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

Good afternoon, everyone. I'm Matt Sykes, life science tools and diagnostics analyst at Goldman Sachs. I have the pleasure of welcoming Anna Mowry, Nautilus Biotechnologies. Thank you, Anna, for being here. I appreciate it.

Maybe just starting off a big picture, can you give us kind of a brief overview of the proteomics market as you see it today? And how Nautilus fits into that broader proteomics ecosystem? Particularly what differentiates Nautilus platform versus what's out in the market today?

Matt, thanks so much for the introduction and for the invite to participate in the conference. We're happy to be here. In terms of proteomics, we think that this is a really exciting business opportunity as well as a key area of research for academic and pharmaceutical organizations. We estimate that this market will grow to $55 billion by 2027, and billions of those dollars are spent in discovery proteomics, which is an area that's really relevant for us. Now proteins are very different from DNA. DNA is the same in every cell in your body from the day you are born to the day you die. And on the protein side, proteins are the key drivers of biology. They are often the difference between whether you're healthy or sick and 90% of drugs target proteins. So despite all of this spend, our ability to measure proteins is very limited. The key workhorse is the mass spectrometer. It is the gold standard today. With that being said, it's very difficult to use, it requires specialized labs and skill sets, it takes weeks of analysis on complex samples. And still, the end result is you can still only see 8% to 30% of proteins in a sample, which doesn't give researchers a clear picture of what's happening in the sample. So newer techniques have come about to try to address these challenges, particularly through affinity or sequencing like approaches. Now the challenges with those techniques is that they typically work on peptides, not proteins or specialized sample types and don't have the scale and dynamic or sensitivity that is necessary to answer the questions that our customers are asking. So Nautilus was founded to meet these challenges head on with a bold new approach to democratizing access to proteomics by bringing to market a new instrument with the aim of comprehensively measuring the proteome from any sample from any organism. So what I mean by comprehensive is that through our in-house developed, what we call multi-affinity reagents combined with our computational approach, we believe we'll have the potential to see 95% of proteins in a sample. Our platform is also designed to measure a single molecule intact proteins over chips that -- or 3 chips that has 10 billion analytes. So that scale and sensitivity, we think will be transformative for those developing new drugs and have the potential to kick off a new wave of precision and personalized medicine.

Got it. Super helpful. And maybe just drilling down a little bit, what types of applications can you unlock with your instrument versus other proteomics tools in the market?

Yes. So because our technology is built to see comprehensive coverage as a proteome, it really makes sense to go after the use cases where customers need to see everything. And in particular, this is helpful in the drug development workflows. In the early stages of drug development, customers or pharma companies don't necessarily know which proteins they're looking for. They're looking to see the rare differences between healthy and sick cells. And then from there, they can -- that helps them identify which targets to go after. Our technology is also really useful in the later stages of drug development because once you apply the drug, you want to measure the response, you want to understand the mechanism of action, you can also dig into later stages like toxicity and even diagnostics over the long term. And through our hundreds of customer conversations, our customers tell us that what they're discovering today is at the very edge of what can be seen by the mass spectrometer. And so we believe that once we unlock greater visibility into the proteome that it can kick off a new set of discoveries.

Got it. And I think I can easily say in my time in covering this sector, you guys are attempting to do one of the more difficult scientific challenges that I have seen. So it's incredibly impressive what you've built and what you're moving towards, and what you're doing is highly innovative and novel and it comes with its challenges. Can you maybe talk a little bit through about what led you to kind of delay your commercial launch to the end of '26? And any color you can provide on the progress you have made to address some of these challenges?

Absolutely. So it's no secret. It's taken us a bit longer than we anticipated. It can be true when you're trying to solve as difficult of a challenge as we have with a completely new approach. But let me give you a little bit more color on what led to the delay. In our platform, there are a few key pillars of our technology. Number one is our instrument and software. We've got instruments running in our facilities in San Carlos, California every single day. We also have our flow cell. And then the third aspect is those multi-affinity reagents I talked about. And over the past couple of years, we've developed thousands of probe candidates and have demonstrated to ourselves that we have -- the antibodies we have developed do, in fact, bind to a diversity of epitopes within proteins and have the ability to differentiate amongst proteins, which is necessary for our platform. Now what we've also realized is that we have defined the specifications of what type of antibody we need or the characteristics of antibodies that will work on our platform. And so some portion of our probe candidates do convert into platform-ready labeled probes. But what we found is that our fallout rate is just too high. And so there's a number of ways we can address that. We could scale up our pipeline -- development pipelines and just keep working that process. It's just a very inefficient way of doing that. So over the past couple of quarters, we've been looking at ways that we can make our platform better align or the specifications of our platform better align to the characteristics of our existing probe candidates and also make our development pipelines more efficient. And in Q1, the path we chose is ultimately one that we feel will result in a more robust assay and allow us to tap into more of our probes that we've developed, but it does take longer. And so that's what led to us pushing our launch time line to late 2026. Now we're still in the middle of that process, although we do have pilot experiments with our new approach that we're working through. It just will take us a little bit more time before we can give any meaningful update.

Got it. Can you maybe just walk us through the differences. You talked a little bit about it at the outset, but just the differences in capabilities across a targeted analysis and broad scale or more basic research. How would you characterize use cases from a customer's perspective across those 2 offerings?

Right. So everything we've been talking about so far is really focused on our broad scale application. This is where we want to comprehensively see each of the 20,000 gene encoded proteins that exist in a sample. As you point out, we do actually have 2 modes of operation in our platform. Both use the same common core platform, but the targeted application is really where we look deeply at a particular protein of interest that's already been well characterized with -- or at least has antibodies -- existing antibodies that have been developed by others. Because our platform is an open platform, we can leverage those antibodies that are raised towards site-specific modifications on proteins of interest and use them on our platform to see what's never been possible. We started this just as an example, with our collaboration with Genentech, where we were looking at the tau molecule. Tau is a highly modified antibody that is known to play a role in Alzheimer's disease. Now there are other techniques out there that do measure how many of a particular post-translational phosphorylation site is present, but there's no technique on the planet that has the ability to say, how many tau do we have that's modified 2x or 3x. In that context, this is where our platform comes into play. And in Q1, the update Parag gave is he said we completed our internal verification and validation of our tau assay. And through that process, we used 12 distinct antibodies to see -- which allows us to differentiate up to 4,000 different unique forms of tau.

Got it. And staying on tau and neurology, you've talked about as being sort of an initial focus for the targeted proteome offering. Can you kind of talk through why kind of existing targeted proteomic methods like mass spec have challenges in the neurology space and how Nautilus differentiates itself, particularly in these low abundant proteins?

Yes. So one of the key features of the mass spec is that it works on peptides. And so the downstream implications of analyzing peptides has some of the ramifications that you're pointing out. So when you work on peptides, your -- there's -- it really eats into your dynamic range. And so if you're measuring one whole protein, but then you -- as we are, that takes up one analyte, you chop that into, let's say, 100 peptides. Now you've got 100 more molecules to analyze. And so that impacts your dynamic range by, let's say, 2 orders of magnitude. You also have to reassemble that data to try to figure out what existed on the whole protein level. And so you need to see hundreds of copies in order for you to confidently say what proteins are present. And so that can lead to you making calls on the most abundant proteins. The other aspect is that you lose the ability to say, as I mentioned before, which whole protein molecules had multiple phosphorylations on one molecule versus a set of independent ones. So our technology was really designed to look at single intact proteins, which allows us to address many of those challenges. So -- and then we combine that with our 10 billion analytes per run, that means that we have -- theoretically, we are now impedance matched to the pharmaceutical workflows where they're looking at 100 to 1,000 cells, we spread all those proteins across the surface of our chip. We can potentially see even the rarest proteins in that sample.

Got it. And what are your expectations for the early access partnerships in terms of either revenue impact or just relationships in 2025. Can you remind us sort of on the economics around the EAP program that you've got? And will you be actively placing instruments during that time?

Yes. So I think what we've said is that we expect to put our targeted assay into the hands of researchers in 2025, and we're working to sign a partnership for those -- for that in the first half of 2025. This is really -- the goals of this are really around helping potential customers get that external validation that we need as well as to demonstrate the power of what it means to be able to see -- to dig into a protein of interest at this level. What we're not doing is modeling revenue associated with this in 2025 because this is really more of that demonstrating the power of this technology and helping us to evaluate the use case. So just in terms of the go-forward business model, that's still something we're evaluating. But I think, as you mentioned, we could do this through a services offering. We could do it through joint development because there are customers that have significant investments, in particular proteins, and they may need some help in evaluating those proteins. And then we could also do platform in the traditional way. So those are all options that we're evaluating and the collaborations and partnerships we're working to establish in 2025 will really inform that business model as opposed to being the source of revenue.

Got it. Understood. And then outside of neurology, there are other applications that you're intending to pursue in thinking about potential partnerships in the targeted offering?

Yes. So a lot of what we talk about is tau related at this point, but the platform itself can actually work with any number of proteins of interest, especially ones where there are a critical mass of well-established antibodies that have been shown to work for site-specific modifications and so on. And so we are at the stage where the tau assay is pretty far along, and we can start to think about what's next. And in Q1, we established a series of customer conversations with more than 30 customers across academic, nonprofit and pharma. And these conversations are really intended to help us understand which type of customer we want to go after, understand their priorities and think through how we go to market. And I think in terms of the possible proteins of interest, it ranges from furthering our investments in the neuro side, but we can also start to look at proteins of interest in immunology or cardiology or oncology and areas where we think that -- or where it's estimated that a particular form of that protein is known to have a biological -- or thought to have a biological impact. And so that would be a really great place for us to go next.

Understood. Turning to the broad scale opportunity. How should investors track progress of your assay reconfiguration in order to gain more confidence in that late 2026 commercial launch?

Let me make 2 points here. So I know I talked a lot about the opportunity with tau, and I think there's a big business opportunity -- or there's a business opportunity we're evaluating on that side. But one of the really great benefits of moving forward with our targeted assay and tau is that it does leverage the same common platform. So it's -- our instrument in software is being used regularly in that assay. And starting with partnerships and collaborations, starts to build our customer-facing muscle and gives us -- as we start to process samples, that's just all aspects that ultimately lead to commercial readiness when our broad scale application comes online. In terms of the broadscale application, what I said is that we need roughly 300 reagents to get comprehensive coverage, and we don't need all 300 to be able to begin our commercial launch. And so what we tell investors to look for is that those initial data where we start to see any meaningful number of proteins in complex samples could be 1,000 proteins. That's the sign that the core pieces of our technology have started to come together. And then from there, I think it's a very different time line to get the remaining reagents that we need to get to comprehensive coverage.

Got it. And then just given the novel nature of the broad scale platform, how would you characterize use cases versus what's out there today? I mean, part of me thinks that you're going to be potentially finding proteins that have not really been known before. And so there's a lot of it sort of like we'll kind of tell you when we get there. But could you kind of talk about maybe what some of the attractive use cases would be for someone for the broad scale, maybe in the academic end market?

Well, what I would say there is that there's a huge amount of research being done today with data that generates a subset of visibility into the proteome. We believe that in the existing market, once customers have access complete picture of proteins in a sample and have this single molecule sensitivity and dynamic range, it unlocks a whole new set of discoveries that customers will learn to incorporate us in more and more places. The second thing I would say is that proteomics is really concentrated in folks that have access or specialized labs and skill sets with a mass spectrometer. And we believe our platform will be more like a genomic sequencer where it's push button simple, easy to use, sample in data out in the cloud. And so what that does is it potentially brings in a whole new set of scientists who want to do proteomics but can't because they don't have those specialized labs and skill sets. And so we think that would be a market expanding opportunity that brings in new folks into proteomics that aren't able to access it today.

Yes. And I wanted to kind of drill down on that point because one of the challenges with mass spec, which you've said already, but we would definitely agree with you that the workflow and the prep is actually pretty challenging. And when you're trying to do something at scale or at speed, it just makes it difficult. So maybe dig a little bit into your kind of sample type workflow for your instrument and how that is sort of sample in, answer out and how simple that is and maybe compare it to what the scientists would do with the mass spec?

Yes. So on the mass spec side, it's a fairly complex process to get a sample through the mass spectrometer. So the first step is you take your sample, you extract the proteins, then you digest them into peptides. You also fractionate your sample. So instead of having one sample, you have many samples and then get -- those proteins get ionized. And then on the other end, you have to go through this very complex process of recombining back your data points to determine what's in the sample. On our side, we consider our sample prep process to be more similar to -- on the genomics side, where we extract -- in our case proteins. Not DNA. But we extract the proteins, we attach them to our proprietary label and then our instrument will take those proteins and spread them across the surface of our chips so that each well will have a single protein molecule. And then from there, our instrument will do its multi-cycling and the data goes into the cloud, very different.

Got it. And maybe just walk through how you're thinking about pricing the instrument and sort of what the potential pull-through on consumables would be in your sort of early estimation?

Yes. So our first opportunity is really going after the mass spec budget. So the -- those instruments today are priced -- depending on the vendor and application, they're priced below $1 million, up to closer to $2 million, especially with some of the more recent instruments being released. And so given our value proposition relative to the mass spec, we've chosen to -- we estimate -- we haven't released official pricing yet, but we estimate we will price our initial instrument package for roughly $1 million. This includes both the instrument itself as well as the initial install and training as well as some support, maintenance and software service contracts. In terms of pull-through, we anticipate pricing our samples at a few thousand dollars per sample. Our instrument is designed to run 12 samples per run. So at modest utilization estimates, we think that our pull-through per instrument will approach $1 million there as well.

Understood. And as you think about sort of the mass spec market, even during times of sort of more challenging macro spending environment, CapEx constraints, there always seems to be a market for cutting-edge mass spec and really specifically in proteomics. And we've actually seen research dollars over time gravitate more towards proteomics because a lot of it has to do with what you've already talked about, which is proteins is the mechanism of action of disease. It's more of the commercial and the biopharma and some researchers are more interested in. So as you come out with this novel new instrument that can give you far more range and depth than what a mass spec can be, do you think that, that instrument will allow you to penetrate into a market that you still might have some CapEx constraints. But given the novelty of the instrument, and you're not -- your pricing isn't off from where those high end proteomics mass specs are, do you feel like that will allow you to kind of penetrate that market? And so therefore, [indiscernible] environment whenever we entered into when you launch the instrument, might not be as sort of a macro headwind for Nautilus?

Yes. I think there's a number of -- we like to think that's true. Yes. We think that for a product that's as differentiated as ours, there will be an ability for us to still sell into these customers despite the market conditions. I think that what we hear from our customers is that the data they're getting today is incomplete. And so we know customers have like one of every tool, right, because they're trying to assemble a very -- a more complete picture of the data that's in their sample, and we think that we can provide the most complete picture. And so of course, we think that there's always going to be a market for that. The other thing I would say is that by the time we launch, we hope market conditions will be different. But our pricing is also market-driven. And so to the extent that the -- and not cost driven. So we have a very different cost structure because we're using fairly readily available components and reagents. And so to the extent that we have to adjust our pricing to meet market demand, we'll have the ability to do so.

Got it. And this kind of like my next question is that despite the novelty of the instrument, there still is obviously well-known constraints within the U.S. academic and government market, which is a large customer base for mass spec and for proteomics. Would you expect to offer maybe discounting for those types of customers, depending on what the environment looks like at that time, just if funding remains somewhat constrained. And maybe sort of as a part of that question like, to the extent you're willing to talk about it, sort of what is sort of the gross margin and the cost structure of this instrument, meaning how much room do you have to be able to do that because it would be an important lever for you to pull initially, at least to get things off the ground?

Yes. I would say that we're already talking with potential partners and collaborators who have well-characterized samples where we can really jointly benefit from because we get the ability to compare our results to their well-characterized samples, and they get the benefit of furthering their studies. And so we're already starting to think through like those partnerships as a way to generate data and publications, which helps them create the flywheel where we can get into some customer accounts where they do need that data in order to move forward. And so academic collaborators is the place where we anticipate starting, some of those more formal engagements. And then in terms of discounting, I think that to the extent that customers help us with the data and publication, that's always something we would be willing to discuss. And then in terms of cost structure, what we've said in the past and it's still true today is that over the long term, we see margins being around 70%, and that's a blended number. And I think that depending on the product type, the instrument, maybe it'll be a little bit less than that. Our software and reagents would be higher than that. I think overall, there's plenty of room there for us to work through market conditions as needed. And of course, we're always focused on finding ways to bring costs down.

Yes. And I think you mentioned earlier, and I think it's -- the services offering is something that a lot of your peers in other areas have done, and it's actually been a big success. One, it allows customers to kind of try it before they buy it. Two, it allows them to see the results. You're obviously very familiar with how the machines work. So there's probably some speed to that. Is that something that you would consider having a services offering, sort of like an internal lab to be able to outsource some of that stuff for customers?

Yes. I think services is something we envision will always be part of our product. Initially, our services is really more intended to help validate our platform, and it gives -- generates that data in publications. It also will be our primary mode of operation for our early access program, which is where our customers send us samples, we analyze them in our own facilities, and from there, that really gives them the data that they need to go write the grant or to get the approvals they need to be able to buy the instrument. So we really think of that services offering as a way to promote lead opportunity generation for the -- to support the commercial launch. Now beyond that, there are some customers that don't have the sample volumes that require them to purchase an instrument. And so initially, we would tap into our services program to support customers who have a need to process a small number of samples or it's their pilot set of samples, which then they can use to go write the grant and support the larger study. And then as we get into the later stages of commercial development, of course, we can use the services mode to be a proof of concept to support instrument sales, but also we'll be coming out with new reagent kits and improvements and enhancements. And so there will always be a need to have some level of services to help get customers access to our data and prior to buying an instrument.

Got it. And just thinking about data interpretation, I mean, you're generating a vast amount of data with the platform ultimately. How do you expect to integrate this data that comes from the instrument into sort of bioinformatics platform or an output that's understandable by your customers?

Right. So in our platform, we think we'll be generating some of the most complete data. It's 95% proteome coverage, it's 10 billion analytes per run. There's a lot of data there. So we do anticipate there will be a need for some bioinformatics tools to be able to make sense of this, to pull out the insights from it. Some of those, we'll be developing in-house and we'll have some data tools that customers will be able to access, particularly through our cloud portal. And -- but we do know that with the rise of AI, there's a need to feed the machine. And so we'll make sure that -- and we're already starting to think through like the data we generate, how does it enable those types of more advanced platform analyses.

Got it. And then you have a cash runway target out to 2027. How should investors be thinking about key priorities of investment through the upcoming years, just to ensure; one, sufficient resources under the commercial launch, but also being prudent with spend, which has sort of been your [indiscernible] for the last couple of years?

Yes. We've been focused on efficient use of our [Technical Difficulty] cash preservation from the beginning. When we went public in 2021, we raised $345 million. And our reported cash balance was $193 million. So we still have over half of the cash we raised on our balance sheet. And this has really been through our continuous efforts to invest in a very focused and disciplined way while still making sure that the company has what it needs. And we've done that also by finding more efficient ways to do things; being innovative, bringing costs down and so on. So I'm really proud of that. I think it's really what put us in a position of strength. In terms of priorities, we're continuing to be very focused on our development and being as efficient as we can. And then we've preserved that cash so that when it is time to go commercial, we have the cash we need to do that. So in our runway that we said through 2027, that does include those initial commercial investments that we would anticipate happening in the months leading up to our commercial launch.

And when do you expect to start kind of layering in the commercial resources over time? I know you made a big hire, someone who's like head of marketing, and so there's obviously been some movement, but how do you think about sort of cash disbursements and spend? At what point do you start layering in the commercial resources?

Yes. So we said that our commercial launch will be late 2026. This is really when we start shipping instruments and reagents. And then in the lead up to that, we'll be doing early access, both internal and external type of programs. And so I think that set of activities will really come when we see those initial meaningful number of proteins coming out of our broad scale. At that point, we know that all the various components of our technology have come together and we're on the path, and we have greater certainty in our lines. We really want to sure we hit that milestone before we start investing in commercial because commercial, as you know, is a very cash-intensive exercise.

Got it. And how should we be thinking about catalysts, data releases? I mean you guys have been pretty regular fixture at HUPO and some of the other industry conferences. How should we think about either partnership data catalyst, data generation over the next year leading up to the launch?

Yes. So in that time line that I was talking about, we've got in 2026, that's when our early access begins or that's when we -- in the lead-up to commercialization, we have our -- mapped out our early access program. So prior to that, is when we would need to see those first set of protein decoding, let's say, 1,000, 2,000 proteins. Once we have that, we would work through publications -- both publications as well as presentation at scientific conferences. There's the various mass spec conferences. There's an Alzheimer's conference. So each of those conferences is when we give updates both on the tau or on the targeted or broad scale side.

Got it. And then maybe -- I mean it's interesting, one of your cofounders, Parag, is a KOL in and of himself in terms of mass spectrometry and proteomics. But what are you from KOLs that you're working with in terms of feedback for the instrument, the process, sort of like areas of feedback to improve but also areas that they're super excited about?

Yes. What I would say is that we have continued excitement and interest from our KOLs, both on the targeted and broad scale side. What we hear from them is that the data they're getting from existing techniques is just not sufficient. I think there's still a gap in what they need versus what they can see. And so they're really excited about both sides of the technology and they're just -- whichever one becomes available, they'll want to get in front of it.

Got it. And then maybe just to wrap up, sort of -- in speaking to sort of the investors, how would you characterize Nautilus in terms of an investment? And what would you have them kind of look towards to evaluate 1 year performance until launch, but then also what something is for them to look forward to in terms of the market opportunity?

Yes. What I would say is just to build on some of the points I made earlier is that this type of data, both on the targeted and broad scale side doesn't exist today. So on the targeted side, we can see a level of granularity on particular proteins of interest that is just not possible with any other technique on the planet. And we've already shown that with tau in our ability to measure thousands of different forms of tau in a sample. And on the broad sale side, look, there's no one that can see comprehensively every protein in a sample. And so we think that the ability to see this data will be game-changing and unlock a new wave of discovery as well as finally kick off -- deliver on the promise of precision and personalized medicine that can't be seen when you are dealing with data that doesn't give you a complete picture.

Awesome. We'll leave it there. Thank you very much. Appreciate it.

Thank you.