Illumina, Inc. (ILMN) Earnings Call Transcript & Summary

Welcome. Day 2 of the TD Cowen Global Healthcare Conference. Wrapping up the day, terrific speaker here. We've got Jacob Thaysen, CEO of Illumina. So Jacob, welcome, and thank you for being here.

For sure. Thanks for inviting, Dan.

Awesome. A lot to talk about. I certainly think coming off of an eventful AGBT, I think it kind of makes sense probably to start there. I'm sure there's a lot of inbounds and questions you guys have been fielding as if we had. So maybe just high level for yourself, what was the message coming out of AGBT in terms of your own product road map? And also if you want to begin to speak to a little bit about Roche and the competitive landscape, and then we'll dig in?

Yes, that's certainly a lot to cover. We felt it was a really good conference. It was a really good show. We have been very much looking forward to show our road map at AGBT. What our focus were for our customers was to show that if they invest in Illumina, we will make sure that we will continue to innovate for them to be successful. And how that is translated in is that we showed different elements of our road map, particularly on the NovaSeq X that we -- it's been a tremendous success so far, but we also showed our customers we'll come up with more flow cells to both address higher capacity applications, but also lower capacity applications. Different customer segments look for different things. We also came out and showed that we will get to a Q70 scores on the quality, which is opening up for especially high-sensitivity MRD, the fact that nobody can do this today, but with the Q70 that we're going to present, we are opening up for very high-performance MRD that -- where the whole market is moving right now. We also showed that we can do faster turnaround time with the NovaSeq X platform. So I think that was very well received, especially in the research academic environment where today, there is, as you know, not a lot of funding. So they really appreciate that we continue to extend the opportunities with the X. So what they have invested in, we continue to make them successful. We also launched the -- what we now call the TruPath, which we had under codename constellation before. It's a technology that essentially eliminate library prep for whole genomes, but it's not only that you can do library prep in 10 minutes, you also get extended insights from your genome. So it's really a technology that combines short and long read, all the power of the short read, but also all the power of the long read. So you can imagine for germline application for hematology cancers that this is going to be the new standard of the genome. We come out and presented this at this point for a price point of $395, which is the absolute lowest and highest quality, but the lowest cost of that type of genome. We also showed a road map that we will do multiplexing on it later. So right now, you can run 16 samples on a run, and we'll get up to approximately likely around 96 samples over the next 18 months, which will also drive down the price point that is going to be very, very attractive. So we saw a lot of customers that are very delighted about that. Effectively, if you have NovaSeq X, you can get started immediately using our TruPath genome. So those were -- then we also showed a lot -- so yes, there was a lot happening on our spatial technology that we again are showing that the spatial we're coming out with is going to be where you can actually look at very large tissue. So instead of looking at very small tissue, which you can't really do biology on, we can now look at actually a whole slide tissues that will open up again, and we're going to price that to a point where we're starting to open up really applications. And it's a lot of excitement. In fact, we were overwhelmed with the interest that were in spatial. It's still in early access, but we see a lot of interest in that. And so we're excited to get that out here during the next 6 months. So that was a lot of the things we presented. We saw a lot of positive feedback from customers. They really love actually the fact that we're not launching an instrument, that we are -- that what they have invested in that they can continue to and we continue to innovate on that. So that was a big part of our confidence, of course. Then of course, this is also a conference where every competitor is putting their best foot forward, and we definitely saw that. Most of our competitors are excited about coming with new instrument launches. A key -- again, a key signal we want to send is not about only sequencing anymore, it's about the highest quality with the lowest end to end. So Roche came out, and we actually have a good relationship with Roche people also. So we're going to compete with them of course, going forward. We were I think what -- the way to characterize it is I think Roche have done a great job of going out over the last 12 months to present all the good things and all the promises around the technology. I think we're now starting to get more of the details about all the consequences of that technology, which appears to be a very, very different workflow, appears to have some challenges in how you actually have to do library prep. And of course, also the pricing model requires you to do high throughput. So it kind of limit the applications that we at least see for now. I'm sure there will be more coming up with Roche, but at this time, we feel very good about how we're going to compete against Roche both short term and long term.

Okay. Yes. So maybe we'll kind of unpack a lot of that. But I had the second question on Roche just because it has been such a focus for investors. You kind of mentioned some of the drawbacks there. But maybe can you just unpack a little bit more when you think about where NovaSeq X sits? Obviously, clinical is very entrenched and that could take time, but even on the research side, maybe where they're going to start? The profile like where do you think the AXELIOS 1 will compete? Where do you think they won't compete? And we can go from there.

Yes. I think -- I mean, there's many details here on the complexity of sequencing. This notion about that sequencing becoming commodity is not true at all. I mean -- and I would say I've been in the life science industry a long time and of that commoditization in life science tools is not really true. It's very few areas you see commoditization. So -- and sequencing is definitely far way from that. But of course, you have to go into the details to really understand it. So there's a lot of complexity on how you run. You have 1 flow cell or 1 [ chip ] with 1 lane on, so you have to think about how you mix and match your different applications. I think there are some challenges on how you do panels, which is especially in the clinical space, very important. But where we definitely saw them put a strong foot forward, it was in the, what we call, accounting application, where the Simplex read is -- we'll be able to address that market segment. From what we can see and when we understand the market and we see that, that is around -- probably around less than USD 100 million of the space that we participate in. So we feel good about that. I think the more important part of that market is less about the cost of sequencing, but really the cost of these experiments in the accounting application that is sitting between the single cell, the bulk RNA obviously also, but -- and of course, the affinity protein part, which is probably the majority of the accounting applications is that the library prep is much more expensive than the sequencing. And thereby, you're not really getting a lot of benefit from a customer perspective to drive down the price of sequencing. Yes, of course, you save money, but it's not like you certainly can do if you drive down the price of sequencing a factor of 2 that you get twice as -- you can run twice as big experiments. You might be able to do 20% more. So the notion of just driving one parameter in that experiment is not helping for any elasticity conversation anymore. And that's what researchers want to do. They want to go from 10,000 single cells to 100,000, from 100,000 to 1 million. And therefore, that's the reason we acquired Fluent, we acquired SomaLogic and other elements to really address the end-to-end applications. So we can drive down the cost of the whole workflow to really make sure that elasticity is working. And so we feel good about our position in that space, even though the headline is a very cost per million read.

Okay. Well, we can come back. I mean we'll have some more questions on that. I don't want to spend the first 10 minutes or 15 minutes on that, but we'll circle back a little bit more. Maybe just kind of switching over to the key tenets of your guidance this year, right, 1% to 3% organic growth ex China. Maybe from a high level, what would you characterize as some of the levers to take you outside the range, either on the upside or the downside? And would you view that the way you set the guide, like some companies will set the guide and they'll say the midpoint is at the low end or the midpoint is at the high end. So would you view that 1% to 3% fair, conservative or maybe on the aggressive side?

Yes, I would start by saying that it's interesting how the conversations are evolving, where we, of course, a year ago, had a very different conversation about NIH funding, our channels in China, tariffs and so on. And today, we have a conversation about, well, you're conservative in your growth rate. So I like that. That's a much better conversation to have than what we had last year. So I appreciate that. We also saw momentum through '25, where, of course, the first part of the year was challenging with everything that was hit from a macro perspective. And the second half, we started to see momentum in the business. So the way we have thought about '26 is to look at the second half of '25 and see that's probably a good proxy for how '26 is going to look like. There is momentum in the clinical business, but there's still some challenges in the research business. And that's how we have modeled '25, and that's how we have guided -- excuse me '26. So if you think about it right now, we have said that our guide for 2026 is, right now, we call rest of world outside of China is between 2% to 4%, with China, it's 1% to 3%. And at this point, we believe we are -- we're always guiding for the midpoint of that. But we still believe there is -- should clinical grow faster, which it could be, we don't see any reason why clinical shouldn't continue to be in a very strong trajectory. And should the research space actually become better than what we are anticipating, that's definitely upside to our guide as we see it right now. If you look at the bottom line, we also continue to expand the -- both the operating margin and also the EPS growth. So on the operating margin, we are -- on the core basis, before the SomaLogic acquisition, we were guiding for 150 -- approximately 150 basis point improvement, which would actually translate into, if we do another 150 basis point next year, we would actually be at a 26% operating margin. Now with SomaLogic, we are within 50 basis points of that. You've seen how we operate. We will do -- we'll work a lot to see if we can close that gap out. And EPS, we guided, at least before acquisition, to a double-digit growth, 10%, and with -- of course, with the headwind from SomaLogic a little bit lower than that. But we feel really good that we continue to grow the EPS also double-digit to teens going forward.

And on the consumable guide for clinical, I mean you get this question, Conor gets to this question, but coming off such a strong back half of the year, 20% growth in the fourth quarter, you're guiding to a healthy growth, call it like lower low double digit to mid-teens. So call it, 11%, 15%, at least what we have. But you're coming off such a strong growth rate, just conservatism, just don't want to get ahead of yourselves or, listen, you've got this transition ongoing, so maybe that truly is going to be where you land. It just feels like it might be biased to the lower end just given where you exited.

Yes. I think -- I mean the transition is still going to happen here in '26. So there's definitely still some pricing headwind. It should start to dissipate over the -- throughout the year. So it shouldn't be worse than '25. The accuracy and how we can predict the growth in clinical is still, I would say, I can't predict it more than within a few basis or within 1 percent growth point. So I think that's why we have said this is probably from a guide perspective. Could that be a little bit higher? Potentially. But we are, of course, very dependent on how the whole environment continues to evolve, how the funding is happening, the macro environment is happening. Just last week, we had another big macro event. So I think that's just why we want to keep some prudency into our guidance at this point.

And in terms of the application, obviously, oncology, women's health, you got screening kicking in now, but what application would you say is the most exciting when we think about the next couple of years for NGS consumable on the clinical side?

There's a lot to be excited about. I think that's the...

I am asking you to pick your favorite child...

No, no. But I'll come back into some of the things -- the dynamic here. But I think the point is still that we are still early -- in the early stages of the opportunity for sequencing, especially in the clinical and the health care. Sequencing is far from standard of care in health care. And eventually, we'll get there. But we also see both the -- many physicians are getting more comfortable of using, of course, genomic insights to make the right decision in their practices. We're starting to see that, that reimbursement is starting to more solidify, so you can actually get predictability on reimbursement. And you see also the more comfortable clinicians and physicians are getting around using sequencing, the more insights they are looking for. So when you previously wanted to have 20 genes looking at, now you want to have 500 genes, and eventually, you want to look at the whole genome. So you're also seeing that happening. So we see a lot of growth in the clinical space for many quarters, if not years to come, also for years to come. If I look at the dynamic is that the oncology will still be the main growth driver in this space, at least for the foreseeable future. And where, of course, most of the business today is still in therapy selection, we're now shifting ourselves into the new type of detections like minimal residual disease and of course, also early detection of cancers. So over the next few years, it will still at least a year or 2, it will still be the therapy selection that will be the main part of the business and thereby also dollar-wise the main growth, but percentage growth, the MRD and early detection is definitely starting to kick in here over the next period of time.

And maybe just on the research side of the equation, just maybe -- and we'll kind of loop it back to some of the new competitors, like you've got the ongoing pressure in the U.S., but hopefully not getting worse, right? When you think about the guide, like is anything incorporated for some of these other -- really, Roche would be the main one, but like even if they're not doing panels, they're still excited on their whole genome. So could that free some of the market? Like have you tried to think about that when you've kind of set the guide?

Yes. We, of course, think a lot about that we are in a very competitive space, and we think about a lot how we can continue to compete and outcompete, and I feel really good about where we are from both where we are from a commercial execution perspective, but as importantly, also our innovation pipeline that continues to be very strong. And I think AGBT was a good showing of that, but much more to come. So when we guide, we try to take as much insight in that we have around what we know about the competitive space, and that's also what is reflected in our expectations. We do believe that even in a highly competitive space, we should easily be able to grow with high single-digit growth over the next period of time as we have said in -- as our goal is in '27. We don't see the market freeze right now. We see that many of the placements we do in NovaSeq X is into the clinical. They have made their decisions. They're going to expand their fleet on this. There will be, of course, individual core lab and others that will test technology, but that's the usual business, if they are not testing one technology, they're testing the other. So we don't see that as a big dynamic change. I'm not seeing any freezing of the market at this point. Could that happen if something big? Who knows? But -- and that's also why we're not going out and talk about instruments in our road map this definition because we don't want to freeze the market.

Is the -- excuse me, is the 35B flow cell that you announced at AGBT, have you priced that yet?

The 35B flow cell?

Yes.

No, we have not priced that yet. We will, as you have seen, we will talk about our road maps, and we have beyond those flow cells also if we need to. But we're not going to price it before we are ready to launch it. As we did with the Constellation now TruPath, we talked about that over the last 9 to 12 months, but we priced it when we were ready to launch it. We don't see any value in coming out too early with pricing. I don't think that benefits anyone.

And kind of Q70, when is that available? And what kind of uptake do you think that could be? And is there a -- what's the financial impact from that?

Yes. So we have -- right now, we went out and saying look, over the next 18 months, there's a lot of cool stuff coming out, including the Q70. We did not put a specific time line on it, and don't take that means that everything is coming out in June '27 or something like that. It means that over the next 18 months, things will come out. We think it's going to be -- as everything nothing is -- of these library prep, it's not a pop, meaning that it's not going from 0 to huge growth from 1 day to the other, especially the technology that addresses clinical application. It takes time. The clinical customer wants to get access to it, build it into their applications, get the clinical validation on it and so on and so forth. So it will take a little bit of time to get growth, but it's something that can really change our -- how you think about MRD going forward. And that's why we think it will be very attractive for many of our customers.

Got it. Maybe just shifting over to some of the new products. You mentioned protein prep. You and I were just discussing spatial and single cell. Just on the protein prep side, just kind of, of the 3, it feels like that 1 maybe is the biggest, but how do we think about -- I think you've kind of lumped all 3 together and get an aggregate impact. But how do we think about maybe the opportunity for protein prep now that you've enabled it on sequencing?

Well, it certainly is the biggest now because we acquired SomaLogic and thereby in itself, it comes with already a revenue base. So that's a starting point that is different from the ones that comes organically from within. We do think protein and using DNA sequencing for looking at proteins also is going to be a big opportunity. We see a lot of interest, especially in pharma right now, but for many studies where you want to look at biology, you want to look at different layers of biology. The opportunity for us is to continue to expand from now 9,500 proteins we can address to -- up to the whole proteome, which is around 20,000 proteins. And no other technologies will be able to do that. Secondly, the workflow on the cost to run the experiment, as I told before, many of the counting, the library prep is much more expensive than sequencing. So us driving down the cost of end-to-end solutions here, I think, can open up for many -- I mean, elasticity again, that is -- that could really open up for many different applications. So that's some of the things we want to work on. We also see for, I think some early indication a few weeks ago is that when you look into early access of cancer is that looking at one dimension might not be enough. You might have to look at both the genome, the methylome and maybe also the proteome at the same time to really get that enriched information. So we see opportunities also for going smaller panels here. And finally, for the proteomics combining that with spatial. Right now, we look at the transcriptome, eventually, you will also look at the proteome and the aptamer technology is very well suited for that. So I'm very excited about the -- now Illumina, together with our new team in Boulder, the former SomaLogic team, are super excited what we can do there. As I mentioned also at the conference, I think we were a little bit overwhelmed with the interest that were in spatial. And now what we are opening up that you can look at tissue full on the slide and also with the resolution we have and also that we will not go out there and come out with a 10% cheaper. We're going to change the end-to-end cost of using these experiments. We talked at the conference around at least a 4x cheaper from an insight perspective. And I think we can do even better than that also going forward. So we want to really open up to these markets so we don't do just 1 tissue and 2 tissue, but you can look at hundreds, if not thousands of tissue. That's where science really steps in. And that's where we need to bring this. And we're going to do this as we did with sequencing. We're going to do that also for spatial and for proteomics.

Have you sized like of the Illumina installed base, how many you think could adopt protein prep, like...

Yes. Not of this. I mean we still think the workflow is a little too complicated to really open up for the huge opportunities, so we will be working on that. But we think that, especially for the bigger experiments, that's where proteomics will come in beginning where you have really run a lot of samples where the cost of building in automation and so on will take care of that. So -- but over time, we want to bring that in so it's going to be a very little, or much smaller capital equipment investment to get into proteomics.

Will you break that out when you report or how -- or just be like in another line like the protein prep just fall into sequencing?

It's fall into library prep right now, but we are still considering how we're going to make sure that all of you understand how great this is going to go.

Got it. And maybe just a question on the spatial 4x cheaper. Is the idea that if you're doing a lot more samples when we do like a per sample would be cheaper? Or do you think actually the flow cell instead of selling $7,000 panel that 10x or a $4,000 panel would be 40% cheaper than that 4x cheaper than that?

Yes, we don't like for space, and we think we don't like to talk about the cost of sequencing, right? We'd like to talk about the cost of insight because that's how our customers are thinking about it. What are they looking at? And then start to have a conversation, if you look at what's the cost -- if you look at single cell, what's the cost of per cell insight? What's the cost on proteomics, is that the protein is that for sample insight? And the same for spatial. Is that per square millimeters or is that per square cell or individual cells? So I think that's a conversation we want to have with customers where we are aligning our pricing model with the insight they're trying to achieve in their experiments. That's how we think about pricing for those applications. But we need to own and have control over the whole end-to-end solution workflow in order to have that conversation. By the way, very important, we will still keep our ecosystem open. So everyone that has a single cell or proteomic assay and so on, we would, of course, make sure that, that works on our ecosystem. We want to make sure that our customers have freedom to choose which panel, which library prep they want to choose, but we want to, of course, come up with a very, very competitive end-to-end solution from Illumina.

So it sounds like spatial maybe is a bigger opportunity than a single cell for you or you wouldn't say that?

I think the way we think about it right now is that from single cell, especially the high-volume applications, the millions of cells is something we're excited about. I think that is something, again, we want to drive that because that drives deep insight, that changes biology. That's the direction we want to go. That's the same spatial, the big experiments with a lot of insights and obviously a lot of sequencing. So we like that, those applications.

Maybe just switching gears to your offering Software-as-a-Service to customers, help them deal. You've kind of talked about this goal, right? SaaS model, deal with the heavy computational load when interpreting data, and you've highlighted BioInsight can easily prep and structure complex data trading a foundational model. Just kind of what's -- can you speak a little bit about this? BioInsight And will we hear more about this? Kind of how important is this for Illumina?

Yes. I think if you think about it over the last years, we all talked about AI, and we do that also internally for how we can optimize our company with AI. But where we sit in a very unique space is that where AI, large language model have changed, of course, our understanding of AI, but also what these models can be used for. But it only became possible because the amount of data that the Internet today have that you can train those models on. And there is a scaling law, the more data, the bigger models, the better insights you get. The same is likely going to be the case for biology. The challenge we have right now is that there's no Internet of biology. There's no big data sets out there that you can train models on. And the data that's out there is very dispersed. We don't have the right standardization. We don't know exactly how those experiments was done. And thereby, it's difficult to have high-quality input into your models. So BioInsight is sit and is put in the world to create those data sets and create that so we can start to really build AI foundational model that can start to predict biology. And Illumina is extremely well positioned. We have, of course, the scale. We have the technology to do that because you need to build -- you need to have billions of insights to do so. you need to have -- if you look at [indiscernible] which we're very excited about, you need, first and foremost, to being able to sequence billions of cells. You need to be able to have the compute power, which we have through Dragon. You need to have AI capabilities, which we have in the company. But also importantly, you need to have a CRISPR library that is optimized across the whole genome, so we can knock in and knock out genomes across the whole genome and the whole pathway. When you have that combination, you can start to build these data sets that we're doing today. Pharma has been very interested to get access to these data sets and actually pay for them. So we will create those data sets. We will sell it to pharma in a subscription base where we're making money in creating the data set. And at the same time, we are accumulating that data that we're starting to help building those foundation model for the individual pharma company to help in that drug discovery, but eventually also build up deep insight that can really change our understanding of biology. And we will make sure that, that's all available for Illumina customers. So if you're part of our ecosystem as a researcher, you will get access to this data, you will get access to this model, you'll be able to build your own model. If you are in a clinical customers, that data is giving you better interpretation than anyone else can provide. So you can imagine a future where you're not only buying sequencing from Illumina, you're buying insights. And this is going to change. And I think we are very well positioned to take being a very substantial player. So that's why we put BioInsight in place and so far, so good.

Okay. So we have a couple of minutes left. I thought it would be worthwhile just to come back to Roche, if you don't mind, just to close it since most of the questions we get. So what do we -- we're sitting here today, they had this launch and there's some missing pieces on them, a variety of things. I guess a couple of questions. First, on the whole genome market. You sized the accounting market less than $100 million. Like how big is whole genome for you today, maybe in research? And do you think that product looks viable from the specs you see there? Just so how do you think your head-to-head with them there?

Yes. We feel really good about it, especially also with TruPath, the Constellation methodology, where you're not only getting short read information now, but you're getting all the structural long readout. So I think actually TruPath is going to change the way you think about a genome. If you look at whole genome applications, why wouldn't you like to get that? And no other technology out there can actually provide the insight that TruPath can and also with the simplicity of the workflow. If you have a NovaSeq X already, you can get going immediately. So we actually think that we're very well positioned for in the genome space, both for research in the beginning, but also for clinical applications. So we feel really good about where we are on that. If you want to still do a short-read genome, we feel good we can address that market also well. I mean remember, it's not only the price of sequencing, it is really the insight. So the whole pipelines, the Dragon pipeline we have, the Tertiary insight also -- Tertiary Analysis Platform we have with the Amgen for rare diseases, but also the ICI for oncology application is also what's translating all that "genome" into insight. So customers want to get that. They're not just excited about the price of the genome.

Okay. And then maybe a quick one, Ultima. Like I know Ultimate Element, fine boxes just haven't really had the traction maybe people expected, although they have had some success. Anything new from them that you saw there in terms of price points or worries?

No. I think that everybody is putting the best foot forward at -- and I was talking to all of them, and so it's a good collegial environment we are in. But of course, we fight every day on the streets, which we should. We feel like they're trying to do what they need to, but we feel -- continue to feel that we have a very, very strong position in the market, and we'll continue to have that.

So maybe just wrapping it up then. Obviously, we covered a lot of ground. We kind of set the message at the beginning. Maybe just in closing, what message you want to leave with the investors then?

I think I want to leave you with the message that Illumina will be able to grow very strongly going forward, at least high single-digit growth even in a very competitive space. I want to leave you with that teams are competing. Illumina has the best team, both in our R&D, but more probably also very importantly, in our commercial organization. I think some of you got exposure to some of our people here at AGBT. I feel -- I'm very proud of being a part of a company that is changing the world as we are and will change health care going forward, and we committed to that.

Terrific. Well, thanks, Jacob, for being here. Appreciate you coming to the conference.

Thanks, Dan.