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

Hey, guys. So we're just going to get started here. My name is Tejas Savant. I'm the life science tools and diagnostics analyst here at Morgan Stanley. It's my pleasure to host Sujal Patel from Nautilus and Anna Mowry who's sitting in the audience here, who's the CFO of the company. Before we get started, I just wanted to read the standard disclaimer here. For important disclosures, please see the research website at morganstanley.com/researchdisclosures. And if you have any questions, please do reach out to your sales rep. So with that, Sujal, thank you for doing this.

Maybe to start for people in the audience perhaps who are not as familiar with Nautilus, can you just provide a brief history of the company, how it came into being? And what are the key unmet needs in the proteomics market that you are trying to address?

Great. Well, Tejas, thanks for the invite to the conference, and it's really nice seeing you again. So... Nautilus is a company that is pioneering a new approach in proteomics. Proteomics is a study of proteins. And proteins are the most valuable source of biological insight. And measuring the genome in blood in cells, that is a conquered problem over the last couple of decades. We've democratized access to the genome, but we really have made very little progress in substantially and reproducibly measuring the protein within those same blood sample tissue samples in cells. And that is a really big problem because your genes don't really change from the day you're born to the day you die. And your proteins do, your proteins make up their cells, that's the cellular machinery that does all the work in your body, and it's the most important thing to measure. I underscore that 90% of our FDA-approved drugs target proteins, most molecular diagnostics, even with this huge revolution in liquid biopsy, most of them still target proteins. And so this inability to measure proteins is a huge impediment to pharmaceutical development, precision medicine. And that's a problem that we're focused on making go away by building what we hope will be the first platform that measures the entire proteome from sample comprehensively and democratizes access to it by making it routine. The background of the company, which was kind of the question that you asked was that my co-founder is a guy named Parag Mallick. Parag is a Stanford faculty. In the last 20-plus years of his career, he is a key opinion leader in the proteomics world. And the problem we're trying to solve is a problem that his lab and his colleagues have struggled with for decades. The state-of-the-art today in protein measurement is using these million-dollar mass spectrometers and a very complicated workflow in front of it. And at the end of that, if you try to analyze a drop of blood, the very best you could do after spending thousands and thousands of dollars and weeks is to measure about 8% of the protein content of samples. And when we're successful and get a platform out there, our goal to make that 95% measurement in make-it routine.

Makes sense. The other question we get every once in a while from investors to job is why now? Like why is the proteomics market finally ready to really invest dollars and scale up on these new approaches?

It's a really great question. The innovation, the need for innovation in the proteomics space is not new. Many companies have been trying to do better than the mass spectrometer for decades, but it's a very hard problem and incrementally move -- improving the max petrometers not the way to solve the problem. From our standpoint, when you think about like what does it take to build a new platform and bring it to market. Our company is based on a unique confluence of computer science technology, data science, technology and machine learning and biochemistry and more traditional biotech types of technology. And it's really the combination of those things, which really is timely today that enables the core innovation behind how we measure in our platform.

Got it. And where do you see your approach fitting within the rapidly evolving landscape, right? So if you've got folks like Quantum SI and Encodia, et cetera, and we've also got targeted platforms like Olink and SomaLogic we're scaling up here. Where do you fit into that ecosystem?

Yes, it's a great question, TJ. I know for a lot of investors, it's a little daunting trying to get involved in the proteomics space because there's quite a few segments of it and a lot of new names that have come public over the course of the last couple of years. The way we think about the proteomics world is that it's bifurcated into a discovery end of the spectrum and a targeted end of -- on the discovery end of the spectrum, customers are interested in taking a sample and getting the most information they can throughput, meaning how many samples you can analyze per day and costs are a little less important. And then the targeted side, customers want to analyze a smaller subset of proteins, but cost and high throughput are really important to them. And on the targeted end of the spectrum, there are companies that have been public for quite some time like Quanterix, there's companies like SomaLogic and Olink, which are new names to the public market. And then on the discovery end of the spectrum, which is where Nautilus lives, there's us, there's mass spectrometer vendors, ThermoFisher, Danaher, Bruker, Agilent. And then there's a small set of other new names that are focused on different aspects of improving the proteomics workflow. And 2 of those -- and really, one of the key categories you pointed out was these companies that are trying to sequence peptides much like we sequence the genome. And sequencing peptides is a very complicated approach. It doesn't yield a very sensitive solution, and we can dig into it more if you're interested -- the approach that we have is a very, very different approach from the other companies that are in this space. It's a technique that operates on billions of molecules in a single run. It's a technique that leaves the proteins intact, whereas everyone else chops the proteins into pieces, which is a 2-log sensitivity loss. And it's an approach that's really focused on high throughput and reproducibility, which we don't get out of those other competing platforms. Now suffice to say, all of us have things to proof, right? The companies that are working on the newest approaches are all still pretty commercial, and we're all buying to be the first platform out there that really makes a dent in biotech.

Got it. And Sujal, sort of carrying on that sort of thought that you just mentioned here. What are the benefits in your mind of a single molecule approach? You've also talked about absolute quantification being an important differentiator for you. And what about sort of the peptide sequencing approach in your mind, sort of makes it sort of incredibly challenging and possibly sort of like fraught with risk of either scale up failure?

Yes, maybe to start answering your question, let's talk about what a single molecule approach is. And it's a term that Marion many investors may not be used to because most of the investment thought pattern is pattern by genomics. And in genomics, every one of your cells carries the same copy of the genome. It doesn't really matter if you can measure one molecule because they're all the same other than oncology, where there's some differences. They're all the same. In yourselves, every single protein molecule accounts. Every one of them has a job to do, and every cell has on average 1 million protein molecules in it. And so for a company like Genentech, which is an early collaborator of ours, if you're trying to look at a population of healthy cells and a population of 6 cells and understand on the cell surface, what are the differences that might be potential biomarkers that I could use to target a drug? What are the different biomarkers that potentially could determine therapeutic response. You have to be able to get down to the rare things that are on the cell surface. You have to be able to get down to each one of those individual molecules. And there really are -- there are no techniques to do single molecule proteomics out there today. And that's really the approach that Nautilus has pursued is an approach that has the sensitivity to be able to measure single molecule differences inside of cells, which is really critical to pharmaceutical development really critical to liquid biopsy as well. All of the other approaches that are out there, whether they be mass spec or these peptide sequencers are what -- they operate on peptide, meaning they take your proteins, they chop them into pieces and then they try to measure. And when you do that, you inherently end up with an insensitive measurement where you really have to see 100 molecules or more before you can ever even say, yes, that molecule, that type of protein is in my sample. And you need to see far more than that if you want to quantify, you use the word absolute quantification. If you want to really know this protein is expressed twice as often as this one, you need to see many, many more copies before you can make that accurate determination.

Got it. One of the features of the platform you've spoken about in the past is your ability to characterize proteoforms and post-translational modifications. Just for people we're not as sort of into proteomics here in the audience. Why is that important? And what is it about sort of your ability to characterize proteoforms, -- like why is that unique to Nautilus?

Yes. It's a great question. So when you think about the proteins that make up your cellular machinery, there's approximately 20,000 of them that are gene encoded proteins. They're in your DNA, your body has the mechanism to make these -- but those proteins alone don't control all cellular function. Those proteins are decorated and modified in various ways with different modifications like phosphorylation. And in addition to that, these proteins come in different forms, where they may be missing pieces of sequence or they may have additional sequence pieces that are inserted. And those modifications and different splice forms have a profound impact on how a protein is functioning, how -- what its confirmation is, how it's distributed in the cell, what is degradation pattern looks like. And so if you want to truly understand the mechanics of what's going on in the cell, you don't just want to know what proteins are in it. You want to understand how is the protein modifies and you want to be able to track that over time. And the thing that Nautilus' platforms designed uniquely enables it enables you to have an entire protein intact. So you can -- after you've identified the protein, you continue to probe those molecules and figure out, well, where are the phosphorylations, what site are these molecules phosphorylated at what form am I dealing with? And for us, the thing that you can do with that is you can build a set of biological insights that gets deeper and deeper. And so if you're a pharmaceutical company, you're building a new drug for some neurological disorder, and you want to go and understand the tau protein, for example, which is a critical biomarker in neurological disorders. If you want to understand what's the phosphorylation pattern, how is that changing over time? Is it related to my drugs response? Is it related to my drug efficacy. Those are all sorts of questions that are unanswerable today and our platform uniquely is able to answer. And it's one of the first use cases that we've started to work with Genentech, Amgen, MD Anderson and other collaborators on...

Got it. And you mentioned this a couple of times, Sujal. The platform doesn't require digestion of the protein, and so you can look at the whole protein, but it does require denaturation. So is there sort of some sort of signal loss in that step in your mind? Or is it still much better than digestion?

So certainly, it's much, much better than digestion. The design criteria for our product was that we wanted to denature our proteins upfront. And so just to define the terms for investors who are less space, digestion means you take the protein and you break it into many pieces and you can never really correlate what piece came from what molecule. -- denaturation is just a process where I unfold the protein so that it's in a more stable form for analysis. The mass spectrometer denatures, proteins, many assays, denature of the proteins. And operating under nature proteins enables us to have a more sensitive output in the end because we don't have to deal with trying to keep a protein in its neat-folded form, which is very, very difficult to do outside of biological conditions. And you asked like, is it important to keep it in its folded form. If we could, and it was feasible, Sure, we might want to do that. But when we talk to customers, and we've talked to hundreds of customers about our platform, the primary question is what gene-encoded proteins are in my platform. The next question is, what are the proteoforms and the splice form. Then they may want to ask questions about spatial Questions about the confirmation and the form. Those questions that are weighed down the... The curve. And yes, those are interesting, and we might want to do those in the future, but they're not important to the customer today. They're not the primary question.

Got it. Now the ability to separate the signal between the molecules is obviously very important to your single molecule sort of attribute that you emphasized, -- how do you ensure good separation between proteins?

Yes. That's a great question. One of the fundamental technologies that we had to invent to be able to build our platform was a technology that could take all of the protein molecules out of sample and essentially array them in an even grid kind of like putting on a chessboard except the chessboard has 10 billion spots on it to separate enough molecules that we could have every molecule in 100 to 1,000 cells, which is the design criteria for the product. In order to do that, we had to develop a specific technology that enables you to essentially land on protein molecule on a landing pad -- and those landing pads are created on a chip, which is built the semiconductor process. And we use chemical vapor deposition to build up a set of functional layers on it that creates an interaction where only one protein lands on those landing pads. You'll recall that I think 2 quarters ago, we talked about it in our earnings call that we have a preprint that's on bioarchive now that describes the entire method in very good detail. And it really shows the ability to occupy over 80% of those landing pads with proteins even at this very early stage, ensuring that we only have one protein per spot. And that spatial separation is critical because we need to be able to put the proteins apart so we can identify each of them.

Got it. And I think Paragon also mentioned requiring about 300 cycles to characterize the entire proteome. Can you run 300 cycles and not have signal loss between washes?

Yes. So just to kind of get some context on your question. So our platform is based on a few technologies. One is the biochip that spatially separates all the protein molecules and then this iterative cycle, 300 cycles of different affinity reagents where we build up that information on a per molecule basis, and we use that information to positively identify what each of those molecules is. And one of the key challenges that we've had to solve is, well, how do you introduce hundreds of different affinity reagents into a flow cell and get accurate measures. And we've been releasing little bits of data over the course of the last couple of years on this roadshow presentation, as we went public, we showed a 300 cycle set of data where we showed multi-cycling and being able to run detection across them. We've been showing internally showing more results where we see there is a little bit of signal loss that builds up over the first 10 to 20 cycles, but then it levels off over time. And this is an area of intense focus for us. Just last quarter, we hired a guy named Ken Kuhn, who spent 16 years at Illumina through the first next-generation sequencer and many, many generations after Ken is the guy who's really focused on this process of building that baseline assay and optimizing it. And these are very common types of problems that companies like Illumina had to deal with in the past. And so it's pretty well worn, and we feel very good about sort of the progress that we've made...

You mentioned sort of affinity probes. What are the features you're looking for in an affinity probe to use on the platform?

Yes. So for us, for each of these 300 affinity agents that we're building, we need to have a pro called an antibody that volumes to a short epitope, and just a few amino acids reliably. We're looking for probes that have good affinity so that we don't need to have a significant concentration of them in order to get a signal section. We're looking for them to have a long off rate so that they can stay down to their target while our optical system goes and does the detection. Those are the primary characteristics of it. And then there's some secondary criteria that are less important.

Got it. And when thinking about sample compatibility, Sujal, I mean, how are you thinking about blood versus tissue versus this is culture sales, you know what I'm saying? And will you focus on each of these at launch? Or will it just be a 1 or 2?

Yes. So when you think broadly about our platform, our platform takes denature proteins as an input. And upfront, there are a wide range of existing sample preparation methods for all those types and for urine and for any type of sample that has proteins in it. It could be glad. It doesn't matter. At launch, when we think about limiting it to a set of things that have big impact and are easy to have standard procedures for the important ones for us are blood, cell list and tissue.

Got it. Makes sense. On the last call, you talked about commercial launch now expected in mid-'24. Can you just walk us through the various factors that led to that modest 6-month push out?

Yes. So for us, I mean, we've talked about this stage is you and I have known each other here at Nautilus. For us, the long pole in the tent for development is building and qualifying those 300 affinity reagents that we need to enable us to get to 95% identification of all the proteins that are in a sample. And building of those affinity reagents comes in a discovery step where we're finding potential antibodies that go and bind to the short epitopes and meet the characteristics that I described earlier. And it comes from a qualification piece where we go and we run it through extensive testing. We put it on our platform. We make sure it operates at our buffers. And at the end of that, we put it into the library, and we're working our way towards getting the 300 that we need to fully ship the platform by the middle of 2024. As you mentioned, we did have a modest push out of about 6 months. And really that's related to the fact that when we went public at the beginning -- or the middle -- excuse me, the middle of last year, what we talked about with Wall Street was that we were going to use the capital to scale up our affinity reagent development pipeline and get to the point where we'd get to our 300. And we're going to do that with a 2-pronged strategy. One was using external partners who are experts in affinity reagent development and the other was scaling our internal initiatives. And I think the biggest contributor to that 6 months is that what we've learned over the course of the last year is that, one, we have one particular internal pipeline, which is doing incredibly well, and we've increased our investment in that pipeline significantly. And we've learned that from an external partner perspective, the external partners really underperformed our expectations. And so we've wound down many of those relationships. Some of those relationships are very strategic to us, and we continue to invest in our relationship with Abcam. But that period there where we were learning what techniques were going to work and what didn't kind of is the biggest contributor to the 6 months.

Got it. And what is sort of challenging about developing these reagents that make it difficult even for these scaled up sort of antibody vendors looking to work with?

Yes. I think that -- the combination of characteristics that we want, short epitopes, long off rates and as well, unlike an idiosyncrasy of the types of affinity agents we want is we don't need them to be specific, whereas every other use case in all of biopharma looks for very, very specific affinity reagents. Those differences in characteristics meant that it was different enough that their standard topics and standard operating procedures just didn't work out that well. Whereas we internally have been optimizing that for 5.5 years and working really hard to build a pipeline that's focused on exactly our needs. And so I think that's maybe a little bit of underestimation of how different that was by our partners is what led to...

Got it. Fair enough. And then beyond Affinity reagents, what are your other sort of 2 does as you work towards the commercial launch? And what sort of gives you confidence that we won't see a further push out versus the time line that you've now laid out?

Yes. So the first step -- the first part of your question is really one of the major pieces, right? We already have a great single molecule array, and it's functioning well, and we're scaling our production of our chips, and we feel very good about that. Our instrument is coming together nicely. There's nothing exotic in our instrument. It's just engineering, and we've got a professional team who has been building instruments of this complexity for a long time. So that's just a matter of time. And then really the last piece is, frankly, all the areas you've just been probing on, which is building affinity reagents and getting the multi-cycling going so that we're at 20 cycles, 40 to 50, 75, 90, 100 and so forth. And those are the pieces. In terms of how do you prevent from slipping again, really, the best things you can do are to have a good plan and to have great people, right? And we feel very good about the plan that's in place now. We've had a lot of time to figure out all of the pieces of science that needs to be figured out. We feel good about the development path is going forward. The last -- the other thing is people, right? We are building an instrument which has a real possibility of being a game changer for pharmaceutical development for precision medicine. And it's the first time somebody is going to build a platform that does full proteomic measurement. It's just as hard as building the first gene sequencer when Flex and Illumina did it. And so we've had to build an R&D organization that has the type of talent that is capable of building something like that. Subra Sankar, who is our SVP of Product Development, has been on board just since we went public just about 2 years now. And Subra was the guy who ran all the instrument development for Celexa and Illumina for the first platform, the GA all the way through the HiSeq and the MiSeq at Illumina. Underneath him and Michael [indiscernible] are 3 key VPs that we've just hired this year. One was a multi-decade SomaLogic veteran, ran all their affinity reagent development. One is the guy named [ Arex ]. He spent his entire career building NGS instruments, Illumina, Affymetrix, Agilent, [indiscernible]. And then Ken Kuhn is the most recent addition I mentioned earlier. Ken spent 16 years at Illumina and almost 5 years at in Encodia, which is one of our new method peer group competitors. And Ken is an expert at measuring assays, developing reagents, building processes and really building high-performance, highly reproducible systems. And so we feel very good that the team is in place and the plan is solid, and that's the best that we can do to help mitigate any risk going forward.

Got it. And on that point, Sujal, are you essentially sort of -- are there any open gaps you're looking to fill in terms of key positions? And how are you navigating sort of the labor markets you're perhaps a little bit more skepticism around joining a pre-revenue company. On the other hand, you do have enough cash on the balance sheet versus the peer group in some cases. So just walk us through that dynamic?

Yes. I will say that the labor pool in general, in biotech, is sort of interesting, I come from the tech world and in tech people are risk tolerant, and they always are looking at what's the biggest potential and getting there as early as possible. In biotech, what I found and what people who've been in this market for a long time now is that the labor pool is actually quite a bit risk adverse and having -- they're willing to take a big bet like Ken took on us recently, but they have to see a big balance sheet, they have to have some stability there. And so for us, being public and having a big balance sheet is actually something that helps us to attract much, much better talent, but it's still hard out there. There's a lot of competition for the top talent, and we're holding a very high bar for the people that we bring in, which means we're not making all of our hiring goals and we're a start-up, right? You have to just -- you have to make -- you have to get the work done anyway and we're doing all of that. But it is a tough environment, but it's also an environment that's full of opportunity because we've got a big balance sheet because we have a big bold mission and people are attracted to that.

Got it. You've talked about early access in early '24. What are your key aims during that phase of the launch? Will customers have a beta instrument at that point? Or will you essentially be running samples in-house?

Yes. Let's kind of walk through what the year before that instrument launch in the middle of 2024 looks like. The first thing that's going to happen is we're going to have a set of data that's broad scale proteomic data. We're going to say, hey, customer, we can get -- we can show you the data right now for 25%, 30%, 40% of the proteome. What we want you to do is sign up to do a services engagement with -- so we can show you what the platform is capable of doing, maybe 24, 48 samples will charge a modest amount of money for that engagement. Those engagements are about showing the customer what the platform is capable of. As we start moving through that 1 year T-minus 1 year before platform launch, somewhere in there, we're going to give some of those customers a physical instrument as well to do early beta testing. And both those services engagements and those early betas, the primary goal of those is not just, hey, look at this great data and how great it is for your business. But our goal is to try to sign preorders for the instrument so that there are firm orders so that when we're ready to ship in the middle of 24, the first thing we're doing is we're fulfilling our backlog and then rebuilding the pipeline as we look towards the second half of the year. And that's really about getting quick lift on the top line.

Got it. And you're probably going to punt on this one, but I'll ask it anyway. I mean, any early sort of like hunches around how you're thinking about price points and then the theoretical maximum on throughput and pull-through?

Yes. So we have not really start detailed pricing at, but we've dropped enough bread crumbs along the way that I can give you some guidance there. Our instrument deal, which is the hardware, 3-year software subscription, service and support, site prep installation, reagents against away, that's roughly $1 million deal. That price point is meant to compete squarely with the mass spectrometers that are used in discovery proteomics today the deal size is somewhere between $800,000 and $1.3 million depending on your configuration and the vendor and so forth. -- that's the instrument itself. Then there's what -- how fast does the instrument go through samples. And we have -- as we finalize the specifications for our instrument, we finalized on a mode where the instrument has a maximum throughput of 12 samples per day. And the different configurations, you could run one sample, you could run 4, 8 or 12 in a day. And each sample is going to cost a few thousand dollars per sample. And so if you multiply that out 12 times a few thousand times 365, it's a really big number. It's approaching $10 million. But we don't think about the world that way. Don't think about maximum because we have a lot of lessons from new instrument launches that nobody ever gets the maximum. We think about customers who buy this instrument and get it operationalized using it twice -- and even at twice a week, it's a significant pull-through number that's far more than most instruments in the DX and tool space, but we think about kind of that as the nominal pull-through once the customer really gets used to it, get through their testing and is using it in real-world applications.

Got it. Scott, half a minute here. So what are the key milestones and metrics that we should be looking to watch as we track your progress towards commercialization? At what point will you perhaps announce a date for the Analyst Day? And I'm assuming it's going to be tied to the unveiling of the instrument. So any color there?

Yes. So let's tackle the first part. I think that the key data that we're going to put out is the same data for the cation leaders in proteomics for our potential customers and for investors. We are putting out data that proves that each of the various components of our system function and are novel and do what we say they do. For example, our single molecule array, for example, our preterm analysis. Then what we're going to do is we're going to put those all together and start showing data that broad scale proteomic profiling on this new approach is possible. And then we're going to show data that says we now are at a point where there are no other platforms on the planet that can do what we're able to do. And as we start building up that base of data, we expect both customers and our key opinion leaders are going to be pretty excited, and we think investors look pretty excited about that data as well. On that road, we will do an Analyst Day. Maybe it will be coupled with the launch of our early access. Maybe it won't. Yes. At some point, we're going to show you the instrument as well.

Perfect. That's a great place to leave it at. So thanks so much, and let Parag know we do expect a magic show with the Analyst Day.

Okay. We will. For those who don't know, my co-founder is not just a professor at Stanford and Founder of a biotech, but he's a professional magician as well. Thank you very much.

Thank you so much. Appreciate it.