BioNTech SE (BNTX) Earnings Call Transcript & Summary

Great. Good morning, everyone. We're going to get started. My name is Jess Fye. I'm a biotech analyst at JPMorgan, and we're continuing the 43rd Annual Healthcare Conference today with BioNTech. We're going to hear a presentation from the company and then go into Q&A. If you're in the audience and you want to ask a question, someone will bring you a microphone or you can send them to me up here through the portal. But with that, let me pass it over to Ugur Sahin, BioNTech's CEO.

So let me -- let's go back to the beginning. So good morning, everyone here and everyone who is dialed-in. Still Happy New Year. I would like today to update you about our status quo, what we have accomplished last year, and I would like also to give you an update about what we are going to do this year. I would particularly focus today on oncology, particularly on 2 programs, on 2 pan-cancer programs, that means 2 programs that we believe will be applicable to a broad spectrum of cancers. So as always, these are the disclosures. Let's get started. From the very beginning, our vision, when we founded BioNTech, was to translate science into survival by building a global immunotherapy powerhouse. And in 2024, we achieved to come closer to this goal. We executed on our oncology portfolio, advanced into late-stage development with ongoing 15 Phase II and Phase III clinical trials. We presented data on our -- on BNT327, our bispecific anti-PD-L1 anti-VEGF antibody in multiple indications, and we announced starting registrational trials in several indications. We strengthened our corporate position by announcing the Biotheus acquisition, and one of the key reasons for doing that is to get a global control of this antibody and ensure that we can execute fast and efficiently. In the COVID-19 field, we maintained our leading position with a market share -- global market share above 50% of COVID-19 vaccines. And we had -- we strengthened our balance sheet by a combination of a strong financial performance combined with a discipline how to spend our cash. So our vision is to transform cancer medicine into personalized cancer medicine or even individualized cancer medicine, and we believe we have built in the last decade unique capabilities for this. This covers deep genomics and immunology expertise and the development of individualized treatment platforms, AI capabilities and capabilities in manufacturing of mRNA, scaling up the manufacturing and ensuring that this type of technologies can be used at a high scale. On the other side, we built a portfolio of therapeutic modalities, which we can categorize into immune modulators, mRNA cancer immunotherapies and targeted therapies. So this is a view to the future, okay? And what is clearly becoming more and more evident, that the future is about an emergence of technologies. It's emerging into the fusion of biology and AI, and we bring into this several key capabilities, key capabilities in genomics, immunology, biopharmaceutical development capabilities and AI. And on the left side, you see more the application approach in using this to develop fully individualized treatments, particularly our neoantigen-based vaccine platform, where we have several ongoing trials, and they are -- we are continuously learning how to improve the prediction, for example, of neoantigens. We are using computational algorithms now for many, many years to identify novel targets for cancer immunotherapy and increase the target space to establish semi-automated manufacturing. But in the last years, we went beyond that, particularly with our -- with the partnering and acquisition of InstaDeep, we got 2 AI research capabilities. We developed DeepChain, which can be seen as a foundational model, which could become a foundational model for any type of lead candidate optimization, regardless whether this is a protein or mRNA-based drug. We invested into optimization of our mRNA vaccine designs and structures. We established an automated dry-wet lab approach, which means this is the foundation where you can use AI to continuously improve and identify targets and lead molecules, and we built in-house computing, supercomputing capabilities with supercomputing cluster. When it was announced, it was among the first 100 worldwide. So we believe that this will help us not only to bring our current technologies, but also step into the future where drug design will be based more and more in increasing the likelihood that computational design molecules can be directly used without much efforts in further selection and validation. So one of the first convictions for cancer immunotherapy is that we believe cancer treatment or future cancer treatment and, particularly, the desire to create cures will depend on combination therapy. And we have a concept. We built our concept based on combining or based on building the ability to combine immunomodulators, which are increasing any type of existing immune recognition. We have to target therapies, and we have mRNA cancer immunotherapies. And we believe that this complementary mechanism of action can be combined in a synergistic manner, giving us the opportunity not only to control cancer but increase the likelihood for cure. And this concept was consequently and in a focused fashion realized by developing candidates. We have multiple candidates in our immunomodulator platform. We believe in bispecific antibodies because we believe that the combination of 2 molecules, not just the addition of 2 targets, but it's more than the sum of the past. We have focused on targeted therapies based on antibodies, antibody drug conjugates, CAR-T cell treatments, and we have our mRNA cancer immunotherapy approaches based on targeting of individual neoantigens as well as shared tumor antigens. So having this broad platform, of course, comes with the risk. What is your focus? And we have a clear focus. We have identified our key platforms or key programs, allowing us to address our vision. That means we want to address the full spectrum, full continuum of cancers across different stage, from resected cancers, which are in the adjuvant stage and high risk for relapse, to early-stage metastatic cancers as well as the late-stage cancers, which are refractory for different type of treatments. So we have 2 high-priority pan-tumor programs, which we believe can be applied to this type of treatments. These are our mRNA immunotherapies, FixVac & iNeST, and particularly suited as we believe for early-stage or low tumor burden cancers. And then we have our bispecific antibody PD-L1/VEGF, BNT327, which has the potential to become a next-generation IO backbone suitable for a broad range of cancers. And of course, each of these molecules can be used as combination partner for our portfolio of molecules this -- complementary mode of actions. Let's focus on our bispecific antibody. We have now data from more than 750 patients, and this data across multiple indications show that this molecule has a high clinical potential and is associated with really encouraging objective response rate, encouraging disease control and also encouraging early data of overall survival. We have, at the moment, 20 clinical trials ongoing, and we started our first registrational trials in TNBC small cell lung cancer and nonsmall cell lung cancer. So I don't want to go into the details of this molecule because we have presented that in several -- at several occasions. But what I would like to state is really, in a very simplified fashion, why we believe that this type of molecules have the potential to change, to replace the standard of care checkpoint blockade that is present now. So this molecule combines as other molecules of this bispecific class of PD-L1 blockade and anti-VEGF, the ability to block PD-L1, PD-1 interaction, and this is true for this molecule as well as for the PD-1 class. And the second function is it neutralizes VEGF. But you might be able to accomplish that by just adding these 2 molecules. So having a bispecific molecule gives the opportunity to have a cooperative effect. So that means that one arm does something or 2 of this -- one specifically does something and the other benefits from that. And we call this cooperative binding. And this is well described for a number of bispecific antibodies, and this is true also for this type of bispecific antibody. So that means binding to -- for example, of anti-VEGF can position the antibody also in a certain tumor location to bind to PD-L1. What we believe is why this antibody is differentiated is that it has not only one tumor binding pattern, but it has also a second binding arm. It's the anti-PD-L1. And anti-PD-L1 binds to tumor in the tumor microenvironment, not only on tumor cells, but also to a number of other cells, here, for example, infiltrating macrophages, infiltrating dendritic cells or immunosuppressive cells that are infiltrating a number of tumor cells. And by this, we are able to position, by PD-L1 targeting VEGF neutralization in the tumor microenvironment, and we believe that this could make a difference. So let's look into the data, why we are excited, so if we just look to the objective response data. What is really interesting about this molecule, and this is an example that we show here in -- from a study in first-line triple-negative breast cancer, metastatic triple-negative breast cancer, is that the objective response rate is encouragingly high. But what is also interesting is that we see this high objective response rate irrespective of the PD-L1 status of the tumor. So we have here now a case where tumors appear to respond high in a strong fashion, even though the tumors are negative. With that, BNT327 has the potential to become a backbone for IO therapy, not only for the PD-L1 positive population, which is treated by so many different type of PD-L1 molecules, but also for -- is my time already over? So I have now to reboot, sorry. And we believe that this opens up really the opportunity to address also tumors that are so far not responding to IO treatment. And this is also visible not only from the objective response data, but also from the progression-free survival data as well as the encouraging OS data, indicating that this type of molecule, regardless of the PD-L1 status, provides a durable control of disease. What does this mean? And if we see now the cancer indications, we can -- one way of seeing that we can divide the world into indications where anti-PD-1, anti-PD-L1 treatment is approved. These are the PD-L1 positive indications. And if you take U.S. and Europe, there are about 1.5 million cases every year in this indication. And even though checkpoint blockade is approved, that does not mean that the medical need is satisfied. Most of these indications have a 5-year overall survival rate of significantly lower than 50%. So there is a truly high medical need in this population. And then we have the other group of indications that PD-L1 or PD-1 treatment is not approved, and this is around -- yes, it turns out that this population is also around 1.4 million cases every year. And the medical need here is even significantly higher. And we believe that BNT327 has the potential to act as a backbone also in this population of patients. So how do we want to address that? So we see that this is -- this, of course, means a huge opportunity, but also a huge challenge. And we designed our strategy in a way that we establish 3 waves of clinical registrational trials. The first wave we already started is the Establish phase. The idea is here to replace the current standard of care by combining BNT327 with chemotherapy backbone and ensure that we can do that head-to-head against the current standard of care. The second wave, which we already started in the second half of 2024 with first exploratory studies is the combined phase. And we want -- in this phase, we want to combine our ADC portfolio with several candidates with BNT327. The idea is here that the combination of BNT327 with ADCs might provide an even longer control of diseases or open up indications in which standard chemotherapy is not so successful. And the third phase would be -- or the third wave would be all kind of combinations. So this is the potential to go into multiple indications to combine with our other IO bispecifics, combine with our cell therapies or combine with novel ADCs that are in development. So based on this concept, we have multiple readouts expected in 2025 of ongoing studies, either as monotherapy or as a combination, which provide -- will provide the foundation for registrational studies. So now come back to the second platform, which is our mRNA immunotherapy platform. And the idea of an mRNA immunotherapy or mRNA cancer therapeutic cancer vaccines is to generate new T cell responses because we know that the immune system is -- or immune activity is depending on immune responses. And even though this task is defined or the concept is defined now for more than 50 years, cancer vaccines, it's incredibly hard to do this because it's not only about inducing some sort of immune response, but inducing an immune response that in the quality and in the quantity is sufficient to deal with billions of tumor cells, which are left in the body. It's a numbers game. So we spent many years in developing 2 platforms that are -- that have an extremely high potency because they are able to induce immune responses against different type of targets. On the top, you see our iNeST1 platform, the individualized neoantigen-specific immunotherapy platform, that follows the idea that everyone, every patient has a different set of neoantigens. And for treatment with this platform, we need to get the tumor of the patient. We do sequencing, we identify the mutations and then we produce a vaccine, which is tailored to the mutational profile of the patient, and we use AI and computational approaches to select the best mutations. And then the vaccine is produced in real-time and procured. And we have a second concept, and second concept is based on shared tumor antigens. That means antigens that are expressed in a wide variety of cancers, and we combine them to cover more than 90% of patients. And we developed a platform that allows us to combine these antigens as cocktail of mRNAs, and this can be used, for example, target tumors like nonsmall cell lung cancer, melanoma, prostate cancer or breast cancer or other type of tumors. So these are our 2 platforms, and these platforms were developed. We generated safety data. We generated immunogenicity data, and now we have multiple trials running where we combine these vaccines with our in-house molecules. So before coming to explaining a little bit more in detail our personalized vaccine approach, I would like to share one conceptual slide that is really important to understand why we believe that mRNA vaccines have a huge potential in the future. So this graph on the bottom right shows different type of indications. And every dot here is a sequencing -- the determination of the number of mutations in an individual patient. And you see on the left side, you have tumors like colorectal, microsatellite instable colorectal or skin melanoma or lung cancer, smoker cancer. And you see a lot of patients having really high number of mutations. And on the right side, you see tumors like thyroid cancer, prostate cancer, ovarian cancer, where the patients tend to have a lower number of mutations. And there is a clear association between the numbers of mutations in individual patients and the response to immunotherapy because T cell neoantigen recognition turned out to be critical for an effective anti-PD-1 treatment. That's the reason why we are seeing that this type of tumors on the left respond so well to checkpoint blockade. But this also gives you an explanation why not all patients in this -- with these tumors respond because you see a number of patients, even though having a lung cancer, have a lower mutation count. And the reason why these patients are not responding, and this is the current concept in immunology and tumor immunology, is that only 1% to 2% of these mutations have a spontaneous immunogenicity. That means if you -- if the patient has 100 mutations, about 1 to 2 of these mutations will be immunogenic. And this 1 to 2 mutations provides the basis for anti-PD-1 treatment to be -- to drive the antitumor response. So our personalized cancer vaccines address the aim that we want to increase the likelihood for getting immune responses in the cancer patients. And the way how we do that is we generate vaccines, which encode multiple mutations. Here, you see our neoantigen vaccine mRNA design where we can place multiple computation selected mutations. The mRNA vaccine is optimized for expression in dendritic cells to stimulate not only the antigen presentation, but also the innate immune response. It is a SAF adjuvant that -- mRNA, which induces Type 1 immune response, and this creates -- really drives T cell response. And we have recently -- a few days ago, we have published our large Phase I study in more than 200 patients where we analyze the safety, identified the safety and the dose for the vaccine alone and in combination with anti-PD-L1. And the observation from the study is really very encouraging. We see the induction of T cell responses across all type of advanced cancers. And this is on the left, so colorectal cancer, TNBC, melanoma, urothelial cancer, nonsmall cell lung cancer, renal cell cancer and other tumors. On the right bottom, there is a little bit complex, but really exciting diagram, which -- where we analyzed the blood samples of the patients before vaccination and after vaccination. And we see that after vaccination, we have 2 groups of immune responses. We have on the one side -- this is the top colored circles. These are T cell specificities, neoantigen-specific T cells that were present before the patient got the treatment, but which were amplified, factor ten to one hundredfold, after -- due to the vaccination. And even more exciting is the circles in the bottom because these patients or this tumor-specific T cells didn't exist before vaccination, but were de novo generated. So that means we have an evidence that, in multiple tumor indications, we are creating now the neoantigen-specific T cells, and the key question is whether this translates really into clinical activity. We believe that doing this in advanced cancers is not yet the right time because we are talking about really a numbers game. In advanced cancers, the tumor -- the patients have tumors in the range of 10 gram, 100 gram, even kilogram-wise and the tumor cells divide before the immune system is built. But we see that in patients with low tumor burden, particularly in the adjuvant setting, this might be an extremely powerful approach. Why? Because the tumor mass and the number of residual cancer cells is relatively low. We have the situation that resistance mechanisms like clonal heterogeneity and immunosuppressions are not yet fully established, and we also deal with a healthy immune system where T cell responses can be created in a higher frequency. So we have identified a number of indications, colorectal cancer, pancreatic cancer and muscle-invasive urothelial cancer where we have ongoing randomized clinical trials. And these indications were selected also based on the high medical need in patients who are supposed to have adjuvant cancer. And I will give you preliminary data from this ongoing trial. So this is a Phase I study in pancreatic cancer already published, where we did the learnings. Pancreatic cancer is known as a tumor type that even after surgery, around 70% to 75% of patients relapse within 5 years. The reason is that pancreatic cancer rarely can be removed completely and the residual tumors go up and even the chemotherapy provide metastatic relapses. So what we have seen in the study, it was a small study with 16 patients, that half of the patients responded with an immune response and half didn't have a strong immune response. And those who responded strongly had a much better control of tumor relapse than those who did not have a strong response. And this is a learning. This is a learning that we have seen also in other studies, but this is pretty impressive because we have now data of more than 3 years follow-up, showing that these 2 populations separate. And we started a randomized Phase II study in pancreatic cancer patients adjuvant stage and did also take the learnings from this Phase I study to increase the likelihood for immune responses in this study. A second study, which we have started already in 2021 is colorectal cancer. And you know adjuvant colorectal cancer after surgery, most patients are cured, but around 30% of patients with Stage II and Stage III tumors have a relapse in the first 5 years. And these patients who relapse have really a bad prognosis, okay? So the key question is how can we address relapse? And the answer in medicine for this is, since 30 years, it's about using chemotherapy, different forms of chemotherapy. And no other treatment has been approved in this patient population. So we are doing a clinical trial in this patient population, but we try to further enrich the patients who are going to relapse. And we are using a ctDNA assay in the blood. Patients who after resection have a test, a ctDNA test 4 weeks after surgery and who are positive for the ctDNA test are actually metastatic. They get -- even though radiologically, they don't have metastatic lesions, they got relapses within a year. And this is the light green curve showing that these patients behave like metastatic patients. And patients who do not have a ctDNA positive, they actually do perform better than the overall population. So we designed a clinical trial for ctDNA-positive patients, and we evaluated in this trial before going into the randomized phase immunogenicity testing. And the really encouraging observation is that in all evaluated patients, so this is a study now in 12 patients with adjuvant stage colorectal cancer, we see extremely strong T cell response, then the T cell responses after stopping of the vaccination, these are the vertical lines, remain visible. And this is the idea of personalized cancer vaccines or -- of cancer vaccine per se to induce a memory T cell response, which continues to stay even after stopping the vaccination. This is a trial design. It's a pretty robust Phase II design where patients receive adjuvant chemotherapy and thereafter go -- are randomized into an observational arm or into an arm where they receive a personal vaccine -- personalized vaccine. We expect data in late 2025 and early 2026 from this study. So this is an overview of the ongoing studies where we have already published data or expect data readouts in 2025 and 2026. So summarizing, what are our priorities for 2025. So I shared the focus on our 2 molecule classes, mRNA cancer immunotherapies. We expect multiple randomized Phase II data readouts, and we will execute the ongoing Phase II studies and start new studies also in combination with ADCs. For BNT327, the highest priority is to advance into registrational studies and generate data for BNT327 + ADCs or other combinations. Of course, we are not forgetting our COVID-19 vaccines. We want to maintain our global leadership for COVID-19 vaccines, advance also here, improve our platforms and do next-generation combination offerings. And we will update even though I didn't talk about our infectious disease pipeline, we have really exciting vaccines for deadly diseases, TB, malaria and other infectious diseases where we generated Phase I data, and we'll share that with you. And in parallel, we are building commercialization. Annemarie Hanekamp joined us last year, and we are now in the process in establishing commercial readiness to advance and, in parallel, advance the development of BNT323, our HER2 ADC, plan to have a BLA submission end of this year and, of course, build the infrastructure and the processes to enable commercialization of our upcoming products. So we follow our vision. From the very beginning, we started with the aim to address medical need in patients with cancer and with other severe diseases. We are good placed with the progress that we made this year. We expect, particularly in the next years, a number of approvals in -- for oncology products across a number of indications. And we believe that in the time frame of 2030, we might be in a mode where we truly can exploit our full potential by becoming a cash sustainable, financially resilient biotech company. Yes. Thank you. Ryan, do you want to join?

Great. Thanks for the presentation. And we're joined on stage as well by Ryan Richardson, the company's Chief Strategy Officer. Maybe to start out, you walked through how your PD-L1 VEGF bispecific could be able to address populations where we don't see optimal results from traditional PD-1, for example, the lower negative PD-L1 expressers. But what about in settings where PD-1 does work well, what's your conviction that a molecule like this can deliver superior survival?

Okay. This is a great question. So far, the data that has been published by us and by Summit indicate that the clinical response data and also the PFS data are highly encouraging and better than the standard of care, which is, for example, pembrolizumab or nivolumab or other anti-PD-L1 treatments. And there are 2 types of concern on this data. It's one concern is that most of the data have been generated with patients in China, and we are addressing this by having started now multiple clinical trials outside of China. And we will present updates on this data in 2025. And the second is, based on the historic observation, that bevacizumab, anti-VEGF treatment, has been often associated with improved PFS but did not translate into OS. And I think many of the colleagues here are aware of that. And the question is, is it really going to translate into OS? We believe so because this is what we are observing at the PFS level. It's clearly different than the PFS improvement observed by bevacizumab or anti-angiogenic antibodies. And the second is we have early data, encouraging data, for example, in TNBC, hinting to the direction that control of PFS, control of disease translates into a better OS. We will get more data on this in 2025 and 2026. But we work under the assumption that everything goes into the same direction, the objective response rate is higher, the PFS is higher. The patient -- the drugs are well-tolerated and first OS signals or first -- even though we do not have median OS, we have OS values for 15 months, 18 months, 21 months of observation time, which hint into a superiority to -- as compared to historical data.

Great. And then maybe turning to iNeST, where we could get the randomized adjuvant colon cancer data by the end of this year, maybe early next year. Would that data set potentially be sufficient to support a filing for accelerated approval? And if it is, would you be ready from a CMC perspective to take it to market?

I think we must be ready. Whether it is approvable, it's always a discussion with the regulators and, of course, a question of the effect size, but at the end of the day, the power of the study. It's a pretty large study with more than 300 patients. And if we see a strong signal there, we would not be reluctant to talk to regulators about this.

Great. Well, I think we're out of time. So thank you.