Nautilus Biotechnology, Inc. (NAUT) Earnings Call Transcript & Summary

Good morning. Dan Brennan, TD Cowen, Life Science Tools & Diagnostics Analyst. Day 1 of the 46th Annual TD Cowen Healthcare Conference. Pleased to be joined here on the stage with Co-Founder and CEO of Nautilus, Sujal Patel. So Sujal, welcome, and thank you.

Thanks, Dan, and I appreciate the invite to the conference.

Terrific. And we have Anna Mowry in the audience, the Chief Financial Officer.

So maybe just to start off, the zooming really far out, and then we'll go into your progress on technology. I thought it would be interesting to understand from your perspective, where Nautilus fits into the proteomics ecosystem, if you will. What do you consider some of the differentiators or things you're trying to solve for, if you will?

Great. Well, that's a good place to kick off. Maybe I'll take a second and back up because I don't know how familiar the audience necessarily is with the proteomic space. One of the things -- just in the story form, one of the things humanity has conquered in the last couple of decades is we've conquered measuring the genome. I can take a drop of your blood, I can tell you what 99.9% of your DNA is. It's accurate, reproducible, it's fast and cheap. The problem is your DNA doesn't really change from the day you're born to the day you die. It doesn't contain the real-time state of what's going on in your body. And because of that, it has limited utility in therapeutic development in precision medicine. Us as an example, 95% of our FDA-approved drugs target proteins, not genes. And so measuring proteins is the next frontier. Proteins do all of the work in your body. They make up the vast majority of the functional parts of your cell. And we, as a scientific community, do not understand proteins very well. Proteins have a lot of complexity. There's 20,000 different gene encoded proteins. We don't have good instrumentation that can measure all those proteins sensitively and sample and reproducibly. And then more complex than that is that once a protein comes out and it's transcribed and it's in your body, it gets modified by lots of different chemical processes, picking up modifications in different forms. And if you don't understand those forms, you don't understand biology well. Nautilus is a company that is trying to develop a brand-new platform to comprehensively measure proteins in sample. That is, what is the gene encoded protein, how is it modified? And we're using an approach that is an approach that's developed by my Co-Founder, Parag Mallick, who is Stanford faculty and is a very unique and different approach that hasn't been tried before. And I'm sure as we continue our conversation, we'll get into it. Nautilus itself is about 9 years old, and we are in the process of building a benchtop instrument that delivers easy-to-use proteomics to any biologist who wants to measure the proteome comprehensively from a sample. That's very different from the state-of-the-art in the proteomics space. The gold standard in proteomics is a complex workflow that sits in front of the mass spectrometer, which is an instrument that is used all over in metallurgical analysis, food safety, chemical purity, but it's used in this proteomics use case using a complex set of preparation steps ahead of it. We sell billions of dollars of mass specs every year into these protein discovery environments, yet that tool doesn't really provide reproducible comprehensive results. And we're out to build this platform that comprehensively measures the proteome.

Terrific. So maybe next, and we can keep going down that vernacular, if you will. So what proteomic applications will the platform enable or unlock today, both on the proteoform side and then on really the broadscale proteome side that maybe aren't possible. So again, speaking to you kind of alluded to some of the drawbacks, but maybe go one level deeper of kind of what you'll seek to do with both of these technologies?

Yes. Let me discuss -- you use the word applications. Let me discuss the word applications in 2 different ways. One, I'll describe our applications, which you mentioned, right. Broadscale, which is what we call comprehensively measuring all the genic proteins in a sample or proteoforms, which is a specific use case. Those are cases [indiscernible] customers have their own use cases for [indiscernible]. Proteomics is used in a wide variety of drug development process upfront to take cells that are healthy, take cells that are sick, you want to understand it in a significant level of detail, what are the differences between them, what cell surface proteins are potentially biomarkers that are indicative of disease, what are my potential targets which I might be able to drug to have a positive impact on a disease? That target identification step and understanding what's going on, that step is already significantly hamstrung by existing technologies, which can't see all of those biomarkers sensitively. They can't see the rare things that are differentiating healthy and sick cells. Next stage of drug development. Once I've got compounds, there's a lot of work that goes into understanding the mechanism of action of those compounds. What are the effects on the proteins in the cell when exposed to a compound? What are the secondary effects on other organs in the body, so toxicity, cross-reactivity types of applications? These are all very critical steps upfront in drug development that would have a massive impact, hopefully, positively, if you could use a platform like ours, to dramatically reduce the cost and efficacy of that drug development process. In diagnostics, that same sort of use case exists, right? How do I find a sensitive biomarker that's going to be indicative of disease or stage of disease? How can I monitor therapeutic response by looking at what's going on inside of the patient's proteins. All of these types of applications are significant applications that our customers have identified as pain points because existing technologies are not adequate. So when you think about what our technology does to map on top of that, right, the primary thing that our customers want to do in a lot of these types of applications is understand. If I have a sample, maybe it's 100 to 1,000 cells, it's like a standard sample size in pharma. I want to understand what are all the proteins in here and what are the proteins and how do they change as different disease states are present. That primary use case is what we call broadscale. And our value proposition in that use case is that we have an instrument that's far more sensitive, that's far more reproducible than what's out there today, which means that you have more actionable results, and the results coming off our system are more comprehensive. The mass spectrometer-based workflows and the other types of products that exist in the market, they still really can't effectively see more than maybe 1/3 to 1/2 of proteins that are in the sample. They can't see [indiscernible] its detection threshold. You see 100 to 1,000 molecules [indiscernible]. And these are critical questions in biology that we can [indiscernible]. The other application of our platform is one that we have begun to take to early access this year. And this is an application that helps to zero in on proteins of interest and look at the modification landscape of those proteins. And so for example, in early access this year, we launched our Tau assay, which is capable of measuring 768 different forms of one single protein, the tau protein. Tau protein is a critical protein to study in neurodegenerative diseases, like Alzheimer's disease. And we have an assay that is capable of measuring 768 different forms of it, which is revolutionary. No one has ever seen all those forms of tau. And no one has understood because they've never seen it. How is that related to your likelihood of getting Alzheimer's disease in the future, the disease progression? How is that related to the therapeutic programs that have been attempted. And how we might -- we'd be able to impact that. This proteoform use case is really interesting because the data that comes off of it has never been seen by the world. It's a use case that's a little different than broadscale. Every protein I want to go after, I have to build a new assay, that takes us some period of time. We did announce on our earnings call last week that we have a second marker that we're working on in oncology. And then we announced earlier than that, that Michael J. Fox Foundation and Weill Cornell, Qatar, our collaborators and Michael J. Fox is funding an initiative to study alpha-synuclein, which is the key biomarker in Parkinson's disease, so another neurodegenerative marker. And so we have more activity going on there as we build this portfolio of proteoforms. And we think in the long run, a single platform, which we showed for the first time last week to the scientific community, a single platform that we're going to release at the end of the year is capable of running all these proteoform assays and our broadscale assays in one single machine.

That's a lot. It is -- yes, if you're successful, it sounds like it's going to be quite exciting. Maybe just go back to U.S. HUPO. You presented some latest updates on the Nautilus platform, I think on the proteoform side. Just speak to some of the key takeaways from the presentations.

Yes, that's great. So HUPO is the Human Proteome Organization conference. They do it twice per year. They do World HUPO, which is generally international, and then they do a U.S. version of it. The U.S. version was last week in St. Louis. And it was a really exciting opportunity for Nautilus because for the first time, we showed the instrument to the scientific community. And so on the earnings call, what we talked about was that we have a number of proteoform assays that are moving through early access to general availability this year. We have an instrument that will reach launch by the end of the year, with generally available placements at the beginning of next year. And we announced that our broadscale capabilities, we expect to launch those in early access in the second half of this year, general availability, first half of next year. So we've got a lot of things to talk about. Scientific community got to see our instrument for the first time, which was really exciting to demonstrate that. And it was a really important proof point because Nautilus is building something that is very difficult to build, very easy for the customer to use, but the task of building what we're building is very hard. And so for the scientific community, this was a massive tangible step where they see the instrument, they could use the touchscreen and operate it. I think at 4:30, there's a proteomics panel, Birgit Schilling, who's our PI at the Buck Institute and in the audience here, will be speaking. Birgit was at our event and saw the instrument. Now she has had the instrument in her lab in alpha form. Buck Institute has the only alpha of our instrument since April of last year. And so she has lots of information that she'll share. And then as well, Birgit at U.S. HUPO presented some really exciting data using her biological samples and our instrumentation and her operators, generating interesting biological data. I'll save that data for her to talk about. But really exciting progress.

Okay. Maybe the data at HUPO, as you just mentioned, was proteoforms, I believe it was proteoforms of tau. And you kind of talked about kind of several new biomarkers, which may become available or in progress. Just could you speak -- and you've already alluded to one, but just how do we think about -- we're not putting the cart for the horse. How do we think about kind of how quickly you might come out with additional biomarkers on the platform?

Yes. So let's just like just separate those 2 use cases, right? So broadscale is a use case where we build it once and sell to everybody. Proteoforms, we're building assay by assay. And the criteria for building these assays today is, number one, an important biomarker where there's significant drug programs and drug development dollars behind it and an area where the forms of proteins likely have a difference in terms of the protein's function in the cell or its degradation or its distribution or any of the sorts of characteristics. So areas of interest are neurodegeneration, among a lot of other areas, neurodegeneration, oncology, autoimmune, inflammatory, cardiac. And so what we've done is we've taken a set of 200 or 300 potential interesting markers. We've mapped on that availability from our partners for antibodies that target different site-specific modifications so that we don't have to build those today. And we've stack-ranked those. We probably have 20 that are kind of on our hit list. I mentioned that we're going to do an oncology marker next. We're actually -- like there's 5 markers that are all great markers. We don't actually even know yet which one we're going to do. We're going to test the antibodies and whichever one is the fastest path is the one that we're going to pick first, and then we'll probably tackle another oncology marker right behind it. And from there between neurodegeneration and oncology, alpha-synuclein will come out the other side. And then we may move to another area. We may continue to double down on those 2 areas. But I think that when we think about this long term, we think about this as a steady road map of proteoform assays. And in the long run, we think that having a large portfolio of proteoform assays plus an instrument that does broadscale makes it a really compelling value proposition for the customer.

And in terms of the first tau, 700 different variations of it, what's been the early interest? I would think that's such a hot area, and there's an established understanding and awareness, and there's a lot of pharma companies and researchers chasing that. So I would think offering this, you would generate a lot of leads. Just any color you can provide on the funnel, what you've heard from customers on that front?

Yes, that's a great question. So once we've started to show this data, which we started showing in a preprint last year, Birgit presented data at World HUPO last year as well, which was very early data off the platform. There's been a tremendous amount of interest from the scientific community. Now a lot of that interest is in early research because this is data that no one's ever seen before. No one ever thought you could measure 768 proteoforms of tau. There's been a raging debate for decades in the Alzheimer's disease research community, is the pathology of tau driven by random hyperphosphorylation, or is there a pattern of how kinases got you to particular forms? In our first data sets, we started to see evidence that there could be a pattern there. Like incredibly exciting, but it's going to take some time to develop, partially because some of these folks have to now apply for grants. Some early innovators like the Michael J. Fox Foundation saw what we're doing and said, "Hey, I have to jump on and do this for alpha-syn." So it's beginning to build. But as a company, we've been running very capital efficiently. Up until today, there was not a single salesperson in the company. So we have one now. And so we're just now beginning to build that capacity. So building the funnel is basically from scratch at this point. And as well, I think, as I mentioned in the earnings call, we got to launching that Tau assay to early access a little earlier than expected as well because it's performing incredibly well. And so with that, we're a little behind on sales capacity, but we're just getting started on that. I think that this year, we'll see some of those early projects build, and then we're going to move those projects into grant proposals and further funding. One of the things I do want to highlight, though, is that the move out of neurodegeneration to oncology is driven by the fact that we are in tau, not because we did some great market research study and said, "This is the best place to go first." We're here because we started working with Genentech 4 years ago, and they really wanted to study this, and it was a great joint learning experience for us. Tau might be a tiny step out of sync with where early drug program development is for the data that we're going to put out. But we think oncology is a really great fit for the type of data that we're going to get off of the platform and where drug development programs are that could be impacted by it. And so I think I'm super excited about what's going on in neurodegeneration, but I'm maybe even incrementally more excited about oncology as we start to get through the second half of the year.

Can you just elaborate a little bit on that, like why you think the marriage between where the market is and what the technology enables, maybe oncology is even like a faster lane, if you will?

Yes. So I think that there's kind of 2 parts of it that I would highlight, right? One is that on the neurodegeneration side, these -- the disease biology is extremely complicated. And we don't yet have the capability to analyze biofluids, CSF, blood. We only are dealing with tissue samples. Tissue samples for a brain that's afflicted with AD, these patients died. And so samples are hard and getting an impact out of what we're doing is going to take a little bit more time, right? In oncology, there is a belief in the scientific community, at least folks that I've talked to and that I know our team has talked to that the modification landscape and the proteoforms of some of these key markers is critical to understanding therapeutic response and biology of these diseases. And our system, this predominant sample type today that we're working with, is cells and tissue. That's an easy sample type to get from tumor biopsy. And so there's a lot of alignment on the sample side and a lot of alignment on the biology. And then remember, if the sample is easy to get and we are able to understand the proteoform landscape in great detail. It's not just about drug development. It's understanding what therapy should I give the person based on what I'm seeing in terms of the proteoform that exists. Like these sort of precision medicine use cases, I think, are much more tangible earlier for us in oncology than in neurodegeneration.

And in terms of the platform, whether the proteoform or the broad scale, in terms of working with different matrices, is there any barrier towards working on different matrices over time? I mean, right now, you're in tissue, but how will that evolve you think?

Yes. I mean, so for our broadscale capabilities, we will have the capabilities to do cells and tissue. We'll have the capabilities to do blood. And then over time, those capabilities will get more and more complicated. Some customers want to do a preparation to only look at cell surface proteins. Some want to do some sample preparation, minimally on blood to reduce albumin or some of the proteins that are really abundant that take up space on an experiment that is unnecessary. So those -- that's the road map on the broadscale side. On the proteoform side, every marker has a little bit of a different sample preparation associated with it. And so when we think about a product for tau, a product for oncology marker, number one, it's a combination of our assay and the sample preparation techniques that go and come together. And so for example, on tau, our internal team has developed a protocol for sample preparation from frozen tissue that enables us to analyze these proteoforms of tau. Birgit's lab at the Buck Institute has been using that protocol. For the oncology marker, we'll have a similar sample prep that's kind of bundled up with it.

I got it. Okay. So maybe one more on the proteoform side, and then we'll get kind of zoom out for the broadscale. But on the proteoform side, you mentioned how well the technology early on is working. Like how would you define -- I guess, even on the tau product, how would you define success? And we'll ask Birgit this later today. But like, what are the features? What are the measurements? Obviously, if there's new discoveries made, terrific. That will take years. But just analytically, like from a quantitation standpoint, maybe, what are the measurements that you'll -- the customers you think will look at to say, "Wow, this really is delivering what we thought, and it's very differentiated and unique."

Yes. Well, let's just address -- you said it your thing, but I'm just going to say it out loud, right? Ultimately, our job is to enable our customers to make discoveries and positively impact human health. And I don't know that will happen in tau. I don't know that will happen in the first oncology marker, but I am certain that out of the hundreds of markers that are out there, many of them will have relevant proteoforms that are significant discovery that positively impact human health. So that's our end goal. It's going to take years, yes, but that's the end goal. One of the things that gave us a lot of comfort around this Tau assay was that we did a lot of validation studies, and we far exceeded our own metrics in terms of what we would want to get this thing into early access. First of all, if you look at the preprint that we had, we did a set of studies to build up to real biological samples, studying organoids, human -- mouse brains, looking at human control patients and AD afflicted patients. And when we did those analyses, one of the things that we saw was we saw incredibly, incredibly tight CVs, very little variation, high reproducibility in the samples. And when we did spike in studies that would take particular forms and increase their ratio and a mixture, we saw exquisitely accurate reproduction of what we expected coming in. Just to give you a sense, if you looked at our preprint and looked at the variation in our data across different operators, different reagent lots, different instruments and different chips and flow cells. So change all the different things in our system, the variability is 5%. Our product management team when they were building the spec for this product set that at 25% because that's what everyone else can do. 25% is like a norm. And we accomplished 5%. It's the [ highest ] variation that you saw on the system. It gives us a ton of confidence that there's really great data quality coming off of the system. And for our customers, data quality is absolutely critical. One of the things you see in proteomics from some other vendors is you see these 10,000, 20,000 cohort studies that are being done. A lot of the studies are done that way because the variation is so wide. You analyze the same sample twice and 30% of your IDs change, like you have to run a lot to go and get data. So if you could deliver more accurate data, it's really transformative to a customer. And that's the most exciting thing that I think not just I'm excited about, but as I was at HUPO last week and I talked to the scientific community, the things that they were excited about in our early results.

Terrific. So maybe just we have 6 minutes left. Maybe I'll ask one more, not big picture question, but talking about the broadscale discovery platform, which you've said throughout kind of the last year or 2 as we've followed the company, like that's really -- the proteoform is exciting, but the broadscale really is where you think the real massive opportunity is. So you've talked a lot about what are the milestones ahead of feeling good on that launch and now you've got that launch in the second half of the year. Just again, remind us in terms of what we need to see between then and now, what your level of confidence is on those time lines?

Yes. Let me just slightly modify your statement, and I'm going to tell you that, that broadscale, I believe, is the inflection point for our top line because it unlocks a sale that is looking for additive data to the mass spec-based traditional workflow in a similar price point, accessing a similar budget pool. We think that's the revenue inflection point. When I was at HUPO, 2/3 of the people I was there said, "I love what you're doing on broadscale, but oh my God, I love even more what you're doing on proteoforms" because this is data that is net new to the world, no one has seen before. So I think in the long run, the proteoform business, particularly when you combine the 2, is going to create an incredibly powerful and sticky business model for us. So that's kind of setting the stage. In terms of broadscale, there has been a ton of complexity over the course of 9 years in getting to the point where we can get broadscale out in the marketplace. And at the beginning of 2025 on that first earnings call, one of the things that we said was, hey, we're going to need another year because we had to go through a pretty significant assay configuration change, which was focused on getting more of the reagents that we've -- proprietary reagents that we've been building to function correctly on the platform. And for those that have listened in on our story, you know that, that broadscale depends on us building a set of 3 -- maybe 350, 400 proprietary reagents that map each molecule. And these affinity reagents are antibodies that we call multi-affinity probes. They bind very short regions of the protein in a nonspecific manner. It's a very specific class of antibodies that we have spent almost 9 years developing techniques to build. Not enough of those antibodies work on our platform. We have thousands of candidates, very little yield. And the reason was that the assay configuration needed to change to allow more of them to work. So we went through a hard process in 2025 that took a little longer than we'd like to get through that assay configuration change. And as Parag on the last 2 earnings call has talked about, we've begun to sort of move through validation steps in our new configuration. We've been able to decode simple mixtures of proteins, 10 proteins, 15 proteins. We've been able to identify proteins that are present in cell lysate, and that's an important step. The next important step for us will be to be able to accurately quantify some reasonable number of proteins, 500, 1,000, 2,000 out of some complex sample like cell lysate. That's not an endpoint by any means, but our system combines these data points together computationally in an exponential manner. So by the time that I have 2,000 or 3,000 proteins, 500 or 1,000 proteins, it doesn't matter what, more than half the work is done. The assay configuration change will have been done and stable. And we are using that marker as kind of the benchmark for which we will say, "Okay, we're ready to get the early access program announced for broadscale, start signing up customers." And by the time that we are ready to analyze the first sample, we'll have a greater number of proteins ready to go. And so that's an important milestone for us and through investor conversations, I know it's a milestone that a lot of investors are looking at as well because that shows the whole thing has come together.

So the goal of that or the plan for that, if there's a second half launch, second half could be December, it could be August, sometime between -- before August or December, we would see this announcement, I guess?

Those are good bookends, yes.

Okay. Just in terms of -- we have 2 minutes left. So how do you -- like how do investors contemplate then kind of the road map then for the company over the next couple of years? Like cash on the balance sheet. You're at this point. Maybe speak a little bit how much you spent to get here. And as you begin to unlock these opportunities, kind of what happens? How targeted do you go just to make sure things are on track? Like how quickly can you ramp? Things like that?

Yes. Yes. I mean you asked a question, how should investors think about it? I would take a more broad view, first of all, like I want investors to think about Nautilus as a company that is building something bold, hard and disruptive, right? And those are companies that when they succeed -- which I am confident we will. When they succeed, they have a transformative effect on markets, right? We're not an incremental sample prep system. We're not yet another assay that looks like Olink, which now Thermo Fisher owns. We are a net new approach that's doing something very different. It takes a lot of capital and a lot of time to do that. And we have been very, very careful with our cash and our balance sheet and very careful with our development so that we have the capital on our balance sheet, which is there today. We ended with $156 million of cash at the end of last year. We have the capital that we need to finish building our broadscale capabilities, deliver on the entire road map I discussed earlier, build a commercial team and launch. We have capital, as we've stated through 2027, not into, but through 2027. And we have what we think is a good plan forward for capitalizing the business as we continue to grow and move towards cash flow positive after launch. So that's kind of how I think about the important markers for investors.

Okay. Well, we've got just maybe a few seconds left here. So I mean, I don't know. How would you wrap it up in terms of -- we've talked about key milestones. We talked about products. We've just talked about kind of the future. How would you like to wrap up from here, the Nautilus story?

I mean I would just encourage investors who want to learn more to reach out to me, our IR team, Anna Mowry, our CFO, is in the audience. We'd love to talk to you about the company and count you among our shareholders. So thank you.

Terrific. Thank you, Sujal. You got it. Thanks for being here. Thank you.