Illumina, Inc. (NASDAQ:ILMN) Evercore ISI HealthCONx Conference November 29, 2022 11:20 AM ET
Company Participants
Dr. Alex Aravanis – CTO
Salli Schwartz – VP, IR
Conference Call Participants
Vijay Kumar – Evercore ISI
Vijay Kumar
Okay. Thanks, everyone, for joining us this morning. A pleasure to have with us Illumina. We have Dr. Alex Aravanis, their CTO; and from Investor Relations, we have Salli Schwartz. I think before we get started, Salli had some disclosures. So, Salli?
Salli Schwartz
Thanks, Vijay. For your review, we may make forward-looking statements during today’s call. It’s our intent that all forward-looking statements regarding our financial results and commercial activity made during today’s call will be protected under the Private Securities Litigation Reform Act of 1995. Forward-looking statements are subject to risks and uncertainties. Actual events or results may differ materially from those projected or discussed. And to better understand the risks and uncertainties that could cause actual results to differ we refer you to the documents that are Illumina files with the Securities and Exchange Commission, including Illumina’s most recent forms 10-Q and 10-K.
Also, and as a reminder, pending the outcome of the European Commission’s investigation into a 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 an independent 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 interest of GRAIL. Commercial interactions between the two companies must be undertaken at arm’s length.
With that, I’ll turn it back to you, Vijay.
Question-and-Answer Session
Q – Vijay Kumar
Great, fantastic. Alex, Salli, thank you for taking the time this morning. I was just joking. I can’t ask any more questions about what is Q4 looking. Alex, when I have a CTO, I need to ask more questions. So, I had to think up for my fireside discussion.
For us, the Analyst Day, you had a lot of big updates, Alex, NovaSeq X, X Plus. I mean, they’re — for us, it feels like NovaSeq went from 1000, 2000, 6000 to now we have an X, but there’s a lot of innovation in that box. Cost came down, throughput increased, you doubled the speed of sequencing. I think from your perspective, right, when you think about those improvements, I think cluster density up at 3x, caught my attention. Like, that seems like a big number, Alex, like, how were you able to increase? Was this like a manufacturing breakthrough or material science breakthrough or what did Illumina have to do to increase cluster density?
Dr. Alex Aravanis
It’s a great question, Vijay, and years in the making, and a lot of hard work by our great scientists and engineers. Just to emphasize the point, right, I mean, when you saw Francis hold up the flow cell for NovaSeq X and X Plus, it’s actually smaller, it’s a smaller flow cell than the NovaSeq 6000, but has 3 times the output, can be sequenced twice as fast. So, I think that really visually highlights the innovation. And what you’re referring to is the miniaturization of the technology, so that we get more data and a customer gets more data in the same surface area. And obviously, there’s significant cost reductions, which allowed for the reduction in price, which we’ve been able to pass on to now our customers, much to their delight, and to over time, maintain the same kind of healthy margins. So, it’s really — our model is innovate, and then bring those innovations to differentiated products, without diminishing our business at all.
Some of the key innovations that allow that, first and foremost, we needed a new chemistry, like in the semiconductor industry, right, when you miniaturize your transistors or miniaturize your clusters, the physics, the chemistry changes. So, we now need more signal out of a smaller cluster with fewer molecules, we need to effectively blast it with higher intensity lasers, very quickly also to get the fast runtimes. And what we found in the early work is that as great as our previous version of the original SBS was, it just wasn’t possible, the actual chemistry and physics couldn’t hold up to that. And so, we kind of did the proof experiment that you can’t achieve this kind of output with the old chemistry that now has been copied by others in the industry.
So I mean, we really needed something that had new properties, which is no easy feat, right, which is why the older chemistry has been copied, because it’s very hard to come up with something better. But our team, which includes many of the original inventors of SBS, really looked at the fundamental physics and chemistry of bonds, they generated thousands of new molecular structures, how to evaluate them. Again, we’ve been doing it longer than anyone else. We have, I think, the best teams in the industry. And they found these breakthroughs in this new type of font.
And then with that new chemistry, it has the new properties that allow it to be scanned faster to get more signals out of it. And so that opened up the potential then to get to denser clusters. But that wasn’t enough, just having the chemistry that could do that. We now needed the material science and the semiconductor processing to be able to make the tinier clusters. So, we used ordered arrays, I mean with wells, right, and that allows us to pack more in, but we needed to make the actual density of the wells 300 times — 300% higher.
And so, when we looked at, again, existing processes for making such small structures, they didn’t exist. And they didn’t exist with the kind of fidelity, scale and cost reduction. So, the team invented a nanoimprint lithography technology, very inexpensive, but way to print these tiny little wells at a higher density, significant effort in how to do that. So you get tens of billions of wells perfectly. We can integrate it into large semiconductor manufacturing processes. And then that began the journey of actually scaling it up, right? And so, over 20 different new semiconductor manufacturing processes that we developed. And then we made the big decision to do that in-house, and to bring that equipment in-house so that we have control over our cost, control over our scale and supply, which we all know has been — can be challenging, over the last few years and potentially into the future, right?
So, I hope that gives you some insights into how we got there, many innovations, many patents between the chemistry and the nanowells.
Vijay Kumar
Absolutely. No. That that was helpful, Alex. And a couple of points, you touched there. Are we — when you look at this chemistry and the advances you made, are we at — have we reached a limit on cluster density? Could this — does this have more room to go down the road? And you did bring up pulling manufacturing in-house, does it raise any risk? Is there any manufacturing risk as you scale up production?
Dr. Alex Aravanis
Yes. Both great questions. So, we have a lot of headroom on the new chemistry. And that was also one of the reasons we made the decision to invest in it, invest in the new ability to manufacture it and to do all of the other inventions that I mentioned, in addition to the chemistry, to be able to make NovaSeq X, because as extraordinary as the product is, again, it gives us significant headroom. So, in the same way that we have been working on NovaSeq X for five years, we are pretty excited about the future, and already working on its replacement in a couple of generations. And so, what’s possible is really breathtaking.
I mean, my perspective is that the headroom on our technology is actually higher than it’s ever been. So, as excited as we were five years ago, this new chemistry opens up just a lot more of headroom and runway on where we can go. Again, we were reaching limits of the previous chemistry. So, if we had stuck with that, we had not made those breakthroughs, we’d be telling you something different. But now, we have kind of as much runway as we could really ever need. So, it’s quite extraordinary that the densities we could get to and the cost implications of that over time.
Vijay Kumar
And sorry, on the manufacturing side, Alex, is there any risk as we move it in-house?
Dr. Alex Aravanis
Yes. So I would say, initially there was some risk, given the scale of the manufacturing, the number of new processes that we had to bring in-house, the facilities, so. But we are beyond that. And now we feel that overall, we’ve reduced the risk a lot, because now it is under our control. And so, everything from the nucleotides to the actual semiconductor wafer processing, we all do that in-house, right? And we’ve never done all of those components before. So again, the goal here was to reduce risk around manufacturing, I think we’ve done that, and to control our costs, right? So, we don’t want to be in a situation where there is a vendor who raises costs and it’s a key component and we have very little control over that. And so, we have eliminated a lot of those risks in doing this.
Vijay Kumar
And sorry, this may be more on the financial side, which Salli feel free to chime in. Has the investments on the manufacturing — setting up this new plan, has that been done or should we expect capital expenditures to step up from current levels?
Salli Schwartz
I mean, I know we haven’t given guidance for next year. But as Alex was alluding to earlier, I mean, this isn’t sort of a stopping point, necessarily, we have the next round of innovations that we’re going to continue to invest in. So I think, you should continue to see us, over time, continue to put money into both our R&D expense budget to fuel the future innovations and continue to drive this forward as we roll everything out and get the manufacturing up to up to full capacity over the course of 2023, and then same thing on the capital side. So I wouldn’t necessarily say it would be a step up, but it’s going to continuously increase over time as we continue to grow and innovate and evolve.
Vijay Kumar
Got you. And then, I’m sorry, Alex, back to you on — I think some of the words you used to characterize is what’s possible is breathtaking. So, what work in this technology? Is this is XLEAP-SBS is the one that gets us to $100?
Dr. Alex Aravanis
Yes. For sure. So, again, we’re just at the start of the kind of technology S-curve, right. So, we’re very early on. So, a lot of additional opportunities for improvements in density, speed, throughput. If you look at like, for example, what we did on NovaSeq 6000, right, even pretty late in its product cycle. We had the v1.5. So, pretty significant improvement in costs and throughput. So, all of those types of things are very possible on the existing platform. Because, again, the $200 genome with the higher output flow cell, that will be available second half of next year. And this is going to be a multiyear product cycle, like what we’ve had with other — with previous platforms.
And that’s just on the existing platform. And then again, as I mentioned, with future platforms, using SBS, we think that there’s — again, there’s a ballpark number, order of magnitude, opportunity for improvement. We’ll require additional investment. But no miracles required given the capabilities with the new chemistry and the infrastructure we built around it.
Vijay Kumar
Absolutely. And sorry, the highest density chip on the X Plus system. That’s the one with 25 billion clusters. Is that sort of end of next year or did the timing put forward, Alex?
Dr. Alex Aravanis
Yes. So, we’ve said second half. I don’t know if we’ve given more specific guidance on the release of that chip, Salli? Yes. So, we’ll be providing more updates early next year, on the release of that. Obviously, we’d like to get it in customers’ hands sooner than later.
Vijay Kumar
Understood. And I know, when we look at innovation, from a Wall Street perspective, we go down to cost per genome. But from a science perspective, I think people care about read length and phasing issues and et cetera. Maybe talk about other areas where this XLEAP-SBS chemistry and new chemistry — it’s been a game changer and what’s been shared by customers?
Dr. Alex Aravanis
Yes. So, a lot to talk about there, Vijay. So, one thing that we knew that there was a lot of interest in and be appreciated, but we’ve been taken aback by the excitement and response is the ambient ship. So, this will be the first SBS that you can ship just through standard UPS, right? It doesn’t require cold chain logistics, doesn’t require disruption of activities at a customer site, doesn’t require the initial freezer storage to receive the reagents and also has a, like a 90% reduction in overall waste and packaging, which for many customers, it’s a big deal for them and their organizations when they have ESG goals. So, again, we knew that that would be tremendously appreciated, a real differentiator in our product, a big first in our industry, but taken aback by the number of customers who have cited that, as — I would say as maybe the second most compelling feature beyond the cost reduction, for many of them.
There’s additional automation in the system. So, on the 6000 to do multiple batches of different sample types, we had a free workflow that we call the Xp workflow, which worked fine. But that added extra steps, opportunity for samples to get stub, errors, additional cost in the lab for our customers. And so, that’s been eliminated and all of that is done on instrument. So, you just put the batches of samples you want and then the fluidics [ph] automates the distribution of those.
So, we also have faster clustering. So we have clustering on board and it’s also a faster clustering than the 6,000, which is part of why we’ve gotten the overall run time down to less than 10 for the — less than 24 hours for the 10 billion read chips, so twice as fast as the 6000.
And then also the onboard DRAGEN. Again, I’m excited about customers out say, again, it’s kind of top 2, 3 exciting features from — we’ve had some very large core — genome core centers tell us that they are excited to redeploy scientists and engineers managing their bioinformatics pipelines to doing more science for them, things that are unique to their institution. And this will allow them to do that in a way that they haven’t been able to do before and couldn’t do on other systems, right? So, they will save money on their processing. They’ll save staff, save compute. And so, we’re, again, really excited about that possibility.
And as I talked at the customer event, that was no small feat, right, because this system has 3 times the output, so the highest for throughput ever at this type of speed. And we simultaneously wanted to make sure that all of that compute was done on board. So, there really is a small supercomputer on board that’s included in the cost, right? So another piece of kit quite expensive that customers won’t need to buy in addition to this that they would potentially have to buy on a system that didn’t have all of the high output compute on it.
Vijay Kumar
I do want to dive into this informatics side, Alex. So, maybe I’m thinking about it the wrong — I’m sure I’m thinking about it the wrong way. So, if I had the supercomputer onboard, it can do an analysis versus previously I was just shipping it to another computer. So, it feels like there was an additional step involved that’s been eliminated but how is this allowing bioinformaticians to now do more science and spend less time on the bioinformatics side? Like, I wasn’t sure just having a DRAGEN onboard, like what’s the big deal?
Dr. Alex Aravanis
Great. Glad to have the opportunity to go into more detail on that. So, one is the time, right? So, if you have your own kind of bespoke system set up, I would say, in general, it takes almost 10 times as much time to process. So, that’s taking longer. Having it on board and integrated into the sequencer, we can make it much, much faster. You don’t have to transfer all this data around. That’s also a cost and time consuming. The other is that the DRAGEN has accelerated hardware, right? So if you use the onboard system, it’s actually just the way the compute is done. It’s lower cost and faster.
So, if you did the exact same thing without the accelerated hardware, it’s inefficient, right? It’s kind of like GPUs right, or a faster way to do graphics than conventional microprocessors, right? So, we have highly specialized DRAGEN and hardware. So it’s literally — compute is more efficient; by about an order of magnitude, it’s faster. And then the other is that it’s automated, right? So for the common workflows, the most popular ones, it’s already configured, right? So, you don’t need engineers to kind of reinvent the wheel for you, get third-party software, configure it, debug it, validate it, babysit it so that if there’s a bug, we take care of all of that.
So, those are very real savings to the customers, and we think will affect how they think about which systems they want to buy in the future when they have that kind of automation, cost saving, speed versus having to bear all the burden of all of that.
Vijay Kumar
That’s helpful. And I think switching back to the chemistry side, I think — maybe the numbers I have in my head is — I think about 85% of your bases in your prior — in the old SBS chemistry, right, 85% of your bases were at Q30 or about. What is that number here for XLEAP-SBS, the new chemistry? And where could that number go?
Dr. Alex Aravanis
Yes. So we — so the overall quality scores are going to be superior. We haven’t released the exact numbers yet. We’ll be doing that early next year for the final implementation of the product at launch. We’ve shared initial data with customers, and they’re quite pleased with it.
If you don’t mind, I want to digress a little on quality scores because we’ve come to appreciate that it’s actually not that — it’s no longer kind of a relevant metric that actually customers care about, right? So, what customers care about and what we really emphasize is coverage of the genome. So, they want to get to the more complex genes, which is, again, no longer a function of quality score when you’re at such high scores, it requires things like the DRAGEN graft genome as an example, or accelerated hardware. And so what we’ve been showing more and more data is that with DRAGEN 4.0, you can now access ever more parts of the genome. So customer is very excited about that kind of capability.
The other is that I honestly haven’t had a customer tell me that Q score is limiting any of their performance. So again, we’re emphasizing what customers are asking for, not kind of arbitrary technical metrics. So, they want to get more output, they want to get more speed, they want to get more difficult regions of the genome. If we wanted to optimize quality scores even further, we could, but that’s not what we’re hearing that customers are interested in.
So, it feels a little bit like a metric that maybe for a company you don’t have anything else to talk about, you’re kind of making a lot of hay of it. But again, just — it’s not the feedback we’re getting.
The other is that when we’ve looked at just some quality claims out there, we analyze it in our hands, we don’t see the data holding up to some of these claims. So, we’re not — again, maybe there is some truth to this, but we’re just not seeing that. I mean, if you look at raw error rates out there, some of these companies, they’re not spectacular. So, that seems inconsistent with their quality scores or as you know, with quality scores, you can kind of cherry pick the regions of the genome to analyze and you can still have an overall error rate, which is not that impressive, but kind of have the quality score turn out as a higher number.
So, again, we thought that this was the key driver. We could emphasize it more, but it’s not what we’re hearing from customers. It’s coverage of the genome, it’s cost, it’s speed, no one is telling us that getting to an arbitrarily higher quality score is that important?
Vijay Kumar
That is helpful perspective for sure. I think related to that, Alex, back at the Analyst Day, you spoke about how you sort of — you have an interesting role, right? It’s just not the R&D portion, but you’re up meeting customers, you talk to them. A lot of their input went into designing the Nova X, X Plus.
I guess, where I’m going with this question is the number one question for me inbound has been, look, pricing is down 60%, I’m afraid there isn’t enough volumes to meet the demand rate. So when you when you were designing the X, X Plus and what the system should look like, right, when you were speaking with customers, clearly, you have the confidence that there would be volumes and that this was net positive, not just for your customers but for Illumina also. Walk us through that process, right? What kind of conversations you had and why are you confident that this volume increase, the historical price elasticity of demand holds true in this new paradigm?
Dr. Alex Aravanis
Yes. It’s a great question, obviously, really important to our business model. So our conversations with customers are that there’s kind of at least three vectors that point to them benefiting from having lower cost, right, in a way that generates more volume for us, right, not just — obviously, they would love to have price reductions and they’re going to get a big one. But one is more samples. So it certainly enables them to do more samples with the cost reduction. So they’re excited about that. It enables them to propose bigger projects, for example, like our future health in the UK, there’s an initial genotyping phase, and then there’s going to be a sequencing phase.
And so, you can imagine — you can think about doing whole genome versus exome or moving in that direction in a way that may have been prohibitive, right? So, that would be a win for that program in terms of the amount of data they could generate, still, overall, probably more money spended but their dollar going a lot farther. So I’d say kind of volume.
The move from — over time in different areas, whether or not it’s oncology or genetic disease testing, right, we’ve seen the industry go from PCR to small sequencing panels, to large multi-hundred gene panels to exomes and to whole genomes. And different customers are in different geographies are in different part of that journey. But what customers have been telling us is wherever they are, this will be an impetus for them to move to that next stage, whether or not it’s small panels to exomes or exomes to genomes.
And so, many customers have been asking about this for years and said, look, if the price just got to X or got below this number, our model says that we should move to a more intensive sequencing format. We get more data. We could offer more differentiated products to our customers. Our science would be better. And so, for a lot of customers, this is going to cross that threshold for them. So, that’s going to be a key.
Along with that, it’s just the more intensive sequencing applications, right? Whether or not it’s going to millions of single cells or doing liquid biopsy, which is, as you know, with liquid biopsy, you have to sequence us much deeper than you do with — when you’re sequencing a tumor from tissue versus blood.
And so historically, the liquid biopsy panels have been smaller than the tissue panels because of the sequencing cost, right? And so NovaSeq X, NovaSeq X Plus allows oncology testing companies think about, gosh, rather than have a 100-gene panel as our standard liquid biopsy test, why don’t we do a 500-gene panel, right, like our tissue panel. Rather than having it be a specialized product, maybe just for pharma, why couldn’t that be our kind of routine clinical test, right? And if one does it, then maybe that becomes then the standard.
And so, you’ve seen this play out over time previously, again, since most of the world is still under sequencing relative to the scientific or clinical opportunity, we think this will move them — we’ll give them the rationale. That will take time, right? And initially, NovaSeq and NovaSeq X Plus and those that enjoy those benefits will be a very small part of our customer base. But that will grow over time. And the bet here is that as out of walls, the applications will move in a way that will be beneficial to us and our customers.
Vijay Kumar
Understood, understood. And I think when I sum all of this together, right, a lot of exciting improvements on the Nova X, X Plus. Clearly, I think from a competitive standpoint, this takes you to another level. I think on your third quarter call, you said Illumina had 50 Nova X orders and maybe another 170 in advanced pipeline. Any update on those numbers and where we are, either from a pipeline perspective or order perspective?
Dr. Alex Aravanis
Yes. I don’t think we have any kind of hard numbers to update yet, but I can just say that qualitatively, the interest continues to be high, and again, even higher than we thought, right, as I think was noted on the earnings call. I don’t know if you have anything to add, Salli.
Salli Schwartz
No. And I checked in with folks yesterday on my team, and even though — the broader team, and even though it’s been certainly a holiday period in the United States, around the rest of the world, there’s — things have continued going ahead, and there’s been strong interest coming into the end of the year.
Vijay Kumar
Got you. And just maybe on the topic, both of you Alex and Salli. Anything on the — from a macro perspective, I think some of your peers have noted a slowdown in Europe or perhaps lab activity or vacations, et cetera. What is Illumina seeing?
Salli Schwartz
So, I think we had noted in our last earnings call that Europe, in particular, and from a couple of different perspectives, one was just the issues with purchasing power parity. So, while budgets — and this is specific to research, the research side and research labs, in particular, but while budgets were holding up, those dollars were just — euros or pounds were just not going as far as they had been historically. And just for some who might be purchasing in U.S. dollars, obviously, it just does — is not carrying as far either from an FX perspective. So, that’s something we definitely saw through the third quarter and impacted our guidance that we revised for the remainder of the year.
And then, we also noted — and this is true for Europe, but also, to some extent, in North America that we had seen some research projects again and in particular, some POPSEQ programs that were having issues collecting samples in this macro environment with people just not wanting to sign up for clinical trials because they were worried about losing their jobs, didn’t want to take time off work or just looking to minimize even gas spend to go back and forth to a clinic to participate in the trial. So, just slower sample collection that we have been seeing earlier in the year, and that seems very macro related as well. And that impacted the remainder of the year’s guidance.
Vijay Kumar
Got you. And sort of — I know the guide was up updated, Salli, but would it be fair to assume that so far trends that we have seen, it’s been largely described when the guidance was updated.
Salli Schwartz
More or less, I mean, it’s only been a few weeks. So, it’s — we certainly didn’t assume that the world would get dramatically better and hasn’t gotten dramatically worse. So, it’s been trending about as expected on the whole.
Vijay Kumar
Got you. And maybe, Alex, one for you. One of the questions I’ve gotten is, look, customers know that there is a new system coming on. So, there is no reason to be buying any NovaSeq 6000 systems anymore. So, this fourth quarter here from a near-term perspective — or maybe not even fourth quarter, from a near-term perspective, whether it’s Q4 or early part of next year, maybe there are zero 6,000 systems being sold, NovaSeq 6000 systems. So if you can shed some color context on who’s buying NovaSeq 6000s, and why we’re not seeing a dramatic slowdown in that part of the business?
Dr. Alex Aravanis
Yes. Great question. So, we are definitely still selling 6000s and there are some good reasons why. I can share those. So for many projects, customers want to have consistency. So again, the data quality on the X, X Plus will be even better. But if you’re two-thirds of the way through a project and you need to do your next 50,000, 100,000 samples and you are planning to complete it this year or early next year, well, you’re not going to hold up your business or your project, right? So for continuity, of the projects.
For some clinical customers, right, as you know, that for the clinical workflows where you’re returning results to patients, there’s a significant additional validation and/or regulatory activities that need to be done. So, if you’re expanding your lab and not what you’re planning to do that and that makes sense for your business, I think many of those customers are not going to hold off in the short term, right, and forgo that business. Moreover, they would — even if they move to an X, they would have to — they have to reduce some of that validation work, right? So that will take time.
We’ve actually been, again, surprised how much interest there is in the X by clinical customers. Because, again, as we’ve talked about in the past, they tend to be later adopters, right, because they’ll buy a couple of instruments, understand the performance and then plan whatever is a natural new validation for them when they’re ready. But, they’re very excited about the benefits of it, even more than we had thought. But it doesn’t change the fact that if you’re running a large oncology testing lab and you have the volume and the business, you wouldn’t put off buying a 6000 in the short term.
So, as we said in the earnings call, there are definitely customers where it has affected them, right, and it is reflected in our guidance but it’s not zero. It’s still a pretty significant number for the reasons that I said.
Vijay Kumar
Understood, understood. And that was baked in the guidance that was updated. Just maybe one of the points you mentioned, Alex, was this — sequencing increasing because now you can do more with the X Plus. Directionally, if I said, look, the X Plus should have a higher consumable pull-through than the 6000 system, would that be a fair comment?
Dr. Alex Aravanis
Yes. So, our goal is to work for that, and we see how that will play out over time. If you look at the overall utilization of the high-throughput systems today, there’s still a significant opportunity for them to be more utilized, right? So, with that kind of — with an instrument that’s faster and is higher throughput, there’s a lot of opportunity for that.
So, I don’t think we’ve put any guidance or quantified that, but that’s certainly in the way things we want to play out, and we think there’s significant opportunity for that given the system.
Vijay Kumar
Understood. And then I guess on the system side itself, Alex, some people have asked me whether — if I had three 6000 systems, NovaSeq 6000, am I just swapping them out for one X Plus, as in the overall pie itself is shrinking? Or is this more of a one-to-one replacement? How should we think about this replacement cycle? Is one 6000 being replaced for one X Plus, or is that down some ratio?
Dr. Alex Aravanis
Yes. I mean obviously, mathematically, there’s kind of lots of combinations. Again, with the elasticity thesis, which we’re strong believers in, we think that for most customers, they’ll be looking at the X for kind of their next wave of work, right, which is they want to build a new lab and they want to expand. They want to do another — a bigger population project. They want to do a bigger cohort for pharma discovery. And so, it’s less about how can they swap out kind of existing systems that are running well and validate it, and they have a business around versus how does this allow them to expand their business, do applications or offer a new service, right? So, like one customer, they offer an exome service, right, they’re going to continue offering that for the near term, right, on their 6000. They’d really like to offer a whole genome service, right? So that’s how they’re thinking about the X. And that’s their rationale that they made for buying new — for buying the Xs. Now over time, if they’re successful, then maybe they’ll stop doing exomes on the 6000, retire them and then go to all Xs. But if they’re doing whole genomes on the Xs, well, they’re actually generating even more pull-through than they were on the exomes on the 6000, right? So I hope that you can follow that.
But that’s the kind of dynamic that we’re seeing, and that’s the kind of dynamic that leads to them buying if anything, more Xs and doing more work, right? Because they can they can command a more significant reimbursement or fee because they’re offering so much more.
Vijay Kumar
Understood, understood. In the last few minutes, I just want to touch on GRAIL, the recent restructuring announcement you guys made. But let’s start with GRAIL. Alex, I was — you were a co-founder of this company but can you — I know it’s being run separately, but can you perhaps update us on where we are on the clinical pipeline? When can we expect some trial data on this UK study that’s ongoing, the 140,000 patients? And I believe that that project is expanding to 1 million patients. Is that — do we know what the pricing is on that expansion project?
Dr. Alex Aravanis
Yes, I can share a few things there. So one is, GRAIL released some information on the study. Operationally, the study is going quite well. So they — with the UK, which is bearing the expense of the actual operations of the test, not the test itself, but all of the other expenses of recruiting people and organizing it and so on, they’ve successfully enrolled 140,000 individuals. So 140,000 individuals have got their first round of testing with the Galleri test. So that’s been successfully executed.
They’ve now started doing the second test. So the trial involves at least two serial tests. And so, there — again, it’s going really well from that. They’re continuing to the study, which means that presumably the data — the initial data is encouraging. A formal readout, I believe GRAIL set in ’24, so I still think that every reason to believe that that’s still one we’ll get a formal readout of the study’s performance.
Now the partnership that GRAIL has with the NHS anticipated a follow-on with success in the pilot. So, if the pilot is successful and those were all — and the criteria was predetermined and GRAIL agreed to that and obviously felt confident that it could achieve that. But NHS is a very rigorous organization as it should be. That success has to be demonstrated in the pilot. With that, though, they anticipate doing 1 million tests, right? So, that’s actually again baked into the partnership that that will automatically happen. And that will be commercial. So that will involve the NHS paying for those 1 million tests — the — I don’t believe GRAIL has shared anything on pricing for that. So I don’t think I can say anything about that. But the intent is obviously to be a profitable endeavor.
Salli Schwartz
And the expectation is that that — those million tests, if that’s where this goes, if the pilot’s successful, would be over 2024 and 2025. So, it’s relatively soon. And to Alex’s point, I don’t think the price point has been publicized yet, but as soon as it is…
Vijay Kumar
And then maybe a last question here for you, Salli. GRAIL…
Dr. Alex Aravanis
Maybe just one other point there, which is that — just want to emphasize that the trial is a big deal for the NHS to put their effort behind this, their expense. Obviously, they’re hoping for success here. With success there and the million, then we’re excited about the opportunity, well then that 1 million becomes 10 million, right, maybe in the UK. And then, it also will be the most significant clinical trial by far, right, I mean, that would take others many years and hundreds of millions of dollars to replicate. So, it will be an exciting piece of clinical evidence for GRAIL, not just in the UK but also in the U.S., right? So, you can imagine how useful that will be, if successful in adoption in just in discussions with regulators about approval and so on.
Vijay Kumar
Got you. And then maybe my last one here for you, Salli. How should we think about spending on GRAIL? Because I know now GRAIL runs independently, right, until we have an EC order. When are we expecting this order from the EC? And how should we think about the plus and minus on margins for fiscal ’23 rate? Is GRAIL going to continue at $600 million of operating loss at this, or could you have some benefit given the recent restructuring?
Salli Schwartz
Yes. Let’s see. So, a couple of things. We expect the divestment order sometime later this year or early next year. That’s the most fine-tune timing we have at this point. Obviously, as we get closer to the end of the year, it looks more like early next year.
Now, for next year, we have to work with GRAIL and make sure that we understand their budget for next year. I know they have been managing their spend over the course of the year — as alongside their revenue in order to manage it against their — the overall operating loss that we had guided to. So, they’ve certainly been doing that.
And just to clarify, the restructuring that we recently announced was only about core Illumina. So, there wasn’t anything in the 5% headcount reduction that we announced a couple of weeks ago that had anything to do with GRAIL. It was specifically about core Illumina and our own efforts to manage our own spend in this interesting and somewhat challenging macro environment so that we could be really well prepared as we get into 2023 to take advantage of the opportunities that are out there in the market.
So, we’ll have more on GRAIL in terms of specific guidance for next year’s spend. But certainly, I think, across the board, core Illumina, GRAIL, we’re all managing in this environment and trying to trying to be prudent with our spend.
Vijay Kumar
Got you. Sorry, that restructuring, just one clarification. Is that a net savings, or will those savings be reinvested in the business?
Salli Schwartz
Yes. So, we didn’t give an overall savings amount, but suffice it to say there’s a few things that are going to effectively offset this. So one, the people that we hired across 2022, we have them a partial year this year. We’ll have them for our full year next year. That’s certainly a drift upward. This year for 2022, we’ve had a — we will have a lower bonus year just given the financial performance of the Company. And as we reset next year and certainly, if we deliver that will be an incremental bump up. And even FX, I mean for the latter part of the year, hopefully, the year-over-year comparisons work decently. But in the beginning part of the year, that’s going to be a trend against us as well.
So, on the whole, it’s several things that are going to drift the expense base upward. I think at a higher level, we are still very much a growth company, and we want to keep fueling all this innovation that Alex has been talking about, so that we can continue to deliver to the customer base. Not only the innovations we’ve already talked about, but the ones that we have next in the pipeline. So, we’ll be managing the expense base carefully but continuing to invest in the business.
Vijay Kumar
Got you. With that, I think we’re out of time. Alex and Salli, thank you so much for spending the time with us.
Salli Schwartz
Happy to.
Dr. Alex Aravanis
Appreciate the opportunity. Thanks, Vijay.
Salli Schwartz
Thank you.
Be the first to comment