Good day, and thank you for standing by. Welcome to the Transgene Webcast and Conference Call. At this time, all participants are in a listen-only mode. After the speaker's presentation, there will be a question-and-answer session. To ask a question during the session, you will need to press star 1 and 1 on your telephone. You will then hear an automated message advising your hand is raised. If you wish to ask a question via the webcast, please use the Q&A box available on the webcast link at any time during the conference. Please be advised that today's conference is being recorded. I would now like to hand the conference over to your speaker today, Lucie Larguier. Please go ahead. Thank you very much. Hello, everyone. Thank you for joining. I'm Lucie, Chief Financial Officer at Transgene. Today, the Transgene team has the pleasure and honor to host Professor Christian Altensmeyer in order to discuss the positive phase I results obtained with our individualized cancer vaccine, TG4050, that were just presented at ASCO last week. Before I turn the call to Alessandro, I'd like to remind everyone that today's discussion contains forward-looking statements, which are subject to a number of risks and uncertainties. You'll be able to ask your question via the conference call or through the internet webcast. In this case, we'll be happy to read your question. Most of you have probably seen Professor Christian Altensmeyer's name in our communications as he's been involved with TG4050 from the very first hours. Christian is a distinguished medical oncologist, immuno-oncology researcher, and lead investigator of the TG4050 head and neck trial at the Clatterbridge Cancer Centre in Liverpool. On the line representing Transgene, you will be also able to discuss with Dr. Alessandro Riva, our Chairman and CEO, and Dr. Emmanuel Dochy, our Chief Medical Officer for NOVAC from ASCO. Today, after a short introduction, Professor Altensmeyer will discuss the data presented at ASCO, the unmet medical need, and the current treatment landscape for patients suffering from head and neck cancers. Alessandro Riva will outline the next step in the development of the company and NOVAC program. Sorry about that. I now turn the call over to Alessandro. Thank you, Lucie. Good afternoon and good morning, everyone. We would like to focus today's conversation on the progress of TG4050, that you know very well is our leading individualized neoantigen therapy that is currently in the treatment of early-stage operable head and neck squamous cell carcinoma. We would like also to put our data in the context of an emerging innovative approach in head and neck cancer. As you know, and I'm on slide six, individualized neoantigen therapies are emerging as potential breakthrough therapies and could be considered game changers for early-setting treatment across potentially a large spectrum of solid tumors. The emerging data with TG4050 in resectable head and neck patients represents a significant proof of principle for this field. As you know, we are just back from an exciting and inspiring ASCO, where significant breakthroughs across different approaches were presented. It was really an outstanding ASCO for the oncology community. Three studies at ASCO have marked this important milestone in resectable head and neck cell carcinoma patients. As you know, the two neoadjuvant and adjuvant pembrolizumab trials, the KEYNOTE-689, and the adjuvant nivolumab trial, the Nivopost trial, support the role of immunotherapy targeting PD-1 as a standard of therapy for early-setting head and neck. Of course, you know the community is waiting for the regulatory authority's approval related to these two important studies. In addition, our randomized phase one study comparing TG4050 as monotherapy supports a potential critical role of our individualized neoantigen therapy in reducing further the risk of relapsing that remains very significant in this patient population despite the very good data from pembrolizumab and nivolumab. We believe that there is the potential for TG4050 to reduce the risk of relapse in a broad spectrum of solid tumor patients in the context of the perioperative treatment with checkpoint inhibitors like non-small cell lung cancer, gastric cancer, and melanoma. A quick snapshot for TG4050 development status is summarized still in slide six. We achieved significant progress over the last few months: positive randomized phase one study in early-setting head and neck, ongoing phase two study in the same setting, a lot of discussions at ASCO on potential future steps to accelerate the development in head and neck and potentially in other indications, and of course, the continued optimization of manufacturing that is critical for an individualized therapy like TG4050. All of these achievements together put Transgene in a very good position to be considered a key player in the field of individualized neoantigen therapies. Slide seven depicts the key unique components of the NOVAC platform, where TG4050, the leading NOVAC-based cancer vaccine, comes from. The approach is based on the right vector, the right target, and of course, the right patient population. The combination of a viral vector based on Modified Vaccinia Ankara and up to 30 neoantigens selected using the ANC artificial intelligence tool, whose sequences are integrated into a vector based on a proprietary tool developed by Transgene, are a critical component of the platform. The randomized phase I data that we presented at ASCO in early-setting head and neck represents a clear proof of principle of this platform. 100% of patients are relapse-free at a median follow-up of 30 months versus 81% in the control arm, and there is a sustained T-cell response against the predefined neoantigen at 24 months. The randomized phase two study, as I said, is ongoing, and there is a real enthusiasm from the community. The study was initiated in May 2024 and is expected to randomize the last patient at the end of the year 2025. The immunogenicity data will be available at the end of 2026, and the efficacy data will be available at the end of 2027, beginning of 2028. There may be an opportunity to accelerate late-stage development with an appropriately designed registration trial vetted by the head and neck investigators and, of course, by the health authorities. The objective is always to prevent the relapse that unfortunately still occurs despite the introduction of checkpoint inhibitors. Lastly, NOVAC platform has the potential to have a wide applicability across a variety of solid tumors with a second clinical program due to start this year. Now, I would like to hand over to Christian, who will walk through the oral presentation made at ASCO annual meeting in Chicago. Christian. Thank you so much. Can you hear me? Yes. Brilliant. I'm a medical oncologist. I look after patients with solid tumors, specifically head and neck and lung cancer, but I'm also an immuno-oncologist, and I have long been interested in cancer vaccine development. It gives me real pleasure to give you my perspective on how these data that Alessandro has already related to sit in terms of the phase one study also, but to give you a bit of a personal perspective on how they sit into the bigger field. Next slide, please. The first step, I think, that we need to take into account is that after many decades of doubt about whether cancer vaccines would have a role to play in immuno-oncology, I believe now we're in a place that we are certain that cancer vaccines can work. I believe, furthermore, that the data are so persuasive that cancer vaccines will become a key element of combinatorial immunotherapy for solid tumors. What we've learned in the last just two and a half years is that the setting matters and that in the adjuvant setting, there is a much lower chance for the tumors to escape the immune system than there is in advanced disease. This supports that cancer vaccines tested in the adjuvant setting are able to demonstrate clinical efficacy, and the Transgene trial is a really good example of how that can be delivered. In advanced disease, I think we still have a fair amount of work to do because we need to understand what needs to go with a cancer vaccine to overcome the adverse tumor microenvironment. It is pretty clear that in combination with anti-PD-1 or anti-PD-L1 or anti-PD-1 plus chemotherapy, there are now multiple examples that these are immune permissive. There are also now data that demonstrate clearly that even in advanced disease, personalized cancer vaccination can turn tumors immunogenic and lead to immune control that is clinically relevant, even in settings where anti-PD-1 alone is not effective. Next slide, please. This takes us to the way the trial was designed, and I think it is a really good example of a really carefully thought-out program of work. At the time, this was very much about establishing the relevance of cancer vaccines and, of course, to demonstrate whether neoantigen vaccines could lead to immune responses. Even at that time, we were very clear that we wanted to test this vaccine in a setting where there was minimal evidence of immunosuppression. At the same time, we were very keen to test this in a setting of high unmet need because that's clearly where the patients will have the largest benefit. That led us to the decision to go into very high-risk head and neck resectable cancer in patients that were disease-free at the time of vaccination. The intent of the study started to be a safety and feasibility study, but I think over the years that this, or a small number of years that this study has been running, we've been very clear that we are able to see early evidence of benefit. The clear path from a clinician's perspective is that this is safe and that it's doable, but also that there seems evidence that this actually does something useful for our patients. Next slide. Let me take you through the presentation that Christophe Létourneau presented at ASCO on behalf of the group of the investigators. Next slide. This was a, I've already mentioned the main takeaway points from this particular piece, and Alessandro has referred to them. It was to find out whether using Modified Vaccinia Ankara as the vaccine backbone could be taken advantage of, building on the enormous experience that the Transgene team has with this approach, and whether this would be feasible and whether it would be safe. Safety at the time was much less clear because we did not really have much evidence or much understanding of what neoantigen-specific T-cell responses might do. We chose to do this in HPV-negative resectable head and neck cancer that had lots of features that predicted for poor outcomes, so locally advanced disease with a chance that we estimated of about a 50% likelihood of recurrence at two years. What we've learned is that the TG4050 vaccine, in even a small study that is randomized, so 16 patients in each arm are ultimately available, improves disease-free survival with no recurrences in the vaccinated arm. For me, as a clinician, this is astounding since some of the patients who ultimately wound up in the study had had multiple and accelerating recurrences in between interventions, surgical interventions. That these patients remain disease-free on the trial is astounding from a clinician's perspective. What we also find is that TG4050 induces very nice T-cell responses against the neoantigens, identifying that the way to assemble the vaccine identifies immunogenic targets, and clearly that is one of the cornerstones of a neoantigen vaccine that actually delivers the kind of immune responses. As Alessandro has already said, these T-cell responses persist to up to two years through the last follow-up. If we can move on to a little bit more detail of the data, we have, until this study has reported and until the most recent two studies that Alessandro has also referred to, remained in a place where unresectable disease is offered chemoradiotherapy in advanced disease, whereas in resectable patients, surgery followed by chemoradiotherapy is the standard of care. Anti-PD-1 treatment or anti-PD-L1 treatment had, by the time up to very recently, really not played a role, but two randomized studies of about 600 patients each have just identified that maybe there is a role for neoadjuvant followed by adjuvant pembrolizumab or by adjuvant nivolumab in the two studies referred here. Next slide. This is important because we now know from a number of examples that this would be a great setting to combine a cancer vaccine because many studies have now identified that this is not only safe, but also demonstrate that adding a vaccine to anti-PD-1 can induce significant extraclinical benefit with neoantigen vaccines, but also with off-the-shelf vaccines. TG4050 is a really good example of this. It's a virally-based personalized cancer vaccine using the Modified Vaccinia Ankara. I believe that this is really clever because our immune system is targeted and developed to deal with viral challenges. To use a viral vector that has all the goodies that the immune system finds interesting makes a lot of sense to me from an immunotherapy and immunology perspective. The vaccine is made personally for each patient, taking resected tissue from the patients at the time of their surgery. It includes up to 30 MHC class I and/or class II peptide antigens that come out of the gene sequencing from the patients. We expected and did actually see that the vaccine induced T-cell responses to these antigens and that this would not be hampered by immune responses against the virus backbone, and the vaccine would be easy to give subcutaneously on six occasions, initially six times weekly, followed by three-weekly dosing up to two years. Please, next slide. The study was focused, as I said, in two stages. We picked patients clinically and radiologically who would be at high risk. We excluded HPV-positive patients because these patients usually have a very good outcome, only 15% or so relapse, whereas up to 60% of the patients with HPV-negative tumors relapse. The patients were pre-screened at that point and then were consented to participation in tissue collection. They then continued to receive their ongoing surgery followed by adjuvant radiotherapy and/or chemotherapy and were then reassessed after the completion of this treatment. If they remained radiologically disease-free and if they had pathological disease at stage three or four and had successfully recovered from adjuvant radiotherapy, they were allowed to be randomized in the trial. Everyone had a vaccine made, so all the patients in this study passed the feasibility criteria, and patients could only be randomized once a vaccine had actually been made successfully. Half of the patients received the vaccine, and half of the patients were followed only with an intent to offer them the vaccine at recurrence. The endpoints were, as I've already mentioned, safety, feasibility, disease-free survival, and then neoantigen-specific T-cell responses. Next. We screened and consented 80 patients. Of these, 33 patients were eligible. There were a number of reasons why patients were excluded, the main one being that the patients had early relapses. We randomized 33 patients, of which in the vaccine arm, one patient progressed very early after only two doses of the vaccine, and the minimum cutoff for efficacy evaluation was at least six doses, so that initial running period of six weeks. That left 16 patients in the vaccinated arm and 16 patients in the observation arm for the comparison. Next slide. The data includes all randomized patients. As I've already said, the numbers are what we've got here. The safety population included anyone who had at least one dose of the vaccine, and the protocol evaluation pre-specified was that they needed to have had at least the six doses to be evaluable for the primary endpoint of disease-free survival. Next. The cohorts are reasonably well balanced, as you would expect for very small numbers. I won't belabor all of the details, but the median age was about 60 years of age. There is the expected predominance of men in the disease, so usually the ratio for head and neck cancer is two to one for men and women, respectively. The HPV-negative tumors were the focus, so that means that there is a predominance of oral cavity tumors, so mainly tongue-based tumors. You see there are a reasonable number of other disease sites, including tumors with a very bad outcome, such as a hyperthyroid tumor. The clinical stages were reasonably well balanced. Do move on to the next slide. That is fine. Thank you. You can see that the key risk factors that we would clinically predict, and particularly the accumulation of the risk factor, to some degree favored the adjuvant arm. In other words, the clinical effect in the randomized cohort favors the effect of the vaccine because that is the group that remains disease-free, but also there are more patients with a higher number of risk factors. Of course, we are aware that this is a very small randomized study, so there is always clearly some variability because you never probe the population accurately in only 16 randomized pairs. It is a multicenter study and two-country study, so the data do represent the European perspective in France, as well as the choices were made in the U.K. Next slide. The vaccine was really well tolerated. There were a fair number of injection site reactions. There was rash. There was one patient who had what looked like an urticarial reaction. We saw all the expected influenza-like syndromes. The incidence was relatively lower than we had expected, and there was not really anything that makes this vaccine stand out as medically difficult. The administrations were easy. There were no significant or durable injection site reactions. For the physicians and for the patients, this was very straightforward. Next, please. These are the disease-free survival data, so the first efficacy analysis. As we've already said, none of the patients in the vaccine arm have relapsed. A number of the patients in the observation-only arm have relapsed. Perhaps overall, the outcome of this cohort is better than we had predicted because we aimed to pick very high-risk patients, but nonetheless, this is what the overall data are. Next slide. These data show some representative examples of immune evaluations. The lines, particularly the dark blue lines, are the cells that are CD8-positive T-cells that are neoantigen-specific. You can see them tracked over time for these seven patients. It's a little bit small to really read the data, and I'd be very happy to take the data and discuss them individually if you were interested. The key takeaway message is that in every patient we looked, we found T-cell responses, and they seem to be durable well beyond the duration of vaccination, which was one year, and to the end of the two-year follow-up. This is quite intriguing because we had assumed, and the literature had suggested, that if you stop and cease vaccination, that eventually the T-cell responses would disappear. That does not seem to be the case, and it will be very interesting to see how long these responses may continue in further follow-up. Next slide. The conclusion is that the vaccine is feasible. It is safe. It induces long-lasting immune responses. Perhaps I should have added that the circulating T-cells had all the right features of the cells that you would hope to induce, so cells that were able to express cytotoxicity molecules, and we would have predicted that these will be the circulating counterpart to what's going on in the tumor. Of course, these patients are disease-free. That's how we picked them, and the patients who were vaccinated remain disease-free, so we cannot answer that specific question in this group. We are very clear that the vaccination improves disease-free survival. We are continuing the phase two part of the study, as Alessandro has already said, and this study is well ahead of its target. The recruitment is now finished, and we're now waiting for the outcome data. Next slide, please. We'd like to thank all the participants, the patients, and of course, all the clinical team members that have made this possible. It's been great fun to work with Transgene, and the NEC team clearly have done a fantastic job at identifying the neoantigens. If we now put this—yep, that's great. Move on. If we put this into the context of the immunotherapy generically, but also in head and neck cancer, then in these last two years, the landscape has really changed a lot. Previously, we were giving immunotherapy only to patients with recurrent disease. Some patients with inoperable disease were also offered immunotherapy if they fitted specific pre-defined criteria, but there were a fair number of negative trials with perhaps a trend for anti-PD-L1 to be useful. Inoperable patients, we had up to now given chemoradiotherapy. You will know that there is an international discord between whether to give chemoimmunotherapy or to give chemotherapy alone. In the U.K., the standard remained up to very recently a single combination immunotherapy with a platinum F5FU-based regimen together with chemotherapy. Now, in the more recent data, this landscape seems to be changing quite quickly, and KEYNOTE-689 identifies that in terms of disease-free survival, there is a clear effect of adding anti-PD-1 pembrolizumab preoperatively, followed by, I think it's 12 doses, if I remember correctly, of anti-PD-1 with chemotherapy and then a single agent ongoing. The post-operative nivolumab has also read out as positive. This was post-surgery, give two doses of nivolumab, and then continue with the nivolumab initially three weekly with the chemoradiotherapy, apologize, and then give six monthly doses of anti-PD-1 nivolumab. In both instances, there is an effect. It's somewhat inconsistent whether PD-L1 expression between the two regimens really makes a difference. My prediction would be that it should because we know that anti-PD-1 activates pre-existing T-cells rather than generating new T-cells. In other words, tumors in patients where there is already an immune recognition, in those patients, you can activate an adaptive immune response and release it through anti-PD-1. These biomarker programs that are needed to understand that better are still ongoing, and it's intriguing that PD-L1 may or may not be a critical feature for this. Of course, it is reasonably well understood now that if you have more PD-L1, it is overall more likely that there are more T-cells. It becomes intuitive and is borne out in many solid tumors that there is a link between clinical benefit. Whether one should pick these patients to add anti-PD-1 or whether it should go before all comments remains uncertain. What is, however, clear is that anti-PD-1 serves as a really powerful way to inhibit the off-switch that the cancer provides to T-cells. The correlate then is, if you now use a vaccine to train new T-cells, it makes real sense to give an anti-PD-1 antibody with the vaccine or maybe following the vaccine to prevent the tumors from turning off the T-cells that you've just trained with your vaccine. That is the overall consensus that is emerging in the literature. It will make a lot of sense to probe that biological question directly with TG4010 only. Please move on. It is really quite astounding that beyond the expected safety and immunogenicity, this is the only example that I know of where anti-PD-1 is not required for immunological and clinical benefit. This suggests then that the TG4050 platform is particularly powerful and stands out among its siblings and rivals, suggesting that if you pick the right vaccine—and I have already given you my personal prejudice on why a viral-based and viral vector vaccine has immunological advantages—then you might even be able to deliver this on its own. However, I suspect that as we move into later-stage disease, we will need to think of combining this with modulators of the tumor microenvironment, of which anti-PD-1 will remain the backbone. I think this is, from a clinician's perspective, so astounding because we really do see patients with bad cancer whose head and neck cancer comes back again and again. Of course, the main predictive feature for developing head and neck cancer is having had head and neck cancer. This offers a real hope to our patients from something that is not only deeply mutilating to the patients, but is also such a threat for their overall survival. Next slide. With that, I'd like to hand it back over to Alessandro to take us on. Thank you so much. Thank you, Christian, for this clear overview of the clinical data that you presented us. Of course, its implication in the context of the treatment for early-setting head and neck cancer patients. As you probably now have understood based on this conversation, we aim to try to redefine the outcomes in difficult-to-treat patients, and specifically in early-setting head and neck cancer patients, in conjunction, of course, with other therapeutic modalities like checkpoint inhibitors that are emerging as a potential new standard for this patient population. We are very excited about the randomized phase one data. We are excited about the rapid inclusion in the randomized phase two. Of course, the conversation that Emmanuel and I had at ASCO this year after the presentation of our data, in order to try to accelerate the program in early-setting head and neck in the context of the emerging role of immunotherapy in early-setting head and neck cancer patients. Now we move to slide 29. As I previously mentioned, there is the potential of expanding the role of an individualized neoantigen therapy in a broad spectrum of tumor types. Transgene is now working very rapidly to implement a new trial in new indications that have a different microenvironment from the head and neck cancer in order to demonstrate that this approach is valid across the board. We are going to select a tumor that has a different profile biologically from head and neck, and we hope to demonstrate that even in this second indication, we can induce the immunogenicity to the antibiotics. More importantly, of course, to reduce the relapse in this new patient population. Last but not least, I would like to note that we are investing significantly to optimize the manufacturing of our platform in order to be able to make this process more scalable and ultimately to serve more patients. Now, I guess we can open to the question. Please, operator, I leave you to manage. Thank you, Sal. As a reminder, to ask a question on the phone line, please press star 1 and 1 on your telephone and wait for your name to be announced. To withdraw your question, please press star 1 and 1 again. Once again, please press star 1 and 1 on your telephone and wait for your name to be announced. To withdraw your question, please press star 1 and 1 again. If you wish to ask a question via the webcast, please use the Q&A box available on the webcast. Thank you. We are now going to proceed with our first question on the phone line. The first question has come from the line of Marshall Decutier from Oddo BHF. Please ask your question. Thank you very much. Good afternoon, everyone. Thank you for this clear presentation, and congrats for your first data on TG4050. Very quickly, maybe two or three questions, if I may. Regarding your phase one, which kind of evaluation are you targeting now? Do you think that the trial is sufficiently powered to have statistically significant? Please. My second question is, could you come back on the potential for TG4050 to be combined with an immunotherapy to improve outcomes and the scientific questionnaire today? My last question is, what are the most important criteria for selecting patients who are likely to benefit from TG4050, and how can this be done in the real-world practice? Please. Thank you very much. Yeah. Okay. Maybe I take the first question. The second question goes to Emmanuel, the role with immunotherapy. The third one goes to Christian in terms of putting this data in the context and the selection of the patient population. The first question in terms of the ads ratio. When we combine the phase one and the phase two study together, because remember, this is a phase one to study, we are going to have ultimately more than 80 patients in the trial, and we will have a statistical power to demonstrate a 20% absolute advantage in disease-free survival. It is very significant for this patient population. That addresses, I guess, your first question. Emmanuel, you want to address the checkpoint inhibitor question? Yes. Thank you, Alessandro. I believe it is an important question because when you are looking to the data, even the very new one coming from the KEYNOTE-689 and the Nivopost, not all the patients are cured, and so there is still a benefit. It will be important to combine immune checkpoint inhibitor with TG4050. We have already some experience in the patients who relapsed. Two of them were receiving combination with immune checkpoint inhibitors. There were no safety issues, but we will need more information in order to demonstrate the benefit of the combination. Thanks, Emmanuel. Christian? The question is slightly more sophisticated than just who will benefit. I think it's going to be a question of how many patients we can offer this to. That, I think, should be stratified according to the risk of recurrence. If you have a cohort of patients where the chance of recurrence is 50% or more, then I think all of those patients will benefit or will potentially benefit at least. I predict that even patients who have PD-L1 low or PD-L1 negative tumors will benefit from vaccination because those are the very patients in whom training new immune responses is so critical for enabling an adaptive immune cell attack. The data that we have suggests that in this particular cohort of patients from the phase one study, the patient cohort was very much enriched for those patients in whom we traditionally thought immunotherapy could not be effective. The patients with very low or absent PD-L1 expression, those with a tumor microenvironment that had no T-cells, that had a rich infiltrate of cancer associated with fibroblasts that we now know actively exclude T-cells from tumors. I predict that the main impact of a vaccine and the time it will be or the group that it will be most easily demonstrated in is the patients with really poor outlook traditionally. I think that that is what our data, I predict, will demonstrate when the randomized data have completed and the whole series is available. I just wanted to add to the point that Alessandro made. If you can see a difference in a small cohort, the effect must be big. I think the sample size calculation allows us to capture a relatively modest effect in oncological terms. I suspect that we will pass that hurdle. Of course, for the patients, a significant reduction would be really quite amazing. I think tumors that are low in T-cells, particularly patients who have low quantities of PD-L1, will be the groups that will benefit the most. It is also likely that patients who already have pre-existing immune responses will additionally benefit from the addition of a vaccine. I suspect that if a resource was not an issue, we would want to offer all of those patients with a significant chance of relapse a personalized cancer vaccine of the flavor we just discussed here. Thank you, Christian. I fully agree. We should not forget that our approach so far is in all countries, right? We randomize in phase one and in phase two all patients that go through the standard chemoradiotherapy. This will be very important to analyze once we have the end of the phase two in order to understand whether there is any trend in favor of one subgroup versus another one. That is a kind of important advantage for us so far. Of course, as we think about the next step, the phase three study, we cannot forget what KEYNOTE-689 did. They recruited all comers and then they presented a subgroup analysis. What Nivopost did, on the other hand, is they recruited only the high-risk patients. It is going to be a kind of decision based on the data that we have today, the feedback from the health authority, and what the health authority will decide very soon on the KEYNOTE-689. Of course, the discussion with the head and neck oncologist to define ultimately the patient population that will be recruited in a potential registration or trial. Thank you very much. Thank you for the question. Yes, operator, please. As a reminder to ask a question on the phone line, please press star 1 and 1 on your telephone and wait for your name to be announced. To withdraw your question, please press star 1 and 1 again. Once again, it is star 1 and 1 on your telephone for any questions or comments. If you wish to ask a question via the webcast, please use the Q&A box available on the webcast. Thank you. We have no questions on the phone line, so I will hand back to you for any return questions. Thank you. Thank you. I have questions. Yes. Thank you. I have a series of questions from Jamila El-Bogroni from All Invest. First question is, in the relapsing patients, are the recurrences limited to the primary tumor site, or do they also include distant metastasis in secondary lesions? Emmanuel, please. Yes. Thank you for the question. For the three patients who had relapse, we had two patients who had locoregional relapse plus distant metastasis. One patient was with lung only and was not treated because of the worsening of his health status. The other one was having lung and bone disease. The third one was having only distant metastasis at the lung level. That is the scenario that we were having in this patient population. Thanks, Emmanuel. Thank you. Second question is, it may be too early to extrapolate, but when looking at the two patients from the observational arm who received TG4050 following recurrence, can we already question whether 4050 at the dose used in a phase one trial may have less efficacy once the tumor has reappeared despite its high efficacy in the adjuvant setting to eliminate residual disease? That's a very good question. I propose Christian, if you don't mind, to take it. Yeah. In every clinical scenario with any treatment, but any immunotherapy, the larger the tumor volume is, the worse is the outcome. Vaccine, immunotherapy, and indeed any other solid tumor setting. The answer in that sense is yes. I think the answer more specifically is completely unclear. We have a tiny dataset on two patients with rapidly progressive disease who were given standard of care. I think one of them had chemotherapy only with the vaccine. I don't think anyone had anti-PD-1 with the vaccine. Emmanuel, perhaps you can add to that. I think what we need is the confirmation in the adjuvant setting that this is really as good as it looks on the tin. Then demonstrate in a formal study the magnitude of the effect rather than be led by the individual two or three events that happened where two patients only got the vaccine. I think it's not a really controlled setting, and I do not think that these data are informative. It would have been miraculous if the vaccine in the first attempt in the palliative disease setting had achieved durable benefit. Of course, the study wasn't aimed or powered to make any such assessments. Christian, perhaps some words on what you think about a therapeutic vaccine in early setting versus metastatic, the different tumor microenvironment, the potential resistance that occurs in metastatic versus early setting, whether the opportunity for a vaccine is more favorable in adjuvant versus metastatic. Yes. There are no really robust human data to answer this question other than the observations from trials that have reported in the adjuvant setting. Of course, we do not know for an individual patient whether they are truly going to relapse or not because we have no tools to measure that. We can only look at cohorts of patients and compare the effect. If we start with the adjuvant evidence, then we know that in a very immunogenic tumor, the addition of anti-PD1 in the adjuvant setting in melanoma in high-risk disease roughly halves the residual risk of patients after surgery and after anti-PD1. The preclinical data that exist demonstrate in a metastatic colorectal cancer model that when the tumors begin metastasizing, they initially travel without their tumor microenvironment, and they innest as tumor cells rather than bringing their supportive niche along. They build that niche when they innest and the metastasis grow. There is really intriguing data that suggests, therefore, that the adjuvant setting is particularly good for targeting with immunotherapy because you do not have to address the immunosuppression of the tumor microenvironment or perhaps at least to a lesser degree. I think it is very clear now that the adjuvant setting is the best place to really test the immunogenicity and the protective effect of a vaccine. I predict that as we become better at understanding the tumor microenvironment in advanced or recurrent disease, we will be able to add immunotherapeutics in combinatorial immunotherapy to address the specific needs in an individual patient. The trials will not initially ask this question because they would have to be very large to assess a stratified approach to combinatorial immunotherapy. I predict the next obvious trial would be vaccine plus anti-PD-1 versus maybe the two of them together. I think we will soon see that in that setting also there is an effect of the combination. The melanoma data in this particular instance with the DNA vaccine very clearly identifies that adding a personalized cancer vaccine to anti-PD-1 in a setting where anti-PD-1 is pretty useless, metastatic and recurrent hepatocellular carcinoma, vaccine turns that into clinical benefit. I predict that that will be also demonstrated in a setting like head and neck cancer, particularly given the information that we already have that the vaccine works agnostic of the tumor microenvironment that is there in the primary tumor. Thank you, Christian. Maybe, Lucie, we have still 15 minutes, 13 minutes. Lucie? Lucie, I do not hear you. Christian, do we hear Lucie? I do not know. No. We do not hear you, Lucie. You can see now. Yes. Go ahead. Follow-up question from Martial at Oddo. Follow-up in terms of competition at the ASCO Congress. You hear me? Yes. Yes. We have more data on the Bicara asset evaluating FISERA first in combination with Pembro. How do you analyze this potential competitor, or do you have a view on this asset? How do you see the next step in terms of standard of care in head and neck cancer first line? Thanks to these positive data on TG4050, what have you learned on the therapeutic vaccine enabling to improve the next step of your pipeline and development? Okay. I guess I can start to answer the first question. Maybe Christian, you can take over if there is anything else to add. I mean, we believe at Transgene, and I think this is kind of across the community, that the Merus and the Bicara data with their respective bispecific that are being tested so far in advanced metastatic setting is very promising for head and neck cancer patients, underscoring the importance of bispecific as a new modality for the treatment of solid tumors. Whether or not this approach will be moved to adjuvant setting, of course, we cannot answer because we do not belong to Merus or Bicara, right? Certainly, the fact that these compounds are active and pending further confirmation from their randomized study, I expect that there will be an opportunity to move the compounds in early setting. Of course, there will be potentially an improvement in the relapse-free survival. Still, we think that our vaccine will play a fundamental role in continuing to boost the relapse-free survival. Christian, do you want to add anything on the bi-specific in head and neck? Yes. I think what is unclear at the moment is what the immunological consequences of the bi-specifics are. So TGF-β, as you know, that's the EGF receptor partner in the Bicara antibody, bi-specific, has pluripotent roles. Blocking TGF-β may also have adverse effects in the tumor microenvironment in terms of inhibiting the development of a key group of protective T cells that are called tissue-resident memory cells. They're the sort of repository in our tissues for protection against immunological challenges. Their number is slightly correlated to clinical outcome in pretty much every solid tumor with the exception of renal cell cancer and juvenile melanoma. It is not clear to me immunologically whether blocking TGF-β in this way will really have the desired clinical outcome. Of course, we'll have to see what the data show. In the end, we can have our smart ideas about what may or may not happen, and the data will then allow us to understand whether we were right or not. If the bi-specifics generate a better adaptive immune response, then it is very likely that a vaccine will boost that effect again. The aim of both products is ultimately to remove an inhibitory tumor microenvironment to enable an adaptive immune response or perhaps to boost what is already there. In that sense, these compounds might very well lend themselves to combination with a product like TG4050. I suspect that the added effect will be amplified by the addition of, will be over-amplified by the combination of the vaccine with the bispecifics. Whether that is finally true will need preclinical modeling and then a careful assessment of immunological consequences of the relevant bispecific. I think that will need to be answered product by product. My overall view is if what Bicara and Merus are trying to achieve leads to better immune responses, it will lend itself to the addition to a vaccine. I think that is what we will be seeing in multiple solid tumors, including head and neck cancer. Thank you, Christian. To your second question about the strategy in our pipeline based on the progress of other modalities. Essentially, we are going to be very focused on the individualized antigen therapy because we think that there is the potential for us really to make significant changes in the early setting of solid tumor patients, not only the head and neck. There is also a great potential to continue to be a strong player in this field, of course, with other companies. We have a clear differentiation. Therefore, we would like to just continue the journey on the individualized antigen therapies. In addition, of course, we are thinking at alternative modalities. Of course, we are looking at the off-the-shelf therapeutic vaccine. We are brainstorming as a specific and specifically potential dark neoantigens that is emerging as a potential interesting field in the context of therapeutic vaccines. Still a work in progress. Our focus so far will be therapeutic individualized neoantigen therapies. We have a lot to do in order to continue this trajectory. Thank you. We now have questions from Romeo from Kempen. Some of them we may have partly answered, but still. Could you comment on the number of cases in the control arm being less frequent than expected or standard of care protocols improving over time? How do you compare the two-year DFS with Pembro, KEYNOTE-689, and Nivo, Nivopost up? I think the final question would be, do you have plans for pivotal trials, and are you still on track to initiate development in the next indication, and how do you go about selecting indications? I start with the second question. I leave Emmanuel to elaborate on the first question, Emmanuel, if you do not mind. I start with the second question. Of course, as I mentioned in the introduction, we are assessing with the doctors, the head and neck doctors and investigators, the best trial design to accelerate the development of TG4050 in head and neck. We hope to have something that can be submitted to the health authorities by the end of this year, right? Then, of course, depending on their feedback, we will decide the potential next step, including the startup activities and the exact quarter in 2026 or 2027 when we will start the potential phase three study. That is the plan. As we also mentioned in the introduction, we are preparing a new indication with a new randomized phase one study that also for us is very important in order to determine the scope of this program that we think goes beyond head and neck. Emmanuel, in terms of the interpretation of our data versus the Nivopost and the KEYNOTE-689 study? Yes. First of all, versus the DFS in general, indeed, the historical data are reporting higher relapse than what we have observed, but it was done after any radiation and/or chemotherapy. Don't forget that in our trial, we were selecting the patient who achieved complete response three months after the end of their radiochemotherapy so that is the reason why we may have a lower relapse rate than what has been seen historically. When you're looking to the data coming from the KEYNOTE-689 and Nivopost, you can see that they are really aligned with the historical data in the arm not adding any immune checkpoint inhibitor. That is really reassuring for us to demonstrate the benefit of TG4050 in this setting and in this very particular population where you do not have any visible disease. Thank you for the question. Thank you, Emmanuel. Of course, as we think about the next step, the idea is to mimic more and more the patient population that has been recruited in the KEYNOTE-689 and/or the Nivopost trial in order really to cover the patient population that has not been covered in the randomized phase one study. Lucie, any other question or? Great question. Question from Lionel Labourbett. From a surgical standpoint, which tumors are the most easily resectable? Likely impact on the vaccine efficacy. Could you tell us when the next data survival update is scheduled? I guess it's for the phase one trial. I answered the second question. Christian, if you don't mind, you take the first one. The next update will be ASCO 2026, right, for the phase one study where we will present the three-year follow-up for disease three and also, of course, disease three survival and, of course, the survival. Every ASCO, we plan to update the community on the randomized phase one up until we have the data of the randomized phase two. Christian. The resection rate depends very much on the stage. You need to have sufficient clearance anatomically to be able to remove the tumor and any nodal disease. We group operability based on initially radiological features, usually based on MRI scan and in the absence of any evidence of systemic disease, typically in the lung. If the disease is surgically resectable, then that is what the primary care is, unless there is a medical reason why surgery would not be feasible. That's, for example, coexisting comorbidities. And of course, you know that the patients who develop head and neck cancer are often heavy drinkers and often heavy smokers. These patients do have a fair amount of comorbidities. The second criterion is whether you can close the hole that you make when you've done the surgery. Because, of course, if you cannot achieve a viable outcome, then that is another reason to not be surgically resectable. The resectability depends almost entirely on the size of the tumor and the degree of extracapsular invasion of the lymph nodes. That's what we categorize in the staging. Stage one disease will be more easily resectable than stage three or stage four disease, which by and particularly metastatic disease, of course. Does that answer the question? I think that's probably as much as I can offer. I think it is fine, Christian. Yeah. Lucie, I guess we are on the top of the hour, right? Lucie? Yeah. I think that we can conclude this very interesting webinar where we try to share our enthusiasm and commitment to the head and neck cancer patients to improve their treatment in early setting. Clearly, what I just would like to say is that we feel more and more, I mean, to be part of the individualized neoantigen therapy community. We are with others, but we think that we have something to say and something to propose to the community, specifically in the neck and tomorrow in other indication. As mentioned, we are going to start a new indication by the end of the year. We think that we have a unique technology based on viral vector, based on machine learning and artificial intelligence to select the right immunogenic neoantigens. We're investing significantly in the manufacturing because we believe that it is one piece of the success of individualized therapy to be prepared ultimately for the phase three and later on for a potential commercialization of this innovative technology. Overall, what is important, I believe, and beyond everything, we think and we want to make a difference for patients. Having said that, we look forward to continue to update you on our progresses and the progress of the scientific community. Thank you very much for having participated in this call, Christian, and of course, Emmanuel, Lucie, everyone that contributed to this call. Talk to you soon. Have a great afternoon or evening. Thank you so much. Please conclude today's conference call. Thank you all for participating. You may now disconnect your lines. Thank you and have a good day.