We're joining us on day two of HealthCONx. I'm Vijay Kumar, the life science tools and MedTech analyst. A pleasure to have with us Quantum-Si. Representing the company, we have CEO Jeff Hawkins, and we also have the CFO Jeff Keyes in the audience. Jeff, thanks for spending the time with us. We're, you know, just the space that you play in, protein sequencing, proteomics. Just give us a little bit of background on how Quantum-Si fits in. What do you guys do? What are your key products?
Sure. Yeah, you know, the field of proteomics is, depending on the market report you read, is about a $75 billion market, and inside of there, there's a little bit larger than $20 billion research market, and we play in that market. And I would say that there are two sort of key applications or two sort of approaches in those markets, right? On one end of the spectrum, you want to screen for a large number of proteins from a single sample, so you're looking at sort of protein throughput per sample. And in those cases, technologies such as the ones from Olink or SomaLogic are sort of the common technologies for those screening studies. On the other end of the spectrum, you want to really deeply interrogate, you know, individual proteins or small mixtures of proteins.
You want to see post-translational modifications. Maybe you want to see variants of proteins. That's when you want to use a technology like Quantum-Si, where we're doing protein sequencing. So we're giving you amino acid level resolution, and we're the only technology today giving you that level of resolution.
Understood. Just within that field, Jeff, what is the unmet need? Like, can't you do this with any of the other systems out there, or is your system coming up with a different approach? Maybe elaborate a little bit on the system.
Sure. Yeah, so we sort of think about customers in two different groups who are adopting our technology. On one end of the spectrum, you have sort of the core proteomics centers, the core proteomics labs. Those labs typically have access to a lot of technologies. They probably have been doing mass spec for many, many years. They'd likely have other technologies as well. So in those labs, we tend to get used to do the types of analysis that are either difficult or aren't accessible with mass spec. So people want to use this for variant identification. They want to use it in post-translational modifications, things that, again, are sort of difficult or not feasible to do with mass spec.
The other end of the spectrum are either labs that do proteomics research and send out to the core labs or are genomics labs that are broadening their focus into a broader multi-omics, and they want to have a proteomic technology. And therefore, the cost points, the workflow simplicities of our technology become bigger drivers, and they'll use the technology in that case for things like protein identification or characterization. You know, less, you know, maybe different applications, not pushing the envelope of the tech capability as much, but it's really more that story of accessibility and cost and ease of use for them.
Understood. And just within that landscape, can anyone else do this? And what is your secret sauce, and you know, what's allowing Quantum-Si to do protein sequencing?
Yeah, sure. So today, we're the only company who's got a commercially available, you know, protein sequencing platform. There are, you know, other companies who are, of course, working on these things, as there are in exciting markets like this, but they're all in the pre-revenue stage. You know, really, we've innovated across every attribute of sort of the technology stack, from the semiconductor chip that we've adapted for this use to the hardware it runs on, all the way down to the individual components of the sequencing reagents, the molecules that actually recognize the amino acids, the enzymes that are in there that cut the peptide one amino acid at a time. So if you look at sort of the technology stack, if you look at sort of the IP around this, it spans that entire range.
You know, we've had to really innovate everywhere. And then, of course, when you have an innovative technology, you've had to build algorithms and the data science over the top of this, all in a de novo fashion, because it didn't exist. You couldn't just pull it off the shelf and, you know, leverage a pipeline from another technology like you can in sort of genomics, today.
What is the workflow on this platform? Is it complex? Like, who's the average user? Is this like a highly skilled scientist who's running this?
I mean, by nature of being in academic research, labs today, you're largely going to have somebody with a scientific background. They could be a postdoc, you know, could be somebody in graduate school. So you have a skilled person. If you ask our customer, what they'll tell you is the workflow from the raw sample to getting it into our device is very similar to what they experience if they use mass spec, right? They have to digest that protein into a series of peptides, and then we have a step to attach a linker that allows it then to go into our platform. So they would tell you that upfront workflow is very consistent with what they do in proteomics today. Once you get on the sequencer, then it runs overnight, and the next day, that's where they'll tell you there's a big delta.
When, you know, at the end of our run, we have a cloud-based analytics tool that does all of that analysis and really gives you much more of a point-and-click user interface to sort of interrogate data and go from summary-level information all the way down to even individual amino acid-level information. Compare that with mass spec. You're going to get sort of an Excel file that has a whole bunch of tabs with raw data, and people are running pipelines or doing informatics over the top of that. So I think the experienced users in proteomics, you know, really see that back-end step, you know, as a huge jump from where they were in terms of the automation and simplicity of the analysis.
I see. And you did bring up, you know, academic labs. Is pharma also a typical user of your customer? How does your revenue mix split between academia, pharma, and how does the revenue model work? Is this a razor razor blade?
Sure, yeah. So at the revenue model level, so we sell the instrument today for $70,000. It's a capital sale. We're seeing very sort of modest discounting in the market at those sort of price points. And then, yes, the customer then buys two things from us. They buy the library prep, so that allows you to digest that protein and prepare it to go into sequencing, and then they buy the sequencing kit, which is both the reagents and the chip. So those are sort of the razor blade in the business model. In terms of customers, you know, 2023 for us has been more of a controlled launch. So most of our customers today are academic labs.
But as we talked about, you know, earlier in the year and throughout the year, you know, starting to get traction, and we expected it to take a bit more time with pharma and biotech and other industrial, sort of applications. So we're seeing more traction there and expect them to become a bigger part, you know, of our user base in 2024.
I see. And, is there, like, a normalized level of, you know, consumable pull per system that would make sense for a platform like yours?
You know, I think today it's hard to say what that number is, being in more of a controlled launch. I also think there's gonna be a delta between, say, an academic lab, where perhaps they are doing work based on a grant, and they use the system very, you know, consistently during that grant. And then there can be a period of time where they're analyzing data and wrapping up their research before the next study, not unlike other technologies in the omics field. I think on the flip side, in pharma and biotech, in other industrial applications, you know, such as antibody QA/QC, I think those users are bound to have a more consistent, and continuous run rate. So we might see higher sort of pull-through per system in those segments.
Again, sort of early days for us to be predicting those, but we do see the likelihood that these two different, you know, segments of the market will have, you know, slightly different sort of utilization patterns.
I see. And why is the system in a controlled launch mode, Jeff? Is that because there are still some technological risks, or is this more for commercial, you know, expansion of commercial footprint and you're being thoughtful, just given the environment we're in?
Yeah, so we use the word control because it is a choice that we made to not just tell the sales force, you know, "Go sell it to every single person who wants it." You know, in the beginning of the launch, at the beginning of the year when we launched, you know, the reason for a controlled launch was really... There were a lot of things we didn't yet know, right? This is the first to market with a technology that can provide this level of resolution. It's our first product to market, so we had, you know, our own thinking around what would be the service and support model, what would it take to help get a customer live. I think those are things we learned in the beginning of the year, and they went largely as expected.
We certainly saw more complexity related to the sample, the biological sample types and sample preparation, during sort of the middle of the year, and that made it a bit more complicated to bring customers on live. And we've taken advantage of being in this controlled launch to say, "Let's really make sure we understand that.
Let's get our technology optimized for this." We've made a series of enhancements to software as customers have said, "Hey, give me this analysis capability or that." So it's really been about trying to be in the market and learn and really make sure everything is the way we want it and optimize the right way so that as we go to a full launch, then we don't have an issue where, you know, the sales force goes out and puts out a large number of machines, and then customers are sort of stuck. We didn't wanna ever be in that position, so we opted for, you know, the more controlled approach.
I see. And what have been the learnings? What's been the customer feedback during this controlled launch phase?
Yeah, I think I mentioned a few of them. I think on one level, some of the early feedback was just, you know, "Hey, can the software give me this? Can I see this?" It was the concept of amino acid-level resolution. You know, we had a thought about what type of data to present to customers and how to present it, but you don't really know what people are interested in seeing or how deep they might wanna go into the data when you have a new technology like this. So, you know, one way to think about it is, when we launched, we sort of gave customers, you know, level one or two of the data, and now we're going down three and four clicks into the data based on that feedback.
The other thing we learned was just how much more complex and how much more variable samples are in proteomics. I think many of us who came up through the life sciences field probably came through one of the big genomics companies. We all got very used to, you know, the value and what you get from PCR, right? You can amplify that sample, and genomics sort of helps deal with the low-abundance things versus the highs. You know, in proteomics, there's no equivalent of that. There's no way to sort of just have a step where you magically raise all of the things you're looking for in a sample to some sort of normalized level.
So that sample complexity really, you know, was probably even more than we expected, and we learned a lot from that. We learned how important it was to work upstream into the sample prep, sort of technologies and make sure they were adaptable. We learned a lot about the library prep step, the step that we provide, and how important that is in the performance of the technology, and things we could do to make it more robust, you know, to that sort of upstream variability that you see in the proteomics world.
I see. And you bring up an interesting point. Because there is no amplification, is there a signal-to-noise in a kind of question, you know, when you're doing this level of, you know, protein, you know, deep level of analysis, looking at proteins?
Yeah, I wouldn't really classify it as signal to noise. I think the challenge that customers are always sort of grappling with is, do you want a more pure sample so you can really have a clean sample and have it concentrated in a certain way, or you want to see as many proteins as humanly possible? And when you want to do that, then what range can you get? Because the dynamic range of proteins in a sample is, you know, about 10 to the fifth or 10 to the sixth. I mean, it's an enormous range of concentration. So depending on which end of the spectrum you're on, if you're on that screening end, you're sort of fighting that huge dynamic range.
If you're on our end, it's more how complex do they want the mixture to be. I think what we see more, Vijay, is that as people are sequencing, they really want to understand, you know, how confident are we in that call. You know, how many times did we see that amino acid? How many times did we see that, you know, that given peptide, or, or how did we infer the protein identification from that? You know, it's the common questions you expect when you're giving out sort of new technology and people are trying to understand, sort of the statistical confidence around these things. That's why we've, over the course of the years, sort of opened up the software so people can see more and more of these details to really comprehend exactly what each of these calls means.
And this space has always been interesting, proteomics. I think we've had some fits and starts over the years. But recently we saw a pretty big acquisition in the space with Thermo announcing Olink. Has something changed within the landscape? Or I was a little surprised to see the transaction. I'm curious, are we at an inflection point? If so, then you know, what is that inflection point?
Yeah, I would... First, I'd agree with you. You know, proteomics for more than a decade, people have been talking about how it was going to be bigger than genomics, and you'd have this rush of new technologies, and that would be sort of the headline for a couple of years, and then it would sort of become more muted, and then now it's sort of back again. You know, I think the...
It's hard to put your finger on the exact thing that says there's an inflection point, but what I would tell you is what we're seeing in the field, and I think it was sort of encapsulated in a couple of the papers that came out in Nature recently. You know, for a long time, there was this sort of message in the field, probably coming largely from companies in the tool space, that you know, you had genomics, and then you could do RNA sequencing. RNA sequencing was, you know, sort of a version of proteomics. RNA sequencing, you know, is a whole topic of its own, and it's a very valuable tool.
But I think a couple of these recent papers where they took RNA sequencing results, and then they ran, you know, the technology from Olink, as an example, and they found that RNA sequencing did not exactly predict what proteins were there. There were differences. Something that I think researchers and scientists in the field have always known intuitively, but it was that first time that sort of in a, in a very rigorous study, you know, peer-reviewed in a top-tier journal, that we started to see some of these studies coming out saying, "Hey, even if you do RNA sequencing, you still need to look at proteomics.
You still need to do a proteomics technology." And I think that really sort of, from a scientific perspective, solidified that, and now you're seeing, you know, some of the large players, as you mentioned, such as Thermo Fisher, who are saying, "Yeah, you know what? We're going to have a direct technology sort of play in this space," you know, for their customers, and that, you know, obviously was a very good outcome for Olink.
Understood. And when you think about your business in fiscal 2024, like, are we expecting a full commercial launch in fiscal 2024? I think you mentioned a launch of a V2 kit.
Mm-hmm.
Is that supposed to happen in fiscal 2024?
Yep.
What does that mean for a revenue step up in fiscal 2024?
Yeah, so the version 2 kit, we just talked about that on our last call. So that kit really is a culmination of things we've learned during the year from our controlled launch, as well as just technology, you know, advancements that we've been making in our internal R&D programs, and really bringing all of that together and going to launch a version 2 kit. And this is, you know, going to represent improvements to every aspect of the technology, from the library prep to the sequencing steps, and even, you know, continuing to advance those analysis tools. So, you know, customers will really see, you know, a step up in the throughput of the technology, the robustness of the technology, you know, various sort of attributes that we've been talking about today.
With that version 2 kit in hand, we believe that will be the facilitator then to move to a full commercial launch. So V2 kit launching in the first quarter, and then, you know, shortly after that, moving to a full commercial launch. We haven't given out revenue guidance on what that means yet, but really will put us in a position to allow our, you know, sales force to really go at full speed and not be, you know, sort of throttled by what we're allowing them to do in terms of the number of machines per quarter.
How large is the sales force right now?
So we haven't given out an exact number of people in the field. It's a fairly small team today. We've always been sort of very disciplined about these types of investments, making them sort of just in time, not ramping up too large and then having to pull back if we didn't get the feedback we wanted. So it's a pretty small team, but a very highly capable team, if you looked at the resumes and backgrounds of these folks coming from, you know, the leading players in life sciences. So we've had... I think our productivity per person is probably higher than it is when you have hundreds or even thousands of sales professionals, because we're fortunate to get these types of people into the company.
But we are beginning to expand that team in preparation for a full launch and, you know, even just a couple of weeks ago, announced our first distribution partner in Europe, that we're again starting to put some of the building blocks in place so that when that version 2 kit comes out, we can really hit the ground running with the full launch in 2024.
Understood. The last, in a minute or so, any closing remarks? You know, when you think about your, your company, what do you think that Street needs to understand the one thing about, Quantum-Si?
Yeah, I mean, I think we talked about one of them. I think the Street needs to really see what's happening here and recognize that, I think this time is when proteomics is taking off. I think Olink and Thermo Fisher sort of made the big headline, but there's a lot of companies, us being one of them, that are really, you know, seeing interest in the technology and seeing the traction. I think for us, 2024 is all about going to full commercial launch and really showing that we can scale the technology in the market. The version two kit is one step on the technological front, and I think we need to continue to show advancements on technology as well during the year. So really, full launch year for us is the message.
Fantastic. I think with that, we're out of time. Jeff, thanks for spending the time with us this morning.
Thanks for having us.