Earnings Call Transcript - NAUT Q2 2025

Operator, Operator

Good day, ladies and gentlemen, and thank you for being here. Welcome to the Nautilus Biotechnology Second Quarter 2025 Earnings Conference Call. I would now like to hand over the conference to Ms. Ji-Yon Yi from Investor Relations. Please go ahead, ma'am.

Ji-Yon Yi, Investor Relations

Earlier today, Nautilus released financial results for the quarter ended June 30, 2025. If you haven't received this news release or if you'd like to be added to the company's distribution list, please send an email to investorrelations@nautilus.bio. Joining me today from Nautilus are Sujal Patel, Co-Founder and CEO; Parag Mallick, Co-Founder and Chief Scientist; and Anna Mowry, Chief Financial Officer. Before we begin, I'd like to remind you that management will make statements during this call that are forward-looking within the meaning of the federal securities laws. These statements involve material risks and uncertainties that could cause actual results or events to materially differ from those anticipated. Additional information regarding these risks and uncertainties appears in the section entitled Forward-Looking Statements in the press release Nautilus issued today. Except as required by law, Nautilus disclaims any intention or obligation to update or revise any financial or product pipeline projections or other forward-looking statements, whether because of new information, future events or otherwise. This conference call contains time-sensitive information and is accurate only as of the live broadcast on July 30, 2025. With that, I'll turn the call over to Sujal.

Sujal M. Patel, CEO

Thanks, Ji-Yon, and thank you all for joining us. Q2 was a milestone quarter for Nautilus. Not only did we continue our momentum across both targeted and broad-scale proteomic development efforts, but we also publicly shared the first preprint to feature novel data generated using the Nautilus platform. We believe that this first-of-its-kind proteoform data across a range of important biological systems can only be generated on our platform. No other analysis method comes close. The manuscript now live on a platform, represents nearly a decade of pioneering work by our team and collaborators. In it, we introduced and validated the application of our iterative mapping method, showing that it can measure proteoforms at a resolution and breadth never before possible. The results speak for themselves. Our approach demonstrated unprecedented dynamic range and industry-leading reproducibility. Even more exciting, early biological insights from this work suggest that iterative mapping may illuminate new mechanisms of tau biology, potentially opening the door to a new generation of neurodegenerative disease diagnostics and therapeutics. Parag will elaborate on these results presented in the manuscript shortly. We believe that this manuscript is both scientific validation and a significant external milestone for Nautilus and that it sets a new bar in the field of proteomics. I want to recognize the herculean efforts of our team as well as our partners at Genentech, the Neural Stem Cell Institute, and Mount Sinai Health System. Zooming out, this quarter, we remained focused on building engagement around our tau proteoform assay and laying the foundation for external collaborations. Feedback from researchers is highly enthusiastic. Many now see Nautilus as a forerunner in decoding the complexity of post-translational modifications with the resolution necessary to drive meaningful biological and therapeutic breakthroughs. This enthusiasm is translating into real momentum. We've continued to deepen our conversations with academic, pharma, and nonprofit partners, and the ability to now reference our manuscript gives us a new level of credibility and visibility. The discussions we're having today are more strategic, focused not only on tau but also broader use cases for targeted proteoform analysis across neurology, oncology, and immunology. To provide more detail on our R&D efforts, let me now turn the call over to Parag.

Parag Mallick, Co-Founder and Chief Scientist

Thanks, Sujal. Good morning, everyone. Q2 marked a major inflection point in our scientific journey. As Sujal mentioned, this quarter, we shared a preprint of a manuscript illustrating how our iterative mapping method enables a unique capability on the Nautilus platform, resolution of proteoforms at the single molecule level at scale. In traditional proteomic techniques, a protein is often treated as a single entity. But the reality is that proteins exist in many modified forms, each with their own distinct structure and function. These different variants are called proteoforms. Just like a single gene may have thousands of variants defined by mutations, a single protein may have thousands of different proteoforms defined by a combination of alternative splicing and multiple post-translational modifications. The prevalence of different proteoforms may ultimately influence the role a protein plays in disease and how best to therapeutically target it. Before I dive in, I'd like to clarify one important point. You'll hear from others that they are measuring proteoforms. However, the reality is that only platforms that look at intact protein molecules and are able to interrogate multiple positions on those molecules are capable of examining proteoforms with the necessary resolution. Existing affinity-based methods such as Olink, SomaScan or LMR are able to report the relative amount of a protein, but they typically do not measure modifications of those proteins and certainly not the co-occurrence of those modifications on individual protein molecules. Likewise, peptide-based methods, such as employed by Shotgun mass spectrometry or even single molecule peptide sequencing methods entirely lose the contextual information required to know that multiple modifications are co-occurring on a single protein molecule. Any peptide-based measurement method cannot measure proteoforms. Consequently, we believe that the Nautilus platform is the only platform that has been designed to readily quantify the thousands of distinct proteoforms of key proteins at scale. With the release of our manuscript, we publicly demonstrated the remarkable real-world capabilities of our iterative mapping method. This is notable for two distinct reasons. First, the manuscript represents an end-to-end validation of our core platform, which is shared between our targeted proteoform assays and our forthcoming broadscale assay. Second, the manuscript demonstrates that the tau assay built upon our platform has the ability to drive powerful biological insights into Alzheimer's disease and related disorders. Diving in, the first part of the manuscript shows that the platform is able to go end-to-end from sample to answer by taking individual protein molecules from complex samples, attaching them to DNA origami nanoparticles, depositing those nanoparticles on nanofabricated arrays, iteratively probing them, and then applying our machine learning-based engine to quantify the proteoforms in the sample. When we began the company eight years ago, each of these challenges represented its own complex scientific frontier. Consequently, demonstrating them fully integrated is an important proof point regarding the scientific foundations of our approach. After introducing the method and how that method is applied to assay the Alzheimer's disease-associated protein tau, we performed extensive assay characterization that serves as external confirmation of the platform's scientific rigor and technical maturity. I'd like to call out two specific aspects of the platform characterization data that will have concrete impacts on our customers. First is the reproducibility data. It is extremely uncommon for first introductions of a new method to perform such rigorous and extensive characterizations of reproducibility. However, we've heard from our future customers that reproducibility is top of mind for them. This is natural as high reproducibility allows researchers to trust their results and know that they are more likely to be replicated by researchers in other labs. We measure the within experiment reproducibility of our platform as having a median CV of 1.5%. Even across multiple instruments, reagent lots, operators, sample preparations and runs, our median CVs were approximately 5%. To put that in perspective, studies of the reproducibility of existing, mature, affinity-based and mass spectrometry-based proteomics platforms that look solely at total protein abundances, not proteoforms, have found median coefficient variation of nearly 40% from run-to-run and up to 80% across labs and operators. The reproducibility of our platform, even at this earliest stage, is a direct consequence of our single molecule methodology, which determines protein abundance not from a single measurement but instead from the aggregate of independent measurements of many individual molecules. Our reproducibility is also a consequence of our incredible team's steadfast commitment to quality. As I mentioned, reproducibility of this type would be considered exceptional for a mature platform. To have demonstrated such world-leading reproducibility at the first introduction of a novel method is astonishing. We additionally demonstrated that the assay is extremely sensitive and able to accurately measure changes to proteoforms abundance across a wide range of physiologically relevant concentrations. For reference, mass spectrometric methods such as tandem mass tagging lose quantitative accuracy when comparing samples in which a protein's abundance changes by more than a factor of 10. Our analysis revealed that our assay could reliably measure how much a proteoform changes even for changes of over a factor of 1,000. Furthermore, the assay is able to accurately quantify extremely low abundances of proteoforms. Forms of tau present in samples at levels approximately 0.1% of total tau can be reliably quantified. This is critical as we know that low abundance forms of proteins like tau can still be tremendously impactful in disease progression. Beyond demonstrating the technical capabilities of our platform, the studies we presented are already providing unique biological insight into Alzheimer's disease and related disorders. Before discussing the paper specifically, I'd like to give a bit of context as to why the findings are potentially so significant. The link between tau and Alzheimer's disease has been established for nearly 40 years. In that time, a huge number of potential biomarker tests and therapeutics targeting tau were developed with the goal of diagnosing Alzheimer's Disease early and stopping or reversing its progression. Unfortunately, these assays and therapeutics have failed in clinical trials. Retrospective analysis suggests these failures may have stemmed from targeting the wrong proteoform of tau. Unfortunately, prior to the introduction of the Nautilus platform, measuring these proteoforms was out of reach. The proteoform resolution offered by the Nautilus platform gives researchers actionable biological insights that aren't otherwise attainable. With the Nautilus platform, researchers will be able to observe not just how much of a protein is present, but which forms are increasing, decreasing, or appearing uniquely in specific states, knowledge that is critical for understanding mechanisms of disease and for identifying precise therapeutic targets. In our study, we examined a diversity of model systems that are used by researchers around the world to develop the next generation of therapeutics and biomarkers. For the first time, we were able to measure more than 130 different forms of tau that were present, some of which had as many as six co-occurring phosphorylation events. Existing platforms would have mushed all those forms together, providing a low-resolution readout of total tau that obscures the critical proteoform information. In addition to looking at model systems, we applied our method to a small human cohort. Within that cohort was a patient with aggressive Alzheimer's Disease. This patient was clearly delineated from healthy controls and even other patients with less advanced Alzheimer's Disease by a form of tau that was quadruply phosphorylated. Moreover, the pattern of forms of doubly and triply phosphorylated tau strongly suggests an order and a timing to how the proteoform came to be formed, an observation that previously had not been possible. This combination of technical rigor and biological insight is why the reaction from researchers with whom we've shared the manuscript has been so strong. They recognize that this isn't just a new measurement method. It's a fundamentally different way of understanding biology. They see that iterative mapping represents an entirely new class of measurement modalities distinct from either mass spectrometry-based approaches or the affinity reagent profiling methods. Scientists are already asking how they can start integrating the method into their workflows. As we continue to expand our reagent panels and data analysis capabilities, we're confident this core capability will remain a major driver of scientific adoption. For an easier-to-understand synopsis of the manuscript, I encourage you to check out our blog where we've tried to distill the manuscript into a form that is more broadly accessible. We believe that these findings validate the full Nautilus platform, not just for tau, but as a generalizable engine for proteomic insight. The core platform and iterative mapping method used in the tau studies is also used for broad scale analysis, and we anticipate the exceptional performance we've observed will translate across applications. Looking ahead, our road map for the remainder of 2025 includes continuing to refine and scale our broadscale assay configuration, advancing multiple external collaborations for tau and non-tau targets, publishing additional data sets and technical white papers to support adoption. With the release of this manuscript and the strong momentum across our platform development and collaborations, we're confident that Nautilus is on track to deliver on its mission to transform how biology is measured and understood.

Sujal M. Patel, CEO

Thanks, Parag. This quarter marks a shift from what could be to what is. We've now shown publicly, rigorously, and reproducibly that the Nautilus platform can do what others cannot, and this is only the beginning. Earlier this week, we presented our data from our tau manuscript at the Alzheimer's Association International Conference, AAIC in Toronto. The conference served as an excellent venue to highlight the capabilities of our platform and get feedback on the potential impacts that high-resolution proteoform analysis of proteins like tau might have on neurodegenerative disease research. Throughout the event, we spoke with a broad range of researchers and potential customers, spanning the academic, nonprofit, and pharma sectors. Conversations consistently centered around the gaps that exist in the understanding of disease progression in Alzheimer's. We heard from several researchers about the conundrum of single PTM measurements of p-tau217. Despite being FDA-approved for Alzheimer's diagnosis, it's been observed that p-tau217 is abundant not only in patients with Alzheimer's Disease but also in young children. Furthermore, p-tau217 tests have a high false positive and negative rates and do not predict disease trajectory. Researchers are excited about the implications of Nautilus' work for generating more precise ways to stage and prognose disease. They're also excited about how understanding the proteoform landscape might inform therapeutic development by helping identify which model systems are most reflective of human disease and which tau species to target. One other surprising area of interest was in using a patient's proteoform landscape to distinguish among various tauopathies such as frontotemporal dementia and progressive supranuclear palsy. The scientific community showed clear enthusiasm for the specificity and resolution our platform offers in analyzing proteoforms that have historically been impossible to quantify at this level of detail. These interactions reinforce our belief that the ability to measure tau at proteoform level resolution could be transformative for neuroscience disease research and more broadly for understanding complex protein biology in neurodegeneration. Building on this excitement, we're currently in active dialogue with several organizations to formalize early collaborations across both pharma, academic and nonprofit research settings. As we shared last quarter, we expected to sign an initial collaboration in the first half of the year, and I'm pleased to report that we've now signed two collaborations with major U.S. research institutes. These collaborations provide the opportunity to demonstrate our platform's capabilities and performance with customer samples as well as enabling new biological insights in Alzheimer's Disease. Though they are not intended to generate revenue initially, these collaborations lay a foundation for driving revenue in the future. While there is strong interest in additional collaborations, we're carefully balancing our resources between our targeted proteoform and broadscale development programs. This balance will guide the total number of collaborations we can engage in at any given time without delaying key development milestones, particularly for our broadscale platform. Turning to our broadscale platform efforts. We continue to make steady progress on the new assay configuration, which we introduced earlier this year. This work is aimed at better aligning our assay design with the characteristics of our expanding probe library, improving probe yield and performance across the platform. Among the most important advances continues to be both in the assay configuration itself as well as in our methods to determine which probes are suitable for the new configuration. These improvements are intended to reduce technical risk and enable higher performance, particularly as we scale towards comprehensively decoding the proteome. In Q2, our broadscale progress was in line with expectations. We began early-stage experiments with the evolved configuration and initial data continues to be promising. This marks a critical step towards achieving robust quantification of a significant number of proteins from complex biological samples such as cell lysate. We've also begun working with our key suppliers to develop updated formats for our consumables, ensuring that they can meet the demands of our platform and future scale. While we're deferring specific updates on probe performance for now, our focus continues to be on maximizing the yield and functionality of both our existing and in-development probe candidates so that we can deliver on the high-quality and proteome coverage our platform is designed to achieve. We expect this optimization cycle to continue over the next two quarters and we'll keep you updated as milestones are reached. Following the technical progress we've just outlined, it's important to put that work in the context of how we're engaging the market and building customer demand. It's worth noting that the two primary applications we're targeting, proteoform analysis and broad-scale proteomics are at very different stages of market maturity. On the proteoform side, we're introducing a fundamentally new measurement capability that hasn't existed before. As a result, we'll need to invest in market development and work closely with academic researchers and pharma partners to validate the impact of this data. Additionally, we need to develop exemplars of how this new measurement modality can be integrated into existing research and drug development workflows and into modern AI-based development workflows. In contrast, broadscale proteomics is already a well-established need. Our target customers fully understand the value of this type of data, have budgets allocated for it, and are actively seeking more effective platforms to generate it. However, even within the broadscale landscape, the data generated by existing affinity-based methods and mass spectrometry-based methods is fundamentally different than the data generated by the Nautilus platform. We believe our platform is uniquely positioned to become a cornerstone technology because our iterative mapping method will provide a resolution and scale exceeding that of existing methods. We specifically anticipate our platform will be unique for its high reproducibility, extreme sensitivity, and wide dynamic range. Furthermore, because we get multiple measurements of each protein molecule, we anticipate being able to provide higher resolution views of each protein rather than simply quantifying total protein or peptide abundance. As we continue to examine the unique and important role that Nautilus will play in the proteomics landscape, we recognize that building deep, trusted relationships with biopharma organizations will be critical to Nautilus' long term success. That's why Parag and I are personally leading those discussions, ensuring that we're not only showcasing the technology but also deeply understanding how to align with the real-world needs of potential customers to unlock a new era of advances in biological insights and therapeutics. I'd now like to turn the call over to Anna to walk through our financials.

Anna Mowry, Chief Financial Officer

Thanks, Sujal. Total operating expenses for the second quarter of 2025 were $17.1 million, an 18% decrease from $20.8 million in the same quarter of 2024. This result is attributable to a reduction in personnel costs from the headcount reduction we implemented in Q1 as well as normal variability in the timing of R&D activities and ongoing cost optimization efforts. We also saw a meaningful decrease in stock compensation expense year-over-year. Research and development expenses were $10.4 million, down from $12.4 million a year ago, while general and administrative expenses were $6.7 million, down from $8.4 million in Q2 2024. Net loss for the quarter was $15.0 million compared to $18.0 million in the prior year period. We ended the quarter with approximately $179.5 million in cash, cash equivalents, and investments and continue to project a cash runway that extends through 2027. While we're planning for a pickup in research and development spending in the second half of the year, we anticipate that total operating expenses for the full year of 2025 will remain below 2024 levels while still supporting critical platform development and early-stage partnership activities.

Sujal M. Patel, CEO

Thanks, Anna. We're incredibly proud of the Nautilus team for the science we're advancing, the platform we're building, and now the data we're publishing. With the public release of our first manuscript, we reached an exciting new phase. The world can now see and evaluate our technology on its own merits. Our foundation is solid. Our belief in the mission has never been stronger, and we're excited about our path forward. Thanks for joining us today. And with that, we'll open the call for questions.

Operator, Operator

Our first question or comment comes from Dan Brennan from TD Cowen.

Daniel Gregory Brennan, Analyst

Great. And obviously, congrats on the presentations at the event and the manuscript. Maybe just kind of starting off on the reaction so far from the field. You discussed the reaction is so strong, and you walked through just how unique this targeted approach is. But at the same time, you talked about the need to kind of build awareness and educate, because it is so different. I'm just trying to kind of reconcile both of those. Maybe could you speak a little bit to maybe the kind of early collaborators who you signed up and kind of what the pipeline of demand looks like and kind of how you think this might manifest in actual projects and/or revenues as we look out over the next 18 months?

Parag Mallick, Co-Founder and Chief Scientist

I was at AAIC talking to researchers in the field, and it was striking to see the tremendous interest from the Alzheimer's community in better ways to understand the disease. We have new markers like p-tau217 that show greater specificity, but they aren't predicting the disease's course or effectively used as surrogate endpoints in clinical trials. They also can't predict which therapeutics might work for different populations or differentiate among various tauopathies. One researcher pointed out that proteoforms are critical for understanding the disease, particularly how tau forms fibrils that underlie many symptoms. This emphasis on the importance of proteoforms in understanding and progressing in Alzheimer's was consistent in every conversation we had, which was exciting. While it is a new measurement, there is a recognition that the current class of biomarkers for tau and the mechanisms for understanding tau do not focus on total tau. Instead, they center on post-translational modifications and truncations of tau. Therefore, acknowledging that it's not just about the whole protein, but all its different variations, is crucial for understanding.

Sujal M. Patel, CEO

I'll take the second half of the question. Your question was about the pipeline and the opportunity for tau proteoforms to develop. I want to emphasize something mentioned in the prepared remarks. The go-to-market activities for proteoforms will differ from those for the proteome. Regarding the proteome, our customers are familiar with complete proteomic data and are seeking better tools for analysis. They want more effective and user-friendly proteomic analysis and have existing budgets for this. Therefore, when we launch our proteome product at the end of 2026, we anticipate a faster ramp-up in revenue and instrument adoption. In contrast, the proteoform opportunity, which I find equally exciting, will take longer to develop. By development, I mean the market activities necessary to demonstrate the potential of proteoforms, as well as our efforts to create assays for each biomarker individually, starting with neurodegeneration and focusing on tau.

Daniel Gregory Brennan, Analyst

Great. I have one more question. While the proteome is clearly the main focus, I'd like to know when you expect this manuscript to be published and in what type of journal you hope to see it. Additionally, regarding revenue, given that the proteome represents a larger and more accessible opportunity, how would you evaluate the potential for the proteoform product if you achieve your goals?

Sujal M. Patel, CEO

I'll pass it to Parag to give you the first part, and then I'll tackle the second here again.

Parag Mallick, Co-Founder and Chief Scientist

We consider this manuscript to be very significant for our future. It has been sent out for peer review, which we believe will enhance it through the feedback received. Ideally, we would like to publish it in one of the top three journals. However, the timeline for publication is unpredictable and will depend on the journal and the review process, so it's best not to guess when it will be published. Our expectation is definitely to aim for a high-impact journal.

Sujal M. Patel, CEO

So Dan, regarding the second part of your question, we are still in the early stages of assessing the size of this opportunity. Looking ahead over the next five to seven years, this could potentially become a multi-hundred million dollar business for us, particularly in proteoforms alone. However, the key question remains whether we will provide access to this data without any benefit to Nautilus based on our findings, or if we will establish partnerships focused on therapeutic or diagnostic development to share in the benefits. Because the data from proteoform research cannot be generated by any other means, it gives us more flexibility in choosing our business models. I don't have a precise answer on how this will develop, but from our initial discussions about neurodegeneration and the feedback we've received in broader areas like oncology and autoimmune diseases, proteoforms are very promising. Although we see this as a significant opportunity, it will require some market development.

Operator, Operator

Our next question or comment comes from Subbu Nambi from Guggenheim.

Subhalaxmi T. Nambi, Analyst

Congratulations on the publication. From the tau manuscript, where did you get the most inbounds in terms of customer profile? And can you share any other information on the funnel?

Sujal M. Patel, CEO

Yes. So Parag, do you want to take the first half? I'll take the funnel question.

Parag Mallick, Co-Founder and Chief Scientist

Sure. I think we've actually seen a tremendous amount of interest from academic groups reaching out, from pharma reaching out and from nonprofit research institutes reaching out. So it's actually been across the board.

Sujal M. Patel, CEO

It's too early to discuss what's in the funnel and its development, especially since our most significant activity just took place recently at AAIC, which had around 8,000 in-person attendees, significantly more than the HUPO conference. This provided a great opportunity to engage with customers. We likely had several times more new conversations this week than in the entire previous year. The initial interest we're seeing aligns with what I mentioned earlier; it's coming from academic institutions, academic principal investigators, nonprofit research organizations, and foundations focused on neurodegeneration. We're also identifying some early interest from pharmaceutical groups looking to conduct pilot studies to explore how this proteoform data can enhance their therapeutic development programs. However, it remains early in the process. Over the next two quarters, we expect to see further developments in this area and will be able to provide more insights.

Subhalaxmi T. Nambi, Analyst

And so, just a follow-up to that. Does having these two collaborations now actually help you show a former proof of concept? Or you think these are independent?

Sujal M. Patel, CEO

So I think that the two collaborations that we've recently signed here, I think, are focused on a few different things, right? First and foremost, both of them are focused on reproducing the data that we have on our samples that we talked about in our manuscript. They're focused on increasing the number of biological samples that we have that have gone through our platform and starting to take some of those insights and showing the world what the power of proteoforms are. And then in addition to that, another key goal for us is that because proteoforms and proteomes share the same platform, each of these collaborations is serving to harden our core platform, to harden our consumables and to work through some of the kinks that you'd have to work through during an early access program on proteomes doing those earlier with proteoforms today. And so I think for all of those reasons, I think this work that we're doing with these collaborations and others to come is really quite strategic for us.

Subhalaxmi T. Nambi, Analyst

Got it. And one last one. I know you guys have decided to not give any additional specific update on probe performance, but any reason you decided to not do that moving forward?

Sujal M. Patel, CEO

Yes. To clarify, I didn’t say we wouldn’t provide updates in the future. We definitely plan to do so in the next two quarters. We didn’t do it today because, as you may remember from our earlier comments this year, we introduced a change to our assay configuration. This change was aimed at improving part of our assay to enable a larger number of the probes we've developed or are developing to function correctly on our platform and in the assay. At the start of the year, many of our probe candidates were not performing as needed and did not meet the specifications to achieve their targets. Instead of creating thousands more candidates, we paused to change the assay configuration and ensure that a larger percentage of them would work effectively on our platform. The assay configuration adjustments are a multi-quarter process. It’s progressing as planned, and recently, we reached a stage where all components of the configuration change aligned, allowing us to begin testing the entire probe library with the new setup. The next couple of quarters will be crucial for completing our transition to the new assay configuration. Once we determine the yield percentage or how many probes are working well in the new setup, we will certainly provide guidance.

Operator, Operator

I’m not seeing any further questions at this moment. Thank you all for joining today's conference. This concludes our session. You may now disconnect. Wishing everyone a great day.