Illumina Inc (ILMN) 2020 Q3 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Third Quarter 2020 Illumina Earnings Conference Call. (Operator Instructions) As a reminder, this conference call is being recorded.

I would now like to introduce your host for today's conference call, Ms. Juliet Cunningham, VP, Illumina Investor Relations. Please go ahead.

Good afternoon, everyone, and thanks for joining us for our third quarter 2020 results conference call. During the call today, we will review the financial results released after the close of market and offer commentary on our commercial activity. After which, we'll host a question-and-answer session. If you've not had a chance to review the earnings release, it can be found in the Investor Relations section of our website at illumina.com.

Participating for Illumina today will be Francis deSouza, President and Chief Executive Officer; and Sam Samad, Chief Financial Officer. Francis will share an update on our business, and Sam will review our financial results.

Similar to last quarter, we are hosting our conference call from a number of different locations, so please bear with us if there are any technical issues or pauses. This call is being recorded, and the audio portion will be archived in the Investors section of our website. It is 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. All forward-looking statements are based upon current available information, and Illumina assumes no obligation to update these statements. To better understand the risks and uncertainties that could cause actual results to differ, we refer you to the documents that Illumina files with the Securities and Exchange Commission, including Illumina's most recent Forms 10-Q and 10-K.

With that, I'll turn the call over to Francis.

Thank you, Juliet. Good afternoon, everyone, and thank you for joining us today. We saw a strong rebound in our business in the third quarter, with a faster recovery than we expected in both our clinical and research customers. Total revenue for the third quarter was $794 million, up 26% sequentially and down 12% compared to the prior year period. I'll share the third quarter highlights, and Sam will provide more detailed financials.

Sequencing revenue grew 25% compared to the second quarter of 2020. Sequencing consumables grew 29% sequentially, with strength across our high, mid- and low throughput product portfolio. NextSeq momentum continued to build, with mid-throughput consumables growing both sequentially and 3% year-over-year, and we expect continued growth in the fourth quarter.

Sequencing instruments also outperformed expectations, with revenue up 24% sequentially. This growth was primarily driven by increased shipments of NovaSeq and NextSeq platforms. NovaSeq purchases have rebounded nicely from Q1, when some customers delayed purchases as shelter in place policies took effect. We have been pleased with the sequential growth in NovaSeq shipments in each quarter since Q1. We're entering the fourth quarter with a strong NovaSeq pipeline and expect the sequential growth to continue in Q4.

In addition, we had another successful quarter for our mid-throughput portfolio, with both NextSeq 2000 and NextSeq 550, exceeding our expectations. Strong demand for these flexible sequencers led to a record mid-throughput shipments in the third quarter. We're especially pleased with the success of the NextSeq 2000 launch, reflecting positive customer feedback and demand. Approximately 1/3 of the NextSeq 2000 placements are coming from new to Illumina customers. Consistent with our expectations, we've seen little cannibalization of NovaSeq conversions among the remaining HiSeq customers.

Sequencing Service and Other revenue grew 9% sequentially and was 28% lower compared to the prior year due to onetime IVD licensing and development revenues of over $30 million in the third quarter of last year.

Total microarray revenue of $86 million was up 28% sequentially as direct-to-consumer customers and other applications began recovering from pandemic-related headwinds.

Turning to our customer segments, we saw solid improvement in both clinical and research activity. As a reminder, we track sequencing run rates as an indicator of general activity. Run volumes are based on sampling and do not directly correlate with revenue. The intent of providing this metric is to help investors understand customer sequencing activity.

Clinical sequencing run volumes increased from an average of 84% in the second quarter to 96% during the third quarter compared to pre-pandemic levels.

In oncology, we're encouraged by signs that point to continued adoption of NGS in cancer treatment. The European Society for Medical Oncology, or ESMO, became the first scientific society to issue recommendations for the use of NGS in oncology. Depending on the cancer type, the society recommends NGS for large multi-gene panels for the determination of Tumor Mutation Burden and to screen patients for clinical trials. To date, TSO 500, which is for research use only, provides comprehensive genomic profiling and has been adopted by approximately 250 customers. We're excited to release a CE-IVD marked TSO comprehensive assay in 2021 to ensure labs can follow ESMO guidelines and continue to improve cancer patient outcomes.

Growing payer adoption will continue to broaden the availability of these tests. Last month, Aetna, covering 19 million lives, announced coverage for large panel comprehensive genomic profiling, for both solid and liquid, including testing for TMB.

We're also encouraged by the increasing availability and adoption for NGS-based liquid biopsy tests. The recent FDA approval for Guardant360 and FoundationOne Liquid CDx demonstrates the clinical utility of these technologies that detects cell-free DNA from the tumor in the blood and can provide comprehensive genomic profiling when tumor tissue is not available. Liquid biopsy panels like these broaden access to more patients who can benefit from NGS in their cancer treatment.

Reproductive health showed strength in the quarter, growing sequentially and year-over-year. In fact, during the third quarter, reproductive health averaged over 100% of its pre-pandemic run rate. We expect sustained growth as reimbursement prospects have never looked more promising.

Recently, ACOG updated its guidance to recommend that NIPT be made available to all pregnant women, regardless of maternal age or baseline risk. This is a major milestone, and we expect payers to continue to follow ACOG's guidance, adding to the already 130 million-plus lives covered for average risk in the U.S. Centene, for example, recently aligned its guidelines, adding average risk coverage to an additional 23 million lives in the U.S. Centene is also the largest Medicaid managed care organization in the U.S., meaning this coverage decision will bring NIPT as an option to more medically underserved areas and populations.

In genetic disease testing, physicians, insurance companies and governments around the globe are increasingly adopting sequencing to diagnose children suspected of having genetic conditions. We're seeing solid progress on the reimbursement front. Payer adoption in the U.S. continues to grow, with more than double the lives covered for whole genome sequencing than at the start of the year. And Australia recently announced that they will reimburse whole exome or genome sequencing to identify the genetic cause of intellectual disabilities. These positive developments give us confidence in sustained long-term growth in this underpenetrated market.

Turning to research. Sequencing runs show continued improvement and increased from an average of 65% in the second quarter to 82% in the third quarter compared to its pre-pandemic run volumes. We're encouraged to see researchers returning to their labs and reopening, and this level of improvement exceeded our expectations.

Within research, we're seeing the role of NGS take shape in infectious disease. We believe the most likely near-term use case for sequencing is in research and surveillance for projects like Australia's first national COVID-19 tracking system, which aims to sequence the virus genomes of all positive COVID-19 tests in Australia and track COVID-19 using genomics across the country.

NIH is also increasing its funding for understanding COVID-19 and the transmission pathway using genomic testing. We continue to believe COVID-19 testing will be primarily served by PCR and antigen tests, with NGS providing supplemental testing capacity. As such, we're not factoring in a material contribution to revenue in the fourth quarter from COVID-19 testing.

On the population genomics front, we're seeing promising signs of progress. All of us received its IDE approval from the FDA in the second quarter, and as expected, samples started to flow from the biobank to partner genome centers. We expect the project to ramp in the fourth quarter and into 2021.

During the third quarter, sequencing for the UK Biobank continued, and we expect levels to gradually return to normal and sequencing to continue into 2021.

And we expect whole genome sequencing services for the NHS Genomic Medicine Service to commence later this quarter. Given the NHS' current focus on COVID, we expect this program to gradually ramp over the next few quarters.

Perhaps one of the most resilient PopGen programs in the face of the pandemic has been the Million Veteran Program, or MVP. The project aims to provide insights on how genetics, lifestyle and military exposures affect health and illness and has recently expanded its targeted enrollment from 1 million to 2 million veterans. The U.S. Department of Veterans Affairs through its sequencing partner, Personalis, expects to cross the 100,000 whole genome milestone before the end of the year.

We believe these and other initiatives will expand our markets and create exciting new opportunities for Illumina. We are committed to continuing to lead through innovation with new product launches to support the markets and customers we serve, including:

NovaSeq 6000 version 1.5 reagents, which were launched during the third quarter. This release offers increased shelf life, improved quality scores, longer indexes and most importantly, it makes the $600 genome more broadly available to help smaller labs be successful and provides additional incentives for labs to upgrade from HiSeq to NovaSeq. NovaSeq 6000 version 1.5 reagents have been extremely well received by our customers.

For example, the Princess Máxima Center, the largest national pediatric oncology center in Europe was outsourcing its genome sequencing to a collaborator, but with the new reagents, they were able to purchase a NovaSeq and can now start to perform testing in-house. Meaning potentially care-altering genomic information will reach children with cancer faster and be performed closer to the patient.

The longer read lengths, combined with high-throughput scale, enabled a Canadian biotech company to adopt a NovaSeq 6000 for the development of antibody products for potential COVID-19 therapy.

We're preparing to launch NextSeq 1000 this quarter. NextSeq 1000 enables lower throughput customers to adopt our newest platform as an entry point with the ability to field upgrade to the NextSeq 2000.

We're also launching the P3 flow cell this quarter. Early access P3 customers, including academic core labs, are already using the flow cell for routine sequencing methods in midsized research projects to enable faster turnaround times. Other customers are using P3 to develop workflows in single cell, epigenetics and spatial transcriptomics.

These are examples of how we continue to execute on our current and future plans throughout these challenging times to advance our mission and create long-term shareholder value.

Now I'll hand the call over to Sam to discuss the third quarter financials.

Thanks, Francis. As discussed, third quarter 2020 revenue was $794 million, a 12% decline year-over-year, but a 26% increase compared to the second quarter of 2020. While the global pandemic continued to impact our business, we saw encouraging signs of recovery within both clinical and research customers.

Sequencing consumables revenue of $500 million was up 29% sequentially and down 5% compared to the prior year period. Quarter-over-quarter growth of $113 million reflected significant improvement across high-, mid- and low-throughput platforms, with NovaSeq pull-through returning to over 1 million per system per year. Our mid-throughput sequencing consumables grew in the quarter versus the same period last year.

Total sequencing system revenue of $109 million was ahead of our expectations with both NovaSeq and our mid-throughput platforms exceeding our expectations at the start of the quarter. The third quarter was also a record quarter for mid-throughput shipments, driven by customer demand for NextSeq 2000 and the NextSeq 550Dx platform. In fact, NextSeqDx shipments represented over 40% of NextSeq 550 shipments year-to-date. Since 2013, Illumina has shipped over 1,000 Dx instruments, reflecting the growing clinical nature of our customer base.

Sequencing Service and Other revenue of $99 million was lower compared to the third quarter of 2019. Total sequencing revenue was $708 million and represented 89% of total revenue. Finally, total microarray revenue of $86 million was down 16% from the third quarter of 2019, reflecting lower consumables and services revenue primarily driven by COVID-19-related headwinds.

Turning to the regions. Americas' revenue of $436 million was ahead of expectations and grew over $100 million or 30% compared to the second quarter of 2020. Sequential growth was driven by both research and clinical customers, as labs continue to reopen and clinical activity increased. In fact, clinical sequencing consumables grew modestly compared to the prior year period, as demand started to recover after delays in testing in the second quarter.

EMEA revenue of $213 million was up $45 million compared to the second quarter of 2020, with strength in sequencing consumables, which grew compared to the third quarter of 2019, driven by a strong rebound in both clinical and research customers.

China revenue grew sequentially to $83 million, driven by strength in instrument placements.

APJ revenue was roughly flat year-over-year at $62 million and increased $11 million compared to the second quarter of 2020, with NIPT helping to lead the recovery with sequential and year-over-year growth.

Moving through the P&L. I will highlight non-GAAP results that include stock-based compensation. I encourage you to review the GAAP reconciliation of these non-GAAP measures, which can be found in today's release and the supplementary data available on our website. Please note that all subsequent references to net income and earnings per share refer to the results attributable to Illumina shareholders.

Non-GAAP gross margin of 67.4% was lower as expected on both a year-over-year and sequential basis. The year-over-year decline was driven by IVD partnership revenue in the third quarter of 2019 not repeating this year and lower volume and higher freight costs associated with the pandemic.

Non-GAAP operating expenses of $365 million were higher compared to the second quarter of 2020 and the same quarter last year. Sequential growth was driven as expected by an increase in R&D project spend and higher compensation expenses. Compared to the third quarter of 2019, operating expenses were higher primarily due to compensation-related expenses and increased project spend, partially offset by decreased travel. Non-GAAP operating margin was 21.4%, up from 15.8% last quarter.

The non-GAAP tax rate of 14.8% was lower compared to last quarter and prior year due to discrete tax benefits related to prior year return adjustments and release of tax reserves.

For the third quarter of 2020, GAAP net income was $179 million, or $1.21 per diluted share, and non-GAAP net income was $150 million or $1.02 per diluted share. Other income was the main driver for the difference between GAAP and non-GAAP net income due to net gains from mark-to-market adjustments on our strategic investments primarily from our marketable equity securities.

Additionally, in the third quarter, cash flow from operations was $153 million, capital expenditures were $48 million and free cash flow was $105 million. DSO of 53 days compared to 55 days last quarter, in part due to increased collections. We repurchased $125 million of common stock in the third quarter, with $295 million remaining for repurchase under our current plans.

We ended the quarter with approximately $3.3 billion in cash, cash equivalents and short-term investments. Our weighted average diluted share count for the quarter was approximately 148 million.

As we continue to navigate this pandemic, there are still uncertainties around the virus resurgence and rate of recovery, and therefore, we will not provide guidance for the remainder of the year. Additionally, we are closely monitoring developments in the pandemic, given the risk of new shutdowns in the U.S. and Europe.

In the fourth quarter, we expect that sequencing run volumes will continue to improve modestly for clinical and research customers, from the 96% and 82% levels, respectively, in the third quarter. Again, we caution that this does not directly correlate to revenue, but data from the third quarter suggests customers are steadily returning to their labs, which will lead to increased activity. Consistent with the third quarter, we expect fourth quarter sequential growth across all regions with China and APJ showing flat to modest growth year-over-year.

For the fourth quarter of 2020, we currently expect continued sequential improvement in revenue, resulting from sequencing consumables being up modestly on a sequential basis, as clinical and research activity continues to gradually improve. Sequencing instruments increasing sequentially, and the fourth quarter is expected to be the strongest quarter of the year for instrument placements. Sequencing Service and Other flat sequentially, and Microarrays up modestly on a sequential basis as we continue to see a gradual recovery in direct-to-consumer and genotyping applications.

For the rest of the P&L, we expect sequential decline in non-GAAP gross margin, in part driven by product mix with higher contribution from instrument sales. Sequential increase in non-GAAP operating expenses primarily driven by R&D project-related expenses, compensation-related expenses and an extra week in the fourth quarter compared to the prior quarter and prior year. Non-GAAP EPS to be roughly flat compared to the third quarter of 2020 reflecting revenue improvement, offset by lower gross margins and the sequential increase in operating and tax expenses.

With that, I'll hand the call back over to Francis for some closing remarks.

Thanks, Sam. As the global leader for sequencing platforms and consumables, we've seen our products enabled groundbreaking research programs and transformative clinical tests that have the potential to shift the paradigm in patient care. One of the largest opportunities for patient impact and revenue potential for years to come is the ability for NGS to detect cancer early.

Last month, we announced a definitive agreement to acquire GRAIL, and we expect the acquisition to close in the second half of 2021. We believe our planned acquisition of GRAIL will accelerate a new era of early cancer detection, transforming cancer survivability and opening up the largest clinical application of genomics we've seen. We look forward to welcoming our GRAIL colleagues to Illumina when the acquisition closes.

Upon close, we expect to operate GRAIL as a division within Illumina. We remain committed to supporting all of Illumina's customers and ensuring that innovators who wish to develop NGS-based tests have continued access to our technologies.

To conclude, the recovery of our business accelerated in the third quarter with significant sequencing consumables revenue growth compared to the second quarter of 2020, and we expect continued consumables growth in the fourth quarter.

Importantly, we're also making good progress incorporating genomics into the standard of care in oncology therapy selection, NIPT and genetic disease testing. And most importantly, we believe that we're laying a strong foundation for Illumina's near and long-term growth.

With that, we'll open the call up to Q&A.

(Operator Instructions) Your first question comes from the line of Doug Schenkel with Cowen.

The quarter was clearly better than most expected, I think even better than you guys probably thought in a lot of instances. I'm just wondering if you have a sense for how much of this was catch-up versus reflecting a new base to build off of moving forward. Based on your prepared remarks on Q4, both yours, Sam, as well as yours, Francis, it does seem like this is kind of a new base to build off of, given the momentum you've described heading into Q4. Is that fair?

Doug, yes, I think that's fair. If we look at how this quarter played out, it certainly played out better than we expected based on how the first few weeks of the quarter were looking. So what we saw over the course of the quarter was the acceleration in the recovery in terms of the sequencing activity in both our clinical and research customers. And so as we exited the quarter, I think we're at a new baseline that we will keep going into Q4. So I think it's fair to say that the recovery accelerated to create a new baseline for us going forward.

Your next question comes from the line of Tycho Peterson with JPMorgan.

Just want to try to hit on 2 things. Just the elasticity on the consumable pricing cut, you highlighted a couple of examples, Francis, of customers that have come forward on the $600 genome, and usually there's a reason to buy NovaSeq. But I'm just curious, as we think ahead and into next year, how much incremental demand do you think that will drive and to what degree will catalyze some of the projects that you hoped it would? And then on the instrument side, instruments were still down 23%. I'm just wondering if you could comment a little bit on when you think they're really going to get back to growth?

Yes, sure. So let me take both of those questions. So first, around the elasticity point associated with the launch of version 1.5 reagents. So what version 1.5 aims to do, and we saw the effect in the quarter, was really catalyze some of the smaller core labs by giving them access to the $600 genome. The larger labs already had access to the $600 genome, so this was really a release targeted to the smaller core labs. And the intent is to do 2 things: one, to unlock the elasticity in that part of the market. And so we saw customers that were thinking about bigger projects, we saw customers that are moving to in-source and I talked about an example of that in the prepared remarks.

The other thing that this release was intended to do was catalyze an upgrade cycle in that segment of the market, the smaller core labs. And so we started to see examples of customers that were either on older instruments or didn't have, so they could be in HiSeq, for example, or they didn't have sequencers. And because of this release, they were able to buy a NovaSeq. Both of those dynamics. So catalyzing the elasticity of demand in terms of bigger projects as well as driving an upgrade cycle are dynamics we expect to continue to play out into Q4 and going into next year.

As you know, typically, when we release a new price point into the market at a lower price point, initially, there is a bit of a headwind. In this case, though, we expect it to be very short, and that's how it's playing out, where very quickly, we're starting to see both of those positive dynamics start to play out.

The second question you talked about was around instruments, and maybe I'll start, and Sam, you can add to it. I talked a little bit in the prepared remarks that we saw really good activity from an instrument perspective. We were really pleased that throughout the course of the year, Q2, we saw an increase in NovaSeq shipments compared to Q1. Q3 was bigger than Q2. And so we continue to see growth in NovaSeq shipments play out over the course of this year from the base of Q1, and we saw strong demand in the mid-throughput instruments, too. Maybe, Sam, I'll turn it over to you to provide more color.

Yes. I would say, in fact, we've been pleased with the sequencing instrument performance, both in Q3 and the resilience that we've seen through the pandemic. You would expect that there would be a year-over-year decline, especially given the pandemic, given some pause on instrument purchases. But as we look towards Q4, we're expecting growth in instruments, our strongest quarter of the year. We're expecting growth in terms of NovaSeq sequentially as well. And when I say growth in Q4, I mean sequentially in terms of instrument revenues. And then the pipeline looks really strong for both NextSeq 2000 and for NovaSeq.

Your next question comes from the line of Derik De Bruin with Bank of America.

A couple of questions. I think the first one, obviously, there's been a big ramp in the operating expense in the second half of the year. As we think about going forward, is this something we should think about rolling into '21, but even higher as some of the T&E comes back in, some of the travel comes back in? And I guess another question, the -- on the NIPT space, can you give us a sense of how many women that were covered under high-risk actually were -- actually were taking advantage -- I mean the women that were eligible for doing high-risk, how many actually went into NIPT testing? And the reason why I ask is, I'm just trying to get a sense of what the penetration would be for average risk, just sort of expanding coverage. Just trying to get some feel for the dynamics and how we can think about the NIPT business expanding.

Yes. So maybe I'll start, and then Sam can add some color, too. In terms of operating expenses, the -- our philosophy over the course of the year has been to review our operating expenses very closely. And one of the things we want to do is make sure we continue to protect the high-priority things that we were working on. So that includes our rich innovation pipeline, and so we've protected our projects associated with upcoming products as well as some long-term investments in the business. So we've protected those, but we've looked for other areas where we could make savings. Some of those sort of naturally played out with the course of the pandemic in the form of reduced travel, for example. And some of those, you're right, will come back next year. And so Sam will comment a little bit more about that.

In terms of NIPT, obviously, it's a very regional story. But in the U.S., for example, you have a reasonably good penetration in terms of the women who are eligible for NIPT going to get it. So the majority of women who are eligible for the test do get it as part of their pregnancy. And so the opening up of that market in the average-risk segment does represent meaningful upside in terms of the actual demand for the test.

Yes. Derik, with regards to OpEx, I would add just maybe a couple of points, and thanks for the question, by the way. The one thing I would say is we've been incredibly focused and diligent in terms of managing our operating expenses, but at the same time, really focused, to Francis' point, about not in any way undercutting our innovation and -- our investment in innovation and in any way impacting our R&D pipeline. But for 3 quarters of the year, up until the end of Q3, I would say we're about $150 million below our own internal budget in terms of operating expense. So we have been very careful in terms of managing operating expenses. We're about 10% below.

And as we look forward, yes, there will be some things that start to ramp back up. Things like again, as employees come into the labs, we will have R&D projects start to scale back up, that will continue into next year. And we'll have certain expenses as hopefully this pandemic eases, like travel, that increase as well. But I would say, in general, we've been extremely diligent and disciplined in terms of managing our expenses through year-to-date.

Your next question comes from the line of Puneet Souda with SVB Leerink.

First question, recently, just this -- in the last 24 hours, we have seen lockdown in France and Europe, and these lockdowns can potentially spread. Just wondering, is that contemplated in your expectations for the fourth quarter? And also, do you expect customers in Europe to be more adept this time around in terms of managing research and diagnostic sequencing operations now versus before? I just wanted to get your thoughts on that, given that sequencing run rate, which you described earlier, is 96% of pre-pandemic levels. Could we see a different standard here? And is this baked into your guidance?

And then if you -- second question, if I could ask around GRAIL, one of the key questions that has emerged last week -- since this week actually, is the approach that Illumina to GRAIL's acquisition in light of another multi-cancer liquid biopsy acquisition this week. I totally recognize that the acquisitions are pre-revenue, but it appears that GRAIL is valued significantly higher versus Thrive. So could you elaborate if there is anything in terms of the overall market opportunity of the pre pan-cancer test? Or technical performance in the Galleri? Or scale of the studies or anything here that gives you a greater confidence and a higher valuation? Appreciate it.

Yes. Thanks, Puneet. So I will start with your question around the potential restrictions in Europe, maybe even the U.S. As we mentioned on the prepared remarks, we're monitoring that closely. Obviously, we all want to see the pandemic get under control, but there are definitely some signs in Europe that there are some potential shutdowns taking place or at least additional restrictions. And so for Q4, we're monitoring that closely. We factored that in as we look forward in terms of our expectations for Q4. In fact, we said clinical sequencing activity in Q3 was 96% of prepandemic levels. Research was 82% of prepandemic levels. And we expect both of those to improve modestly partially because of the fact that we're really tracking kind of how this pandemic evolves in Q4 and not to improve more than that.

With regards to your other point about customers being more adept, I do think they are getting a bit more adept in terms of learning to work and live with the virus. Not to say that they're not obviously very careful and cautious about that. But at the same time, when we look at, for instance, the number of labs that are now, based on our Q3 data, that are either completely open or in sort of some level of operation, it's about 90% of the labs compared to about 50% back in April. So that doesn't mean that there -- that 90% is all full systems ahead and they're all open, but there are either fully open or in restricted operations. It's about 90% of the total labs globally. So they are really gradually ratcheting back up, but in a cautious way. And that's what's factored into our expectations.

And then, Puneet, if I could comment on your GRAIL question. We've talked about the fact that early detection of cancer represents, by far, the largest clinical application of genomics we're likely to see over the next decade or two. And we also talked about the fact that we expect many players in this market. And because it's such a large opportunity, and we expect players to take different approaches, whether it's single cancer versus multi-cancer or different technological approaches. And certainly, we have a number of customers including Thrive, Exact, but also Freenome and Guardant and other liquid biopsy players like Foundation Medicine that we expect to play a role in this market going forward. And we will continue to support all of those customers going forward.

What attracted us to GRAIL was its unique position in that market in terms of being the closest to having a commercial test, and in terms of having the performance characteristics consistent with being closest to having a commercial test. So GRAIL is looking to launch its test next year, and that will likely make them the first player into the market. They've also undertaken, by far, some of the largest clinical studies in that space. And so they have over 100,000 people enrolled in the studies that they've done and generated quite a remarkable bolus of data associated with their test. And then the performance characteristics of their tests are very strong. So for the 12 deadliest cancers, they have a sensitivity in the high-60s percent, with a false positive rate of less than 1%, which really does represent a best-in-class product at this time in the market.

What we also like, too, is that they're the only player at this point that has FDA approval in terms of being able to run an IDE study to return results to patients. And so they are working with some of the premier health care systems, including the Mayo Clinic, Cleveland, Oregon Health System, the Intermountain Health system, Dana-Farber, Sutter Health, and so they already have their gallery test in the hands of doctors returning results to patients as part of this study. And so having that FDA approval was also very attractive. All of that obviously factored into the price associated with GRAIL.

Your next question comes from the line of Dan Arias with Stifel.

Francis, on the All of Us samples that should begin to flow this quarter, is there a material consumables assumption that's built into that work for 4Q? And assuming that things do get going there, are you thinking that, that should be all sequencing? Or might there be some genotyping that makes it into the mix there?

And then if I could ask a quick follow-on. It sounds like you're thinking modest growth in China is in the picture for 4Q. Did you say year-over-year, that would be a sequential step-up there that we haven't seen in a while? So what's driving the confidence in the move, I guess, back up to the mid $90 million range?

Yes. Let's start with All of Us. So for All of Us, we do expect the All of Us samples to start flowing to partner genome centers in Q4 and to start to see both sequencing and genotyping activity. From our perspective, the actual revenue contribution in Q4 we're expecting is modest, but it's a very important step because it then sets us up well for All of Us to ramp up more meaningfully going into next year. And so it means that the [actual] process has been set up, the workflow has been set up and the machine is going, if you like. And so again, modest contribution in Q4, but that's a big step and sets us up, we believe, well going into next year. And then, Sam, maybe you can talk about China?

Yes. So Dan, thanks for the question, first of all. With regards to China, we're expecting, I would say, flattish year-over-year performance in Q4. So we said on the prepared remarks, for China and APJ, flat to slightly increasing. So for China, specifically, it's flattish performance year-over-year. In fact, China, the clinical performance has been strong year-to-date and growing year-over-year.

Research, as we've talked about on previous calls, has been impacted definitely by the pandemic, repurposing some of the work towards PCR testing and just the general impact of that on research. But we're pleased with the clinical performance that we've had in China. Q3 had strong instrument placements as well in China, which bodes well for future activity. And so pleased with the clinical performance there for sure, and that's what gives us confidence.

Your next question comes from the line of Tejas Savant with Morgan Stanley.

So a couple of questions for you, Francis and Sam. So on the sequencing consumables line, you had some destocking in the second quarter. Did some sort of a restocking dynamic here play into that 29% quarter-over-quarter growth at all? And then in oncology, how are you thinking about sort of a bolus here in the fourth quarter and in the first half of '21 as the missed screenings come back online?

Yes. Thanks, Tejas. With regards to stocking levels and restocking, yes, in Q2, we had some customers that destocked, specifically in AMR. So that was an impact on our performance in Q2. We have not seen from our data that there's been an increase in inventories or restocking back in Q3. So essentially, we've seen, and we've really looked at this with a lot of our customers, we've seen that customers are really purchasing to the demand that they have but have not restocked back some of the destocking that occurred in Q2, nor are we expecting in Q4 that there is restocking either, that they're going to continue. Based on the activity levels that we track and triangulating that back with revenues, we can see that essentially, they're buying to their demand.

If you look at oncology, I mean, I think you bring up a really important point. I think one of the tragedies associated with the pandemic is that there are people who frankly aren't going in to get screened for cancer. And so they're walking out with cancer developing. I think there are sort of 2 dynamics that will play out. One is there will be, at some point, to catch-up, right? Unfortunately, for those individuals, it means that cancer has progressed. But whether that plays out in Q4 or next year, there will be a catch-up associated with people who haven't had their diagnosis yet.

In addition, the other dynamic that's playing out is we're seeing that blood-based tests are actually doing better through the course of the pandemic. So if we look at how the market is playing out, the service providers that offer blood-based tests have a proof that their business is more resilient through the pandemic than the service providers that do tissue-based analysis.

And so I think one of the things that's playing out through the course of the pandemic is you're starting to see an acceleration of blood as a sample type for cancer testing. So you're seeing an acceleration in the growth of liquid biopsies, which we think is a durable dynamic, and it's not just as a result of the pandemic, but we think that will continue to play out in the coming years.

Your next question comes from the line of Vijay Kumar with Evercore ISI.

I guess 2 quick ones. One, maybe I misheard this, but the GRAIL deal close, is that a second half of next year? And I'm curious how we should be thinking about investments here. And then when you think about the system placements here, given this razor blade model, maybe perhaps elaborate on how the lower systems this year should play out next year? I do realize the comps for consumables are easier. Should that offset the system declines for this year? I'm curious.

Yes. So let me please touch on both, and then Sam can add some color. If we -- so on GRAIL, yes, we are expecting the deal to close in the second half of 2021 as we go through the regulatory process. We've talked a little bit about the dilution associated with GRAIL. But frankly, we'll have more information for you as the deal closes. And again, that's the second half of 2021.

In terms of the instrument placements, we -- obviously, you're right, it is a razor blade model. And that's why it's important -- and that's why we're so encouraged by the recovery we're seeing in instruments placements after we got out of Q1. And so the strength in NovaSeq placements through Q2, Q3, and certainly, the pipeline as we walk into Q4, is very important.

Similarly, the strength that we're seeing in the mid-throughput part of the portfolio, specifically, NextSeq, both the new NextSeq 2000 as well as the 500 and the Dx boxes are a good sign for us as we look going into next year.

Your next question comes from the line of Catherine Schulte with Baird.

A 2-parter on the GRAIL studies. What's the timing next year in terms of when we can expect to see data from PATHFINDER? And do you expect any performance degradation as you move into an all-comer screening population for that study? And then second, for the STRIVE Study, I believe GRAIL completed enrollment in late 2018 and was following patients for 30 months. So is that something where we could see a readout at some point next year? Or is it likely into 2022?

Yes. So let me talk about the GRAIL studies. The -- the answer is we don't expect the PATHFINDER study to provide substantially different performance data than we have seen so far. And the reason for that is that GRAIL has done a number of other large-scale studies. So the CCGA study had 15,000 people enrolled in the study. The second phase of that study was a prospective study, an all-comer study. And so given the size of the studies they've done, with the same test, the performance that they've published externally is based on statistically significant data. And so we didn't expect the PATHFINDER data to reveal anything different. And in fact, before we signed the deal, we had a small group of people look at some of the early results -- of the midterm results associated with PATHFINDER. And while we won't talk broadly about it, they're generally consistent with what we expected. And so we don't expect, when the final study results get published next year, them to be significantly different from the performance characteristics that you've seen so far. What was the second part of that question?

With STRIVE, Francis.

Yes. So STRIVE is a prospective study. It's 100,000 women, and it's fully enrolled about 18 months ago. So it's, again, pretty far down its path. And again, similarly, given the size of the studies that have gone before, we expect this to be further validation in terms of the performance characteristics and not significantly different.

Your next question comes from the line of Patrick Donnelly with Citi.

Great. Maybe just one on the POPSEQ side, it's encouraging to see some of those bigger projects gaining momentum to get flowing here in the near term. Obviously, the time lines have slipped a few times now. I guess can you just talk through your confidence on these taking off on time, I guess, with the new 4Q time lines, particularly all of us? And are the trajectories similar to kind of what you've talked about before, All of Us, I think, for example, at the beginning of the year, you kind of talked about 60,000 whole genomes in the first 6 months. Are we still thinking that the trajectories are the same once that do get launched?

Yes. It's a good question. And as we look at where we are with those population studies now, I'll touch on sort of the 4 big ones, right? So it's the UK Biobank, the Million Veterans, All of Us and the U.K. NHS. And where we are now is that, walking into next year, the majority of them have already been stood up, and it's a matter of just either getting back to where they were, which is, for example, the UK Biobank, or just ramping up volumes of samples. That's a much better place to be than we were last year, for example, where all of us still had to get FDA approval and then work out the workflows to get the samples flowing to its partner genome centers.

And so if I look at all of them, the UK Biobank, it actually -- they entered the year sequencing at high capacity. They pulled back in March because of the pandemic, and they're starting to ramp back up going into next year. And so assuming conditions are good, we have high confidence in their ability to ramp up quickly.

We talked about the fact that the Million Veteran Program has been remarkably resilient over the course of the year, and in fact, has increased its scope over the course of this year.

All of Us, we see samples will start flowing in Q4. And so really, it is all about just scaling up next year, which is a pretty good place to be in terms of actually getting it going.

And then the U.K. NHS, for this year, has been on hold primarily as the NHS has been dealing with the pandemic testing that they're doing. But they are expected also to start sequencing later this quarter. And so that puts us, we believe, in a good position to start to see the scale-up going into next year.

So across all 4, we feel really good that as we end this year, all of them will be in a position where they are processing samples and ready to scale up and deal with the pent-up demand that they've had.

Your next question comes from the line of Dan Leonard with Wells Fargo.

Just a couple on the expense side. One, can you elaborate on what's happening on the gross margin line, both the driver of the sequential downtick and the further downtick expected in the fourth quarter? And then secondly, can you talk about -- does the anticipation of the GRAIL acquisition philosophically change any of your thinking around base Illumina budgeting as you roll into 2021?

Yes. Thanks, Dan, for the question. So with regards to gross margin, first of all, the downtick from Q2 to Q3 was really expected for our expectations. So gross margin came in at slightly better than our expectations. And that was driven by some additional -- we had some additional freight expenses, some additional compensation expenses in our operations organization and also some of the ramp on the projects as well had some additional impact on gross margin or cost. So that's our expectations.

The Q4 downtick sequentially versus Q3 is really mostly driven by mix. And the fact that we have an instrument quarter that's strong, that outweighed some of the benefit from the consumables, mix is impacted negatively by that.

With regards to the expectations for '21, really, when we think about the core business and Illumina, there is really not a material change in terms of our budget philosophy and approach. We are absolutely focused on continuing to invest in the core business and also generating the shareholder returns that we were committed to in the core business to continue to drive towards that over time. But at the same time, we have a really compelling opportunity with GRAIL, and we're going to continue -- we're going to invest behind that opportunity. It's going to be dilutive in the next few years. But we're committed. So not a major change in terms of our budgeting philosophy for '21 on the core business.

Your next question comes from the line of David Westenberg with Guggenheim Securities.

So we appreciate kind of your long-term expectations of double-digit growth in consumables -- sequencing consumables. Can you quantify how much of that is clinical and specifically oncology as a component of growth? And then as kind of a follow-up to that, can you quantify or maybe even just qualify the opportunity of moving from panels -- to panels from hotspot testing in terms of just growth there?

Yes, sure, David. So maybe I'll -- I'll start with [single]. As we entered this year, we talked about the fact that the clinical business represents just under half of our sequencing consumables. But we also talked about the fact that, that part of our business is growing significantly faster than the research side of our business. And so you should expect those lines to cross. And once they cross, they aren't going back. That sequencing consumables from the clinical business is going to be the majority of our sequencing consumables, both in dollars and in growth going forward.

Of those sequencing consumables from the clinic, oncology testing for therapy selection represents the biggest part of that market. And so we haven't -- you should expect that to continue going forward. We haven't given long-term numbers there, but you should expect that to continue to be the biggest part of the sequencing of tools in the clinic going forward.

We do expect, for a number of reasons, to continue to see the migration from hotspots to larger panels. And frankly, from not just us, but the smaller panels to larger panels, too. And that's driven by a number of things, including the increasing use of TMB as a biomarker for therapies. And so if you look at what happened with Keytruda, for example, in the last few months, that continues to add momentum to say it's not enough just to look at gene hotspots to prescribe a therapy, but you need to look at things like TMB. And that will require the use of larger panels. And so you pick that, you take the fact that you're seeing -- and I talked about ESMO, for example, you're seeing momentum in guidelines to use biomarkers like TMB, and all of that will continue to drive momentum to the larger panels, which will then drive more sequencing and we'll make oncology testing grow even faster and therefore, be a bigger component of the sequencing consumable revenue going forward.

Your next question comes from the line of Dan Brennan with UBS.

Great. Francis, could you just give us a sense of the upgrade opportunity that still remains ahead of you for your smaller customers that are on the HiSeq? I guess we haven't gotten an update in a while, and particularly in light of the price cuts and the comments this quarter about the strong demand elasticity that you saw. So any color regarding the size of that opportunity? Maybe what the early traction look like this quarter? And kind of how do you think that upgrade opportunity evolve as we look out?

Yes. So maybe I'll start with sort of a qualitative view, and maybe we can work our way through that. So when we talked about sort of the multiyear upgrade opportunity for NovaSeqs, there is this next segment that we're activating starting with the version 1.5 reagents, which is the smaller core labs. They are the ones that are still primarily still using the HiSeqs. And frankly, they haven't had the volume to justify the upgrade because you need volume to see the benefits associated with the consumables' drop in cost. So 1.5 now changes the economics for them, and that's the segment of the customer base that we're looking to activate. So in a while, in terms of the number of instruments there, maybe I'll turn it over to you, Sam, to give us some color on that.

Yes. Maybe the color I'll give, Dan, and thanks for the question is, as we think about the NovaSeq placements, I know we used to give more specificity around this when we first launched NovaSeq and the number of HiSeq customers that we still expect to take on NovaSeq but we haven't provided that for some time. But the metric I will give you is for NovaSeq placements year-to-date, we're still seeing roughly 40%-ish-or-so coming from new-to-Illumina customers. So really a great sign around the democratization of sequencing that NovaSeq has done. We're seeing about an additional 40%-or-so that's coming from NovaSeq capacity upgrades, so existing NovaSeq customers that are just upgrading or taking on more capacity. And then roughly 20% or so from those HiSeq customers that are continuing to transition to NovaSeq. So that's still going on. We're going to continue to see that. The v1.5 upgrade that we talked about will also help catalyze that as well, will drive that as well. So still very encouraging and good room to offset.

There are no further questions at this time. I'll turn it back over to our presenters for any closing remarks.

Thank you. As a reminder, a replay of this call will be available on our website in the Investor Relations section as well as through the dial-in instructions contained in today's earnings release. Thank you for joining the call today. This concludes our call, and we look forward to updating you on our progress in the fourth quarter.

Thank you. Ladies and gentlemen, this concludes today's conference call. You may now disconnect.