In Vivo Antibody Discovery with Humanized Transgenic Mice

In this episode of the Biologics Brief, Mosaic Biosciences' Chief Strategy Officer Tracey Mullen leads a conversation on in vivo antibody discovery using humanized transgenic mice. She's joined by Dan Rohrer (Chief Technology Officer, AbTherx), along with Mosaic's Eric Firfine (Chief Scientific Officer), Maria Lo (Head of In Vivo Antibody Discovery), and Stacy Capehart (Director of Data Sciences).

The panel traces the evolution of transgenic mouse platforms from first-generation models through today's third-generation Atlas Mouse, discusses what "full human diversity" really means for downstream drug development, and covers how Mosaic decides which discovery platform, phage, yeast, single B cell, hybridoma, or transgenic mice, fits a given target. They also walk through a real CD22 discovery campaign, covering immunization strategy, hit identification, kinetic characterization, epitope binning, and developability profiling, and close with a look at where AI-generated antibody workflows fit alongside in vivo discovery.

Topics covered:

·       The evolution of transgenic mouse platforms and what makes the Atlas Mouse third-generation

·       What "full human diversity" means and why it matters for antibody discovery

·       Freedom to operate considerations often overlooked by bench scientists

·       How Mosaic selects the right discovery platform for a given target

·       A real-world CD22 discovery campaign: immunization, screening, and hit identification

·       Binding affinity, epitope binning, and developability results from the CD22 campaign

·       Strategies for managing large antibody hit panels without losing good candidates

·       Where AI-generated antibody workflows fit alongside in vivo discovery

If you're evaluating an in vivo discovery campaign or want to talk through the right platform strategy for a difficult target, reach out to the Mosaic team, we'd love to help.

🎧 Listen now and subscribe wherever you get your podcasts.

mosaicbio.com

Transcript

Welcome And Agenda

Tracey Mullen 0:04

Welcome back to the Biologics Brief from Mosaic Biosciences. I'm Tracy Mullen, Chief Strategy Officer at Mosaic. Today we're going to be talking about in vivo antibody discovery and specifically what it looks like to run a therapeutics discovery campaign leveraging humanized transgenic mice and trying to reach difficult targets. So this is a space that's matured a lot over the last 25 years from the first fully human transgenic models all the way through to the platforms that we work with today. And of course, it sits kind of right at the center of how a lot of therapeutic programs get started. Of course, there are a number of ways to generate antibodies. We use most of them at mosaics, so things like phage display, use display, single B cell screening, hybridoma discovery. And every platform provider will tell you that their approach is the right one. But of course, we all know that the right platform really depends on the biology of the target. So today we're going to talk about where transgenic mice fit into that picture. You know, what makes a modern transgenic platform different from earlier generations, you know, what a full human diversity model actually gives you, how you should evaluate results from a campaign, and where in vivo discovery sits alongside some of the newer approaches like AI-generated workflows. We'll also walk through a real campaign, a CD22 discovery effort leveraging the Atlas full human diversity mice to make some of this a bit more concrete. And to do that, I'm joined today by a very strong panel. So we have Dan Rohrer, Chief Technology Officer from AbTherx, whose team has been building transgenic platforms and running therapeutic discovery for over 25 years, could really speak to the Atlas mouse and how it was designed. From the mosaic side, I'm joined by Eric Furfine, Chief Scientific Officer, Maria Lo, head of In Vivo Antibody Discovery, and Stacy Capehartt, Director of Data Sciences. Thank you all for joining me today.

Maria Lo 1:48

Thank you. Nice to be here. Excited for the conversation.

Tracey Mullen 1:52

So before I get started with setting some questions, I think it might be helpful just to set some context. Mosaic, as I mentioned, works through

Why In Vivo For Hard Targets

Tracey Mullen 2:00

a lot of different discovery modality. And we are kind of on principle platform agnostics. So when we decided to make the vivo discovery and specifically the Atlas Mouse a priority strategically, it wasn't because we thought that one technology should win. It was really because we kept seeing that programs coming through our door required additional, more sophisticated approaches. More and more clients were bringing us harder targets, complex cell surface receptors, GPCRs, ion channels, other multipass transmembrane proteins. And for these types of targets, it can be challenging, although not impossible, but challenging to leverage alternative platforms like in vitro display for antibody discovery purposes. The other thing that we were noticing was how often time and costs were lost in the handoffs, starting with native wild-type animals. So when you start from a fully human transgenic model, you get to skip that humanization step entirely downstream. And that takes a lot of risk and a lot of time out of programs and really lowers the immunogenicity questions as well that you sometimes carry into development. And then pairing that with some of our single B cell efforts, you can go right from target concept in general to naturally paired development-ready sequences in very rapid timeline, with of course fewer detours along the way. So the Atlas Mouse, for that reason, was very additive for us. It gave us a modern in vivo option that fit the targets that our clients were actually struggling with and bringing to us real time, and also fit very cleanly into the downstream science that we already run here at Mosaic. So that was sort of the market read that made this the right choice for us and a big part of why this collaboration with EvTherx exists. So with that as the backdrop, let me bring in Dan because I think the natural place to start is really with the platform itself. So Dan, AbTherx has been building transgenic platforms for a long time. Can you give us a quick sense of where the Atlas Mouse sits in the evolution of this technology and what makes it the latest generation and why that distinction matters to someone running a discovery campaign?

Dan Rohrer 3:56

Yeah, sure. I will start with some of the historical background here. I think that's that's important. But together with my colleague Peter Brahms, we have been working in this space, transgenic platforms expressing human antibodies, for over 25 years. And we were fortunate enough to have worked under Niels Lomberg, who was really kind of the pioneer in this space. And the first gen platforms really were transformational in advancing therapeutic antibody development. And that particular platform we worked on at Metterex over the years has actually resulted in 13 different approved therapeutics, which are still generating over $30 billion a year today. The relevance to therapeutic antibody discovery is still quite high. But despite that, the first generation platforms actually were subpar in terms of their immune performance and uh certain characteristics about how much repertoire they captured. And so that really ushered in in an early 2000s platforms such as represented by Regeneron or ChiMab, so-called second gen platforms. And and there they actually greatly improved the mouse immune performance and also captured a much broader uh diversity of uh human variable regions. And so those are also quite important in the evolution of this tech overall. Now, going into the 2010s and 2020s, we actually saw an opportunity to even go beyond what that second gen had been brought in. And really the areas that we wanted to improve were in actually making the model itself being a little more streamlined. And so to that end, we essentially created a single-step knock-in technology, whereas those prior platforms are multi-year, multi-step genetic engineering undertakings. And so we greatly simplified that step. We also, in terms of gene content, we removed all of the pseudogenes and non-functional variable elements from these constructs, uh, making more compact transgenes and more efficient in terms of B cell recombination. So those aspects kind of contribute to what we consider to be third gen. But in addition to those aspects, we also are bringing on new model types altogether that are really addressing uh more modern therapeutic needs. And to that end, we're talking about a bi-specific binary fixed light chain model, which assists in making either bi or multi-specific antibodies, as well as our long CDR3 model, which is actually leveraging longer CDR3s to get at difficult targets like GPCRs and ion channels. So, you know, as taken as a whole, improving on the full diversity standard offering as well as the other two models, we really believe we're kind of in this third gen space and can help address almost any kind of target type that might come to you guys.

Tracey Mullen 7:02

Thanks, Dan. So Maria, let me come to you for the execution side, kind of now feeding off of what Dan already shared from upstream. So, from a scientific standpoint, what does it actually look like when a client brings an in vivo campaign to mosaic? You know, what are they handing off? What do they get back?

Maria Lo 7:18

Yeah, thanks for the question, Tracy. The one

Third Generation Atlas Mouse Design

Maria Lo 7:20

good thing about Mosaic is that really no one size fits all. We're we're very flexible and clients can engage with us at pretty much any stage of the in vivo discovery process, depending where they're at in their program. We have some clients who are just getting started and then they just have a target, broader target class in mind or a target ID. Some have the protein. And really, based on the target, we can recommend the most appropriate immunogen format since different targets often require different strategies to maximize the chances of success. I think we'll talk more about that today as well. Some clients already have reagents, screening tools, some even have the sequences. We can really take it wherever you're ready for us to start. And a reminder here that, you know, having these reagents is really important for our screening. So having these tools in place is critical, and we are really here to help you generate these if needed. Um, again, our workflow is really flexible, so it can hand off at whatever point makes most sense. Some clients really want us to run the entire discovery program while others are starting in at a very specific stage and want us to hand off after a certain milestone is reached. And then another important point I want to make too is that we really are a sequence for discovery platform. And so one of our primary deliverables is these sequence high-quality sequences, and we try to get these as early as possible in the process and hand them off to you right away. And from there, we can take it as far as people want. We have capability to express antibodies and even validate them at the larger production scale. We also offer downstream characterization services, including functional validation and developability assessments. So all of this really allows us to have our clients receive the candidates ready for the ready for their next step of drug discovery.

Tracey Mullen 9:01

Thank you. Dan, back to you. So knowing all of that, you know, if I'm a scientist evaluating transgenic platforms, what's one thing about the Atlas mouse that I probably don't know but should to help me make better decisions?

Dan Rohrer 9:13

We will get into a little bit about kind of the diversity measures that come out of the mice and so forth. But you know, one aspect that I think might be underappreciated here and not typically part of someone's decision-making process is really around freedom to operate. It's a strange concept, but uh as you mentioned at the top, this is a fairly mature field. It's been around for over 25 years. And with that maturity, there's a lot of intellectual property and a lot of space to navigate in there in order to make sure you have freedom to operate. You know, it for a typical scientist, if they're working on a particular program, they can get some fantastic panels, fantastic hits. And essentially, if they move from that early discovery space into actually development, there comes a time where that sequence actually gets exposed to the rest of the world. And it's not only the sequence, but the method under which it was discovered. And one can run into issues there. I think this is probably underappreciated by most bench scientists, which is that if for whatever reason your candidate or the approach uh is actually infringing, this can completely stall these discovery programs. And so looking at this up front is probably a good idea. And Aphtheryx has always placed a very high premium. We have a philosophy of IP first in in terms of understanding what the landscape is and doing our best to ensure that the platforms we've made really have that freedom to operate pathway. So it's probably a strange topic to mention in this context, but uh it's an important one because it can actually mean go or no go for a project that can have a lot of great scientific input.

Tracey Mullen 10:55

Yeah, it's an important consideration too. I think it's something that's often overlooked. So glad that you glad that you called that out. I I think just staying with you for a moment, you know, the Atlas MUS is described as a full human diversity model. What does this actually mean in practice for those listening who don't fully appreciate that yet? And why does the diversity of of this matter so much downstream?

Dan Rohrer 11:16

Yeah, another great question. And a kind of interesting one from a biological perspective. As most of us know, the immunoglobulin genes are an area of the human genome which probably has some of the greatest diversity across the entire genome. So that's a first point. And typically when one talks about antibody diversity or diversity possible in the immune system, it's typically provided as a range. And the reason for that is there are unique evolutionary adaptations across different ethnicities, geographical areas, exposures to pathogens. So we don't all have the same complement of immunoglobulin genes. And so our approach was really to survey a number of uh well-published repertoire studies to understand both the commonalities and the differences there. The first priority is let's not bring something into the mouse that might be immunogenic in certain populations. So we're excluding rare variants. But at the same time, we really want to get the most common set of V genes, as we'll describe them, into the mice to be sure we're being as broad as possible and capturing good diversity. So I'll just give you one vignette here, and that is we make use of a Tera SabDAB, which is a hosted by University of Oxford, and essentially it's a compendium of all of the therapeutic antibodies and antibody-like uh molecules that are that are out there. And if you kind of stratify that database and look at the fully human offerings, now I'm talking about the heavy chain variable region uses in this case. And our mouse has captured 40 of those 42 in terms of expression. We feel that that's very wide capture of the diversity that's there. We've done an analogous exercise with the kappa-light chains as well. So what we're trying to do really is balance this idea between getting as much diversity into the mouse as possible by while also not using variable regions that might actually be immunogenic across uh broad human populations. We have seen, as you'll probably report, you know, very broad epitoke coverage. And I think that sequence diversity clearly translates to epitopic diversity. So the mice are are delivering those two aspects uh right now.

Tracey Mullen 13:51

Thanks, Dan. So with that background information, let's open

Running A Flexible Discovery Workflow

Tracey Mullen 13:55

this up to the mosaic side a little bit. Uh maybe going over to Eric, you know, we have access to multiple discovery platforms at Mosaic. Again, phage, yeast, single B cell, hyperdoma, et cetera. So when a client comes to us with a new target, how do we decide whether transgenic mice are the right starting point for us?

Eric Furfine 14:13

Yeah, it's a it's a very good question. And it's something that actually we think hard about. We know what the benefits are of each of our platforms and they're overlapping in in their qualities, but sometimes distinct things can come out of one versus another. For example, if your target is highly homologous to the mouse, it can be more challenging to for the mouse, as good as the ephtherics mouse is, to generate the diversity that you'd want. And those might be places where you might favor using in vitro methods. As you go in after harder targets and things like GPCRs and ion channels, that's where you know the in vivo methods and and especially the ephtheryx mouse are are really going to give you a bang for your buck. So I guess the the point that I'll make is really that we think hard about this and we think about your target and we do it with you. And we come up with a plan together with you, telling you what the pros and cons are of of each of our platform approaches. Maybe one last nice thing about the diversity of platforms that we have is that we can blend them. And maybe you want to start with an in vivo platform. Maybe you start with the apterix mouse. And while one might sometimes go with uh B cell screening, you know, you can actually make an immune library and put that, port that into the yeast display platform and get potentially an even larger diversity there. And so the idea to be able to blend things together and take advantage, essentially full advantage of the diversity of approaches we have can be really helpful to improve your probability of success.

Tracey Mullen 15:53

That's that's helpful insight. Thank you. Yeah, I think what would be really interesting for the next topic for discussion here is maybe pivoting from that and making this a little bit more concrete. So we we do have a CD22 campaign that we promised we'd be talking a little bit about today. Dan, maybe we can, you know, have you set the front end of this, tell us a little bit about the target, the immunization approach, how we got to hit identification, and then we have Stacy here who can walk us through what the data showed across finding characterization, developability, epitope coverage, and ultimately, you know, what we learned about the platform.

Dan Rohrer 16:25

You know, we wanted to pick a therapeutically relevant target and one that others had worked on so we could have some basic idea of benchmarking. So human CD22 was actually chosen. And in this case, it was from the in vivo side. Essentially, what we did was we carried out the immunizations and the primary screening of the B cells that came out of it, and at that point handed it off to mosaic. But the immunization aspect was fairly standard. You know, we're doing a lot of sub-Q immunizations to really engage lymph nodes. That is our primary source for B cells in this particular case. It was about a five to six week immunization period. We then harvested the lymph nodes and extracted the CD138 high cells from that exercise. And working with our partner single cell technologies, we actually went to a single B cell screen with that material. And from one of their essential microchips, we identified 6,000 IgG secreting B cells. And from that, we identified 541 CD22 specific unique B cells. And that is the point where we share data with U at Mosaic to actually go deeper.

Stacy Capehart 17:45

Yeah, I'm happy to take that one. So yeah, once there were 541 sequences identified, we wanted to be able to express and purify and characterize a small subset of those sequences. And so we tried to capture as much diversity as possible. So we identified sequences, a small subset of sequences

Risk Proofing With IP And Diversity

Stacy Capehart 18:10

around 20 that had unique clonotypes, diversity in terms of their CDR3 lengths, and then we filtered out obviously any sequences with liabilities such as you know oxidation or deamidation or glycosylation and unpaired cysteines in the CDRs. And so we selected about 20 sequences to express as IgG human IgG1s and purified those. And then we took those into a full kinetic characterization by SPR, and we were able to identify a nice variety of Ks, KDs, and big KDs, so the equilibrium dissociation constants ranging from micromolar to subnanomolar affinities, which is really exciting to have that diversity. And then after that, we also did epitope binning for these 17 antibodies against CD22 to see which antibodies had blocking or non-blocking interactions to identify unique clusters. And we were able to identify eight unique bins for these antibodies, which is really exciting out of a small number, just 17 sequences. And then from that point, we were also interested in looking at not just their affinities and not just their diversity in terms of their epitope binning, but we also wanted to see the developability profile of these antibodies. So we assessed that by doing a suite of developability assays that we perform regularly here at Mosaic, which included hydrophobic interaction chromatography to look at the hydrophobicity of the antibodies. We did ACSINS to look at self-interaction of the antibodies, BVP ELISA to evaluate polyspecificity, and then we also evaluated thermal stability using TMTEG. And from that, we were able to classify each of these antibodies in terms of their overall favorability, uh in terms of their developability profile. And what was really exciting is that not only do we have a nice variety of binding affinities, nice coverage and epitope thinning, but we also have a really overall nicely favorable developability profile for this small subset of antibodies that we expressed and purified. So overall, the results were were really exciting to see for this CD22 campaign.

Tracey Mullen 20:57

Yeah, that's that's really great to hear. And thank you for walking us through those results. I think one thing that stood out to me was that you know you were starting with a panel of 500 plus sequences and really had to go through the process of narrowing those down. I think, you know, one thing that discovery teams always ask about is how many good antibodies they could be missing out on in the sea of hundreds of sequences. So maybe Dan, I'm curious if you could speak to, you know, how to handle large hit panels so that you know unexplored hits don't get left behind and you know, whether you can speak to the candidates that weren't characterized, at least for this campaign as an example.

Dan Rohrer 21:30

Well, I I look at this in kind of two different ways. And the first is kind of what Stacey already mentioned, which is, you know, starting out by filtering for clear, you know, bioinformatic or in silico, you know, liabilities. So it just takes a number of them off the table, probably a small percentage of them, but it does take some off. And then really you start drilling down into clonotypic families, right? And you want to make sure you're surveying those clonotypic families and really making certain that you test as many. Of those families as possible. And if you find something interesting within those, you can actually go back to the clonotype family, drill down deeper, and interrogate in that particular way. So there are ways to actually deconvolute such a large panel effectively. I think Mosaic is champion at that type of approach. But there's other things you can build in too. You know, you really have to have a good target product profile to start with. That is, what are you the characteristics you're looking for in these antibodies? And if you can build those screens in up front, you know, it actually takes these large panels and makes them more manageable in terms of their size and so forth. So if it's possible to start wide and narrow down in a logical, sensible way, that's really the way that I would address it. And I think you guys are quite good at that. I would take it back to you in terms of how you strategize such a panel, whether we're doing it in the same way.

Eric Furfine 22:57

You know, one of the things we did after Stacy outlined her her approach, which showed that there were unquestionably great antibodies in the mixture, was to use two different AI ML methods to select sort of different ones. And what was amazing, while there was some overlap in what the AI methods selected based on our first computational approaches, there was a really most of them were not the same and were not overlapping. And what I was really impressed with about this mouse was that no matter how we selected the diversity that we wanted to this representative, we got really good antibodies all three ways. And it just really speaks to the quality of the candidates that that come out of this that almost, you

CD22 Campaign Setup And Screening

Eric Furfine 23:45

know, you you can't miss because no matter how you pick them, there are good ones in there. And it wasn't an accident that we we got lucky in that first experiment.

Tracey Mullen 23:55

Yeah, Eric, maybe we'll we'll stay with you for this next question. I'd be curious to hear, you know, if if you were advising a biotech that has never run a transgenic mouse campaign before, what would you tell them to pay attention to, you know, when they're reviewing results? And what should they be asking their CROs to enable the best chance of success?

Eric Furfine 24:12

Yeah, I think one one thing you want to ask them is, you know, what what differentiates your platform and why is this platform good for my target? I think those are the key things. And I think we have a lot of really great answers to those questions. One is this may be the best humanized mouse on the market in terms of the quality of hits that come out of it, as I mentioned before. And, you know, for those tougher targets where sometimes you you want to really dig deep into diversity, we have the ability to do that by porting into the yeast display platform, as I mentioned before. But sometimes actually you want to do function first. And so using our B cell screening platforms allows you to do that. So I think the combination of the quality of this mouse with the methodology that we have to screen what comes out of the mouse really makes us special. And I think that's the question you should ask.

Tracey Mullen 25:10

Thanks, Eric. I agree. Definitely. So, Maria, maybe this one's for you. So we've built out Mosaic's capabilities to handle N vivo and NVTRA Discovery, uh, including some newer tools. And just would love to hear from you as head of NVivo Discovery here at Mosaic, where you feel the Atlas Mouse fits into that broader toolkit. You know, is it is it additive or does it replace something in our current workflow?

Maria Lo 25:34

Yeah, definitely. And Eric definitely touched on a lot of this already, but just to reiterate here, this is this really is expanding our toolkit rather than replacing any existing platforms. Like, like Eric has said, each technology has its own strengths. I mean, our goal here is really to apply the right platform to the target biology and the client needs. Obviously, a huge benefit of the Atlas mouse is that we're getting fully humanized, fully human antibodies through our natural immune response of the mice. So because they're fully human from the outset, we don't have to do a separate humanization step, which really allows programs to move directly into the downstream characterization. And we're still getting, just like wild type mice, we're still getting the in vivo affinity maturation and sagrable developability characteristics from the mouse. And yeah, like Eric has said, sometimes phage or yeast are sufficient and fast and suit your needs. But for some of our more difficult targets, this mouse might be a really good choice and provide a high probability of success of these maybe rare candidates. So, and like we've been saying, no one size fits all. Um, one of the benefits of Mosaic is that we have all of these different options and we can really work with the clients to decide what the best fit is and leverage all of our different platforms and you know choose different strategies to complement each other. So it's really not replacing anything, but it's improving our flexibility and making us able to tailor each campaign very uniquely. So I think that's one of the main strengths of Mosaic that I see.

Tracey Mullen 27:04

Thank you, Maria. Yeah, and I noticed in your response there one thing that we didn't really touch on too much yet, although it has come up, is of course the AI aspect of antibody discovery. Um and I know there's you know a lot of AI workflows that are sort of enabled, if you will, at Mosaic. Looking forward, you know, you can't get through a podcast without asking this question. You know, as AI generated antibodies become more common.

Characterization Affinity Binning Developability

Tracey Mullen 27:30

Where do we see human transgenic mice fitting in? Maybe this is a question for Dan. You know, are they are they complementary? Are they competitive? Something else entirely? I'd love to get your insight, Dan.

Dan Rohrer 27:41

Yeah, thanks. I would love to come back to this in five years and see how our answers today compare to what is actually happening. But clearly AI in a more general sense, not just with reference to making AI generated libraries, but in a more general sense, it is totally revolutionizing antibody discovery. Uh, there's no doubt about that. And it will continue to do so. What you have talked about already really resonates with, I think, where AI fits in, and that is that you guys can pull a lot of levers. And uh AI should be one of several. And of course, I'm here representing the in vivo side, so I'm gonna be a little bit biased there, but I realize I'm part of an ecosystem, and it's a very well-validated part of that ecosystem, but still I'm just kind of prefacing where we fit into this, you know, larger scope, but I do have a bias towards the in vivo system, and it it really starts with some basic biology. You know, the humoral immune response is essentially something like 500 million years old in evolutionary terms. So that is has extreme power behind it. And an in vivo-evolved immune response is actually simultaneously solving for affinity, expression, a number of other parameters, self-reactivity, things like that, that is really hard to replicate currently with AI tools. So this thing that's happening in vivo really is quite amazing, I think, when you take it in terms of what it's able to deliver. So I want to say that, but I also want to say to Eric's earlier point, there's gonna be approaches where in vivo just doesn't work that well. And certainly highly conserved targets are a great example of that. You know, if you've got something that's 100% conserved between mouse and the human target, um, you're gonna want some other parallel or differentiated technology to try and get around that breaking tolerance type of issue. So I'm really excited to see where it will play its biggest role. I think it already is playing a great role in terms of this kind of hit picking, which Eric referred to, you know, developability characteristics, a number of other aspects. And I think getting to the point where an AI-generated library is the first thing you choose, we're a ways off from that. But I know that it's coming. And I think a client that comes in and is completely agnostic, they just have an interest in the target and the biology and getting a therapeutic, I would say you're well advised to be as broad as possible. I would definitely make use of in vivo. Maybe you try AI on the side, maybe you try that as a complementary approach. You're certainly going to be using AI tools in either process to kind of mature hits, move forward. It's exciting times, and I really do despite the fact that it may uh be the death knell for in vivo systems in say 10 years' time, I think uh important to pay attention to it now and really use it where

AI Hit Selection And Outlook

Dan Rohrer 30:50

we can.

Tracey Mullen 30:51

Absolutely. Yeah, no, I and actually I think that that's a really good place to stop too, because we we just don't know. We don't know the future, right? And so we'll we'll have to wait and see. So before we close, I just would love one takeaway from each of you, all the panelists that we have here, one thing that maybe this collaboration demonstrates why it's important. Maybe Dan, you can start and then we'll go around from there.

Dan Rohrer 31:13

I think we were just super impressed with, you know, in terms of if this was a handoff like in a football game, you know, what you guys did with the ball after we gave it to you, and it really spoke to what you've all talked about here today, which is the breadth of your biological experience and ability to direct things where they need to go to address a client's needs. And I I just think uh, you know, that breadth part and your scientific expertise is very refreshing. And, you know, I think we're uh fortunate to be a partner with you.

Tracey Mullen 31:50

Thanks, Dan. Eric, over to you.

Eric Furfine 31:53

Yeah, I would just maybe reiterate a little bit of what Dan said, which is that the mammalian immune systems and even non-mammalian immune systems have an incredible track record of creating great antibodies. And while there are limitations to in vivo methods, I think to the extent that you can incorporate that approach as part of what you do is wise. And it's clear that the aptheryx mouse is is a really good one to choose for that.

Tracey Mullen 32:20

Thanks, Sarah. Maria?

Maria Lo 32:22

Yeah, I'm very excited about this collaboration in general. From the lab itself, we've been doing some pilot studies with these mice and getting some really exciting results. So we'll be excited to share more about those soon.

Tracey Mullen 32:34

Great. Thank you. And Stacey?

Stacy Capehart 32:36

Yeah, just to reiterate, I think just the performance of the results from the full human diversity mouse in terms of the CD22 case study, where we saw, you know, a really nice diversity uh of epitope bins and affinities and really favorable

Closing Invitation To Reach Out

Stacy Capehart 32:55

developability profiles. It's really exciting to see see those results.

Tracey Mullen 33:00

Great. Thank you, Stacey. And I think yeah, that this is a great place to end. I just want to thank everybody here for joining us today. Thank you to Dan and the app Vheric team, to Eric, Maria, Stacey, you're at Mosaic. If you are thinking about an in vivo campaign and you want to talk through the right way to structure it, this is exactly the type of conversation we'd like to have with you. So please reach out. Uh in the meantime, just want to thank all of you for listening in to the Biologics Brief. We'll see you next time.