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

Good afternoon. Welcome to Day 1 of the Cowen Health Care Conference. I'm Dan Brennan. I'm one of the 3 tools and diagnostics analysts here. Recently joined, really thrilled to be here and thrilled to have CEO Francis deSouza, President and CEO that is, of Illumina with us here on the virtual stage to present. So Francis, I wanted to say welcome and thank you for being here.

Dan, thanks for having me, and thank you, everyone, for attending this. Maybe before we start, I should get the safe harbor statement out. So let me just start with that. I have been asked to remind you that my comments today could include forward-looking statements. You should refer to our SEC filings for a discussion of the risks and uncertainties that could cause results to differ materially from our current expectations. It is our intent that all forward-looking statements regarding our financial results and commercial activity made during today's discussion will be protected under the Private Securities Litigation Reform Act of 1995. Also, as a reminder, pending the outcome of the European Commission's investigation into Illumina's acquisition of GRAIL, the commission has adopted an order requiring Illumina and GRAIL to be held and operated as distinct and separate entities for an interim period. Compliance with the order is monitored by a monitoring trustee. During this period, Illumina and GRAIL are not permitted to share confidential business information unless legally required and GRAIL must be run independently, exclusively in the best interests of GRAIL. Commercial interactions between the 2 companies must be undertaken at arm's length. Okay. With that out of the way, let's proceed then.

Awesome. Thanks, Francis. So just a road map, I'm going to have a couple of introductory questions. We'll dig through Chemistry X, touch upon Infinity, go through the clinical market and then discuss competition and GRAIL. So we'll see. Hopefully, we can get to everything, and we'll do our best. So maybe just to kick it off, Francis, fourth quarter capped off a really strong 2021 and the guide for this year was better than people expected. Last year's results, I think, far exceeded initial guidance. So maybe give us a sense as we look at your guidance for this year, what are the levers by which if we were to stay in here at year-end and say, wow, we came in better than expected again. Like what is it that might have the potential to go better? And then within that, I'd be interested to get some color from you on the NovaSeq pull-through, which is an important part of your business and the guidance reflects kind of flat to down pull-through. And just wondering, what could possibly bring that pull-through above last year's level?

Yes. It's a great place to start, Dan. So let me talk through some of the considerations as we thought through the guidance for this year. The first point to consider is that the business is incredibly strong as you pointed out. Every quarter last year, we beat our guidance and then we raised guidance and then we beat it again. And so we're coming into this year with a lot of momentum. If you think about where we finished 2021, we finished the year with record revenue, and that revenue was more than $0.5 billion ahead of what we expected to bring in at the beginning of the year. So we closed with $0.5 billion more revenue than expectations at the beginning of the year. We also had shipments that were the highest in Illumina's history. We saw 40% consumables growth, sequencing consumables growth. And very importantly, we added 50% more customers last year than we did in either 2020 or even 2019 pre-pandemic. And so we're adding customers at a faster rate, significantly faster rate than we've ever had before. And so we're coming into this year with a huge amount of momentum. We're also starting this year with a strong backlog. So our instrument backlog coming into this year is almost double what we came into 2021 with. And we're also coming in with utilization at its highest point than it has been over the last 5 years across our platforms. And so we're walking into the year with not only adding a lot of customers, shipping a lot of instruments, but then the utilization of that instrument is at its highest level to where it's been in the last 5 years. Also, given the nature of our business, we have significant visibility into the revenue stream. So I think one of the underappreciated facts is that more than 80% of our total revenue is recurring in nature. The majority of that obviously comes from our consumables business. And then you add our services business, and now you're talking about 80% of our revenue being recurring. And so all strong momentum walking into this year. In addition, the external lead indicators for our business are strong. Research budgets tend to be strong walking into this year, and we've seen an increase there. We're seeing momentum in the consumables side of the business, from the clinical customers we have and the big expansion that we saw in reimbursement for oncology testing, for genetic disease testing is really playing out. And so we're seeing momentum there. And so walking with a strong momentum, good visibility into the revenue, lead indicators are strong as we walk into this year, both on the research side and the clinical side. Now if we look at where could there be more upside. There are a few places where there could be more upside. One is on COVID surveillance. Obviously, we have been proud to play our role in helping fight this pandemic and helping do surveillance for mutations in over 100 countries now around the world. Now as we looked at this year though, we took a $70 million to $90 million haircut in our guidance compared to last year. And so we're hopeful that we are sort of seeing the pandemic wind down. And so what we baked into the plan is that we will see $70 million to $90 million less than we did last year depending on a couple of things, depending on how COVID plays out, but also depending on how fast countries start ramping up surveillance beyond COVID, that could be a source of upside. In addition, we derisked the population sequencing programs by $60 million this year because the UK Biobank finished in the third quarter of last year. And so we took that down by $60 million for this year and didn't really assume that any one of a number of programs that could come in would come in and compensate for that. So that could be a source of upside. And then you touched on the third one, which is NovaSeq pull-through. We saw really strong NovaSeq pull-through last year, and we sort of had this interesting trifecta where we had huge backlog, a lot of shipments, record shipments, and yet pull-through per instrument was off the top end, right, over 1.3. So all 3 were red line in terms of positive numbers. Now coming into this year, what we've modeled, though, is less than that. We've modeled 1.2 to 1.3. Again, no reason to believe that it will be less than last year, but we thought it was prudent to model that range. Now that's still higher than the range we had before, which was 1.1 to 1.2. But if we continue to see what we saw last year, that represents, as you pointed out, a source of upside.

Great. Thanks, Francis. So let's maybe switch over to Chemistry X, if you don't mind. So not a lot has been disclosed. Obviously, there's been a couple of slides with some metrics that you provided in terms of the performance enhancements. But net-net, to the extent you're out there and doing your normal meetings with customers and your sales force, I'm just wondering, to the extent those customers that are aware of this, like how are they reacting? Just any feedback from customers yet or is it just way too early given there's just really not much out in the public domain?

No. There's already a huge excitement, as you can imagine, because the reality is Chemistry X is the most significant upgrade in terms of our chemistry that we've done since Solexa first came out with SBS chemistry a dozen-plus years ago. And so there's a huge amount of excitement. And the excitement is also driven by the data that they've seen so far. So I've been sharing data earlier this year, earlier last year as we've been making progress with Chemistry X. And so that's generating a lot of excitement around what could be possible, right? So what I have shared is the significant step forward that our customers will be able to see in terms of lower prices enabled by Chemistry X in terms of a much faster turnaround time associated with the chemistry. And that all adds up, right? So I've talked about the fact that because we'll have all new dyes, new linkers, new blocks, new polymerases, we're going to enable 5x increased density on the flow cell, 2x faster cycle times, 2x longer reads, 3x accuracy increases. And that's at the beginning of an S curve. And so customers are starting to imagine what's possible. I was talking to somebody yesterday. I know it's Sunday, but I think it was about what they could do. And they were imagining experiments that were 3 orders of magnitude bigger than the experiments that they could do today. And they were saying, look, we're really optimistic that the price points that you'll be able to drive down will enable those experiments. And so there's a huge amount of excitement building around what could be possible across a whole number of different segments that will benefit from vastly improved price points, turnaround times, accuracy. And so definitely a lot of excitement building.

Interesting. So you mentioned the price reductions. And kind of looking back at like prior cycles when you've gone from, whether you went from the HiSeq to the X and then the 3000 and the 4000 and then from that product portfolio to the NovaSeq. The price decline looked to be somewhere in that 30% to 70% depending on the product or the flow cell. I'm just wondering, is that kind of the ZIP code of how we should be thinking about what Chemistry X could usher in somewhere? Again, I'm sure there'll be a range of outcomes depending on the customer. But is that like an expected reasonable range to be thinking about?

I think you should be much more ambitious than that. And here's why, because you're comparing the chemistry to a specific product. And so the right comparator, in my opinion, is think about with the last generation of chemistry enablement. When we first launched our sequencer in 2007, the price for a genome was $150,000. Today, depending on your size, you can either be at $600 a genome or less, right? So think about what that last chemistry enabled, $150,000 a genome to $600 or less a genome. So what we're saying is that this is the most significant upgrade we've done since that generation of chemistry. And so that will enable a whole suite of products. And initially, you may see something of the range that you talked about, but the ambition is very broad to say, look, we're going to try and enable customers that want to do experiments that are order of magnitude bigger. To do that, we know you'd want to take prices down over a generation much more significantly than you talked about.

Right. Right. Yes. I was certainly thinking first iteration back half of this year for 2023, but certainly, I would assume there's a lot more headroom. I mean, your math, if we got to the same level, would be $2.40 a genome at the end of this iteration of this chemistry. So demand elasticity, obviously, it's such a critical factor in your business. The topic came up on the fourth quarter call. Just kind of can you give us a sense of the latest thinking from your research and clinical customers, maybe starting with research. Price drops 50%. Like how do customers react? It's such a critical component to the Illumina business model.

Yes. Huge amount of excitement because it enables research in a number of areas. And I'll just give you some examples, right? If you think about the whole area of single cell research, for example, our customers are telling us that the reality is they need to do single cell experiments that are much, much bigger, like 1, 2, 3 orders of magnitude bigger than they're doing today, both to understand the difference in expression profiles across different types of cells, but also to understand DNA heterogeneity across, for example, a tumor. And so to do that, they're saying, look, we need to do sequencing that's 3 orders of magnitude more than we can do today. And so they are starting to say, look, what gets enabled and they're plotting a road map to say, okay, initially, we'll take this step and that step. So single cell spatial drug discovery, especially for complex diseases like neurological conditions, like Alzheimer's and Parkinson's, where because those are complex diseases to really uncover the genomic drivers of those diseases, you need to do population cohorts that are in the hundreds of thousands, maybe millions of people. And so those cohorts now get enabled from a research perspective. It opens up potential in cardiovascular disease as well. And then the whole area of high-throughput proteomics, for example, is another area where because we could be the readout for it, depending on the platform you pick, we enable, again, much, much bigger experiments, which are necessary to really uncover the biology, and that gets enabled as you start to drive prices down.

Interesting. On clinical customers, on the fourth quarter call, you suggested that they may not be offered the lowest cost at least initially. And demand elasticity is interesting given the sense of if a price is fixed today for a certain application, even if the price is reduced for the input, they may not decide to do more samples. So I'm just wondering how we think about the impact of this next product cycle and this whole Chemistry X on clinical customers.

Yes. So clinical customers today have a choice and a lot of them choose to buy an end-to-end solution, a complete workflow. And so they're saying, look, the sequencer is an important part of it, but they want to buy an end-to-end workflow that includes the library prep, sample extraction, automation, all the way down to reporting. And they want to buy a product that's cleared. So for example, we have the VeriSeq NIPT product that's a CE-IVD mark solution in Europe. That is an end-to-end solution. It's cleared and you pay for it on a per sample basis. And so because it's an end-to-end solution, the price point is more driven by sample rather than the price of sequencing. Now you'll see more of that from us to say, whether it's NIPT, as I just talked about, or in areas like cancer therapy selection for cancer patients, they'll want to buy an end-to-end solution too, and that will be priced per report. Same with genetic disease testing, again, we're looking to provide an end-to-end solution, and that will be priced by report. And so what you'll see increasingly is that in the clinical segments, the price point will be driven by the clinical utility and the economic value provided by the test, and that will be used to obtain reimbursement levels to make that test viable for labs, but it won't be connected to necessarily the price of sequencing. And so we'll price those end-to-end solutions on a per sample or per report basis, and that price will be determined by the value that is generated. And that price can stay fairly consistent once the model works for a lab even as the price of sequencing changes.

Interesting. Okay. So in terms of other markets potentially being enabled, screening of babies, obviously, population sequencing kind of related to that. Just maybe speak to a little bit about, the question was asked on the fourth quarter call about these application-specific approach of maybe Chemistry X, and you've already alluded to some of the end markets. Maybe could you tie it together as we look ahead and obviously we'll find out a heck of a lot more in September, October. But in terms of this idea of specifically tailored towards certain key areas that it does perfectly enable or as well as some of these other big categories, do you think Chemistry X could really open up some of those markets?

Yes. I think Chemistry X is going to be fundamentally enabling a number of very large markets. And because it will enable sort of the next generation of our portfolio that delivers the lower price points, you'll see a large amount of research that is now able to be done in areas like one big area, for example, I think that we're seeing a lot of promising work and could have clinical implications even nearer than people thought, is cardiovascular disease, for example. Now we talk a lot about oncology today because genomics can make a big difference in the lives of cancer patients. But cardiovascular disease is actually responsible for more deaths each year than cancer. And so that's another big area where you're seeing people realize the impact that genomics could have. And so you will see cardiovascular disease emerge as another area that could be the next oncology for us. Where, first, you'll see and you're seeing today sort of large research projects associated around the genetic drivers of cardiovascular incidents and not only risk stratification, but stratification around the different types of cardiovascular incidents that show up. You're seeing research around the emergence of therapies that are keyed off genomic biomarkers just like we saw in cancer. And so that's an area that I think will get fundamentally enabled as we enable big research projects. But also other areas I talked about are neurological conditions, which are also really big areas of unmet need where you'll see big research projects, but then you'll also see potentially diagnostic tools. You will see tools that are used around therapy selection associated with those patients. Each of those areas that I just talked about is very large. Some of them could be larger than the cancer markets we serve today. And then there are still big areas of research beyond those that are enabled by the new modalities, spatial single cell, both on the RNA side, which is happening today, but on the D&A side, which is really not yet happening and needs more powerful tools. High throughput proteomics is another area, too, that you will see more work on it.

Great. How about like, I think you've talked about the first application or the first tool, if you will, would be the high throughput part of the market. But I'm just wondering, particularly as we maybe move into some competitive questions later on in this discussion. How about the mid throughput part of the market? The NextSeq new rollouts haven't been that long ago and those are still doing fabulously well in terms of placements. Just how do we think about the ability to put Chemistry X on the desktop? And is it something more like '24, '25 or we'll have to wait and see what comes out this fall?

Yes. We haven't announced specific time frames, obviously, for when Chemistry X is coming out where. But what we have said is absolutely any new platform, you should assume it's going to be based on Chemistry X going forward. We're at that stage now. And then we have said that Chemistry X can be used in some parts of our existing portfolio. And so we still have that in front of us. It will generate the benefits I talked about in terms of enabling lower price points, faster turnaround times, higher accuracy. And so as you pointed out, we're still very early in the NextSeq 1000, 2000 rollout cycle. We're really excited about the reaction from customers. It's really emerging as a clinical workhorse and opening new parts of that mid-throughput segment. And at the right time, we'll consider whether it makes sense to introduce Chemistry X on it. But that remains a lever that's available to us.

Okay. So maybe switching over to Infinity for a sec. The company is obviously very excited about this opportunity as well. And I think there's less transparency or understanding yet of this, of like really what it is. Is it phasing? Is it not? Is it single-payer discrimination? So we'll have to wait and see how it comes out. Maybe could you just give us a little bit of color there? I know the company has talked about it's kind of an, I think you used the word like it's a new approach. It's a new type of sequencing. So net-net, maybe what have you shared with customers to date? Early feedback we heard has been pretty constructive. And is this actually going to be competitive with the existing long-read technologies?

Yes. So I think the feedback, and I've shared a little bit of it earlier in the year has been very positive from customers that we have done an early preview with. I was trying to think about it because I get asked a lot about compared to sort of previous approaches to long read and short read. And I guess the only metaphor I could think of is to say, look, think about when Steve Jobs introduced the iPhone, right? What he said is, look, the previous world, you had a phone and then you had a music player and then you had an Internet browser communications device. And they were 3 just different markets. And what he did in that launching was say, those aren't 3 different markets. What's happened now is that the performance of the phone, the technical breakthroughs in that phone have now allowed a single device to do all of those things. And everything changed after that. There was no market for separate music players and a separate Internet communications device. And I think we're on the precipice of something like that happening here, where historically, because there were technology limitations in what was in the market, you had one platform that was really, really good for high accuracy, low cost, short reads, and we're on those and the leading one of those. And then there were the long-read players. And that there was no technology that were able to really do both, deliver very highly accurate, both short and long reads, at low price points with low DNA input. And so we were forced from a technology limitation perspective to define the market as 2 separate things. That's never what customers wanted, though. If you talk to them, what they said is they want a single platform that has high accuracy, low DNA input requirements, the cost profile that you can get from short reads, and that's what we're able to deliver, right, with Infinity. We're saying with Infinity, you'll be able to take your existing platform. So the 20,000 instruments we have out in the field, you'll be able to take your existing instruments and be able to scale up and down. The super short read, all the way up to 10,000. And that's actually what the market wants. And I think we'll move away from defining a market based on the technology limitation because we now have the technology breakthrough. They've happened now in the last couple of years, where increasingly customers will just expect a single mobile device that is a phone, a music player and a communications device all in one. And I think you'll see that here where they won't tolerate now multiple platforms for different parts of this.

Interesting analogy. So one of the long-read competitors talked about, maybe some of the inability to call tandem repeats, methylation, dark regions of the genome. Like I mean, we'll have to find out the technical capabilities of Infinity. But like to those particular areas, can you say whether or not this technology will be able to make accurate calls there?

You'll see more and more data come out, and so we'll let the data speak for itself. I think you saw some of the data in January maybe, very powerful data. So we'll let the data speak for itself. But I feel really confident that our approach will be able to address the regions of the genome that historically we've had technology limitations to have a single platform that goes up and down.

And would you think, like if we think about the market opportunity, assuming like the product obviously comes out as you expect. Is this something that 15% of your customer base may want, 80%? I mean, we'll find out more, but without putting the cart before the horse, just maybe some introductory thought process about the applicability of this technology.

Yes. I think you'll see sort of 2 vectors of growth associated with Infinity. One is companies that or labs today that were considering buying a long-read platform. And so now they'll say, look, if I already have Illumina or if I wasn't, but I can look at Illumina, it gives me both. And I think we will show up very well in those situations. So if you were considering buying it, and you're going to spend multiple hundreds of thousands of dollars on a dedicated machine for that segment of the market, I think there'll be a lot of excitement around saying, I don't have to do that. I can use the Illumina machine for that and save those hundreds of thousands of dollars. So I think we'll definitely be a very viable, very strong alternative for customers today that are looking at spending all that money on a long-read dedicated platform. But you know what I didn't expect, and we've been hearing now on incoming customer support calls is the many smaller labs that frankly weren't considering long reads at all. And they were saying, we just can't afford to spend hundreds of thousands of dollars on another machine. We have a MiSeq or we have an NextSeq or MiniSeq. But this opens up that segment to them. And they're saying wait, oh, this is just a library prep on top of our existing instrument. Oh, well, that could be an interesting add. And so I think that's the other segment that it opens up. That frankly, I was a little bit surprised on the upside to hear more about from, again, and mostly it's coming from inbound customer support calls from existing customers. So I think it opens up that segment as well. People who, frankly, were priced out of the long-read market today, but now it's been made much more accessible with just being enabled on their existing sequencers.

Great. Maybe flipping over to competition. Illumina's had a formidable position and maintains that today, really remarkably over the last 12, 13 years. But we're at the onset of what appears to be certainly some new competition coming into the market with Singular, BGI, let's see, and with Element and Ultima, who we don't know much about yet. We're going to find a lot more from Element this upcoming Monday. So maybe, particularly maybe as we think about the desktop part of the market, again, we'll wait and see what specs look like, but Singular is talking about now maybe high single-digit dollar pricing per gigabase, which I think the NovaSeq P3 might be somewhere $15, something like that. So how do we think about the desktop part of the market? And you've done remarkably well there and you're still early in the, I guess, the launch, but you are going to have some competitors coming out maybe with some interesting, whether workflows, accuracy and maybe really potentially very competitive pricing.

The only thing, Dan, is we've had waves of competition for our entire existence. In fact, when we first came in, we were the underdog in the sequencing market. There were some big players that had strong dominance in the market. And so we fully expect to be competing vigorously for our customers' business. And now the way we win is provide the best value proposition in terms of price, performance, quality, end-to-end offerings for our customers as well as a commitment that we're going to continuously innovate. And we're going to continuously set a new bar in the market. We spend, as you know, north of $600 million a year in terms of R&D, and most of that is focused on driving SBS forward. And so whether it's in the mid-throughput part of the market, we provide today the most powerful, best price point, highest quality product in the market. In addition, because we have DRAGEN technology built into it, we provide the most comprehensive end-to-end solution from a sequencing perspective on the market, and we provide the ability to seamlessly move up or down the portfolio. If you start in the mid-throughput part of the market, very often as those customers grow, they then move to our high-throughput instrument. And so what we offer is the ability to say, you can seamlessly move as your business gets bigger, rather than having to replatform your entire lab. And so our approach is going to continue to be to say, look, deliver the best promise to the market in terms of price, performance, quality, end-to-end solution, a commitment to an ongoing road map like Chemistry X, like Infinity and the flexibility of having access to our entire portfolio.

Interesting. Okay. Yes. So I guess we will certainly hear more about. I'm sure there'll be more information in the marketplace maybe coming next week as well on that front. Maybe switching over to GRAIL. We have a few minutes left here. So the GRAIL guidance was ahead of expectations. I believe most of the $70 million to $90 million was attributed to GRAIL and it's from health system customers. You talked about smaller contribution from concierge medicine and self employers. Maybe can you just walk through a case study of how, whether self-insured employers or some of these health systems are beginning to adopt Galleri and as a vantage point for what the opportunity could be?

Yes. It's really exciting to see what's happening with Galleri since they launched the blood test in June of last year in the U.S. and what's happening with the NHS since the NHS started rolling it out for its pilot in October to 140,000 volunteers. They are seeing good traction in each of the segments that they talked about. So in the employer segment, I talked about it at the conference earlier this year the traction they were seeing. What's exciting to see is, look, there are companies that are on this call and we expect it to be some of the first to jump on, financial services companies that tend to be more forward facing in terms of benefits to their employees. Obviously, tech companies also, we expected to be on, and they are on as you'd expect. They also tend to be more forward-facing. But what was interesting was to watch companies from industries that aren't necessarily known to be forward facing like logistics, for example, and transport and big companies there that are now rolling it out. I can give you my experience. We rolled it out at Illumina. And this is one of the few times as a CEO where you can roll out a benefit that is beloved by employees because it truly can save somebody's life. But at the same time, it actually lowers the cost. We're self-insured. And I know not only the tragic human cost of cancer among our employees, but it's also very expensive from a company perspective. And I also know that catching a cancer early means it's sometimes easier and cheaper to treat and can result in a better outcome. So it saves us money as well. And we even heard from one of our employees that because of the GRAIL test, he found an early-stage breast cancer. And now men typically don't screen for breast cancer. And he was reflecting on what kind of tragic outcome that could have been for him. And so it's not surprising to me that we're seeing, they're seeing the traction that they're seeing in the GRAIL team from employers who recognize that it's a beloved benefit in the war for talent, but it also saves you money. From a payer perspective, some of the surprises we expected to see traction among the concierge medicine providers. What we were surprised to see was them signing up the first Medicare Advantage payer or what they were able to do with Point32, which is a combined organization from Harvard Pilgrim Health and Tufts Health Plan, to be one of the first commercial health plans in the U.S. to work with GRAIL. So I think that's a little bit earlier than maybe I had expected to see traction there. And so we're very optimistic and excited about the traction that they're seeing.

Great. Well, I think with that, Francis, we're out of time. So really appreciate you being with us here today, everyone on the video. Great discussion, and I hope you have a great rest of the day.

Thanks, Dan.

Okay. Bye-bye.