Hi everyone and welcome to our ADC and Bioconjugation Tech talks about advancements in process and payload development for Adc's and next generation Bioconjugates. My name is Patrick Zahing and I will be guiding you through the introduction today before handing over to our moderator. The tech talks will be moderated by Kaden Wilkie. Canaan is the Associate Director of ADC and Biocongregation Services, leading the strategic marketing initiatives for the growing ADC CDMO landscape within the Life Science Services organization. Prior to joining the services business, she led chemical biology initiatives and product commercialization into the Sigma outreach portfolio. Canaan received her Bachelor of Science and Biochemistry. At the University of Saint Thomas, St. Paul, MN and her pH. D in Chemical Biology at Indiana University. Before I turn things over to Kayden, I'd like to cover a few housekeeping items. At the bottom of your screen are multiple application widgets you can use. All the widgets are resizable and movable, so feel free to move them around. To get the most out of your desktop space, you can expand your slide area or maximize it to full screen. By clicking on the arrows in the top right corner, if you have any questions during the webinar, you can submit them through the Q&A widget. We will try to answer them during the webinar, but if a more detailed answer is needed or if we run out of time, we will answer later via e-mail. Please know that we do capture all questions. You will also have the opportunity to participate in a couple of quick poll questions throughout the session. I encourage you to take part in these. If you are watching this webinar on demand, you can still submit poll responses as well as questions. The webinar is being streamed through your computer so there is no dial in number for the best audio quality. Please ensure your computer, speakers or headset are turned on and the volume is up so you can hear the presenters. An on demand version of the webinar will be made available afterwards and can be accessed using the same link you have already received. So that's it from my side. It's my pleasure to turn things over to Kaidan. We will take you through today's agenda and introduce our speakers, enjoy the tech talks. And over to you, Kaidan. Thank you very much, Patricia. Hello everyone and welcome. I am joined today by my colleagues Melissa Salmiyan, Huawei, Jake and Ross and we are so glad to have you with us. We have an excellent lineup of topics to present to you today. That relate to process and payload development for a DC's The way we have organized the session is I will give a short introduction to our business and then hand over to the experts. A brief reminder that the life science business of Merck KG A a Darmstadt, Germany operates as millip or Sigma in the US and Canada and this is who we are. We are a part of the. The life science services organization with Milipore CTDMO services that being contract testing and development and manufacturing services. While many of you are here for a DC's, we are actually a vibrant multimodality service organization built to serve biotech and pharma drug developers with end to end best in class services. And we do this for viral vectors including gene and cell therapies. Biologics with monoclonal antibodies and recombinant proteins, small molecules including a PI's, highly potent API's and activated pegs. ADC's including diverse bioconjugates and MRNA&LNP formulation. For these modalities we partner with you on your drug journey from preclinical to clinical to commercial. We provide a single organization through our vast global network with coverage of all stages of a molecule's value chain. As you can see by this map, our organization is a diverse one with centers of excellence spanning many countries. Coming back to a DC's, what we can offer you other than this webinar is expertise, capacity, and innovation. We are an established CDMO with over 35 years bioconjugation experience and 15 years in contract manufacturing. This has instilled in us a broad, excuse me, a broad range of skills that we can apply with you and our customers. For example, we have produced ADC's and bioconjugates with a variety of components. We can also help you assess these through our ADC Express service to generate a DC libraries to identify lead candidates. We use both established conjugation platforms as well as novel or customer specific ones and rely on myriad expert teams to develop and control your process for manufacturing. This is also a really fantastic group of people at our St. Louis, MO facility. We also have capacity for your evolving drug substance needs from clinical to commercial. And we're very excited about our recent opening of 6 brand new single nanogram kilo labs at our Madison Verona site in Wisconsin that develops and manufacturers linker payloads for a DC programs. On the topic of linker payloads down at the bottom here in innovation. We have recently launched a series of products that enhance our CDMO capabilities, including keto sensor and a DC cores that you'll hear about today. We continue to onboard advanced technologies, which is also a topic for today's session. Additionally, we recognize that you and other customers are continuously creating novel methods and molecules and as your CDMO partner, we love helping you bring your innovations to the clinic. Now what we've all been waiting for, I'm happy to introduce our lineup for today. We'll begin with a talk from Melissa and Salmion on platform technologies for next generation conjugates. Continuing in the process space, Jake, and how we will highlight the integration of PAT into bio conjugation processes. And thirdly, Ross will give an overview of some of the recent linker payload technologies he and our teams have launched to adjust challenges. On the small molecule side. Each talk will be approximately 20 minutes with five minutes for question and answer directly after each presentation, a reminder to watch for the polls and just submit any questions using the Q&A wedget. And with that, I'd like to introduce our first set of speakers that will present on expanding the scope of established platform technologies to adapt to next generation Conjugates. Melissa Ritchie joined our process and development team in St. Louis, MO in 2018 and has a PhD from the Wake Forest School of Medicine. She started her career as a fellow scientist at Oak Ridge Research Institute for Science, Science and Education. And was also a postdoc fellow at UNC Chapel Hill with a strong background in protein biochemistry and bioconjugation. She has over 7 years experience in the industry. Samiyan Srinivasa Srinivasa Raghavan joined our process and analytical Development team in St. Louis, MO in 2019 and has a Master's in Science from The Ohio State University. Before his current role, he optimized mab downstream processes with an msat operations and developed novel processes for dairy proteins and lipids. He has six years of experience in industrial biotech and food service. And with that, I send it over to you, Melissa and Salmian. Thank you, Kaylin. Thank you very much for the kind of introduction. Hello, everyone. Today we'll be talking a little bit about. What the ADC platform is at our company and how we've adapted it to include novel non ADC conjugates as well. So before we proceed with the novel conjugates, let's talk a little bit about the ADC process platform itself here at the Center of Excellence for bio conjugation. We over the years have developed a platform for conjugating your traditional Adc's, your antibodies and your Hvapi's and purifying them into final BDS. So we have developed a variety of conjugation chemistries optimized through our DLE and no fat studies for control of bar for removal of free drug and things like that, and our purification steps as well. On the left you see the conjugation happening and throughout the bottom right you see our various unit operations of purification and formulation. So generally with Adc's downstream processing involves one to two purification steps. It's either TFF or chromatography step after conjugation and then a TFF before our formulation. We also try to maintain a 10X progressive scale up so that we I can go into GMP processing without any issues. So talking about downstream processes, DFF or tangential flow filtration is one of the most commonly used downstream processes for a DC purification after conjugation. So we do. Concentration of TFF to hit our formulation target we buffer exchange it into our final buffer before adding recipients for formulation and we also used TFF to remove unconjugated free drug or free drug linker. So typically in a traditional ADC process we use a 30 kilodalton membrane cut off. The reason for this is our Mavs are 150 kilodaltons and are in. Free drug is one to two kilodalton, so it's much less than a membrane cut off. So that works out well for us. We use a variety of membranes, membrane Mlcs like PS and ultracellular those depending on the conjugate that you're purifying. In the top right, you see a time course of free drug clearance by TFF. And you can see as the dye volumes increase the free drug going down and this is manual sample at the bottom you see this the same thing but this is inline PAT with our flow VPE instrument. The yellow line is the free drug and you can see that across the dye volumes the free drug is going down the blue line being our conjugate concentration remains relatively constant before our final. Concentration stuff. Another important and much used purification step in ADC downstream processing is chromatography. We traditionally use hydrophobic interaction chromatography for dark species isolation since our free drug and consequently our ADC is highly hydrophobic. We use ion exchange chromatography sometimes for removal of unconjugated drugs, so in a typical bind and elude method. And we use CHC or other mixed mode chromatographies for removal of aggregates as well. So on the right top right you see a typical preparative hydrophobic interaction chromatography trace where. The application of a step gradient. You can see the different dark species coming off as peaks. The bottom is a traditional bind and elude ion exchange chromatography where the first red peak is your free drug being washed off as your conjugate is captured onto the column. And with an increase in our salt concentration we are able to elute our target conjugate purified target conjugate as well. So this is our traditional platform and of course we have 15 years of experience modifying it for traditional Adc's to a plug and play mode. But what about emerging novel conjugates? To talk about emerging all the conjugates and how they're different to traditional Adc's, I'm going to take the example of all the GO conjugates. To contrast them against traditional Adcs with an all the conjugate, your payload is much bigger than your traditional ADC. It's 15 kilodalton or more and it also has a better hydrodynamic dynamic radius since an all ago is more linear. So using a 30 kilodalton filter just for filtration just for removal of the free all ago might not be possible. We might need to use different membrane curves. Also with an ADC, it's mostly protein by mass, but with an oligo conjugate, up to 35% or more of your conjugate could be a mile ago. So this causes trouble with traditional binding, a little chromatography because your dynamic binding capacity goes down on the column, so because of the increased molecular size. So that means we need to adjust for that as well. Another issue is traditional ADC's in the process development space. The biggest batches that we do are 40 to 50 grams for preclinical and our GMP manufacturing is on at 300. But with all the other conjugates with increased doses strategy we tend to see north of 300 grams in just the preclinical non GMP batches and up to two kids in the clinical batches. So this increased scale. Means we need to intensify the process because we can't just increase the scale of the chromatography or increase the scale of the TFF infinitely. Another issue. With all of the conjugates as compared to a B c's is that they're charged and or neutral. So the traditional HIC mode of dark species isolation might not always work. We might have to switch to ion exchange which traditionally has only been used for clearance of free drug for isolation of dark species. All the conjugates also like to bind to the columns and they like to bind tightly so. That restricts our resin loading further the dynamic binding capacity issue and we need to use organic solvents on the resin and some of these resins might not be rated for organic solvents, so we are further limited by by these factors. The other issue is because at really low skills, especially for non dari species, there's mixing issues. Are really low scaled down model might not reflect the actual at scale mixing. So we need to develop some mathematical adjustments and ratios to make sure that these scaled down models accurately predict our scale up. So that was the example of oligo conjugates. There are other kind of conjugates that we have worked with for example. On traditional fabs, fragment antibodies, polysaccharide conjugates haptons, some chelator and radio conjugates as well, and some dyes. And each of these novel conjugates come with their own set of issues for which we need to modify the existing platform because we don't want to reinvent the wheel every single time we develop the process and of course each of these process issues. Lead to novel solutions, for example modifiers and purification. We do plate based resin screenings to accurately screen hundreds of resins. We have mathematical models like I mentioned before SO and we do parallel process development to keep to our timelines as well. And before we move to the next section, analytical section, where Mel will talk about some analytical issues, we have a poll question here. What time of bioconjugate is your team working on? So I'll give you guys a few seconds to go through the options and answer. All right. Moving forward, we will see more answers. This poll question is open, so you can answer further as well Moving forward now. Yep. Thank you, Sam. So again, reemphasizing that we have established our expert knowledge on the ADC platform, but that has. Provided a foundation, but it is also helped us realize those gaps and what we need to understand and how we want to build our expertise to then be able to do business with these next generation conjugates. And so as an analytical support. My question is how do we define the process development success and from an analytical perspective that requires robust analytics to inform every process operation as we are developing for customers. That can be challenging, especially when we're dealing with different physical chemical properties, which I will discuss with an all ago example. Typically requires a lot of upfront method development so that we can make sure that we are confident in what we are evaluating and what I'm actually reporting to my process colleague and we need to define those quality attributes as soon as possible. And so again that analytical ADC platform has. Paved that way for us to be able to to expand into the bio conjugation portfolio and as Sammy has mentioned, some of the challenges are going to be redundant that I'm going to present in this next slide. However, sorry let me get here. What we want to highlight is where those challenges arise for the analytics. So again, redundant. Looking at the ADC, we understand very well how the linker and the high potent payload looks like within the ADC space. We know what the weight percent looks like. We know what the physical chemical properties are. We know how to look at this analytically within the UV Spectra, right, if we move forward into something like a Linker plus chelators. Again, we start to see that weight percent creep up. We have changes in the physical chemical properties which impacts both the process and what we're able to use for analytical characterization. One of the challenges, anything that adds significant charge, that's always going to be a challenge that we're going to need to overcome. But then also looking at something that has no UV characteristic, how do you distinguish A? Map from your linker key later if you don't have anything to distinguish and that is amplified with the oligo conjugate space where again we see that increase into the 10 to 40% weight of what the oligo is. Adding into this five conjugate again limited hydrophobicity. So that does change the approach for what we do. We can have significant charge, different charge distributions. But really what we what we see is that there is a lack of EB specificity for all of my biochemists in the room or online. You know that we have protein specificity at 280 nanometers and all of those are at 260 nanometers. And so that creates a lot of interference when we're looking at key quality attributes. And so all of these different challenges again you know looking at vaccine Bioconjugates, we see that again because we have. Protein, but then we're adding these in novel format conjugates that are enough distinguished from high potence that we're really having to look into the Gray and so. So moving straight from a DC's into the key later conjugates, we've had to develop additional strategies. You know, really go back and look at what tools do we have and how else can we look at the data. And so one of the things that was really interesting was to see how we're able to use something like Secmols to evaluate the hydrodynamic size of the molecule and be able to back calculate the amount of chelator that's being added. So that was you know one of those little nifty physical characteristics that you wouldn't typically think you could get like out of Secmols experiment. And then if we look at all it goes, which is going to be the leading example moving forward, again we have very minimal bandwidth to look at protein and all it go between 260 and 280. The complications are how do you measure accurate protein given the extension coefficient, how do you accurately assign your DAR, how do you effectively choose the right signal and so going. My deep dive here is going to be the challenges of how we have developed the expert understanding of all of the conjugates. And so it gets complicated and I'm going to try to do this as quickly as possible through for this format. So my apologies if this does feel rushed, but what we learned is that accurate conjugate protein concentrations and are are challenging because of that UV, that lack of specificity. And So what we see is that being able to determine an extinction coefficient to measure protein is dependent on the Hick or the DAR evaluation, which is what we use Hick for. And so trying to understand how these two interplay the DAR dependence and everything it really this loop feedback and so we have to have a really good controlled understanding. And so I just wanted to highlight the complexity of what we're looking at. So if we dive in. To the first example, how do you measure protein? So this is just a nominal example. If we look at the table one on the left, looking at assisting conjugate, you have a total of eight possible conjugation sites to add an oligo given the strong UV characteristic of this payload. You can see that with the addition of each oligo, we have a range of 1.6 to 7.5. Extinction coefficient. So with every added conjugate, we could be within 5 to 10% error on our protein measurement and that directly impacts my process colleagues with what they're looking at for loading on their chromatography and their TFF. So we really needed to be in control of it. And again, just to highlight, if we have a DART 8, you know we're looking at 35% weight percent. So how do I use this information using these physical chemical properties? Well, I can look at my Hick chromatography. And by applying each discrete extinction coefficient to the DAR distribution which is shown in Table 2 for every batch and for every DAR distribution, we can derive a unique extinction coefficient. And so just this example of looking from batch 1 to batch 3, you can see that the the slight change in distribution of what was done for process development. We have that extinction coefficient from 4.7 seven to five point O2. So even within the batch we're looking at 5% variability. And so this was the first key strategy of at least looking at protein concentration. But again, back to that previous figure of the circle. The Hick in the dark distribution directly impacts this protein assessment, this final measurement. So then how? How are we able to confidently derive the diary evaluation? So now we can look to this example and if we start with what we know the limits of the system. The limits are that we only have 20 nanometers of bandwidth. So if I have no conjugate and I have just an antibody, we know the Max signal will be at 280. And if I had all all ago, then my Max Spectra would be 260. So that was at least a nice framework. I have 20 nanometers to work with. What can I what can we do with that? So then using the same math, and I'm trying to keep this very brief, so I apologize. But if we use the math exercise of looking at the extinction coefficients and understanding how much of that is the antibody versus how much of that is all ago, we can derive a theoretical weighted Max nanometer. In the Spectra. And so I was able to create this ruler from 280 to 262 nanometers and then go look at the Spectra under my hip hermatography. And that provided me a way to confidently assign each peak so that that so that again that dark distribution that was feeding into my protein extinction coefficient, we were confident in the accuracy of that measurement. So now it gets just slightly more complicated because again this example of kermatography is at 220 nanometers. And so maybe some of my analytical chemists are asking why? Why are we looking at 2:20 and not 280 or 260 because that the, the choice of signal directly impacts the dark species distribution. And So what we did, let me get to my next slide here. So the the choice and signal to evaluate our distribution was also key. So again there is this math exercise very directly tied back to the previous slide. So that if we were to look at the 280 signal, we understand that there is going to be all ago interference. We're going to have inflated peak area percent and we actually see that if we look at this middle column. If I were to just evaluate the 280 signal, we would overestimate the DAR, we would overestimate our extinction coefficient, we would be we would not hit our target our peak quality attribute. If I correct for these values based on the math in the previous slide, we can show that look at 280 corrected peak area distribution matches the 220 signal and so for the purpose of being A. Manufacturing site and thinking about our quality colleagues and just what it means to tech transfer into a quality space. It's really nice to have streamlined methods that don't require a lot of math manipulation removes points of error. So this is a really great tool to be able to identify a single wavelength that we could then derive all of the information to push for those key quality attributes. So I apologize if that was a little bit quick, but I will give it back to Soundman and if you guys have any questions, we'll be able to address those at the end. Thank you, Mel. And so this kind of deep dive has enabled us to develop more and more novel conjugates and fit them into retrofit our ADC process platform and because of that. We have considerable experience, so you can see here over 80 different constructs and over 700 batches in development. The development batch scale here, it's at this one make to 400. It's a little less than one, little more than 400, but who's counting? So we've transferred over 55 products manufacturing and supported 55 Imd's experience with cartography, as we mentioned before, with different kinds of conjugation. Traditional and novel site director conjugations as well and also at different kinds of payloads and linkers. The key thing here is that seventeen of these have been next generation conjugates and that number is only going up more and more and we think for a novel conjugate we need a novel process and. To aid that, we have been developing a single use reactor since 2017. I believe it's in it's, it's just completing its alpha stage and should be in its beta stage soon. So what about next generation processes themselves Right now this is a a a slide from our bio continuum colleagues in the map space, but I think this applies to Adcs as well. Where are we today? We're in batch mode, we're mostly in stainless steel or glass. We have standard methods, stand alone controls and it's pretty digital, the plan. So where we need to be is continuous processing, single use of course real time KT which Jake and how we will talk about in the next talk. We would like to have predictive process control and not just reactive process control. And we would like to be fully digital in our plan. So in those initiatives we are doing novel process development as in developing novel processes as well, some of which are here. So continuous processing, full chemistry for example. So this is basically getting rid of our batch mode, getting rid of our big reactor. So we. The limitations of traditional scale go away if you do continuous processing and we can more accurately control the stoichiometry, so we're working with some colleagues to develop as you can see on the first figure on the left of flow chemistry in line flow reactor system, which would accurately give good stoichiometric control and theoretically unlimited scale, which is exciting. We are also working on solid face conjugation, so conjugation on chromatography and so conjugation inside of a reactor. So this will also aid in our continuous process development. We can either do traditional solid face conjugation where we immobilize the MAV or the protein and recycle the liquor payload, or if it's a novel conjugate with a charged molecule we can mobilize the novel payload. And recycle the reduced mat or recycle our protein of interest. At the bottom you can see an example of that in early development age, which is really exciting. We also have PAT. The next talk is gonna be about PAT. So to tune our flow EPE system, I've already shown an example measures protein in line. We also measure impurity clearance in line because of that. And of course there's the patrol system which does in line dark estimations. Finally, we also want to explore conjugation more upstream of what we're doing. Currently we get fully processed, fully purified map and then we conjugate and do another whole set of purification process development and the downstream processing colleagues know. We take a hit on yield with every single unit operation. So what if we can reduce the number of unit operations, do the conjugation more upstream in our process development had have an integrated map to final novel conjugate or final ADC in the same single stream. So we reduce cost of goods and services, our yields go up, our throughput goes up, so. These are some of the proactive development strategies and initiatives that we have going on and hopefully we will be, we will bring out the next generation processes and we will have much better results in the future. And that's the end of our talk. I will head it back to Kaylin. Great, Thank you both for a lovely talk. I especially liked how you compared the the different types of conjugates and then developed this correction factor for a challenge that arose when working with a nontraditional antibody conjugate. So we do have a few questions that have been submitted to kick us off and we will be monitoring for others that come in. So let's let's go to the first. Given the slightly more complex nature of novel conjugates, do they take longer to develop and take to the clinic? Thank you. It's a great question. I think the we have the tools now to integrate novel conjugates into our ADC. Platform. So I call it an ADC platform. It's really now a conjugation platform that includes not all conjugates as well. And because of some great work, some deep dives like Alyssa talked about, we now have the tools to bring in our conjugates, even novel conjugates and develop them within the same time frame as a traditional ADC and transfer them to our manufacturing site. Thank you very much. Let's move to a second one. For charged or hydrophilic illegal conjugates, what method is used for DAR determination? So that question kind of gets at what the customer won what, what is the quality attribute that the customer wants but. Right now so we our primary approach would be these pick or something like a reduced reverse phase, but there are also development opportunities around I'm an exchange chromatography again those highly charged molecules they do pose more challenges because just anything with extra charge you get it's more challenging for resolution but it's not impossible and so we really just at that point we really. Collaborate with our customers to make sure we understand what all of the quality attribute that they need and then what else we may need for defining that process success. So we we do them at the development and we move forward. Makes sense. Thanks. We will take one more our non. Highly potent conjugates or highly potent novel conjugates manufactured in the same suites as traditional highly potent Adc's. OK, same suites, I I take that as same manufacturing suites and yes given that our. Motor manufacturing is largely single use or dedicated equipment and we do, we have cleaning procedures between batches for perennium cross product and contamination. We do run non HPC conjugates in HPC suites as well. Makes sense. All right. We're going to move on. So please thank me virtually, Melissa and Samian for a great presentation and a great discussion on next generation conjugates. So moving on to our second talk will be applications of Pat for bio conjugation processes and this talk will be given by Jakes Bees. Who has been with our company for eight years developing and manufacturing ADC's ranging from preclinical to commercial. During this time he has developed, optimized, characterized and scaled up a DC processes. Presenting with Jake is how a song who has been with our company for five years, first in process and analytical development then in the novel modalities R&D department. He has over 20 years of experience in analytical R&D, new product development, project management, tech transfer and supporting GMP manufacturing across academia and industry. With that, I am turning this over to you, Huawei and Jake. Thank you Kaylen and thank you for the nice introduction and make me feel proud of myself. And I better you enjoy the half hour time with Soman and Mel and the talking about the next generation bio conjugations and Jake and I will keep you entertained and by talking about the next generation bio conjugation process. So we focus on the process and the featured on PAT and you will hear a lot about the instrumentation, automation and statistics. And in all days and everybody like you know have some math, mathematics and we will show you what we did. So first is agenda and it will give you a clear picture what we're going to tell you in the next 25 minutes or 30 minutes and the first we like to you know tell you what is all understanding. About the pros and cons or the inland testing and particularly process analytical technology for bio conjugation process. In the second part we'll introduce different pet technologies has been implemented or is underdeveloping in you know a company and the 3rd and the Jake will take over you know to show several very successful case study. And to demonstrate the benefit and the great half of you can obtain by applying those PAT technology in the Biocongregation process particularly for ADC and we're focusing on the HPLC based technology. So first and so many other advantage to apply inline testing and particularly PAT for ADC Biocongregation process. And we just list the four of them, the four major one you know man. The first one is you will have a deep dive for your process, obtain so many different informations and a very very very quick turnover time and the frequency and the way those deep dive, the information and the better understanding of your congregation process and you will have a. Definitely a better controlled process and the and the to assure a better quality for your ADC. Control the quality and of course in the ways of the PET and the inline analytical application implementing and you will have a great reduction for the risk of batch failure and the and the with the better control the process. The third one you will have a you know, very significant increase of efficiency and I know many of you are bio congregation or ADC experts and just think about you know how long you need to wait. If you do have a critical sample and either in process analysis how long you need to wait. Tell your QC colleagues to you know take the sample and analyze them, get the result back to you. I'm not saying the QC scientist is not competent. Even with the best scientist and I think easily you can get two or three hours and the whole time and particularly if you have very sensitive construct chemistry and then you you will really want to reduce that whole time and. Real time release, you know we all talk about you know can we have a real time release and then to reduce the the best successful rate and then is not a dream anymore particularly you know with the development with the other ones in electronics instrumentation and the software development. You know of course any any technologies you know have with this advantages also has this challenges. PAT cannot escape this fate. So there's a couple big challenges of involvement for applying PAT implementing PAT in ADC process. The first one is the implementation. Because it is a new is a novel technology and it is more complicated than your routine in process analysis and how to implement it is the key. And no matter, you have a really, really great method and then you need to be simple and easy to use, and you also need to train your colleagues in the manufacturing floor and to be ready to adapt it. And the second you do introduce another piece of equipment in the in the process into the process into the manufacturing floor. So a cleaning validation is needed and it need to be successful. And the third is the master qualification and if you want to do the real time products release, you want to bring in the GMP floor and then you have to figure out a way to qualify the method That's might be the most challenging part for applying implementing PAT. A DC bioconjugation and before we jump into different examples in the real technology details how we apply different P t's in different step or the bioconjugation process, I would like you know to rehearsal the understanding the definition or the process analytical technologies to make sure we are in the same page before you. The technical details, So on the top this is the definition FDA gave what is the PT process analytical technology it is and to break it apart and it actually you have four key features for PT suspect PTS, particularly for ADC process, the most important one is your method. You need have a really, really quick turnover time. As you all know, a typical ADC process, the conjugation, the reaction usually two to three hours and then a TFF filtration usually two to three hours. Compared to the upstream cell cultural processes, it's really quick. So you analytical method need to be held. You have a really, really quick 10 hour time duty cycle to fit this quick reaction time. Secondly is. Your sample you are dealing with is really complicated. You have really complicated matrix at different step and then your matrix will be different. Or your matrix could be changing and during the progress or your conjugation or your. And also you are dealing with a really really broad dynamic range for your analyze. You know from really really low concentration to really really high concentration. And and again also you are dealing with multiple analyze and not just your final, your ADC itself and you may deal with impurities and call solvent so on and so forth. And from the two first two features actually as you can see really quick 10 overtime really complicated sampling and you will generate a tons of datas and those requests and you need have staticities. Software and in place to help you do the data interpretation, data mining and it's beyond a manpower you know to do all the work anymore. And the last one is you. We're not just documenting those bit Pts, document those data for your Ind filing and you need to translate those status, those great results into a suggestion on process optimization process parameters. And the process control only by that and then you will make the PTA really great app for your back communication process. And before we jump to the different technologies and which we are developing or we are implementing, you know our company, I would like to have a pool and to learn, you know other experts are the experts from you. What you are using, what kind of PAT technology you are using in your process and then in that way we can learn from each other and really simple pool. And so are you using HPLC based or spectroscopy based or anything other and or you're not using PAT at all, at least not yet. I bet you want to use it. And how long we need to give for those pool questions to be Kayla, I think maybe you control the progress for the pool question, we can just give it maybe 10 more seconds, 15 seconds, Okay. So I I will go to the next page should be the the result and hopefully I can see something. Wow, that's a bumper 100%. You're not using PAT Well, I will just persuade you with the next couple slides and you should use PAT because it will have. Particularly with the data my colleague Jake will present and it's a great app. And then, you know, must have tools for your bad conjugation process. All right, so first this is a snapshot for you know what different pet technologies we are developing or implementing for bad conjugation process in, you know, our company. And for the reaction step and usually we you know separate to the, you know the ADC manufacturing and it's 3 steps the reaction and then the Chrome separation could be optional and then the potential flow filtration also the master health step and for the reduction the conjugation and then you can have you know patrol you PLC based which is the most. Western technology and the most powerful and technology can provide you know a lot of options for you and for different target and and also it's more complicated hard to implement and the prosthetics Raman that's a Raman analyzer developed by our company and then it provide an inline analysis for different targets. And of course flow VPE is most simple and easier technology if you want have a quick quick applications and we are not going to into the chromatography purification. For tangential flow you can use Prosthetics Raman and it's an inline analysis while you know fitting to your tangential flow process. And of course you can implement Flow VPE. And the Pendle tax is a system provide different sensors and may not be that accurate, but also provide the automatic control for lab scale TFF and if you are TFF system don't have the capability for automated control so. So in last four slides, we introduce flow VPE, the Pendle text and then the Raman Analyzer and the Patrol HPLC separately. And the first is the Flow VPE. As you all know, the Flow VPE is a using a variable path line technology which allow you compare the traditional UV spectroscopy. And allow you determine the concentration in the really, really broad range or dynamic range for protein per se is from 1 Meg per meal to 250 Meg per meal without dilution. So it is a great technology, you can adapt it into your process as a PAT. As you can see on the bottom, you can implement the flow VPE in your feed line which give you the average concentration you can map. Monitoring the average concentration in your retentate tank and if you put the flow VPE in your retentive line not the feed line and that it will give you a real time snapshot what the concentration is coming out from. Remembering on the right and we're illustrating, we are illustrating a typical bifiltration ultra filtration process from the. Initial concentration, recirculation, difiltration and the final concentration step. So four steps other you can see your product signal is keep consistent and but it does increase it during those two concentration step. But your impurity signal only decrease during your difiltration step keep constant and all the other steps. Like I said, Pendle text is really nice and easier. System allow you automated your lab skill or development skill, TFF and provide PTS and automatic control for your TFF process. On the left is all set up. I know it's a little bit crowded, but it works really well and really charming. And as you can see the number three that's the flow VPE and we do using flow VPE other than the UV sensor for the from the Pendle tags because it does provide a really, really broad dynamic range also accuracy compares those UV sensors and. On the right, they can see those, You can control the membrane pressure automatically, and at the desired point you can set it. And then at the same time you can have a pH and the conductivity you know analyzed or displayed in real time to control your process. We're not. Talk too much about you know the Pendle tax because there's so many different systems that you might using different one and for your application and they all works really nice and we spent some time, we're going to spend some time on the prosthetics Roman Analyzer which is the star products from our company is a Rama analyzer which with integrated statistics you need. For Raman or the PAT you need house in place because like we we talk about the dynamic range, the matrix complication, you cannot use a single Raman shift to do the quantification anymore. In most case showing you here instead of single Raman shift, you are using not of the whole range but different section or the Raman spectrum and then use all the information together. And by either partially secure a statistics or PCA analysis and to achieve the modeling and prediction and another feature or the OR prosthetic Straw Man Bio Analyzer is that you can have a configuration over 4 different probes and that allow you to monitor four different reactions at the same time. On the bottom as you can see you know we with the. Integrated statistics package and the predicted value and the reference value are so close you know so match to each other really really nice demonstrate a really really broad dynamic range and accuracy So we save the best for the last you know for the technical or you know master the introduction for our talk and then the patrol you plc and. Like I said is the most powerful one, also the most complicated one and you can tell from the diagram on the right and then it is UPLC. But the feature is a process sample manager. With this process sample manager and you can do automatically online sampling and which is the key and and also you can do automatically online dilution from one to 100 fold that will allow this. Wide dynamic range suit for PAT application. This is my last slides before I hand over to my colleague Jake and to do the case study to show you the case study and I will briefly introduce the method behind those case study and we use patrol SEC size chromatography and combined with DTNB as a post column derivatization. Which achieve the real time monitoring for ADC process particularly for 16 chemistry and you have a reduction step and also have a conjugation step. We can monitor both and the principle behind this method is using element Reagent DTNB. As you can see free cell can react with the DTNB and so generate the TNB. TNB have a yellow color. And the features Max absorbance at 45 nanometers by monitoring 45 nanometer absorbance and we can monitor both reduction process as we can see here. So your map will reduce typically by T SAP. After the reduction you will generate a free cells and those free cells can react with TNB and to produce react with DTNB, produce TNB into how? Full 10 nanometer absorbance and over here they can see with the reduction progress and then you will see the full 10 nanometer absorbance go higher and higher when you do the conjugation. Actually that it's opposite. So your free cell generated by the T cell reduction will be capped by your drug linker and the drug linker capitals free cells. Those free cells won't be available anymore for the TNB. And then you cannot generate A-Team dtmv, tmv anymore and you will see a reduction. Your photon nanometer will be, you know, getting lower, lower with the progress or your conjugation. And on the bottom is our setup is a simple diagram how we set up a system. We use the prostatic pump to circulate your reactions through this sample loop. So when it's time for sampling, this sample loop will switch offline and in line with HPLC and allow injection of the sample to your SEC column. SEC column serve with a separation. Separation of the matrix could interference with this reaction and from your ADC you know, allow we can accurately precisely monitoring this free cell on the ADC itself. And as you can see on on the bottom and then with this technology you can obtain a really, really nice reduction connected curve and the real time and and then you can watch those reduction process just where you're doing the reduction process or doing the reduction. That's all I have and to introduce this method, this technology and then I will hand over to Jake with my colleagues. And he will show you what the great benefit you know this PAT technology can bring into your bad conjugation process. Jake. So you are on stage. All right. Thank you very much. How we and hello everyone. In the following slides I'll show you real case studies for the patrol based application for monitoring the cystine reduction and melamide conjugation. The first two slides will investigate the technique at different process scales and in different reactor types using a model antibody. And in the following three slides we apply this technique to a real world clinical ADC project. And So what you see here first to Orient you with this graph, the X axis is process time and the Y axis represents free style content and it at the bottom is sort of a like a quick visual to track the reaction progress. The circle here represents the the reduction start. The square is the conjugation start in in in a couple slides forward here the triangle will represent quench start and so this experiment is at the 1.5 gram scale in a 300 milliliter class reactor and after the T step is added you can see a fast reaction period that slows down gradually and then plateaus and you can also see. After reduction when the when the drug linkers added you see a a steep drop off and that indicates is how a briefly discussed indicates the drug linker conjugating and quenching the free tile and you can see that the conjugation was much faster than than the reduction. And if you're familiar with cystine chemistry you know this isn't really new but this is the first time that we've that we've captured this data real time. High quality data with with good resolution. All right. And so the the the next thing we wanted to do was to prove the concept at a larger scale. And So what you see here this experiment is at the 40 gram scale in a 10 liter milli pour single use reactor and a in as you can tell you know we have a similar react similar kinetics curve. Was recorded here. The reduction appeared to start slower but reach steady state faster than the small scale batch that might. That could be some sort of insight to scale up and and mixing. But again, we see here that the that the conjugation is much faster all right, and so the in those previous two slides. The work with the model A DC demonstrated the real time free file monitoring could be used to track the progress of the reaction at different reaction scales. And so we wanted to explore this technology with a clinical ADC process, a real world clinical ADC process to ask questions like what is the kinetic profile, can we eventually improve the process and as always when we're looking at the process in a new way. Will there be questions that arise that we haven't thought to ask yet based on this on this rich data that we're gathering And so this is also cystine reduction in my conjugation and the the X&Y axes are the same as before. So this in this first exploratory run it's at a, it's at a relatively smaller process scale and so and you can see sort of the pictures on the left here. So from this run we learned that the reduction steady state appeared to be achieved prior to the defined target reaction time in that the conjugation reaction appeared to be complete well before the quench which is denoted with the the triangle in that bottom that very bottom diagram. So these data suggested that reaction times could be optimized in this first exploratory run. We wanted to see you know potentially could we optimize the reaction and so these data show we potentially could optimize the the reaction time. And so for our second run using the insight from that first run data, we ran a similar scale reaction with a shortened reduction in conjugation to see how the how the kinetic profiles compare with one another and to see if the offline DAR or other product qualities. Are impacted between between the two batches with the short and reduction in conjugation time. And so the blue trace here is run #2 and you can see with the shorter reduction in conjugation time, run two had a kinetic profile that was very comparable to run one and all of the key quality attributes were were equivalent and so that you know that was that was very exciting and so moving to. Run #3 run runs one and two suggested that the process was repeatable and that we could potentially shorten the reduction in conjugation, you know compared to the initial targets that that we had set. And so run #3 we scaled up into the 10 liter single use reactor that you can see in the in the slide on the on the left here. So we wanted to ask the same questions around Kinetic profile and offline product quality, but this time in relation to scale up. So how does the scale up have an impact and again you can see clearly the the comparable reaction profiles and for run #3, the offline key product quality attributes were were not impacted and so. So now we have this high quality data set that our process is consistent at various scales with with respect to kinetic profile and key quality attributes. And so one benefit of the data set is now this can be used to set reaction time targets in ranges in GNP manufacturing, which of course derisks the process and can help avoid unnecessary process deviations. All right. And so to summarize here regarding PAT, if we can, if we can better characterize the ATC process real time, we can better, we can better detect, predict and control the process variability, which is which is the key. And also that different ATC unit operations require various forms of PAT. So the application of PA, TS and different process steps need to be fit for purpose. So, so for PAT what I like to compare it to is the is the camera. So with the camera initially you could take effectively one picture at a time, right. And and as the as the technology got better the amount of frames per second increased and today we have high speed cameras that can detect the slightest motion. And so I think I think we're we're sort of on the on the same track with PAT we're we're really on the on the front tier. Of being able to to really see the ADC process in real time and of course when you can see the process very clearly in high resolution, you can better understand and control it and so and so for me it's it's really not a question of if PAT will be used to release a bio conjugation product but but when you know will that happen. So with that I'd like to acknowledge. Everyone that's been involved in this, in this undertaking. Of course this doesn't happen with just how we and myself. You know we had a large team working on this. And with that I'd like to thank you for your attention and spending the time with us. Thank you Jake and how we and for speaking to a topic that will not only be key for. Greater process control, but it's likely to be at the forefront of future digital plans. So this is a really exciting area to watch. Given the time, we're just going to take one question and then move on to the final talk. The question that will answer how would this be done if not targeting total saturated DAR but instead DAR 224? Jake, I I will start and if you have anything you can add and then if I didn't address it clearly and the first yeah it it is a really great question expert question and then you know then I think the Dark Age is most more popular recently and then in old days we always target the two to four in in. And also the redox, the reduction is a very dynamic reaction and keep involving keep progressing. But keep in mind and we are as we introduce, we introduce in the DTNB as a post column and derivatization. If you can see that post column reaction and the basically you're captioning every moment during your conjugation and all your reduction process and you take that. Whatever the free cell content exists in that moment will be react with your DTNB and then analyze it right away because you are you. So you analyzer is right behind your post column derivatization cell and then captured in the real time exactly a snapshot on that moment. So no matter you have a full conjugation or you're targeting dot 2.4 and it doesn't matter. The patrol SEC DTNB technology will give you a snapshot on them on the on the progress of the reduction and conjugation in real time because the SEC separate all the TSAP all the interference away from the ADC and the yeah the nature of the ADC is is monitoring right after the column separation and then really I will think you know in two or three seconds and then the the the derivatization. Yeah, Highway I was. That's exactly what I was going to add is the separation of the T step happens very quickly. So we don't have to worry too much about, you know, the reaction progressing while we collect that sample and then analyze it. Yeah, well said, nothing to add. Excellent. So for for the purpose of time, we will move on. Let's please virtually thank Jake and Hawaii again for a wonderful talk on P/E T and we are moving to our third talk today which will be given by Ross Bamowski. Ross oversees the analytical development for new API products focused on supporting ADC's at our site in Madison, WI. He is also the technical lead for the Keto Sensor Technology Program. Prior to this, he was an analytical development scientist supporting the CDMO offering from our company. He studied synthetic organometallic chemistry and received his PhD from the University of Iowa. His talk is titled Innovative Kaito Aligo Saccharide Based Solubility Enhancer, Gerlinker Payloads, Improved Efficacy and new life for challenging payloads. Over to you, Ross. Thanks Karen and thanks for the wonderful introduction. I'm happy to have the last talk of this session and and my talk is, is is a round of this new technology that we have around you know increasing solubility for for some challenging liquor payloads. And we we started to go down this path because we saw a number of of trends in the ATC community. One of which was a trend towards more hydrophobic or and harder to solubilize payloads and initially moving to to higher dollar and and both of these things kind of go together and and make you know solubility and and aggregation a a challenge for for customers and and a EC developers So so from from that if if you end up with this you know the problem with solubility or or aggregation with your. With your ADC typically the the first step to to solve that is is is you're lowering your DAR and and the reduction of of DAR often requires then an an increase in dosing. And this increase in dosing will often lead to both a a narrow narrower therapeutic window and also might increase the side effects of that therapy and so. You know to if you run into these issues you know trying to solve them is is is challenging you might addition to changing DAR you might want to you know maybe change your formulation or your payload and and all these things these would take you know additional investments both in time and money and then this also increases your your development risk and then in potentially running into you know some IP or or or FTO constraints and you know if you can't solve these things. That can lead to project termination and by by our count, about 22% of all ADC terminations were caused by by poor ADC solubility. So that leads to my first whole question, you know what, what kind of challenges do you face in using and using solubilization, solubilization technology these days? And you know, Bill, Peter, pick whichever ones you know apply to you. And we'll keep an eye on the whole responses as they come in and so so our efforts to solve this problem revolves kind of came out in in the form of Keto sensor. So keto sensor is our technology which is shown here in the orange box. It is a pentosaccharide based on kitten and it can be attached to your. Drug linker is shown here with a with a example with using MMAE where it's attached at the PABC portion. And the idea here is that the highly polar molecule will significantly increase the the, the solubility of your molecule. It should be noted that that while this is based on chitin, we do not do not isolate it from animal sources sources. It's actually isolated from a bacterial fermentation. So there's no worry about, you know, animal origin problems. So you know as as an idea this is great but you know how does it, how does it actually work in practice. And and the first thing we wanted to test was you know using hydrophobic interaction chromatography. So this is a a direct measure of of how hydrophobic something is And and you can see on the top trace we are comparing a bare antibody in this case tretuzumab to an ADC, in this case we made it a Darfur ADC with BCMMAE, so pretty standard ADC. And you see, so the blue trace is the antibody by itself, and then the red trace on top is just the standard ADC and you can see it has shifted to the left indicating it has become more hydrophobic, which is exactly as expected on the bottom trace. We did the same thing using Contusa MAP and this time we made an MMAE based payload, but we used a keto sensor and now you can see it only barely shifts off of where the antibody was by itself. Indicating and only indicating of you know and a reduction of the hydrophobicity. So that's exactly what we wanted to see. This is you know it's really good good proof of concept show that we're that where you know we we're doing what we thought we could do. But Darfur, MMAE based ADC's are not exactly something new. So we want to see you know how far can we push this what, what kind of constructs can we make that were not previously accessible to us. And for that we look to Durkamycin for those of you who are not familiar with that particular payload, it is rather troublesome and in terms of it is very hydrophobic and and often leads to some serious solubility issues like the last example we have using TMABS, so the top is just bare Tmab in the middle we did made a DAR 8 Durkamycin using just a standard linker. You can see the the peak there looks fairly terrible and that's because it's about 50% aggregated. So that so the DAR aid molecules is really of no clinical use and is is would that that project would be terminated, terminated at the very bottom. We made the same construct except now we use ketocenter as part of the linker and you can see again only a slight shift off of the antibody itself. Indicating that it is now only slightly more hydrophobic than than the antibody by itself and this one only had very minor or or insignificant amounts of of aggregation. So this still a very, very useful ADC. So you know from there we wanted to also you know kind of see how far you know what kind, what kind of range can we use, keep a sensor with what what technologies is it, is it. Compatible with and so we we tried you know various linkers, things like those disulphides get these and B type linkers about fits, melamide, basic luconorides, linkers, all of which are perfectly compatible. And we also switched various payloads you know of the of the you know most kind of ones those statins, methansines, durkamycins and on and on. All of which don't don't seem to have any issues. Same thing can be said with various types of antibodies from you know various IgG formats to to engineered by specific antibodies. And initially the same thing goes for the conjugation technologies, you know the standard chemical couplings or enzymatic couplings, Things like transglutaminase can can be used you know to site specific chemistries and stochastic chemistries. And and really what this goes to show is that Kilo Centauer is is widely applicable across many different types of chemistries and that allows that you know a DC developers to be very flexible with their development times or development of their stuff and they can be you know tailor fit to to to whatever needs to be used. So all this flexibility is great you know but you know how does this really work in vivo is it this how does this you know this actually you know become a a good therapy. And from there we wanted to we used an SKO V3 xenograft. So this is a human ovarian cancer xenograft. And we made two different ADC's again both were based on tratuzumab. The first one was the standard Darf 4B CMMAE based ADC and then we also made the Ketocenter based MMAE based ADC and as you can see here from from the graph we did A1 single. Treatment one single doser should say at day 31. The purple line on there is just the the control, that's just the you know PDF buffer. The blue line is the standard ADC that is one that does not have keto sensor and you can see it it it did slow down the growth of the tumor, but it did not stop the growth of the tumor And then in comparison to that you have the keto sensor based ADC. And that's the line in pink where you see saw a C rapid and complete tumor regression and this was you know very exciting, very, very, very excellent data for that to see that. You know we have a very massive increase in in that in efficacy and beyond that some some data but not shown here is that if you dose the the ketocents are based ADC at about the with the 1/3 of the dose. So that in this case 2 megs per gig. You will get about the same result as as the standard ADC. So both of which you know go to show that you're seeing a a very significant increase in in in efficacy here which is which is great but efficacy isn't everything. You also have to look at tolerability for this we went we looked at you know body weight of of the mice and the studies you know at at various. Dosages and it also at you know various stars do so in this case you can see the line on top we have a the Darfur study at the the highest dose which is 6 migs per gig on that's a line in blue. You could see there's really no no effect on body weight indicating that this this therapy is is well tolerated by the mice and that the same is is true for at the bottom for a dar eight study. Same thing, even at the highest dose you could see no, no loss in body weight which is great. So you can see, you can see that that these these ADC's are well tolerated. Tolerability isn't isn't everything. We also have to look at at toxicity. So here we we looked at the Histology of various organs and mice. So these are dosed at again at various at various dosages and for Dars 4:00 and 8:00. And you can see we checked many different organs and and the the the story was always the same. There was there's no, no negative toxicity effects on on any of the organs which is great. So, so you know not only do we have you know good tolerability, we also have no toxicity as well which is wonderful. Beyond that we also wanted to kind of look at immunogenicity. So we we tested some human PBMC's. From three separate donors in this case with Ketocentar and and for none none of those donors we could see any Cytokind release. So that's that's indicating that also Ketocentar does not have any immunogenicity problems which again is wonderful. So going back to that that first study that we had where we saw that you know the dramatic increase in in efficacy what we started to have to tell us why what's what's you know. We can understand the increased solubility, but then what's what's the source of this increased efficacy. And one of the things we started to think about was you know maybe Ketocentar is is increasing the the stability of the PVAC portion of the molecule that is kind of slowing down the release. And this is specifically a problem inside of mice because the captation type captation B lingers are also sensitive to carboxy asterase 1C. Which is present in mice, so you can get some nonspecific cleavage sometimes in mice. And so we think that you know maybe maybe this this ketocenter is also increasing you know beneficial PK properties. So we looked in in an in vitro study in mouse and human serum again with the MMAE based tratuzumab ADC in Ketocenter and. We could see that the stability at both 72 and 96 hours was was very good. So this also this leads to you know there are leads from credence to to the theory that you know maybe this improved improved efficacy is at least in part due to improved PK properties. But you know in in vitro studies are are great so but you know we we just do a little bit some in vivo studies again to to to back that up. And for this we we collaborated with our our colleagues in in healthcare and we did, we made a molecule in this case that's proprietary payload Durkamycin base and proprietary antibodies. So I can't disclose the targets, but it's using a a DART 2 Durkamycin with keto sensor and for this. We did this with human FCRN mice and we took samples of plasma at both 4 and 168 hours, which is 7 days. And we treated these against some antigen positive cells to to see if there was still efficacy against these antigen positive cells. And we could see at both 4 and 168 hours that the results were the same. They were still both very effective at killing the antigen positive cells. So we can show that this also in vivo shows that you know this we still have efficacy after after one week which shows that you know improved stability. So that's great using the same ADC we took it out, took it out to A to a full mouse study in this case we use a xenograph and using a medium expression level. So a little bit harder target to get and and we see the same results as we saw last time. Where the the the non keto sensor are labeled ADC. In this case the the blue line is not nearly as effective as the keto sensor based ADC. In this case again in pink at the bottom where we saw same same results as last time. The the rapid and complete regression of of the tumor and this one this case this study was actually taken out beyond the 80 days to 100 days and eight of the 10 animals or completely tumor free. So this is again this a second, a second example of of how well of Ketocenter is working you know with a completely separate antibody, separate payload and still the the the story is, is the same which is is very exciting. So seeing how this is, you know this is very exciting stuff we want to be able to get you know Ketocenter out into the to the hands of of our customers and and others who want to try it. So we started making some some examples and we're just starting to do that now. So here on on the screen I have a few examples we have in our freezers right now. They're available. If you want to try them, they're free. You can either try them in your own lab or you can try them through our A/C express service. You need some help with conjugation and we're continually going to be adding on to to the the, the, the samples that we have available. So if there's something else you don't don't see here, just just. Maybe give us a call and we'll we'll see what, what we can do. And so beyond that you know this is again great, but then what, what happens when you want to use Kilocentr on the clinic, How how do you access it as from a on a more you know production scale. And and to that we you know want to kind of look towards our our CTDMO services which Caitlin really touched on at at the beginning of these talks. And and I think to to talk about those, it's best to talk about them in the in the through the lens of an example. So let's say you have a proprietary antibody and you want to make an ADC with it using a a linker keto sensor and that's just a a proprietary dolostatin or statin based payload from that we would go. Use our, you know, our global CDMO footprint. So looking over to our our Martiac France site or they can do the the bio development and CGMP production of our of MABS. And then we also have our Sheboygan site which is here in Wisconsin where they make the, you know, critical starting materials and intermediates and then over to the Madison Verona sites. Which is where where I'm located. And then and we here make the highly potent HHBAP highs, you know, payloads, linkers and some of our advanced CGMP intermediates. And then we can also send finally send out all all the stuff down to our our colleagues at Saint Louis at the ADC and Bioconjugation Center of Excellence that you've heard of a number of my colleagues today. And then they can, you know kind of put everything together to make the final ADC package. But taking a step back to our Madison Bronocite and you know linkers and payloads, this is another spot where we saw some struggle with our with our customers and and that's in terms of you know it takes a long time to develop and and synthesize some of these these linker payloads. So shown here on the right is the. The literature synthesis of MMAE also 16 chemical steps and it's you know it takes a very long time to do that. So you to do this you have to find you know qualified CDMO, the one that can handle potent compounds. Often times development of this type takes you know between two and four years and upwards of $5 million per program and you know that sometimes you have to deal with royalty and licensing fees and FTO issues. And and and these are all things that we saw our customers struggling with and we thought you know what what can we do to to help solve this. And that's where we came up with the concept of of our AD course and that is these are advanced GMP intermediates to to kind of accelerate this process that you kind of off the shelf available compounds. And so as an example, so this is using our Delco intermediate, so to the same exact molecule in this case of protected MMAE. It would take from, you know, 15 or 16 steps now down to just three chemical steps. So that's gonna be a much faster time for both developments and for synthesis. Excuse me. So, so you know, this is gonna be a very big advantage to our customers and so these days we've made. A a few of of these AD core products, so we have may core for candenoids, dual core for for dual statin or statins and and PBD core which is for our PBD based payloads and and very shortly we'll be launching actually T can the Q3Q4 time period to have that on the shelf and and all of these compounds you know are are very versatile so we can use them to make. You know plug them into your synthesis wherever it makes sense. So it's really to to gain that you know the speed that that that allows really cutting time a lot of time off of the of the of the synthesis. And then because your your supply supply chain now is much shorter you have reduced risk in in your supply chain and all of these compounds have been synthesized you know in for for high quality and also for you know CGMP in in mind. These are all CGMP intermediates. And and we offer the the regulatory support to to support that CGMP quality. Additionally all of these intermediates are are offered royalty free. So just this on a program sold on a program basis and you know this this will allow you to get to the market faster and and you'll kind of get your ADC to the market faster for that for more exclusivity. And really the whole point of all this is that you know this increase in speed is really going to help save more patients lives you know get. Get the drugs to the patient patients that you know as as fast as possible and to that leads towards my second question. Whole question is you know which which payload class are you most interested in and obviously feel free to check more than one if you're investigating a few of them. And like before we'll keep an eye on the poll responses and that leads to the end of my talk. And so you know just just a touch base with a few things just so keto cents are you know it allows for you know a new and higher DAR AD C's with higher hydrophobic payloads. We've seen you know this, this efficacy boost with Keto cents are across you know multiple AD C's you know that are totally unrelated. And then we can couple our, you know, our keto sensor with our with our AD cores and our CTMO services to kind of have the full package from from start to finish. And for that I would like to thank everyone for their time and I would be happy to take any questions. Thank you, Ross. And and just to add, I think that the, the keto sensor has been a really inspiring story to watch it unfold especially for challenging payloads. And we've also been recently discussing how some of these ADC cores could be useful in the design of new payloads through chemical diversification. So perhaps another interesting area for discovery. So we are a bit short on time. So maybe just one, maybe 2 questions and we will follow up on all the others through e-mail. So the first Ross, does Keto Sonsar affect? The binding affinity of the antibody, it does not. So we've done a number of tests ourselves and then we've also talked to a number of customers who have tried it. And in both cases, there's been no effect on the binding affinity, which is great. Okay, yes, yes, that is great. And lastly, what scales are the ADC core products available at? It, it depends on which one right now we're looking at mostly you know the the few hundreds of grams but all those indices have been designed to be easily scalable. So we can make you know upwards of of of kilos of of each which is you know kind of kind of the idea of the whole point of the the 80 core is kind of we we make them to be to fit whatever the customer needs. So we can get we can supply large quantities if if if necessary wonderful. Thank you, Russ. Let's all please virtually thank Russ again for telling us more about Keto Sonsar and the ADC course we're running a little over. So we'll be closing today's webinar. Thank you for all of the questions. So we will follow up shortly. Before we close, would like to take another moment to to thank all of the speakers, all of the organizers and each of you for joining us today and participating in the questions. If we did not get to that, we'll answer later. As a reminder, the webinar will be made available on our website soon and on demand. All participants will receive an e-mail notification when it is available for viewing. To register for future webinars or to access that webinar library, please visit our website. I hope you all have a great rest of your day. Thank you. Goodbye. _1732521122303

       _1732521122566