Welcome, everyone. At this time, I would like to welcome you to the Innate Pharma Industry and Analysts meeting. Today's conference is being recorded. All lines have been placed on mute to prevent any background noise. After the speaker's remarks, there will be a question-and-answer session. At this time, I'd like to turn the conference over to Jonathan Dickinson, Chief Executive Officer of Innate Pharma. Please go ahead. Good morning, everybody from here in New York City. We're delighted to be here on behalf of Innate Pharma this morning to basically share with you the latest and greatest from a Nectin-4 ADC perspective. Before we do that, I'd like to, if we can go to the next slide, just take you through a couple of things. First of all, we have our standard disclosure statement. If we can go to the next slide. This is the agenda we have for you today. I'm going to give a quick introduction. I'm then going to hand over to Yohann Loriot, who's joining us today from the Gustave Roussy Institute in Paris. He's an expert in the Nectin-4 space, and he's going to take us through some of the latest from a Nectin-4 perspective. I will hand over, we'll go on to Yannis Morel, who's our Chief Operating Officer, and Yannis, will take us through all of the latest preclinical data from an IPH4502 perspective. We will hand over to Sonia Quaratino, and Sonia, is our CMO. Sonia will take us through the clinical development plan for IPH4502 and the latest in terms of where we are today. If we can go to the next slide. Just to remind you, before we get into Nectin-4, we have three pillars to our strategy at Innate Pharma. We announced this at the beginning of the year in terms of refocusing our strategy. Those three pillars are based around our anchors. They're based around ADCs, which is what we'll be focusing on today, and then our current late-stage assets. If we move to the next slide, you can see where we are overall in terms of takeaways from an Innate Pharma perspective. Today we have actually eight products in the clinic, a mixture of partnered and proprietary assets, a mixture of late-stage and earlier phase assets. We're driving forward our ANKET platform, our NK Cell Engagers. Recently, we've initiated a collaboration with iFly, which we see as a validation of what we're doing in the lymphoma space with IPH6501. Also, our ANKETs, are moving forward with our partners, with Sanofi. We are in phase I and phase II with the lead asset IPH6101. We'll talk about ADCs later, so I'm not going to go into any more detail. In terms of late-stage assets, we have Lacutamab, which went through a recent FDA interaction. We have an accelerated path to approval potential. We're working on the confirmatory study, which is advancing very nicely. We're looking to establish a partnership to take that particular asset forward. Finally, we have Monalizumab, which is in partnership with AstraZeneca. We have a big milestone coming up in 2026, which is the primary endpoint, big clinical milestone in 2026. We're looking forward to that. Cash position of the company, we've communicated, audited finances to the end of this year. We're working on basically extending that runway well into mid-2026. That will be announced after we have our year-end results and the audited accounts. On that point, I'd like to hand over to Yohann Loriot, and he can share some of the latest from a Nectin-4 perspective. Thank you so much. Good morning or good afternoon, everyone. I'm Yohann Loriot, I'm a medical oncologist, a professor of medicine here in Paris at Paris Saclay University. I'm the Deputy Chair of the Department of Early Drug Development at Gustave Roussy. One of my interests, I would say, is the gene oncology. I was involved very early in the development of many ADC, including enfortumab vedotin, which is relevant for today. As you know, ADC is a new technology to administer chemotherapy, but also any other agents. There are basically three different pillars, three different components of ADC, the antibody, the linker, and the payload. Historically, most ADC were constructed based on tubulin inhibitors. There is now a new trend. Since 2022, the numbers of topoisomerase inhibitors based on ADC are going up, and now it's even superior to tubulin inhibitors. Especially now, there is a growing number of exatecan-based ADC, so one component of the topoisomerase inhibitors class. Nectin-4 is a validated target. And why? Because, as you know, enfortumab vedotin, an ADC developed by initially Seattle Genetics and now Pfizer and Astellas, was approved worldwide in one specific disease, in urothelial carcinoma, in metastatic urothelial carcinoma, based on two phase II trials that I will show you in a second. The first and the only one ADC targeting Nectin-4 approved across solid tumors. It provides some huge interest in targeting Nectin-4. As you can see on this slide, there is a lot of new ADC currently under investigation. Basically, there are two classes of ADC targeting Nectin-4. The antibody that includes tubulin inhibitors as a payload and the other class, as I mentioned earlier, the topoisomerase I-based ADC. At the bottom of this figure, you can see here the numbers of ADC with tubulin inhibitors as a payload. Some of them are currently in phase II, phase II trial, like one from China MY2821 and one from Bicycle Therapeutics, the BT8009 or Zelectated Pivototin. Other ADC with tubulin inhibitors payload are currently in phase I and phase II. Now we have more and more, as I said before, ADC with topo one inhibitors, ADC in clinical trial. In a phase I trial from Lilly, from Merck, but also, of course, from Innate. There is only one phase II trial ongoing with the ADC and the topo one payload in a phase III trial. Why there is some excitement and commitment now to develop more and more ADC targeting Nectin-4, as mentioned earlier, we have positive phase II trial with ADC targeting Nectin-4. The first in this class of ADC is enfortumab vedotin, which is approved in metastatic urothelial carcinoma as single agent in patients previously treated with chemotherapy and PD-1 inhibitors, in second line, but also in first line, upfront in combination with immunotherapy. At the top here, you can see the phase III EV301 trial. In second line, patients were treated with EV monotherapy versus all chemotherapy, I would say, like vinflunine, docetaxel, and paclitaxel. Of course, the trial met the primary endpoint. The median overall survival was around 30 months versus around nine months with chemotherapy. Importantly enough, response rate, 40%, for patients treated with enfortumab vedotin versus only around 18%, for the patients treated with the standard chemotherapy. There was a rationale to combine enfortumab vedotin and immune checkpoint inhibitors. It seemed that ADC can induce some immune activations. A phase I trial was conducted five years ago and reported at ESMO 2019, showing that in around 60 patients, the response rate in first line, so in naive patients, response rate could be as high as 70%. Based on this data, a phase II trial was conducted, EV302. Again, we are here in first line, metastasis setting for patients with urothelial carcinoma. Patients were treated with either the combination of enfortumab vedotin and pembrolizumab or the standard of care, so the chemotherapy followed by maintenance treatment with immunotherapy. Here you can see the dramatic improvement in overall survival. As I know, the ratio was 0.47. It is a new standard of care in many countries in North America, in Europe, and worldwide. You can see that there is still some problem. The first problem is the primary resistance. Around 20%-25%, of patients do not benefit from this combination upfront. Number two, the acquired resistance, secondary resistance. You can see here, the shape of the curve, meaning that some patients will develop resistance to this drug, even if it is very active initially. The drug is in daily practice in the first line and second line. We are now very experienced with the use of this drug. It is a drug that is tolerable, but there is some problem in terms of safety profile. The first problem is the skin reactions. Why? Because around 15%, of the patients will develop serious skin reactions, grade three and four. It is a main problem initially when we give this drug to the patient. The second problem, and maybe more importantly, is the peripheral neuropathy. A vast subset of patients will develop neuropathy after six or seven months of treatment. Sometimes it could be quite serious. You can see here that grade three and four occur actually in around 7-8%, of the patients. Even a grade two is a problem because you have to discontinue the treatment or reduce dose. Sometimes the patient is not willing to be treated with this drug anymore. The most common treatment-related adverse event leading to discontinuation of EV is indeed peripheral neuropathy. One priority in terms of clinical research as of today is how we can de-escalate this drug because of the risk of neuropathy and how we can develop new agents without this risk of neuropathy. Of course, if we can develop new ADC with low risk of neuropathy because maybe we can use a payload that's different from MMAE, it could be something very important in daily practice. Enfortumab vedotin, is approved in bladder cancer. It is validating the role of Nectin-4 as a target for ADC. You have to know that beyond bladder, Nectin-4 is expressed in many other tissues, in lung cancer, in breast cancer, in GI cancer. The thing is that there is no validated sign of activity with Nectin-4 ADC so far. The response rate, is quite disappointing in many other tumors, including again, GI, breast cancer, or lung cancer. There is clearly opportunity and admitted need for all other tumors beyond bladder cancer. There is a need of development of next generation of Nectin-4 ADC, to overcome some challenges that we are facing in daily practice. The first challenge that we have is the low and heterogeneous Nectin-4 expressions in other tumors and also in bladder cancer. What we need here, it's a new ADC that could target Nectin-4, but with a higher bystander effect to address this limit, so the low and heterogeneous level of Nectin-4. Number two, the second challenge is how we can overcome acquired resistance and the resistance to the approved ADC, with MMAE as a payload. Meaning that now what we have to do, try to do, is to develop different payloads and to investigate different payloads to address this question of primary resistance and secondary resistance to enfortumab vedotin. The last challenge, as mentioned earlier, is the toxicity and how we can deal with the risk of peripheral neuropathy and skin toxicity. Meaning again that we have to provide patients with better ADC or ADC with better safety profile, limiting this risk of neuropathy. Thank you so much. Hi, everybody. I am very pleased to be here with you today to present to you our preclinical data package for IPH4502, which is our drug candidate, anti-Nectin-4 ADC, conjugated with exatecan. Next slide, please. First, I will not be very long on that slide, which is more a reminder on the rationale why Nectin-4, is a very good target for ADC. It's obviously a validated target as Padcev, is approved, but just to remind you that it's a molecule which is part of a family of different Nectin molecules, which are adhesion molecules. Nectin-4, has a very limited expression in normal tissue. Here you have two examples of Nectin-4, staining in normal lungs. There is no expression. There is little expression in some other epithelial tissue like esophageal or breast. Where, as was mentioned by Professor Loriot, there is the highest expression in the LC tissue is in the skin. Interestingly, and I will come back to that in a moment, it's overexpressed in a lot of different cancers, like urothelial cancer, but also breast and lung cancer. It has also been described in the past as being of poor prognosis. Here you have a retrospective study in lung showing that when there is a high expression of Nectin-4, it's also of bad prognosis for these lung cancer patients. Next slide, please. What we did in order to address the limitation of the current Nectin-4 ADC compound and especially Padcev, we looked at how we can really increase the potential of a Nectin-4 targeting ADC. We looked first at where Nectin-4 is expressed. I would say that urothelial cancer, is a bit the outlier because it's really where the prevalence of high expressor is really the most important. That's where Padcev, is approved. It's also approved in many more solid tumors, but at low to medium expression level, you know, like especially esophageal cancer, breast cancer, lung cancer as well. We think that there are really here an opportunity to target all these non-bladder tumor types, which are showing a lower expression of Nectin-4, but also, as mentioned by Professor Loriot in bladder, also to have a molecule that is overcoming the limitation of Padcev, both in terms of therapeutic window and also different toxicity profile, but also addressing the relapse after and the resistance mechanism after Padcev. Next slide. What we've done is that we have designed a proprietary ADC based on a proprietary antibody, which is binding a different epitope than Padcev. Again, I will come back to that in a moment. It is not competing with the binding for Nectin-4 with enfortumab. It also has a competent FC, which is able to mediate immune mechanism at some point. In terms of linker payload, we chose the exatecan, which is a very well-known topoisomerase I inhibitor, which has the advantage of having a very good bystander effect. We conjugated this payload to a drug antibody ratio of eight, so really sticking a lot of exatecan to the antibody. We managed to do that by using, and this is a bit our secret sauce, a linker that is very hydrophilic, which is also very stable, that is allowing us to basically mask the hydrophobicity of the exatecan. Next slide, please. Here on the left, you can see the epitope that we have identified for our drug candidate. It's at the hinge between the distal C2 domain and the V domain. Actually, enfortumab, but also almost all the antibody that we have looked at in different patterns are really binding the V domain. We are really having a very unique binding epitope that all the others don't have. Also, an important feature of our drug candidate is on the right, which is the hydrophobicity. It's hydrophobic chromatography, here that you can see, which is basically a measure of the hydrophilicity of the molecule. The more the molecule is on the peak is on the right, the more hydrophobic is the molecule. You can see that enfortumab is far more hydrophobic than IPH4502, which is rather in the same range as trastuzumab deruxtecan. This is a very important feature because the more the drug is hydrophobic, the worse PK it will have. Having a hydrophilic molecule here is important also for the manufacturing first, but also for having a good PK property. Next slide. We characterize the efficacy of this molecule in very, I would say, classical in vitro and in vivo experiments. We are having a very good sub-nanomolar killing activity of this drug candidate. We have characterized also the in vivo activity with a minimal effective dose of 0.5 microgram per kilogram in xenograft model. It is on the next slide, we have also, very interestingly, benchmarked the internalization capacity of our antibody compared to others. You can see here in gray, the E curve, it is the enfortumab, so the parental antibody of Padcev. We have made also a nomab version of what is called LY here, which is the parental antibody of the Eli Lilly molecule. You can see that we are having a far better internalization rate compared to this, I mean, better than enfortumab, but really far better than the Lilly compound, which is also another molecule currently in phase one with the exatecan linker payload technology. What we have done, and we have on the next slide, generated again a homemade version of the Lilly candidate using the parental antibody and the linker technology that they have published in their paper and poster. You can see on the right that we're having a better killing capacity. We're having approximately one log better potency compared to their compound. We are really comparing favorably compared to the Lilly topo-1 ADC. Next, we characterize the efficacy of the IPH4502 in a panel of PDX models. These patient-derived tumor models. These are small pieces of primary tumor that are implanted in the mouse. We have looked at this series of 14 different bladder cancer models in the mouse. You can see on the left the efficacy of enfortumab vedotin. You can see that in some models, there is a very strong tumor regression, but some of them are resistant to Padcev. On the right, you have the efficacy of the IPH4502. You can see that all the models are responding to 4502. We looked at understanding this difference between the two compounds. On the next slide, you can see that one important factor is the level of expression of the Nectin-4. On the left part of the slide, you have a model, a PDX model, where the expression is considered as high with an H score, which is a way to quantify the expression, which is by multiplying the frequency by the intensity. You can see that in this high expressor model, basically the efficacy of Padcev and IPH4502, are rather similar. When we are exploring the comparing the efficacy in medium to low expressor models, and this intensity of expression is decreasing, you can see that Padcev, is not working at all. These models do not respond to Padcev, whereas IPH4502, can induce complete regressions in this model. We can really potentially address all the patients that are having a low to medium expression level of Nectin-4 where the Padcev, is not working. Next slide. We then looked also at another important factor, which is the heterogeneity of the tumor. Because you have the expression level, but also in some tumors, the expression is not completely homogeneous. What we've done here is that we have recreated artificially a model where the Nectin-4, is expressed only on half of the tumor. You can see that on the left side, so we took a Nectin-4, expressing cell line and treated with IPH4502. It's completely resistant. When we knock down, we CRISPR, out the Nectin-4, from the cell line. The cell line, is becoming completely resistant. There is no non-specific toxic effect of the ADC. You need the expression of Nectin-4, to have the toxicity efficacy in this model. When we mix in a one-to-one ratio and recreate a tumor model which is expressing 50%,, of Nectin-4 and 50%, of the clones do not express the Nectin-4, we have complete regression of the tumor in this model, really showing that the antibody needs to be internalized by the Nectin-4, expressing clones and then release the payload and kill the neighboring non-expressing cells. It is really very important because it is helping us to address the tumor type where the Nectin-4, expression is more heterogeneous. Next slide. It is an ongoing effort that we are having at Innate. We are exploring also the different type of solid tumor where IPH4502 is working. Here you have one of our first examples, which is in TNBC. Again, the previous slide where on the panel of bladder cancer, we have very good efficacy here in TNBC. We are also having a very good efficacy. Now we are working at accumulating this data in other tumor types. Next slide. Also, one important aspect that Professor Loriot, underscored was the emergence of resistance to enfortumab. There has been a recent paper on a very limited number of patient cohorts, which is also a retrospective study. I mean, it's something that is known that you have one of the mechanisms of resistance to chemo is the upregulation of this ABC transporter like the MDR1 molecule. Here you have an example on the left of one patient having bladder cancer with a primary tumor expressing Nectin-4. There is no MDR1 on the primary tumor. When they looked at the metastasis, there is a slight decrease of the expression of Nectin-4. It is becoming, I would say, moving from high to medium expressor. There is also a very important increase of the MDR1. This is very important also because many chemotherapies that are approved, like Taxan or Cisplatin, are known to upregulate the MDR1. They have correlated retrospectively the expression of Nectin-4 and MDR1. Again, I will not enter into the detail because it is more a hypothesis generating than really conclusive data. It is pretty clear that when MDR1 is expressed, and here it is the PFS, of the patient treated with enfortumab, enfortumab, is not working at all in the patients that do have this upregulation of MDR1. On the next slide, you can see what we have replicated, tried to replicate that in preclinical model by using a very aggressive mouse tumor model that we have engineered to express Nectin-4, and which is known to express this MDR1. We have taken this MT38 Nectin-4, expressing cells. As you can see on the bottom line, we have treated these cells with enfortumab vedotin, so Padcev. I would say, as expected, it's totally resistant. We have here a model of primary resistance to EV. Interestingly, when we treat with IPH4502, there are some responses in this model. We even tried to treat mouse that were exposed to Padcev. When they were progressing, we treat them with IPH4502. We can induce tumor regression even under the exposure of Padcev, in this tumor model. Really showing that we could really address this potentially primary resistance mechanism. On the next slide, we actually, so this is one paper who is describing the expression of the Nectin-4, under enfortumab vedotin treatment. It looks like the Nectin-4, I mean, even if it can be modulated, it's still expressed upon treatment with EV. We then generated a model to address the secondary resistance, the acquired resistance to EV. That's what we have done on the following slide. Next slide, please. Where we have treated, so we have used a PDX model, which is on the left, which is sensitive to Padcev and IPH4502. When we take this PDX, we can induce complete regression in all the mice. It has a pretty medium level of expression of Nectin-4, with an H score of 200. By having this selection pressure by multiple injections of Padcev, we are able to generate a model that is becoming resistant to Padcev. The expression of Nectin-4, is conserved. We are still having an H score of 200. This model is now relapsing when we are treating the mice with EV, whereas it is totally sensitive to IPH4502, really showing that we can address this secondary resistance mechanism by switching the payload to a topoisomerase I inhibitor. Next slide. We then further looked at the potential application of the IPH4502. As Professor Loriot, highlighted, now EV is approved in first-line beta in combination with Pembro. We really wanted to look at whether IPH4502, could at some point also eventually be combined with PD-1. You can see that this model is not very sensitive to EV. When EV is combined with Pambro, there are some complete responses. When we look at the combination of 4502 with the PD1, we are really having a very good complete regression in this mouse model, really showing that we can combine potentially 45 with PD1, at some point in the development of the drug. We looked at the toxin PK of our antibody non-human primate. Basically, the findings were, I would say, what we expect with this class of drug, mostly related to the exatecan from GI and hematological toxicity. We had this skin hyperpigmentation, but really milder than what was described with enfortumab. We do not have this peripheral neuropathy issue, which is really related to the MMAE payload. What is really important to keep in mind, we treated this monkey at up to 30 mic per kg. We are having a very good exposure to the drug related to the hydrophilicity of the ADC. What is really interesting is on the bottom of the slide, on what is in red, we have a very low release of free toxin or free exatecan. You know that this is really one thing that you want to avoid, to have this release in the circulation of the payload, because it will limit the therapeutic window of the ADC. We think that our data are good because we have benchmarked them. It is on the next slide with what the others are showing. On the left, you have the data that are published for enfortumab, which has a very, I would say, a more narrow therapeutic window. Because you know that with MMAE, we cannot dose really high. The approved dose for enfortumab vedotin is 1.25 mg per kg. When we benchmarked our data with what we think our closest competitor has shown, which is the molecule from Lilly, and the data are coming from their poster, we are having a far lower release of free exatecan in the non-human primate. Potentially, it is also another element of differentiation with the internalization, the better killing capacity, and also the more stable linker payload system that we are having. Next slide. This is to conclude that I hope that I managed to demonstrate to you that with this IPH4502, we are having a very novel and very differentiated Nectin-4 exatecan ADC, with a high internalization capacity and bystander effect. With the linker payload that we are using, it translates into a very good exposure in the non-human primate and very low release of free exatecan, which is also an advantage for our drug. The efficacy preclinical data package is showing a superior efficacy over enfortumab vedotin, with a broader spectrum of activity, and especially in the low Nectin-4 expressing model, as well as in some model of primary resistance, but also acquired resistance to Padcev. We are now showing efficacy also in other tumor types. We are now having data in TNBC, but as I said, it's something that we are continuing. We are also having, but it's more for the long term, a very good combination potential with PD1 blocker. I will stop here and hand over to Sonia, for the clinical development plan. Thank you, Yannis. Here in the next slide, please. I hope we managed to convince you that we have a very differentiated asset here. All the data that Yannis, has presented at CITSI, at AACR last year, and the new data that have been presented today form the basis for us to move in the clinic with this differentiated asset, where we see that we really have an opportunity, an opportunity to have a broader therapeutic window, an opportunity to have a better safety profile than what has been shown for Padcev, and correctly highlighted by Professor Loriot, an asset that, yes, produced some efficacy at the cost of high discontinuation rate and peripheral toxicity. With exatecan, as a payload, we should not have any peripheral toxicity because exatecan is an old molecule that's been used before. We know what the toxicity is for exatecan, and peripheral toxicity has never been reported. We also have the potential to address not only the high expressor tumor like bladder. We really have the potential to address all the tumor types that address any level of Nectin-4, moderate, but even low. Not last, we have the potential also to address tumors that have been previously treated with Padcev, and they become resistant. We can address tumors that have a primary resistance because perhaps they did not have enough expression to be responsive to Padcev, or they upregulated the multi-drug resistant mechanism and became resistant to payload. For all these reasons that are highlighted here in the green part of the slide, we believe we have a differentiated asset. We are very proud to have, let's say, started our first in-human trial at the end of January. In the next slide. Thank you. We really capitalized, the strategy of the clinical development is really capitalized on the two main factors. Indications like bladder that are resistant to Padcev. In that respect, we can address a growing medical need in urothelial carcinoma and eventually quickly move to phase I, sorry, quickly move to first line, also in combination with anti-PD1 in this indication, but also expand in all the different indications that express Nectin-4, and are currently not responsive to any of the Nectin-4 ADCs because, as Professor Loriot has shown, many tried to develop in different indications, lung, head, and neck, but the response rate has been at best disappointing. In the next slide. Here we are talking about unmet medical need in many different indications. You can see there are very common tumor types like non-small cell lung cancer that you may argue is not really unmet, but still there is room to improve even in this indication, breast or colorectal cancer, where we are still using old-fashioned chemotherapy, prostate, head and neck, ovarian, gastroesophageal junction, cervical, esophageal, and also melanoma, even though fantastic immunotherapies have reduced the medical need in this indication. The next slide is the study design for the first in-human trial that has started recruitment at the end of January this year. I'm also happy to say that we already treated the first three patients on this trial. This is a dose escalation that is guided by Bayesian design. We are going outside of the classic 3 plus 3, that is a very rigid design, to adopt a more flexible design using the Bayesian statistical method and also have the possibility to enrich cohort of interest with backfill cohorts. Therefore, we will not pause the recruitment during the dose escalation, which is always a problem when we have, let's say, patients that may be lined up for treatment and also have the possibility to expand our knowledge of indication of interest. In general, we think we will recruit in total around 45 patients between the dose escalation and the backfill cohort and then move in a single indication to the dose optimization. Of course, we can do a dose optimization in more than one indication. The primary objective of this first in-human is, as usual, safety and tolerability. The secondary endpoint is assessment of PK, immunogenicity, and signs of activity, preliminary efficacy. Sorry, I keep moving the next slide. We also have, as part of the first in-human trial, a biomarker plan for two reasons, because it is very important to assess the level of Nectin-4, expression at baseline in these patients. We do not prospectively select patients for expression of Nectin-4, because this would require a validated assay, would need a threshold of sensitivity for the assay, and would take too long for patients that are critically ill that usually end up in a first in-human trial. It is extremely useful to understand what type of tumors are we going to have effect upon because there are, apart from the negative that are described here, but let's say the easy one are what you find in the top at the, okay. This one, high and homogeneous. These are the ones that usually should respond well to Nectin-4 ADC, let's say even like Padcev, high expression, homogeneous expression, let's say it's a given. Where we are going to test our hypothesis and where we consider we have a winning strategy is about tumors that have a low and homogeneous expression, as you see on the top. Like most of the times, we know that tumors are highly heterogeneous. We may have patches with high expression and other patches with either low expression or no expression at all. This bystander effect that Yannis, has very nicely described basically means that by binding to the areas that express, for instance, even a low or high expression of Nectin-4, we have the possibility to, let's say, diffuse the payload to the adjacent cell that may not express the Nectin-4, and therefore kill anyway, even, let's say, these highly heterogeneous and patchy tumors because this is really the reality of patients. In the next slide, we also look, and this is, let's say, easier to do for soluble markers like soluble Nectin-4,, by having different samples during treatment. Soluble Nectin-4 may be a useful parameter to follow because it has been shown, let's say, highly upregulated in cancer and versus LT donors. It may also decrease over time, mirroring the clinical efficacy of patients. In the next slide, here is where we are at the moment. As I mentioned, we just started the study. We have completed the first cohort. Since we have the sites that are highly engaged with this program, we consider we may have data around preliminary safety and activity between the end of this year and beginning of next year and immediately start with the dose optimization that, since it is going to be in selected indication, will take the time depending on whether the indication is more or less rare. At the same time, depending on emerging data, we have the possibility to start a more robust trial in combination with standard of care with anti-PD1 or else, and have an expansion cohort from selected indications as single agent. This is going to be in 2027. Next slide. Thank you very much to all of you for attending the meeting today. Okay, thank you. What we'd like to do now is to open up the questions. We have Professor Loriot, on the line. Maybe we'll start in the room, Henry. Is that okay? Yeah. I think we have a question here. If you speak loud, maybe I don't have to repeat it, but if not, I'll repeat the question. Sure. Thanks very much, Jonathan. Just a few questions here. Maybe firstly, aside from Pfizer's compound, obviously, there's a few other developmental ADCs already in Nectin-4. Could you highlight just how easy your agent has differentiated versus the ones you developed currently? Do you want to take that one, Yannis? Yeah. As As you may have seen in one of the very first slides, there are, I would say, the first wave of second generation Nectin-4 ADC, which are more based on the MMAE payload, like the Bicycle compound, but also the Corvett and Labwell. All these agents will at some point have the same kind of limitation as Padcev, because it will have the same kind of maybe a different dose, but the same kind of toxicity profile because the peripheral neuropathy, it's related to the MMAE. It will be sensitive to the same kind of mechanism of resistance because MMAE is sensitive to MDR1 upregulation. There are the topo. In the U.S. and Europe, the only company that we are competing with is Lilly. They have these two compounds, one which has started the phase one in March. We are a little bit behind, but as I showed during the preclinical portion of the talk, we think that we are having a better asset because internalization and killing capacity is really key for the therapeutic index of the final ADC. Understood. Yannis, could you speak to any differentiation that you might expect on the dose or dose administration profile compared to these other agents? Potentially with less, again, it's potential that with less release of free exatecan, we can potentially dose higher than the Lilly one. Also, one key differentiation, so it's a linker payload. We are not using the same linker. We think one is more stable. The Lilly compound is the Fc silent, so that it does not mediate in any immune mechanism. So it's a pure targeting. That we are targeting, but we can also have the effect on the ADCP and ADCC, mediated by the antibody. That is something that our antibody can do. Thank you. Jonathan, if I may, maybe one question for Sonia. I think we have to repeat the questions apparently because it is not coming through loud on the. My apologies. apologies. Maybe I can repeat it after you. Yes. My question is, I understand that you first dosed your first patient recently. Congratulations. When might we see the first data from this program? Can we just repeat the question? The question was that we understand that the first patient has been dosed. The question is, when might we see the first data? As I mentioned previously, we hope to have all the safety and initial, let's say, antitumor activity by the end of 2025. As always, the caveat in first in-human is that anything can happen. Even a serendipitous DLT, may slow down the recruitment. The sites are highly, highly engaged. As I said, in one week, we have filled the first cohort, which was great. I cannot guarantee that it's going to be the same all the time. I would say that we are optimistic on this. Clinical sites, just how many would you have? We are recruiting in the U.S. We have opened two sites in the U.S. in December already. We will have four sites in the U.S, and some sites in France. Thanks very much. I think was first. The hand went up a little quicker. The question is, are we trying to enroll patients who had Padcev or both? I mean, Sonia, if you want. We do not have any exclusion for patients that have been previously pretreated with Padcev. On the contrary, we are welcoming, let's say, the enrollment of patients who had primary resistance or secondary resistance to Padcev. The only thing is that there may not be so many patients available. There is no exclusion, let's say, in the protocol. Another question. Nectin-4. Are there other Nectin, like a Nectin-3 and Nectin-2? Are there other ones that you can go after? No one's talking about them. Yes, it's a family that is Nectin-1, Nectin-2, Nectin-3. Not from our initial analysis. Can you maybe pull or you can talk about the benchmarks that the other ADCs, have seen outside of urothelial carcinoma? Can you speak to whether those benchmarks were enriched for Nectin-4 expression? Without any enrollment for Nectin-4 expression, what gives you confidence in these 45 patients you'll be able to see a signal that's more attractive than what those competitors saw? If you end up enrolling, let's say, 30 CRC patients, with very low Nectin-4, maybe like the worst case scenario. Of course, we will try to prevent any skew of a particular indication in the study. It is, let's say, highly unlikely that we are going to have the situation of 30, let's say, CRC patients, in the study. This is the reason why we have excluded, for instance, pancreatic cancer from the list, despite pancreatic cancer expressing Nectin-4, because the moment you have pancreatic cancer in the study, you are loaded with only pancreatic cancer. We would like first to have a good signal before eventually opening to pancreatic, which is for sure a high unmet medical need, but extremely difficult to treat for different reasons. In terms of patient selections, there are different reasons for not going at this stage in a patient selection. I appreciate your concern, which is actually also our concern. Retrospective analysis may lead to a higher than envisaged number of negative patients. In that case, we have, let's say, time to optimize the method for selection. Also, the issue, especially in first, sorry, in first in-human trial, where we have patients that need to be treated as a matter of urgency because they have exhausted by protocol any standard of care, and they are extremely, let's say, ill. The time for turnaround, having first a positive trial screening that basically means they are eligible for the study, having the biopsy, analyzing the biopsy, and have the result will take so long that the patient will progress. On this, I actually would welcome Professor Loriot's, comment because he has been an investigator on the trial. Johann, how do you see this topic? I think it's very challenging now to envision some biomarker in the field of ADC, and especially for, let's say, EV. The data that we have from EV301 and even 302, trial showed that actually in bladder cancer, most samples express Nectin-4, at different levels, and the level of expression is not clearly correlated with the response. There is no need based on this data to select patients on Nectin-4 expression. As you know, as discussed earlier, ADC, has three different components. To envision a biomarker, a good biomarker in the field of ADC, probably we have to take into account the expression of the target, of course, but also maybe the internalization, the migration of the ADC, in the metastasis, and the sensitivity to the tumor cells. That's why, except some very limited ADC in oncology, like in breast cancer with trastuzumab deruxtecan, there are very, very few ADCs, that are approved with a good biomarker. Obviously, the molecules don't have great activity outside of bladder. Is that impacted by Nectin-4 expression? Do you mean in different indications? Yeah, the competitor ADCs, just to help us contextualize that bar. Yeah. To my knowledge, there is no, like Professor Loriot said, to my knowledge, all the data that we have wrapped together in this trial, there is no selection of patients based on the Nectin-4. It's just that you have certain types of tumors where the high expressors are more prevalent, and bladder, like I said, is number one, but in other tumor types, the expression is really more variable, and the proportion of patients with low to medium expressing level of Nectin-4, is greater. You have a higher probability of showing lower response rate in these tumor types. We have never seen really a direct correlation that has been presented, at least by others, that, let's say, even the 10%, response rate that they observed was due because of a higher expression compared to the patient that did not respond. To my knowledge, this has not been ever presented. Okay. Is there a question for Professor Loriot, and Yves here? Even though topo and isomerase inhibitors have been used as a chemotherapy agent for a long time, why has trastuzumab, not been really looked at as a payload when MMAE, has always been a payload of choice? Yeah. I don't know whether you heard that, Professor Loriot, but the question was, why have topo isomerase inhibitors not been used as a payload or in the past or looked at in the past, and it's always been MMAE? I think you're on mute, Professor. Yeah, it's a very relevant question. Actually, it was based on the profile and the clinical profile of this patient treated for metastatic urothelial carcinoma. Usually, it's not the case, but historically, these patients were quite old with a lot of comorbidities. This patient was previously treated with platinum-based chemotherapy, and then there was nothing. Historically, there was some clinical trial with vinflunine, and they were not able to receive anything further. The clinical trial that could investigate topo I inhibitors was very limited due to the rapid progression of the tumors, number one. Number two, they were not fit enough to receive topo I inhibitors historically in third line or in fourth line because performance studies were quite bad and with a high risk of toxicity, and especially with regard to the neutropenia. I was going to say, was there any rationale not to go to an exatecan payload earlier versus MMAE? In the context of ADC, no, there was. I mean, the first coming in the field of urothelial carcinoma with the onco tumor veloute, because we knew from the past that tubulin inhibitors could be active in bladder cancer because we have some clinical trials suggesting that MMAE could be something important for bladder cancer. That is why the first ADC, so EV, was associated or was built on MMAE payload. As I said before, in pre-clinical model, we know that bladder cancer models are sensitive to topo I inhibitors. Why we do not have any clinical trial with topo I inhibitors historically, it is just because of the difficulty and the challenge in the past to conduct trials with such agents in bladder cancer. It does not mean that topo I inhibitors are useless in bladder cancer. If you look at the ADC technology field, for many years, the paradigm was to go with ultra-potent payloads. Before Daiichi and DXd, everybody was either looking at more potent toxins with other lower bar, potentially like some PBDs. It's really the DXd showing that it's better to stick more molecules with lower potencies and less molecules with high potencies that could work. I think this is one turning point, the success of DXd. The second thing is also a technical point, that as exatecan is very hydrophobic, it's not that easy to stick eight exatecan on an antibody without having aggregation of the product. You really need to work on the linker to mask this hydrophobicity to be able to have something that is soluble and monomeric and not just create aggregates. Okay. One One other quick question. What gives IPH4502, the ability to target low-expressive Nectin-4 tumors, where EV doesn't seem to be functioning? Yeah, repeat the question first, Yohann. Yeah. was asking why we think that IPH4502,is able to target low-expressor Nectin-4, where EV, is not working. I mean, we have some hypotheses. I think first, the epitope, which is different, might be an explanation. We still don't understand exactly why, but it might be an explanation. Also, the fact that with different sensitivity of the payload might also explain why we can capture this low-expressor with the IPH4502. Thank you. Thank you. Sorry. I think Kelly Shubham, Jefferies, and thank you both for your presentation. I have two questions. First, for your PDX model studies, maybe I did not catch this because does IPH4502 also respond well in low-expressing tumor types other than bladder cancer model? Secondly, for your phase one data readout, do we anticipate or do you prefer to have a sizable patient sample who had a prior pathologic experience? Thank you. Thank you. I will take the pre-clinical question. The question was whether there is an activity in PDX model, other than bladder in low-expressors, and this is an ongoing work. We are currently accumulating data, and we did not disclose yet, but it is an ongoing effort. Thank you. Regarding the clinical data, the question was whether we prefer to have PATSF, pretreated patient rather than other indication. Let's say we do not have, at the moment, a preference. We do not exclude any, as I mentioned before, we do not exclude patients with urothelial that have been previously pretreated with PATSF. Actually, this may constitute, let's say, a quick way to move forward if we had enough patients that are experienced on Padcev, and they respond to IPH4502. This would be great to have. The idea is really to move fast on this first in-human and only in a second time produce, once we have evidence of safety and hints of clinical efficacy, really to try to shape the clinical development in one indication rather than another, depending on business case or fast speed for approval. We would leave this at the second step. Yes. All right. Two quick ones from me. This is Doug Matthiasson, from HC Wainwright. You spoke a little about your enthusiasm for the epitope and the high affinity that you have for targeting. Of the different cancer subtypes that you're investigating, are there differences in epitope expression of Nectin-4? If so, have you done any modeling or preclinical work that would indicate your targeted epitope? There There might be higher affinity for one certain tumor type. The question was, I will shorten it a bit. The question was whether the epitope targeted by the IPH4502, is different. I mean, there is different expression of the epitope in different tumor types. So far, we do not know exactly. We have looked at that, but we do not see any difference. It looks like it is the same Nectin-4 molecule, expressed in every tumor type. That is why I am saying that we do not understand exactly the molecular advantage of this. Okay. A second one, you talked, a couple of other questions were already about patient selection, and you sort of described that threadly. Are you trying to, of your 45 patients, trying to weight at all towards a particular indication, like Nectin-4, UC? Of those 45, let's make sure there's 10. Like another cancer subtype where Nectin-4, is not an approved target, do you think there is a lot of possibility there? Let's make sure we have five there so we have statistical significance moving forward. Okay. The question I'm going to summarize here, whether we are trying to shape the recruitment in the dose escalation. Again, here, the main objective is really to go as fast as possible. Therefore, we don't want to really hold at those levels because we are not having, I don't know, the melanoma that we had in mind. The idea is really to move fast. We have this backfill cohort that is, let's say, optional, so to speak. We can really shape. For instance, ask the investigator, "Can you please get two patients with head and neck?" In that respect, we can perhaps wait, but not in the dose escalation as otherwise. I'm sure that when we are at the end of 2025, we won't be happy that we don't have results. Right. Sorry, the back of the room up in the end. Hi, this is Jeff Walkhout, of EY. Based on your PDX model, we're saying that the IPH4502, will be dosed once per two weeks for the first four weeks, and then will be dosed once per four weeks. Your one-man dosing basically would be translated to a Q3W dose regimen compared to EV. If you're talking about the dose regimen that we use in the study, we go for a Q3W. An administration every three weeks is from the beginning until progression or toxicities. Stronger anti-tumor activity than most 40-day model. Can you give some color on the weight loss effect on this anti-tumor model? There was no particular toxicity on the weight loss in the PDX. I mean, we are using doses that are tolerated by the mice. Hi, this is Kim from HC Wainwright. My question is, is there a current curve suggesting overlap between Nectin-4's regimen and HER2 and FGFR3, other targets in UC? How do you think about competition of your ADC versus other targets? Please learn exactly. Josh? Yeah, not yet to my knowledge, but I may add something. There is no association with this receptor. Yeah. Second question. You have shown that your ADC and the PDX2, in the sub-1 connective tissue cell line, that yours is superior than the other critical ADCs. I believe that cell line is a Nectin-4 high cell line, but we do not need to have analysis in low Nectin-4 cell line. Are you saying superior origin? Yeah, it is a question. The question was whether we have benchmarked the IPH4502, against the Lilly antibody in the cell line that is a high expressor, whether we have similar data in low expressor. I think it is something that we are currently doing. Hi, this is a couple in the phase one trial enrollment. Ask it in kind of a different way than someone asked before. I know speed is sort of part of the objective. I don't want to bias it too early, but how many patients do you need to be comfortable in selecting a tumor type for expansion? Hypothetically, you have 11 tumor types, four of each would be about 45 patients. They could have different Nectin-4, different dose levels. I mean, is that enough to have confidence in your tumor type selection? That is sort of getting to like, would you want to bias that or restrict that to sort of have that confidence and to get that signal? My second question is sort of the biomarker. Bicycle reported it's been early, but potentially intriguing correlation with Nectin-4 gene application, and activity in the basket cell tumor context. Have you considered looking at that in your biomarker plan as well? Do you want to start, Sonia? Sure. Sure. Regarding how to define the signal, considering that the number is relatively small, first of all, it's not a given that we are going to have all the patients that are listed. It is very likely every time that there is a, let's say, a basket trial where you have, let's say, 10 indications, in reality, you are going probably to get five because plus, let's say, one patient in the other indication, which does not constitute, let's say, a significant number to make any decision in that respect. We always have the possibility based on, for instance, pharmacodynamic changes, to have some hypothesis. In that respect, use the backfill cohort. Bear in mind that there is always the possibility to amend the protocol to expand. I don't need to, let's say, quote, for instance, the initial PD-1 trials, the first in-human, they ended up with 1,300 patients. We are not in this situation, of course. To have a protocol amendment to have another, let's say, 40 patients in case, we want to expand would certainly be not an issue. Anyway, we can really expand in this indication where we think we might have something and see whether that materializes or move to a different indication. In terms of this gene amplification, this is something since we have, let's say, peripheral sampling that is taken on treatment, this is something that we can always test. Perhaps you want to expand on this. We may not think it may be really significant for our study. Yeah, it's something that we follow. I mean, we start showing a correlation between the ROM1 expression, and this gene amplification. It is shown in this German paper. It is shown in urothelial, but also in other different solid tumor types. We will follow that. Again, we think that we may be active in lower expressors. That is why it might not be needed. One quick follow-up. Where do you think investigator bias is from an enrollment perspective? I know that sometimes it takes where they want to look towards a certain product or drug for clinical trial perspective. Do you think it will maybe that is why it could go bladder? Or do you think investigators are kind of trying to look to see how Nectin-4, can be brought into other tumor types right now? Are you asking if there is a bias from the investigator perspective? Yeah, like in which tumor types they might want to enroll or otherwise. I think, and here, I would also like to ask Yohann Loriot, for this because he is an investigator. Let's say from my 20 years' experience, of course, very often there is a kind of bias because if you are asking if your investigator is a specialist in GU, it is very likely that it's going to be a patient in the GU space, and not a patient with lung cancer. It is quite rare that, let's say, the phase one unit is totally, let's say, indication agnostic. Please, Johann, expand on this because. Yeah. You're right. I think there is a risk, but to me, it's very minimal because the different phase one unit that has been selected are quite agnostic. Just for example, our unit here, in our unit, we are treating all the tumor types, including hematology. Lung, breast, GY, GY, bladder, of course, prostate. We will not have any bias. Of course, in this context of this specific drug, if a patient has been, for example, treated with EV, and referred to our center to find in the trial, I think it fits very well with the trial because there is a strong rationale to give this new ADC, for a patient who has been previously treated. Generally speaking, in departments where all the tumor types are represented, there is no bias or very limited bias. More questions? Can you talk about this is Dana Graybosch, from Leerink Partners again. Can you talk about the soluble Nectin-4, and whether your ADC binds, that or there's differences based on your epitope of binding that from the other ADCs? Yeah. So Dana, is asking whether the IPH4502, is binding to soluble Nectin-4. Yeah. I mean, to our knowledge, it should bind because it's cleaved near the membrane. So the epitope is we didn't check on the conservation. It's something that we will do. But in all likelihood, the antibody should bind the soluble Nectin-4. Please keep in mind, if you looked at the dosage that we are presenting, they are in the nanogram per mL. Given the dose that we are considering, we will be four to five log. We will have four to five log more drug than the soluble Nectin-4. It's unlikely that it's acting as a sink for the drug. It may be, like Sonia presented, an interesting biomarker to follow to look at the, I would say, the tumor load in these patients. Okay. Another question. Just one. What kind of antibody do you think is called the most appropriate or do you think bispecific antibody would be by some of the newer option for you? Yeah. Your time, Your time, Jim. As you know, we do not have an antibody one. Therefore, we have to rely on, let's say, approved drugs. It is very difficult at this stage to say that the preference is for one or the other. There is no reason why, let's say, one antibody one should be more effective than the other in combination with the ADC. As you know, I think this all depends where we can find the collaboration. Okay. We've got some questions online as well. Chang Li, at Oppenheimer. Do you think there's an accelerated approval path for 4502 in Padcev, refractory urothelial cancer patients? For non-UC tumor types, which tumor are you more excited about? Yeah. I think the other questions have been asked. Sure. I'll take this. Certainly, for the Padcev, refractory patients, IPH4502 efficacy, in this patient would constitute a fast track for registration. The first part of the answer is yes. What tumor types we are most excited? There are many with high unmet medical need. I would say, for instance, colorectal is certainly one where we are still using kind of old-fashioned therapies. No real breakthrough recently in colorectal. I wouldn't exclude any indication. There are all unmet medical needs at the end of the day. Yuan, do you have a preference? No. As we mentioned earlier, there is no Nectin-4 agent, approved in other tumor types except bladder cancer. There is a need for many other solid tumors, and especially, for example, lung or breast cancer, and especially triple-negative breast cancer, where PD-1 inhibitor, could be something very important to consider. I think here it's probably two important solid tumors to consider. Regarding the first part of your question and the possibility to get an accelerated approval, definitely yes. As you know, there is no standard of care for patients previously treated with EV and Pembro. The level of evidence is four. It's platinum-based chemotherapy or elder fitness. Clearly, there is a bias here. Okay. Thank you. A follow-up. Sorry to cut you in. You mentioned with the dose escalation, you're not doing the 3 plus 3 typical. You're doing the phage and adaptive. What's the least number of patients you could have to enroll per dose? How much faster could you go to get to your recommended phase II dose with the phage versus the standard? The question is around the dose escalation and the statistical method. The number of patients that we recruit for any dose is three. In that respect, it may look similar to the 3 plus 3, with the exception that in the classic 3 plus 3, the moment you get to the DLT, this is the end of the escalation. With the Boeing design, you have the possibility to expand the cohort and still escalate to the next level. Basically, it prevents stopping the dose escalation due to serendipitous DLT. A follow-up question from Chang Li, at Oppenheimer. Will Nectin-4 receptor, cleavage affect the binding of 4502 or other Nectin-4 ADCs? Are you planning to use prophylaxis in a phase one study for potential hematomas? Yeah. Yeah. I mean, yeah. I mean, if Nectin-4 is cleaved, there will be no binding to the tumor cells, but binding to the cleaved Nectin-4. I mean, that's something that has been shown in the literature that Nectin-4, is highly expressed in primary tumor. It's still expressed on metastasis. It should still bind to tumor and to metastasis. Yeah, there is no prophylaxis, really used so far. We need to build a safety package to understand if prophylaxis is needed. A couple more questions online. Is the primary catabolite of after linked cleavage pure exatecan, or is it a chemically modified version of? It's exatecan. I think you covered this a bit, but what sort of drug sync effects are you anticipating given Nectin-4's, high expression in skin and other healthy tissues? Might this effect be more pronounced or reduced with different epitope and Padcev? We expect to have the same in theory same thing effect than with Padcev, except that we don't use the same linker payload or a different mechanism. In the TOX study, in the non-human primate, we didn't really observe, I mean, except some hyperpigmentation in the skin, which has been described with other, and which is reversible after the recovery period. We expect less toxicity than what is described with Padcev. In the clinical trial, on the other hand, we have a full list of, let's say, expected toxicities and the skin toxicities one. We just need to monitor. Okay. I think that's it from a questions perspective. I would like to thank everybody here in the room for attending, everybody online for attending. Thank you for your time. I hope it's been useful and you've got some valuable information. We will be looking forward to moving forward on our Nectin-4 journey with you and keeping you informed as we go through the next steps. Thank you.