serious diseases. This is very similar to the sensing of RNA-based immunocomplexes and other signaling in autoimmunity, particularly in cutaneous and systemic lupus, but also other autoimmune indication where blocking TLR7 and TLR8 will have a very strong potential to have a positive effect. So I stop there, the Phase I part of the trial is completed. And now we are, next slide, 31, in the first of the Phase II trial. So we are planning obviously to test these in immunological diseases, as explained briefly before. But the similarities between COVID-19 pneumonia and other diseases in SLE and lupus, it's very striking in terms of the cytokines that are affected and in terms of mechanism triggering the cytokine where TLR7 and TLR8 could play a very, very critical role. So we decided to go forward with a randomized Phase II trial designed to test the hypothesis that blocking TLR7 and TLR8 with M5049 will have a beneficial effect in diminishing the consequences of COVID-19 pneumonia in patients hospitalized for this disease. The trial is going on. We have several sites activated and recruitment is picking up as we speak. I'll go now back to Rehan for some conclusive remarks.

Thanks, Luciano. So let me just touch on Mavenclad. So there was a lot of data that was also at ECTRIMS that further characterized and, I think, substantiated the growing belief in the benefit-risk profile of Mavenclad. I'll touch on a couple of the most important parts, and that is, one, that there was a study that basically showed how rapid the impact of Mavenclad is on the disease course. This is really important in the context of choosing higher efficacy agents because if a patient comes in with a very hot MRI, he has had a series of bad relapses, what you really want to choose as an agent that doesn't just work profoundly, but an agent that works rapidly to profoundly. And I think -- so the data sets that we presented there really support that. And what you see is a very profound effect and a very rapid effect within month 1. And this adds to the body of evidence from the earlier clinical trials. From a post-approval safety perspective, we presented a very robust cut of data from 20,000 patients in the post-marketing setting. This is really reassuring to the community that we are doing this with this data because of its transparency, and it's further educating them, I think, on the profile. And in particular, I think it's really reassuring to see that viral respiratory infections are really low percentage in terms of their recurrence. And there was also data cuts around malignancy and the rates that we've seen so far in the post-marketing setting are below what has been indicated in the various different labels. So again, really reassuring to the community. The other part, I think, of the data release, which was really important is how do patients on Mavenclad actually do if they are actually infected with SARS-CoV-2 and develop COVID-19. And so far, I think we've been able to provide data -- growing amounts of data that really suggests that at this stage, there's nothing to worry about. Now this is really important because this is an area of decision-making where there's not a lot of evidence, but at the same time, it's become the primary driver for decision-making, particularly so in March, April and May. And so it's reassuring that this is the case with Mavenclad. But it's important to recognize that, that's not the case with every DMT for which there's accumulating evidence. And as a good example, there was a note sent out by Roche to all physicians in the United States saying that there may be an increased risk, in particular, of severe COVID complications if you are undergoing treatment with Ocrevus. So this is reassuring, and I think the data will continue to evolve. Finally, a piece of data from the NHS, where the health system is really useful in being able to understand patient persistence. And what it shows here is very much what we expect and what we've been telling you, which is that Mavenclad is very effective, very well tolerated. And that if patients initiated treatment course with Mavenclad, they tend to persist on Mavenclad over the term. And here you see persistence rates over 2 years, basically above 90%. So now if we move towards what about the progress. When we talked about Q2, we told you very clearly that there had been an impact basically on the prescriptions, purely really driven by COVID and the logistical issues that, that presented because physicians were not seeing new patients, physicians were not seeing switch patients, and they certainly weren't going to switch patients onto agents or start patients on agents if they weren't actually able to see them in person. So what we've actually seen now, since basically the low point, is a real recovery, progressive recovery. And our new prescriptions are really regaining momentum. Now it's important to recognize that we're regaining momentum in the context of a higher efficacy dynamic market that is still depressed about 20% to 30% versus the pre-COVID baseline. So the market is still depressed. Our momentum is returning. And a big part of what is continuing to drive our momentum is the fact that despite the COVID pandemic, Mavenclad has continued to grow share versus other higher efficacy agents and grow share versus other oral agents. And you have here basically the share point increases from the pre-COVID period to the post-COVID most recent period. The other point to mention is also that access. Access was a point that we continue to educate you on basically over the last 12 months. And I think we just wanted to draw a line in the sand here now and say that the access position for Mavenclad is great. It's very much on par now with basically what you'd expect for the established DMDs. And we have not at all paid for that access. So we're in a very good position from an access perspective. If we turn our eye to the ex U.S. performance as well, it was similarly impacted by COVID. And what we see here is new initiations have really started to rebound and potentially the leading market in terms of the recovery is really Germany, where the [ Merck ] really did such a good job to control basically the pandemic, where you still have a lot of in-person physician engagement for the industry and also obviously then for patients. And so that's really taken off most significantly, and the other markets have been following suit. What we've also seen is very good performance in terms of year 2 patients. And again, all of this really underscores the continued confidence in Mavenclad and the fact that Mavenclad will continue to be a really solid choice in this pandemic setting. From a safety perspective, the continued data and the continued experience is really showing that Mavenclad's perception on safety, which was an overhang at the time of introduction, is improving. And then finally, what's really important to us as well is that we finally cracked France. So we received a positive recommendation from the Transparency Commission. And in H1 2021, we will basically have our full commercial launch because we will have our price and access and everything, but basically, the recommendation is positive. And so you can look forward to that being a growth driver in 2021. So with that, we're going to trend over to -- well, we're going to move over to the Q&A, but just a few words in summary. Now that you've heard from Peter, from Klaus, from Luciano, we really have momentum. We really have strong momentum across the Innovative Medicines pipeline, and we think this is not one-off momentum. This is growing momentum. We have all of these NMEs with significant potential to differentiate that are now in the clinic and making progress. Since 2017, we basically had 3 medicines approved across 6 integrations across the major markets. We have growing contribution from new product revenue. We have increasing momentum around that, and we have real confidence that we're going to be able to deliver the EUR 2 billion by 2022. It's important also to us that you recognize that we're not just about the EUR 2 billion. We have growth momentum, we think, far beyond the EUR 2 billion by 2022. And I hope that you see that that's underscored by the pipeline that Peter and Klaus so elegantly spoke about. So with that, let me hand over to Abby, the moderator, so that we can get into your questions.

[Operator Instructions] Our first question comes from Jo Walton with Crédit Suisse.

Jo Walton here. I just have a few questions, please. At ESMO, we saw some excellent Amuvatinib data. I just wonder if you could tell us how you think tepotinib is going to fit in with that? On your COVID study at 2Q, you said that you were hoping to have some results by the end of the year. This slide here seems to be a little vague depending on rate of recruitment, et cetera. I wonder if you could just flesh out a bit where you're doing that study, where your latest best guess of when you would have some data surrounding that? I have a question on your MS franchise. And you talk about dynamic share. I wonder if you could just refresh us on what you think the proportion of the market is that is genuinely changing. We hear things about 30% change dynamic opportunity. But given that there are still an awful lot of people who are taking older interferons, it looks like it's a 30%, but the same 30% that changes every year rather than anything more addressable. So just a little bit more update on your view of that, please? And then very finally, I wonder if we could ask Klaus to give his perceptions of the organization and the opportunities that he sees at Merck. He's a person that we used to see quite a lot over at Astra and clearly, you've moved from one oncology department to another. So it would be very interesting to hear your comparisons of what you're finding.

So Klaus, can you go first on the various oncology question and then Rehan.

Yes. Absolutely. I'm happy to go first on the fifth question, Jo, because I must admit I did not exactly hear your first question. But let me go on the comparison to -- one company to another. It's very obviously that Merck is having a very, very significant sustainable pipeline. I think that the whole aim as of today was to show that there has been a very clear transformation in the company if I were viewing it from an external perspective from building some very significant late-stage potential, 2 significant approvals within the last -- number of 8 months. But as Peter very clearly spoke to today, there is, when I take Merck's size into consideration, an absolutely fantastic possibility from the early-stage pipeline that is in the forefront of the science. And I'm not saying that other companies that I have worked for are not in the forefront of science, absolutely not. But there is an attraction for me in absolutely joining a company where there is that focus, ambition, not being pushed by being considered a little broader, but actually being bold and ready to make some significant contributions, some bold risks. So all in all, very impressed. I expect you would likely not have me to say anything different from what I was just saying, but I am actually in the company, I have been voting with my feet and what I saw when I was speculating joining Merck has absolutely been coming to full fruition. That was the fifth one. The first one, I heard that you were saying, you were seeing some excellent data. How does that fit with tepotinib? Sorry?

Sorry. No, it was the amivantamab data I was thinking of and whether that made any difference to your thoughts of positioning your assets in multiple myeloma?

No. I don't think it does. I think it actually just substantiates what we are doing. I think it's very important to say that the amivantamab data is a little bit of a different approach because it is actually adding on to see whether going in a treatment setting can get into, first of all, higher efficacy and, obviously, also, hopefully, a longer duration of that efficacy. Whereas our positioning of the tepotinib program, if you speculate in non-small cell lung cancer, it's much more a positioning as a rescue medicine, if I can speak like that. And that is by no means to say that, that's not a critically important element because if you look at the number of patients that would be in need of a rescue medicine, irrespectively of whether you have amivantamab in the equation or not, is going to be significant in years to come. I mean if you look at the Tagrisso FLAURA data with median PFS of around 24 months, it is about now that you would start to see patients in need of a rescue medicine and that's why we are proving such a significant effort into trying to understand whether the tepotinib combination would be the right thing to do there. And as I hopefully showed you, we kind of have that proof-of-concept that tepotinib is a very clean, very safe molecule to use in that setting based on the INSIGHT 1 data. I hope that addresses what you were after.

Yes. Klaus, that seemed very comprehensive. And let me just add on to the point around the MS dynamic share. So I think -- let me make sure I've tried -- I've gotten this right, but I think the question was really how big is that pool? Is it really going to continue to be a big pool? And therefore, it's a good source of business and can we count on that really for growth? And I would say the key point is that the dynamic opportunity continues to actually be significant and continues to actually be relatively stable. Traditionally, when you basically had new introductions, what you saw was a pretty big increase, actually in the dynamic opportunity because it would drive a lot of switching. But what we've seen more recently is because there are so many drugs, that there are so many different mechanisms that new introductions don't tend anymore to really drive up the dynamic opportunity. But what they do is basically sustain it. So we've seen a pretty good stability, I think, in terms of the dynamic opportunity. What we are seeing, though, is continued increase, of course, in terms of the high efficacy share. So when we say high efficacy, we're thinking the S1Ps, Mavenclad, the anti-CD20s, Lemtrada, Tysabri. We're seeing an increased share of the higher efficacies in terms of utilization in the dynamic segment. So that's why we look importantly at how are we doing relative to other higher efficacy agents in terms of dynamic share accumulation.

Let me just -- I just want to complete the final question very quickly. The COVID-19 trial with TLR7/8 inhibitor just starting, we are recruiting in several countries, just recently activated Brazil and we started to see good activity in Brazil and some U.S. societies. So it's too early for me to speculate on how quickly we'll go. I just want to make one quick point. As you probably understood from the brief presentation, we believe that if the TLR7/8 is a center for RNA viruses and trigger these inflammatory responses, in particularly cytokine storm, this is for us another opening for an additional venue for this novel first-in-class compound that goes beyond out immunity. Thank you for the questions.

We will take our next question from Richard Vosser with JPMorgan.

Three, please. First question is on the differentiation of your TIGIT relative to the Russian Merck TIGIT. Second question is on the Merck 1 EGFR ADC. Could you remind us of the prevalence or the percentage of lung cancer that is expressing Merck 1? I think we had that in our minds sort of seeing these few years ago, but what is that? And then just maybe your thoughts on how this might look efficacy wise, positioning wise versus there's a couple of ADCs that are slightly ahead, the top 2 potentially the cCAM wise. So just your thoughts on positioning here? And how quickly you can move it to avoid having to do trials against another ADC rather than chemotherapy? And then final question just on Mavenclad. And just your -- if you could give us an idea of the proportion of patients that are naive to treatment at the moment versus those returning and how that has developed through the COVID pandemic crisis? And one aligned question, just if you could give us an idea of other launches beyond France that you might have through 2021 and '22? And how big the French market is relative to Germany?

I'd like Peter to start, and then we go to Rehan. Peter, please.

Yes. Let me take first the TIGIT questions. And rightly so you asked the MSD and Roche question because we believe those 2 are the relevant competitors also with ADCC activity, which we believe is important field of TIGIT. Let me say so much that we, as I mentioned, always profile wherever possible our molecules against competitors, and we feel very confident about the robustness in preclinical models, obviously. How this translate into clinical development, we just have to see. We are in -- we started the clinical trial, but I think it's too early to speculate about the differentiation in the clinic. But we have good hypothesis based on preclinical data. And then maybe let me take the first part of the question on the expression of Merck 1 EGFR. Actually, of course, the answer, it depends on which level of expressions you need for this bispecific antibody. And actually, that is relatively low in terms of receptor density on sales for both targets. And therefore, we believe, actually, it's a significant part of non-small cell lung cancer and, of course, also other indications that express enough Merck 1 and EGFR to trigger this internalization process that we believe is so important for the specific activity and selectivity of this bispecific ADC. And maybe on the second part of that question, comparing versus chemo or ADCs in the clinic, I would hand over to Klaus.

Yes. I mean very clearly -- I mean you mentioned yourself is that the main competitor landscape here is like the cCAM. And I think -- I mean, we do not have any clinical data yet from this program. But we kind of know what we are up against. I think what we are up against is very clearly, very efficacious cCAM from Sanofi, but only the cCAM very high expressions that would actually narrow down that patient population pretty significant. So I think there are absolutely room to play there. In a chemo combination, I think the aim there is primarily to see whether you, first of all, in that setting would be in a position to develop some new innovative ways of not having to use chemotherapy. But I mean, as Peter alluded to, it's early, it's a very interesting new concept of selectivity and uptake for EGFR-driven bile site tumors, that clinical data drivers. But I clearly think that there is room based on what I have seen up to now, especially the cCAM.

Before we go to Rehan I want to -- this is -- I want to add, Richard, that there is an excitement about having the bispecific, in this case, driving potential better therapeutic index and much higher potency because of the selective internalization. The biomarker that you select and, therefore, the cutoff for both Merck 1 and EGFR within the tumor is going to -- that cutoff is going to determine the prevalence. I think Peter explained that, but I wanted to clarify. So the potential is very high. We did a lot of work preclinically to understand what was the ideal biomarker, but we're not going to reveal in detail how the selection would be done for that trial. But it's an exciting, early but very innovative approach to ADC to improve therapeutic index and increase potency in the selected population. Let's go back now to Rehan.

Sure. Thanks, Luciano. So Richard, there's 2 questions, really. One is sort of the split of patients in terms of where we're sourcing them from. So just as a reminder, in the U.S., basically, all of the patients that are actually initiating Mavenclad or most of them and, anyway, all of the patients that are actually commercial patients have received at least one prior treatment and that is consistent with the label. If you go ex-U.S., we see a pretty common pattern, which is that about 30% of the patients are sort of treatment naive and about 70% of the patients have previously received treatment. We haven't seen really any significant impact at the moment or change in that based on COVID. All we basically saw was that there was just a general reduction in terms of new initiations basically across the board, whether they were new patients or whether they were switch patients, it was basically the same because, again, it was fundamentally logistics, being able to access the clinic, being able to be subjective to neurological exam, being able to be subjective to MRI. Those are the things that were really the barriers. And it wasn't, therefore, driving sort of a push towards patients that are on treatment versus new diagnosis. So I hope that gives you some perspective. And if you think about the build of Mavenclad over time, we're continuing to gain share. We're continuing to build up new patients. But of course, over time, the importance of the returning patients also to the revenue build increases across all of the markets. And -- but I would say that we're still very much in sort of the ramp-up growth phase, and the new patient acquisition is a really important contributor to the revenue model. I think the -- yes, the other question was France. Yes. So just to give you a sense, I think the best thing to do at this stage is just to index France, the Germany based on population. And of course, there are fewer people in France than there are Germany that's a better way to look at it right now. And maybe in the future, we can give you a little bit more precision, but that should give you a good enough sense of the opportunity. And of course, the rest is going to be driven by the share that we managed to achieve there. And then beyond France, I don't think we comment on some of the specific opportunities, but I will say that we have made Mavenclad accessible and it is approved in many markets. And in total, those markets do start to contribute meaningfully, I think, as we go forward. And of course, the access dynamic in those markets can be different. Some of them come online later, like Italy, like Spain, some of them maybe coming online more in the time of France. So you can expect definitely some good tail contribution, I think, from outside of the major markets as well.

We will take our next question from Marietta Miemietz with Primavenue.

Two questions, please. The first is on Bavencio in bladder. There was some suggestions at ESMO that patients who get IO in the frontline should still be getting IO in the maintenance setting. Is that actually something that you're hearing from the field? And if so, would you expect patients to get switched to Bavencio in the maintenance setting or maybe even receiving Bavencio upfront to avoid switching later? My second question is on tepotinib. And I just wanted to try another tack on Jo's question. I mean MET amplification is obviously a very frequent resistance mechanism in NSCLC and maybe some other cancers. Do you think at a very high level that the community would be willing to give a MET inhibitor before MET amplification even appears, which is something that, for example, J&J is -- seem to be trying with their MARIPOSA trial? Or do you think that, that's really an overkill and the preference would be to wait for MET amplification to appear before an anti-MET is given? And what -- if you get the sense that it's very much turning towards giving MET inhibitors in the frontline setting, would you then actually restructure your tepotinib program to make sure that it can be given upfront? And then my final question is on bintrafusp alfa and the biomarker program for the monotherapy. I was just wondering can you give us a bit more of a flavor as to how you actually select patients that you think will really benefit from the monotherapy in the sense that the cancer is really driven by TGF-beta as opposed to other factors. Or ask differently, given all the TIGITs are coming, do you realistically think that bintrafusp alfa as a monotherapy can be standard of care in some cancers ahead of TIGIT combinations? Or do you think that realistically, you need the bintrafusp-TIGIT combination to compete with other TIGIT cell combinations?

So we'll start with Rehan on the question regarding sequence. Obviously, our indication for Bavencio is going to be in the patients that do not progress after initial chemotherapy. But...

Yes, let me just add a point to that because as Luciano said, our indication is distinct. Meaning that you are going to be treated with platinum, have stable disease or better and then basically receive Bavencio. So in a way, when you're thinking about the treatment you've actually got this being a different segment of patients, meaning the segment of patients that weren't put initially on IO, but instead were actually put on a platinum-based regimen. Now one point that I do just want to underscore in terms of ESMO is because, of course, we've engaged with a lot of advisers post ESMO to really digest the data. And actually, what's really interesting coming out of that is the growing belief that there may be too much single-agent IOUs, basically in the frontline. And as you know, this is -- the initial approvals were based on single-arm studies. And we now have the benefit of additional data sets with single-arm immune therapy and combination immune therapy in the frontline setting. And this is leading people to really question whether or not PD-L1 testing is really useful, and that's part of the indication for the single-use monotherapy. And the other point is, given the availability of avelumab following basically platinum induction and the significant improvement that, that has on overall survival, the median 50% increase from platinum alone, the question is, from all of the data sets now, why would you not be doing that in the vast majority of patients? And why would you be using off-label single-arm immune therapy when it has no advantage at this stage over platinum in terms of improving the median overall survival in randomized controlled trials? So I think what we may see is actually a progressive use of platinum. And of course, this could potentially increase the size of the opportunity potentially, particularly in the U.S., for avelumab. Now back to you, Luciano.