Good afternoon, and thank you for standing by. Welcome to Xencor's first quarter 2023 conference. All participants are in a listen-only mode. After the speaker's presentation, there will be a question-and-answer session. Please be advised that this call is going to be recorded at the company's request. Now, I would like to turn the call over to your speaker today, Charles Liles, from the Head of Corporate Communications and Investor Relations. Go ahead, Charles. Thank you. Good afternoon. Earlier today, we issued a press release which outlines the topics we plan to discuss today. It's available at www.xencor.com. Providing abbreviated comments on the call is Bassil Dahiyat, President and Chief Executive Officer. Afterward, we will open up the call for your questions. We will be joined by Allen Yang, Chief Medical Officer, John Desjarlais, Chief Scientific Officer, John Kuch, Chief Financial Officer, as well as Nancy Valente, Chief Development Officer. Before we begin, I would like to remind you that during the course of this conference call, Xencor management may make forward-looking statements, including statements regarding the Company's future financial and operating results, future market conditions, plans and objectives of management, future operations, the Company's partnering efforts, capital requirements, future product offerings, and research and development programs. These forward-looking statements are not historical facts, but rather are based on our current expectations and beliefs and are based on information currently available to us. The outcome of the events described in these forward-looking statements are subject to known and unknown risks, uncertainties, and other factors that could cause actual results to differ materially from the results anticipated by these forward-looking statements, including but not limited to those factors contained in the Risk Factors section of our most recently filed annual report on Form 10-K and quarterly report on Form 10-Q. With that, I'll pass the call over to Bassil. Thanks, Charles. Good afternoon, everyone. At Xencor, we use our array of modular protein engineering tools to create a broad internal development portfolio of antibodies and engineered cytokines in oncology and autoimmune disease. We target novel biologies with our candidate designs, including CD28 costim bispecifics, more tumor-selective CD3 bispecifics, and potency-reduced cytokines. This broad portfolio lets us take multiple simultaneous shots on goal in the clinic and use the proof of concept data from our early-stage studies to guide which programs we advance, which we terminate, and which we partner so that we use our resources on programs with the greatest potential for success and make room in our portfolio for the next wave of XmAb bispecifics and engineered cytokines. We continue to enroll patients in phase 1 and 2 trials across our wholly-owned portfolio of 4 oncology and 1 autoimmune XmAb candidates. Last month at AACR, we presented on our preclinical portfolio of XmAb CD28 costimulatory bispecific antibodies, which are the focus of our research now and are a very exciting new area in immuno-oncology. XmAb CD28 bispecifics are designed to activate T cells only in the presence of tumor antigen and thereby drive signal two activation to amplify and sustain T-cell antitumor cytotoxicity. Our poster highlighted our platform's rapid candidate generation and the broad opportunity for CD28, using data from five CD28 bispecific antibodies targeting a variety of solid tumor targets like CEA, Trop-2, and STEAP1. We've initiated preclinical development of a second internal CD28 program with a plan to file the IND next year. Recall that our lead clinical CD28 program, XmAb808, targets B7-H3 and is in phase 1. Our next data presentation will be later this quarter when we expect to present data from our regulatory T cell-targeting cytokine, XmAb564, at the EULAR Congress of Rheumatology in Milan. We plan to present updated biomarker data from the single ascending dose study that we initially presented in November 2022. For my last comment, I'd like to welcome a new member of the leadership team at Xencor. Last month, we announced the appointment of Nancy Valente as our Chief Development Officer. She'll be responsible for all of Xencor's clinical activities and will join our CSO, John Desjarlais, and me in the scientific leadership of the company. We're delighted to welcome Nancy to our team and look forward to benefiting from her deep expertise in developing drugs like Gazyva, Polivy, and Venclexta and Hemlibra. She was in a unique position to see our development programs, technology, and people up close for the last eight months as a member of our board of directors, from which she's resigned, and chose to commit to work full-time here. Welcome aboard, Nancy. Now, with that, we'll open the call to your questions. Operator? Thank you. At this time, we'll be doing the Q&A session of the presentation. To ask a question, please press star one on your telephone, and you will then hear an automated message advising that your hand is raised. To withdraw your question, press star one one again. All right, the first question comes from Mara Goldstein from Mizuho. Mara, your line is... Hold on just one second. Hi. Thank you. Mara, your line is open. Hi. Hi. Thank you. This is support from Mara. Thank you for taking our question. I have a question on XmAb564. Congrats on getting the slot at EULAR. Just curious, I know there will be additional biomarker data, but I was wondering if you can guide in terms of what we should pay attention to. On XmAb564, I'm just curious if competitors, data sets and development, you know, including Astellas, Rezpeg, and Amgen compound, have given you a positive on this mechanism and why atopic dermatitis and so psoriasis are selected as indication for the match study. Thank you. Any other questions? That was about four. Thank you. That's all. Great. Let's get started. The kind of biomarker data that we're going to present is really an elaboration in more detail on the phenotype of the T cells that the regulatory T cells in particular that were amplified in our single ascending dose study that we presented in November. Really characterizing how those look in more detail. I think we delivered the punchline in November, you know, remarkably durable with sustained T-reg increases out to 21 days. You know, we're excited by the opportunity to try to extend the dosing interval beyond what sort of the class has gotten to, which is every 2 weeks. We're really quite enthusiastic still about the mechanism. I think the data from the competitor that you mentioned, you know, has to be put in the context of their less than ideal selectivity for CD25. You know, John, do you have anything to add to that? No, I mean, just based on our in vitro comparisons, you know, looking at various molecules and what's been available in various posters. You know, we like our selectivity profile better. Ours is a little bit more surgically engineered to, we have preferential binding to CD25 positive Tregs and very carefully potency reduced to maximize the, you know, pharmacokinetic and pharmacodynamic effects. We're still quite enthusiastic about the potential of the mechanism of action. I think it's really about making sure you've engineered what you think is the right molecule to attack that MOA. You asked about our indication selection in our Phase 1b multiple ascending dose study that we also started treating patients in November of last year. We picked atopic dermatitis and psoriasis, and there's 2 goals of that study. One is to rapidly determine a good dose and schedule with multi-dose. We wanted to pick indications that we could rapidly enroll, and that we could also directly view clinical outcome quite easily. These skin autoimmune diseases allow you to do that, look at actual clinical response, correlate that to Tregs and your safety profile. We think in particular for atopic dermatitis, there remains a strong opportunity for good agents that can have long dosing intervals and good safety. Got it. Thank you. Thank you. The next question comes from the line of Edward Tenthoff from Piper Sandler. Your line is open. Great. Thank you. Can you hear me okay? Yeah. Hi, guys. Can you hear me okay? Yeah, great. Hey, I apologize if I missed this, but what, if anything, are you guys gonna be showing at ASCO, and when should we get the next vudalimab update? Thanks. Yeah, we're not presenting any new data at ASCO this year. As you recall, our tempo of presentation has usually fallen towards the late in-year conferences. We'll guide on that, specifics on that as we get a little further on into the year and closer to whatever data updates we're gonna do. We're not guiding full year data. We're really doing a little closer like we did just now for XmAb564. Awesome, guys. Looking forward to EULAR. Thanks. Take care, guys. Thank you. The next question comes from the line of Brian Cheng from J.P. Morgan. Go ahead, Brian. Hey, guys. Thanks for taking my question. Maybe just one question on XmAb564, your IL-2-Fc. You previously mentioned that psoriasis is a great starting point for you to get some proof of concept. You know, as we look forward to some data, early data in psoriasis, you know, in early next year. I'm just wondering, you know, how do you think about the market opportunity for this molecule, you know, given that this may not be the ultimate market for you to move forward with XmAb564? Any guidance on, you know, just from an analyst standpoint, how should we think about the TAM here for XmAb564? Thank you. I'm sorry, how should we think about the what for 564? I missed that word. The total addressable market, for XmAb564. Got it. Got it. TAM, sorry. You know, I think we made a point of adding atopic dermatitis into the phase 1b mix, and we will be dosing patients in cohorts there as we, as we step up a little bit in dose because we do think AD has still a significant unmet need, both for just the breadth of the number of severe patients as well as the need for longer acting agents. We know that there's a lot of excitement about lebrikizumab coming on with a monthly, which is better than dupilumab, which is every 2 weeks and is really bound by the target mediated clearance there. You know, we think there's We try to be thoughtful about adding things that can both have proof of concept as well as a meaningful potential for differentiation into our phase 1b. Remember that this goes to your total addressable market. There's a wide range of indications that you could potentially reach with a broad autoimmune drug. We know that some of our competitors are developing ulcerative colitis. There's development going on in other kinds of diseases like type one diabetes, like just a wide range. As for total addressable market, I think you could be sort of incredibly broad right now, or you could wait to see how we define it further as we pick additional indications as we come out of this phase 1 later, say in 2024. Great. Thank you. Thank you. Our next question comes from the line of David Nierengarten of Wedbush Securities. Go ahead, David. Hey, thanks for taking my question. I just had a quick one on the claudin-6, six CD3, just with the way the ovarian cancer market is evolving, kind of where you would see it slot in potentially. I know it's early stages, but I was curious on that. Thanks. Ooh, that's a tough one. You know, I think, I think it's a market in flux, but still, even with the agents that are coming on, even the response rates are still not that dramatic and people relapse fairly quickly. We think with a distinct target, you know, you would be aiming to be the first time the patients see anything hitting this target. I think you're gonna start with the later line in the clinic, but try to move forward pretty aggressively. There's nothing that slammed the door, certainly, and I think that a cytotoxic agent like this could have a lot of potential. I know that there's competitive landscape with ADCs, with other CD3s, but I still think there's a wide, big unmet need. Yeah. I'll add with a completely different mechanism of action than some of those other molecules. Mm-hmm. I don't think there's, you know, any emerging resistance against those therapies will really read too much on a CD3. Could you remind me at least, maybe not all of us on the phone, but, is there a differential or a requirement or a likely requirement for testing of expression here, or is it broadly expressed? I just can't remember off the top of my head. Yeah. We think it's definitely gonna be prudent to very closely monitor for consistent expression. Okay. Got it. There's definitely some heterogeneity. Okay. Thank you. Thank you. The next question comes from the line of Dane Leone of Raymond James. Hi. Thanks for taking the questions. Can you just give us an update on where you see the CTLA-4 program getting to this year? Thank you. Sorry, Dane, you were choppy. Where we can see the CTLA-4, did you say CTLA-4 PD-1? Yeah, CTLA-4 PD-1 bispecific getting to this year in terms of clinical updates and progress. You know, like as we said earlier, we're not guiding specifically on data timing until we get a little closer to events. You know, we've traditionally updated on that program late in the year, but we're not saying anything definitive yet. We are enrolling well in our, both of our phase 2s, our combination chemo phase 2 as well as our monotherapy phase 2. The first being in prostate only, the second in prostate as well as gyn tumors. We'll update, we'll give guidance on data timing a little bit later. Okay. Thank you. Thank you. The next question comes from the line of Kaveri Pohlman from BTIG. Kaveri, your line is open. Hi, good afternoon. Thanks for the updates and for taking my questions. For IL-12, how do you think about this field given that there have been some disappointments in this space? Can you tell us about your approach and how it could be different from others? My second question is regarding vudalimab. If you could provide more color on your trial amendment for chemo combinations for CRPC. And you are enrolling 20 patients per cohort for different subtypes. Can you tell us what you would like to see from these cohorts to move forward? Thank you. Maybe for the IL-12 question, John, if you want to address our differentiation there. Yeah. Yeah. Well, first of all, I mean in terms of, you know, disappointments, all we have is a, you know, something about, you know, assets being moved around. We actually don't have any of the data from that study yet, so it's really hard to interpret what that means. You know, our molecule is very different. It was specifically designed, you know, following off of what we learned from our IL-15 and IL-2 programs, it was designed to have reduced potency, to the tune of around a 100-fold reduced potency. You know, what we've seen in our preclinical studies, very similar to what we've seen in the human studies for the IL-15 and IL-2 programs, is that potency reduction actually gives you a dramatic improvement in pharmacokinetics. It seems to also impact tolerability and we hope it'll positively impact therapeutic index. We definitely saw evidence in the non-human primates that we have a much more gradual dose response of the pharmacodynamic activity with our IL-12 than, for instance, compared to a wild type IL-12. We think we'll have a lot more flexibility in phase 1 to really, you know, zero in on the best dose that gives you the highest therapeutic index. I'll note there's a lot of noise around other programs that weren't actually engineered cytokines being injected systemically, things that are very complex like oncolytic viruses and localized delivery programs. I just don't think those are relevant comparatives. I don't think those teach you anything, no. Now on to your vudalimab question number 2 about the more on the trial amendment to the chemo as well as what we think the bar for success is roughly. Maybe, Allen, if you wanna touch on that. Yeah, sure. Just to remind people on the 717-04 study, the design, remember we had several groups in there. We had an aggressive variant group. We had a PARP-naive homologous recombination deficient group, homologous recombination deficient PARP-treated group, MSI-H group, and then a biomarker negative group. It's really dependent on the group. Several of the groups have chemotherapy, depending on the chemotherapy regimen, you'd expect a response rate of about 40%-50%. For most regimens, you would wanna see numbers above that. However, some regimens like the aggressive variant or the PARP experience group, the bar will be much lower. Got it. Thank you. All right. Our next question comes from the line of Charles Zhu of Guggenheim. Charles, your line is now open. Hi, this is Edward on for Charles. Just a question on 564 please. How strong is the mechanistic link between. You know, inducing Treg levels, as you've shown pretty clearly, to actually having, sort of a beneficial effect on the disease. You know, there's two sets of data that read on that. One fairly recent coming from actually one of our competitors, the Rezpeg program that Lilly presented last year, showing clear and convincing activity at reducing psoriasis as well as atopic dermatitis, signs and symptoms and in particular, a really profound durability post end of treatment. I'd say the other line of evidence comes from really the last decade and a half of work with low dose IL-2, wild-type IL-2, which sort of mimics a CD25 selective molecule because of the high expression density and selectivity for Tregs for very low doses of IL-2, modest selectivity. There's efficacy that's been shown in, you know, prior intractable autoimmune disease in, you know, like ulcerative colitis, like type one diabetes, like RA, like lupus, like, I mean, lots of other indications. Those are the two threads of evidence. Yeah. I'll add, you know, just going back to real basics. Remember there's that disease IPEX, which where, you know, patients have a defect in the FOXP3 gene, which is critical for producing Tregs, and they have a whole array of various autoimmune type diseases from having that disease. From a very basic perspective, Tregs are critical for preventing autoimmunity. Great. Thanks. Maybe just a follow-up, if I can, on the 104, the PD-1 ICOS. I think you mentioned in the press release that you are going forward with the colorectal MSS stable expansion. Just wondering, are you also considering other potential expansion cohorts? Kinda what pushed you to focus on that one at least for now? Thank you. Right now, we're focusing on the MSS colorectal cancer, really looking for a larger number of patients than we had in our expansion cohorts to see if, you know, we have the kind of activity that would suggest a relevant path forward in MSS CRC. You know, we don't have other indications right now. We're focusing on the CRC. You know, we'll disclose specifics at a later time, but we saw in our expansion cohorts things that suggested MSS might be viable for this agent, in particular in combination with ipilimumab, which is how the study is designed in these additional sets of patients. We're eager to enroll it and gather the data and see. Great. Thank you very much. Thank you. The next question comes from the line of Gregory Renza from RBC Capital Markets. Go ahead, Gregory. Hi, can you hear me okay? Yes. Great. Hi, this is Yin Wang for Greg. congrats on the progress, and thanks for taking our questions. Maybe just on a high level, just with your current cash position, wondering how are you thinking about business development as we have seen M&A heating up in the space, and how would you balance internal innovation and capital preservation in this cash-constrained macro environment? Thank you. Yeah. We really do try to focus on having a high bar for putting molecules into the clinic, because the clinic is where the spend really hits you, right? Our research is incredibly efficient on a both cost and people basis, and that's what's been able to drive so many partnerships, so many license agreements over the years. Right now it's about being stringent about the molecules we put in the clinic and then getting even better, more aggressive and faster about making quick decisions from our early clinical data to either advance quickly or to stop, suspend by partnering. I think you can see from our plamotamab deal a year ago, that program's still progressing, but now that's really in Janssen hands. We try to balance that holding on to commercial rights for molecules of our own so we can build the value of the company with the spend, and we just wanna be as fast and as nimble and as quick about the decisions we make because it is a, as you're right, a capital-constrained environment. You know, partnerships have sustained us over many years, but partnerships require another party involved, and so that's unpredictable. I think it's we're being doubly stringent and strict about advancing programs and the design of our studies to be really efficient. Thank you. The next question comes from the line of Zhiqiang Shu from Berenberg. Go ahead, your line is open. Zhiqiang Shu, your line is open. Okay. Maybe we'll go to the next question, see if we can circle back with him later. Yeah. Wonderful. The next line to be brought up on stage is going to be Jonathan Chang from SVB Securities. Go ahead, Jonathan. Hey, guys. This is Matt Cowper on for Jonathan. Thanks for taking my question. I suppose just to piggyback on that last question on cash, noticed that your R&D burn was a little higher this quarter. I was just wondering if that's something we could expect, you know, going forward as you sort of launch these early-stage programs. Second question is just, I was wondering if you could provide any mechanistic rationale for the development of XmAb306 in multiple myeloma. Thanks. Maybe, John, do you wanna talk about the mechanistic rationale of 306 in myeloma? Yeah. I mean, it's... You know, there's actually two components to that. It's extraordinarily good, as we disclosed, what? About a year ago, at expanding natural killer cells. The combination with daratumumab, it turns out that daratumumab actually takes out some of the NK cells, as it's working. It was sort of a natural idea on Genentech's part to, you know, try to explore the combination with XmAb306 to expand the NK cells further, to sort of mitigate that sort of fratricide effect. Hopefully, it'll go beyond that and even further potentially, daratumumab. You know, the other combination with the CD3 engager cevostamab, that's also just motivated by some of our preclinical work. Recall natural killer cells are generally more sensitive to IL-15 because they express higher amounts of IL-2 receptor beta. It turns out we also were able to see preclinically that when you activate T cells, such as with a T cell engager, you also upregulate IL-2 receptor beta. It's really a natural question. We admire the fact that Genentech, you know, wants to explore three oh six with multiple different kinds of modalities to see, you know, where it has the most utility. Maybe on the cost question, I wouldn't read too much into the quarterly fluctuations of the reported R&D spend. I mean, John, I think can you guide on our cash use this year? Cash is pretty consistent. To address your question, it can be lumpy. The early-stage R&D, I mean, there's preclinical studies, but it's not gonna be consistently increased quarter-for to the extent that it was over the previous year. Got it. Thanks for taking my question, guys. Thank you. Our next question comes from the line of Boris Peaker from TD Cowen. Go ahead, Boris. Great. Thanks for taking my question. This is Nick on for Boris. I just have a quick one on Ultomiris. You guys mentioned in the press release that AstraZeneca announced the phase 3 study initiation. I was wondering if you guys are expecting potential milestone payments from this phase 3, like whether it comes to the end or whatnot. Will you potentially receive royalties from this additional indication if it is approved? Thanks. We receive royalties on any indication from any country that there's valid patent claims worldwide, regardless of the mode of administration, IV subQ. I'm sorry, what was your first question on their phase 3? I missed that. Yeah. If you would potentially receive any milestone payments on the phase 3. No, there's no more clinical milestones left. There's only a sales milestone left, which, we could get this year. We'll see. Okay, great. Thank you very much. Thank you. The next question comes from the line of Michael King from EF Hutton. Go ahead, Michael. Hey, guys. Thanks for taking the question. Most of my asset-specific questions have been answered. I just wonder if John might give us his thoughts on CD28 versus CD3 bispecifics, especially as applies to solid tumors, and, you know, whether the odds of success in solid tumors are gonna be greater for one or the other, or is there no real biology to predict, you know, what might succeed and what might not? You asked me my favorite question. In fact, I almost anticipated that this morning, thinking about it. You know, the easy, the glib answer is we don't really know yet 'cause there haven't been enough CD3s versus CD28, so there certainly hasn't been any sort of head-to-head studies against the same tumor-associated antigen. There's nice validation of CD3s in solid tumors, you know, and a couple different programs, and there's really nice validation, recent validation for CD28 in solid tumors. Really, to answer your question, the way I think of it, how differences might emerge. A CD3 engager is basically gonna take any T cell that it can find, either in the periphery or the vicinity of the tumor. We don't know if those T cells are tumor-reactive. They could just be bystander T cells, and it's gonna utilize those, and it, you know, takes advantage of having all those T cells around. CD28 really has to build off of an existing signal one. If you're doing a combination with, say, a checkpoint inhibitor with a CD28, the signal one actually comes from the T cell receptor recognizing neoantigen peptide, you know, MHC complexes on the tumor cell. I'm guessing, and it's just pure speculation, that CD28s would have potential for longer durability of response, just because you're actually building up a memory T cell response. You're specifically expanding those tumor-reactive T cells. Okay, thanks for taking the question, John. Thank you. Our next question comes from the line of Peter Lawson from Barclays. Go ahead, Peter. Hi, this is Shayon for Peter. Thanks for taking our question. Just quickly first about vudalimab. For the second study in the gynecologic tumors and high-risk prostate, could you maybe speak to a little bit whether you're thinking of just pursuing monotherapy, or is this also going to be ultimately combo therapy? For 808, the B7-H3 bispecific, could you maybe speak to where you see the highest chance of success in solid tumors and whether that might be prostate based on prior data and how you might see that fitting into such a competitive space? Thank you. I think with vudalimab, we expressly ran the phase 2 monotherapy because we think there could be potential monotherapy in those GYN indications and in that slice of CRPC. I think, you know, Allen can address some of the data we got in expansion cohorts about why we believe that. Yeah, I was just gonna add that, so remember this is a XmAb717-05 study, so it's our second shot on prostate as a monotherapy. We defined the high-risk prostate group clinically as opposed with molecular markers. We're looking at GYN signals as an early marker in the expansion. One thing is that we're looking at cervical endometrial as well as ovarian cancer, which is an unmet need, and checkpoint inhibitors have shown activity, and there hasn't been a lot of activity. I will also add one of the neat things about the XmAb717-05 study is that it studies a better dose and schedule. We're doing a flat dose and a every 3-week schedule, as opposed to every 2 weeks. The other question was around B7-H3. You wanna have that? Yeah. Yeah, Yeah, I'll take that, Allen. Yeah, so, it's a great question about B7-H3 for XmAb in a week, which solid tumor indications are we most excited about? Prostate for sure is right at the top. You know, there's certainly nice validation targeting B7-H3 in prostate cancer. Of course, we've got nice validation of CD28 bispecifics actually working in prostate cancer. Other histologies that we, you know, we would certainly think about would be small cell lung cancer based on some proof of concept data there targeting B7-H3. You know, B7-H3 is expressed across a lot of solid tumors and we're gonna wanna take a look at a lot of them. Great. Thanks so much. Thank you. The next question comes from the line of Zhiqiang Shu from Berenberg. Go ahead. Great. Thanks very much for taking the question. I also wanna ask a question on your CD28 portfolio. Obviously you present some preclinical data at AACR. I wonder your criteria in terms of selecting the other targets for your future molecules, and also how you decide to move one or more targets into the clinic. The criteria is really based on where we think CD28 could have the biggest impact. We think that's in tumors that have been cold for immune checkpoint therapy. We think the example given by the PSMA CD28 data presented by Regeneron last year suggests that this CD28 pathway could turn on to response tumors that would be cold. They did that in prostate cancer, albeit in a small, very small number of patients. There's a huge unmet need and a lot of patients that would be ecstatic if you could turn on turn on tumors that are nominally pretty cold, like small cell lung cancer, like colorectal cancer. Ovarian cancer. Ovarian cancer. Like, I mean, there's. The list is longer than the list of tumors that actually respond well to immune checkpoint therapy. Even the ones with some labeled checkpoint inhibitors are pretty weak, right? Like gastroesophageal and gastric, HCC, et cetera. we're focusing on targets where there's a big delta possible, right? It looks cold to checkpoints, and where you could really maybe change the landscape, but also where you could clearly see a signal early in clinical development if you start seeing response. I think that's important because we don't wanna be lost in a sort of a swamp of trying to deduce whether it's the checkpoint or it's the combination agent, our CD28 driving the activity, like you might be if you were enrolling patients in melanoma or RCC, which are pretty warm tumors. Got it. Just quick follow-up on the technology side of things. It appears all your CD28 and CD3 bispecifics you are using 2 plus 1 format. I guess, Is 2 plus 1 your preferred format to go to other targets as well for CD3, CD28? Yeah, I would say, you know, is the 2 plus 1 is a tool. It's just one of the tools in our toolkit. Yeah, every target's different, right? There's all kinds of considerations like density. You know, how different is the expression on tumor cells versus normal cells? You know, we like the 2 plus 1 for the CD3s because it gives us a little bit more, we believe, room to design in the therapeutic index to preferentially kill tumor cells versus normal cells. CD28s are a little bit different. Where we think we want or need the 2 plus 1, we'll utilize that. 1 plus 1s are also on the table for the CD28s. Great. Thanks very much. Thank you for your questions. I would now like to turn it over to Bassil Dahiyat for closing remarks. Thanks very much, and thanks everybody for joining us this afternoon. We look forward to giving you further updates later in the year. Goodbye. Thanks a lot. Thank you for your participation in today's conference. This does conclude the program. You may now disconnect.