Illumina Inc (ILMN) 2018 Q4 法說會逐字稿

Welcome to Illumina's Fourth Quarter and Full Year 2018 Conference Call.

My name is Adrienne, and I'll be your operator for today's call.

(Operator Instructions) Please note, this conference is being recorded.

I'll now turn the call over to Jacquie Ross.

Jacquie Ross, you may begin.

Thank you, Adrienne.

Good afternoon, everyone, and welcome to our earnings call for the fourth quarter and full year 2018.

During the call today, we will review the financial results released after the close of the market and offer commentary on our commercial activity, after which we will 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 provide a brief update on the state of our business and Sam will review our financial results.

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 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 will now turn the call over to Francis.

Thank you, Jacquie.

Good afternoon, everyone.

Illumina ended 2018 on a high note with record revenue of $867 million, up 11% from the fourth quarter of 2017.

For 2018, Illumina reported revenue of $3.3 billion, up 21%, reflecting strong demand for sequencing and array systems, consumables and services.

2018 represented our 20th consecutive year of growth as genomics continues to enable an increasing number of research, translational and clinical applications across a broad range of customers.

In the fourth quarter of 2018, sequencing consumable revenue of $466 million grew 8% compared to the fourth quarter of 2017.

Normalizing for timing of stocking orders, sequencing consumable growth was 16% and NovaSeq consumable pull-through was roughly flat with the record level we reported in the third quarter.

For the full year, 2018 sequencing consumable revenue grew 23% or more than $300 million to $1.8 billion.

Growth was visible across our high-throughput, mid-throughput and low-throughput categories, and Illumina continues to uniquely support the broadest range of customer applications and throughput requirements.

Within the high-throughput family, HiSeq consumables continued their expected decline as customers transition to NovaSeq.

For the full year, NovaSeq consumable revenue increased by a factor of more than 5. For 2019 and as part of our strategy to support a multiyear transition from HiSeq, we're catalyzing the next wave of upgrades among our lower-throughput HiSeq customers.

We're doing this in 2 ways.

First, we're launching S Prime, our newest NovaSeq flow cell in February; and second, we reduced the list price of the S1 and S2 flow cells by approximately 25% and 10%, respectively.

We believe it's important to help customers of all sizes unlock the power of the genome and these changes will enable even greater flexibility, allowing lower-throughput customers to sequence on demand rather than waiting to batch on a larger flow cell.

There have been no changes to the pricing of our S4 flow cell, which continues to represent the majority of NovaSeq consumable revenue.

It was another record quarter for our mid-throughput sequencing consumable revenue, in large part due to growing adoption of oncology and NIPT applications.

Once again, NextSeq pull-through per system was at the high end of the $100,000 to $150,000 per year range.

And the low-throughput family delivered strong year-over-year growth, with pull-through for both the MiSeq and MiniSeq within their respective ranges and a modest contribution from iSeq consumables.

Moving to sequencing systems.

Revenue of $159 million was up 21% year-over-year, representing a new quarterly record for Illumina.

As we anticipated, the fourth quarter was the strongest of the year, with robust performance across the portfolio, including more than 100 NovaSeq shipments.

Since launch, we had shipped approximately 600 NovaSeqs by the end of 2018, up from the 285 installed at the end of 2017.

And we continue to see strong demand from both existing and new customers.

About 75% of the HiSeq installed base has yet to take even its first NovaSeq, so we believe that we're in the very early stages of a conversion cycle that will span multiple years.

In addition to the HiSeq upgrade cycle, we continue to see strong adoption from new-to-Illumina and desktop conversion customers.

This group represents about 30% of orders since launch, expanding our footprint of high-throughput users.

In the fourth quarter, new customers included one lab that is using a NovaSeq to build out a full-service hematology pathology lab, bringing clinical trials to community hospitals.

Another new customer, who is bringing sequencing in-house, had initially planned on purchasing a NextSeq but decided to purchase a NovaSeq to prepare for higher output in the future.

And one of our desktop customers is now using a NovaSeq to support multi-omic translational research to drive precision medicine efforts.

NextSeq placements were strong and, in fact, NextSeq shipment revenue was the second highest in any quarter since launch.

Once again, we saw a good mix of existing and new-to-sequencing customers, with oncology and NIPT the primary drivers of system acquisition.

NextSeqDx was approved by the PMDA in Japan, establishing Illumina's first IVD-registered instrument in the country as we work to bring NGS applications into clinical care.

Building on this approval, I'm pleased to announce our partnership with Sysmex Corporation to commercialize the first NGS-based oncology IVD panel in Japan.

This test will be run on the NextSeqDx system and be based on a PMDA-approved, 114-gene panel developed by Sysmex and the National Cancer Center Japan.

Within our low-throughput system family, both MiSeq and MiniSeq shipments were up sequentially and iSeq ended the year with approximately 350 units shipped in its first 2 quarters of launch.

About half of these iSeq shipments were to new-to-Illumina customers.

Collectively, these customers are utilizing the iSeq across a broad range of applications, including library quality control, metagenomics, targeted re-sequencing and in field disease and outbreak monitoring.

For example, one customer is now using the iSeq for plasmid characterization and synthesis confirmation in place of CE sequencing.

Another customer is using the iSeq to sequence bacterial and viral genomes as well as disease vectors from mosquitoes.

Moving to sequencing services and other.

Revenue of $104 million was up 20% from the same quarter a year ago, largely driven by GeL.

Revenue was down sequentially, primarily due to oncology collaborations payments in the third quarter that did not repeat in the fourth.

In total, our sequencing revenue grew 12% in the fourth quarter of 2018.

For the full year 2018, sequencing revenue grew 21% with solid growth across research, translational and clinical applications.

Across these areas, we continue to be encouraged by a myriad of developments that we believe will enable genomics to continue the progression from research to the clinic and, over time, to standard of care.

We continue to see a broad range of research efforts leveraging NGS.

For example, just last week, Cancer Research UK announced 3 new international initiatives to tackle some of the biggest research challenges in cancer: the microbiome, chronic inflammation and why cancers grow in some tissues but not others.

These research programs were collectively awarded GBP 60 million.

One of the programs is led by a team at the Dana-Farber Cancer Institute and Harvard Medical School that is exploring the relationship between the microbiome and colon cancer.

Moving on, population genomics initiatives have multiplied from a handful of trailblazers to dozens of efforts intended to improve human health.

We are tracking around 50 global initiatives and expect several to begin ramping using Illumina technology in 2019.

Despite GeL's completion of its 100,000th genome and gradual transition into clinical practice, we expect population genomics to contribute to revenue growth this year.

To give just a few examples, All of Us will begin its journey to genotype and sequence 1 million participants.

France will launch its pilot program for whole-genome and exome sequencing as well as RNA-Seq to ultimately sequence 235,000 genomes a year.

And Singapore will continue the first wave of the 10K Project, which is a 3-phase initiative to sequence 150,000 whole genomes and genotype 1 million samples by 2028.

We also continue to see excellent progress in the evolution of our oncology, RUGD and NIPT markets.

Next-generation sequencing continues to redefine the way clinicians are approaching oncology with the regulatory and reimbursement environment evolving rapidly to support innovative approaches like comprehensive tumor profiling and liquid biopsy.

Illumina is extremely proud to have been granted Breakthrough Device Designation for the TruSight Assay that is in development.

We join a short and prestigious list of leading innovators in oncology diagnostics and look forward to bringing our distributable 523-gene tumor profiling panel to market.

The assay has been designed to detect known and emerging tumor biomarkers, including MSI, TMB and gene fusions and variations for targeted therapies.

The hope is that this pan cancer assay will deliver clinically actionable data to ensure patients are matched with the optimal therapeutic or clinical trial.

While we continue to drive IVD development with partners, the assay is based on the content of TSO 500, which is now available as an RUO.

We're pleased with initial customer response from our early access sites around the world and we expect the vast majority to add the RUO assay to their lab offering.

Other early adopters include China's ChosenMed, which is working in support of the Cancer Genetic Atlas of China.

ChosenMed will use TSO 500 as part of a $15 million grant to build a bank of 10,000 tumor samples.

Moving to NIPT.

Coverage for average-risk pregnancies continues to increase and is now at 46% in the U.S., while high-risk pregnancy coverage is around 96%.

More than 40% of all births in the U.S. are to families eligible for Medicaid.

And with the addition of Pennsylvania and Ohio in January, the number of states covering average-risk NIPT increased to 5.

Looking abroad, France recently announced that they will reimburse approximately EUR 360 for trisomy 21 NIPT.

Screening will be offered to women with contingent risk between 1 in 51 and 1 in 1,000, representing nearly 8% of pregnancies.

Additionally, in the first half of 2019, we will ship VeriSeq NIPT Version 2, which adds karyotype resolution across the genome and increases the number of genetic diseases that can be detected.

For RUGD, Aetna expanded coverage to include whole exome sequencing, which brings total lives covered to 147 million.

This, coupled with CMS' final CPT code pricing of over $5,000 per genome for RUGD patients, enables more patients to access NGS, helping to end their diagnostic odysseys.

And just a few weeks ago, Priority Health, a health insurance plan with 1 million lives in Michigan, issued the first positive coverage policy specifically for WGS for acutely ill patients.

While utilization today remains nascent, we believe that genomic testing is poised to become the standard of care for RUGD patients.

Illumina is working to help accelerate adoption, for example, recently partnering with Victorian Clinical Genetics Services, a not-for-profit providing RUGD services in Australia.

Moving to arrays.

Revenue of $132 million was up 7% from the same quarter in 2017.

As we expected, revenue was down sequentially with lower system revenue following the record third quarter and a decrease in array services revenue.

This sequential step-down in array services is in line with the recent revenue trend from Q3 into Q4.

That said, array growth is expected to slow in 2019.

We have factored this into our guidance but expect revenue growth to reaccelerate as health-based apps are more widely adopted and as genealogy and other tests are introduced in new markets outside the U.S.

Moving to regional results.

Americas revenue grew 12% versus the prior year period and grew sequentially for the eighth straight quarter with strong performance across the business.

EMEA delivered record revenue quarter with 12% growth from the prior year period, including a strong contribution from Genomics England, which announced last month that it sequenced its 100,000th whole genome.

Other growth drivers included robust sequencing system placements, notably NovaSeq shipments, and sequencing consumables.

Greater China grew 8% year-over-year with sequencing growth moderated by the effects of tariff-related stocking in Q2 and Q3 of 2018.

When adjusting for the tariff impact, the region grew 28% from the prior year period.

Finally, APJ revenue of $70 million was up 9% from the fourth quarter of 2017 and up 21% sequentially.

NIPT continues to make headway in the region as we signed up our first partner in Korea, BioGeno, and look to expand further in 2019.

With that, I'll hand the call over to Sam for a review of our quarterly financials.

Sam?

Thanks, Francis.

As discussed, fourth quarter revenue grew 11% year-over-year to $867 million, driven by 12% growth in sequencing and 7% growth in microarrays.

Sequencing consumable revenue of $466 million was down slightly on a sequential basis and up 8% from the same quarter a year ago.

As discussed in previous quarters, we experienced some shifts in the timing of customer orders associated with China tariff-related stocking.

In the fourth quarter, this included approximately $5 million of sequencing consumable revenue that was previously expected in the first quarter of 2019.

Normalizing for the China-related stocking in 2018, sequencing consumables revenue was up $19 million sequentially in Q4.

And after also adjusting for a sizable stocking order in the fourth quarter of 2017, sequencing consumable revenue grew 16% year-over-year.

Array consumables in Q4 grew $13 million sequentially and were up $14 million or 17% versus Q4 of 2017.

As Francis noted, sequencing service revenue of $104 million was down $5 million sequentially and up 20% year-over-year.

Array services was down $9 million sequentially and down 22% year-over-year.

Combined, total consumable and service revenue represented 80% of total revenue in the fourth quarter of 2018 and 83% for the full year.

This compares to 81% in the fourth quarter of 2017 and 81% for the full year of 2017.

Moving to systems.

Sequencing systems grew $21 million sequentially and 21% versus the same period last year.

Array systems were down $5 million sequentially and up $3 million versus Q4 of 2017.

Instrument revenue, therefore, represented 20% of total revenue in the quarter and 17% for 2018.

Moving to gross margin and operating expenses.

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 69.1% decreased almost 200 basis points compared to the third quarter, primarily driven by product mix.

Year-over-year, fourth quarter non-GAAP gross margin decreased 180 basis points, primarily due to product mix and lower margins in our services business.

Non-GAAP operating expenses of $388 million were up $32 million from last quarter, largely reflecting the timing of OpEx spend weighted toward the end of the year, specifically higher R&D and marketing spend.

Non-GAAP operating margin was therefore 24.3%, down from 29.4% last quarter.

Excluding Helix, operating margin was 27.1% compared to 32% last quarter.

The non-GAAP tax rate of 16.3% was down from last quarter due to a prior year tax return adjustment.

For the fourth quarter of 2018, GAAP net income was $210 million or $1.41 per diluted share and non-GAAP net income was $197 million or $1.32 per diluted share.

Cash flow from operations was $300 million.

DSO of 54 days was higher compared to 46 days last quarter, driven by less favorable revenue linearity and geography, in addition to collection seasonality.

Fourth quarter capital expenditures were $65 million and Q4 free cash flow was $235 million.

This quarter, we repurchased $98 million of stock, leaving $49 million available for share repurchases under our current plan.

We ended the year with approximately $3.5 billion in cash, cash equivalents and short-term investments.

As of January 1, the 2019 and 2021 notes were no longer convertible.

However, approximately 1 million shares remain in our Q4 diluted share count because our share price has exceeded the conversion price.

As such, our weighted average diluted share count for the quarter was 149 million.

The 2019 notes will become convertible again on March 15, 2019, ahead of maturity 3 months later.

We expect to repay the $633 million principal in cash and issue stock for the premium, which is expected to approximate around $150 million, depending on stock price.

Moving to guidance.

As we previously shared, we expect full year 2019 revenue to grow in the range of 13% to 14% or $3.77 billion to $3.8 billion, representing an increase of approximately $450 million at the midpoint.

This guidance allows for a less favorable FX environment that reduced our full year growth expectations by approximately 1%.

Regarding revenue linearity, we expect 2019 to be more back-end loaded than it was in 2018.

As a reference, revenue split in 2018 was 48% in the first half and 52% in the second half.

We expect the proportion of revenue in the second half of 2019 to exceed 52%.

We expect sequencing revenue to grow in the mid-teens and sequencing consumables to grow above 20%.

And we expect array revenue to grow in the low single-digit range, reflecting a cautious view on the consumer opportunity.

In addition, in 2019, we expect sequencing systems to grow in the mid-single digits with NovaSeq shipments expected to be flat to slightly up compared to 2018, consistent with our expectation for a steady multiyear upgrade cycle.

We expect full year non-GAAP gross margin to be down slightly from 2018 with a higher revenue contribution from lower-margin population genomics initiatives and our oncology collaborations.

Compared to 2018, we expect operating expenses to be down slightly on a percentage of revenue basis.

We expect the 2019 tax rate to be approximately 17%, up from 2018 due to a onetime tax benefit related to our Helix investment in 2018.

Therefore, we expect GAAP earnings per share in the range of $6.07 to $6.17 and non-GAAP earnings per share in the range of $6.50 to $6.60.

This includes approximately $0.20 of Helix dilution.

For the first quarter of 2019, we expect total revenue to be down on a sequential basis due to lower sequencing system revenue, which is consistent with the trend we've seen over the past 4 years.

Following our record fourth quarter performance, we therefore expect sequencing system revenue to decline approximately $50 million compared to the fourth quarter of 2018.

Offsetting this in part, we expect sequencing consumables to be flat to slightly up sequentially from the $466 million reported in the fourth quarter of 2018, sequencing services and other to be flat with growth offset by the expected decline in GeL revenue and we expect microarrays to grow sequentially in the mid- to high single digits.

We expect non-GAAP gross margin to be up slightly compared to 69.1% in the fourth quarter due to favorable mix, partially offset by higher expected array service revenue.

Non-GAAP operating expenses are expected to be flat on a percentage of revenue basis compared to the 44.8% reported in the fourth quarter of 2018 and we expect the first quarter tax rate to be meaningfully lower than our expected full year rate due to a onetime benefit.

Finally, we don't expect any meaningful change in share count this quarter compared to last quarter.

Francis?

Thank you, Sam.

2018 was a spectacular year for Illumina.

We saw genomics inch closer to broad clinical adoption with an evolving regulatory environment and incremental reimbursement in NIPT, in rare and undiagnosed disease and in oncology.

These developments can only further invigorate research and translational work in genomics as academics, institutions and even nations around the world invest in the promise of precision medicine.

As with many of the most exciting scientific discoveries before us, the path isn't perfectly linear.

What is clear is that Illumina remains in the earliest innings in our journey to improve human health by unlocking the power of the human genome.

A wave of population genomics initiatives will start to ramp this year.

A growing number of cancer patients will gain access to genomic testing and oncology diagnostics.

More pregnant mothers than ever before will be screened with a noninvasive prenatal test and more children will have access to genomic tests that could end unnecessary and costly diagnostic odysseys.

And above all, Illumina will continue to innovate and to push the limits of sequencing.

We will continue to listen and partner with our customers to turn today's impossible into possible.

With that, we'll open up the call for questions.

Operator?

(Operator Instructions) And our first question comes from Tycho Peterson from JPMorgan.

I want to start with guidance for sequencing consumable growth.

Greater than 20% is certainly very impressive.

I'm just wondering, Francis, if you can provide a little bit more color on how you're getting there.

Is that really all driven by upside from NovaSeq?

Or are you seeing any slower ramp-down in HiSeq utilization?

And are you factoring in contributions from some of these expanded POPSEQ initiatives?

And then just as a follow-up, I'm wondering, can you confirm whether you actually hit guidance for NovaSeq placements for '18?

I think you talked about around 600 installed, but we had a few people asking about whether you actually hit the 320, 330, 350 guidance.

Sure.

So let me start with the question around what will drive the consumer growth in 2019.

We said the consumer growth to be driven by customers across the entire sequencing portfolio, from iSeq all the way to NovaSeq and across the range of applications.

If we look into the rest of this year, we talked about the fact that we're seeing a number of these populations in genomics initiatives ramp up over the course of the year.

We talked about the fact that we saw record systems placements happen in the last quarter of last year.

And so we expect to see growth from the NextSeqs that we placed.

We expect to see growth from the record number of NovaSeqs that we placed in Q4.

And then we expect to see continued growth from the uptake in NIPT, driven primarily around the growth that we're seeing in Europe.

So the growth in sequencing consumables is very broad based, across our portfolio, across the applications.

In terms of NovaSeq placements, I said in the prepared remarks that we exited 2017 with approximately 285 NovaSeqs placed in 2017 and we exited 2018 with about 600 NovaSeqs.

So the number we placed was short of the 330.

And frankly, that's driven by just timing in terms of our customers and when they choose to take systems.

As we take a step back and look at how the NovaSeq upgrade cycle is evolving, it's really playing out generally as we'd expected.

We've started last year with the S4 building momentum to start to see some of our large NovaSeq customers begin the transition from HiSeq into NovaSeq.

And then this year, with the launch of S Prime and the pricing adjustments on S1 and S2, we expect to start to see some of the smaller high-throughput labs begin their upgrade journey.

So as we planned, this was a multiyear upgrade cycle for HiSeq customers and it's going to play out that way.

And our next question comes from Derik De Bruin from Bank of America.

A couple of questions.

So the first one is on the array guide.

Can you sort of walk through what you're sort of seeing in the DTC genomics market?

Is this a question of basically people buying kits and not turning them back where it's like all the people that are sort of interested in ancestry testing and heritage testing basically done with that and now you're looking for new expansion in that market?

And then also, can you sort of do a comment on where we are with the PacBio transaction and just sort of an update on the regulatory there?

Are you getting any incremental -- you got another request information.

I'm just very curious, can you update us on that process?

Sure.

On the array side, as you know, we had really spectacular growth in 2017 where we genotyped about 7 million consumers and then again in 2018 where we genotyped about 12 million consumers.

So in working with our direct-to-consumer customers and looking at 2018, they felt that the right answer was to think about growth moderating given the size of the numbers they've seen over the last couple of years.

And so our forecast is just based on working closely with them.

What we believe jointly, them and us, is that we will see that growth reaccelerate as we start to see this market show up as a health market as well as a genealogy market.

And then for us, we will see another vector of growth kick in as we start to see the customers outside the U.S. ramp up.

So we had lots of interest from DTC customers, especially in Asia.

And so we started to engage with those customers and we expect those customers to start to ramp up playing out starting this year and then going into the future.

So we expect to see a reacceleration of that market, but for this year, we've modeled it down into the single digits.

PacBio.

And then on the PacBio side, so where we is we've got the second request from the FTC as we expected, and so we're in the information gathering phase.

And so the process, in general, is going as we expected and we still expect it to be a midyear close.

And our next question comes from Ross Muken from Evercore ISI.

So maybe just talking about -- going back to Tycho's point on both the NovaSeq and just sort of kind of the cadence of consumable growth, obviously with all of the stocking orders, it's sort of a little bit hard to tease out some of the underlying, but if you look at sort of where utilization has gone and as you can see sort of the project starts and then the project stops, I mean, clearly, the guide for the year on sequencing consumables is strong.

I guess, I'm just trying to get a sense in terms of the back half weighting, how much of that is sort of the dynamic we saw this year where you had instruments that really picked up in the back half, but it doesn't necessarily seem like that versus maybe project starting over the balance of the year and then cycling through some pieces that are coming off and so that will naturally pull up kind of the back half growth.

I'm just trying to get sort of a little bit of color on the -- what drives the cadence.

Sure.

So if you look at the cadence in our business in any year, our business tends to be back half-loaded.

So we talked about the fact that last year, we saw about 52% of our business done in the back half of the year and that seasonality is driven primarily because you have -- there's 2 end-of-year effects playing out here: the end of the U.S. federal -- you have the government year that ended in Q3 and then you have the fiscal calendar year, obviously the close at the end of the year.

And so that's always driven for us not only strong systems orders, but also strong consumables orders.

This year, we expect it to be a little bit more pronounced than usual because, as you pointed out, we'll see all the usual effects, but then we start to see things like some of the population sequencing efforts that will be ramping up over the course of the year and really start to hit their stride over the back half of the year.

As an example, if you look at GeL, they just wrapped up their 100,000th genome in Q4 and they're in the transition phase and really moving into an NHS-driven sequencing service, and that will take some time.

And so that will really start to play out more in the back half of the year than in the first half of the year.

Similarly, if you start to look at All of Us in the U.S. or France, you'll see that same dynamic play out.

And so you take our normal seasonality and then you layer in this effect, then you'll see this year would be more back end-loaded than the usual.

And our next question comes from Doug Schenkel from Cowen and Company.

I guess, a couple quick ones.

I'm not sure if this relates to Helix or not, but the redeemable noncontrolled interest balance on your balance sheet dropped from $218 million in Q3 to $61 million in Q4.

I'm just wondering what drove that.

And then going back to the placement number for the year for NovaSeqs, you've now acknowledged that you've missed the low end of your target for 330 placements in the year, yet you said you shipped over 100 in the quarter.

So I'm just wondering, first off, what exactly were you thinking you were going to do in the quarter?

Because 100 is pretty good.

A lot more than that seems, on the surface, pretty aggressive.

So I'm just wondering if there's some disconnect between timing of shipments and when revenue is rec'd or if you just really were that aggressive on your quarterly expectation.

So with regards to the redeemable, Doug, that's related to the valuation that we perform usually on an annual basis regarding Helix.

So that's a result of that annual valuation that we do.

And Francis, you have the...

Yes.

So if you'd look at the NovaSeq number, we did come short of the 330, as I said earlier.

And what drove our thinking is, as always, the customers that we are working with in the pipeline and what we are expecting to close over the course of a quarter and then over the course of a year.

As I said, we still have 3 quarters of our HiSeq customers yet to purchase their first NovaSeq.

So you can imagine, we are working with a lot of customers in the pipeline.

And at any given time, we're working with them to estimate when they will take their NovaSeqs.

And the way it played out is, in terms of timing, fewer took their NovaSeqs in Q4 than we were anticipating, but that's just a timing issue.

We're continuing to work with those customers and we expect those customers to come in, in the coming quarters.

And our next question comes from Puneet Souda from Leerink.

Francis, if I could ask about the broader 600-plus customers.

You have S Prime on the market.

You are discounting S1 and S2.

I was hoping to get a sense from you in terms of how do you expect that market to uptake NovaSeq.

There -- I mean, on one hand, you have a shared number of these labs, but on the other hand, they have a little bit less capital budget flexibility compared to the larger customers you have given their grant cycles and projects and so.

So I just want to get your view of uptake there.

And then just on published and sequencing, what needs to happen in order to bring the 5 million plus or the sort of 6 million plus genomes closer to the sequencer?

Sure, Puneet.

So let me jump in on the NovaSeq upgrade cycle first.

So as you pointed out, it's the smaller labs that typically have less flexibility around quick access to capital.

And so as we were planning the multiyear NovaSeq upgrade cycle, our strategy was really to target the larger labs first.

And so what you saw, what has come out with the S4 a while ago and really around 2018, we engaged with a lot -- larger customers, a lot of the larger high-throughput customers that has more ready access to capital and because of the high volumes of sequencing that they do, so we can have a shorter payback time.

And so that was the story for us for 2018.

Yes, we did some NovaSeqs across the board and we did sell NovaSeqs to new-to-sequencing customers too, but really our focus was really taking the S4.

And you can see that in the remarks around S4 being the majority of the NovaSeq flow cell revenue in 2018.

In 2019, we now had time from the smaller labs to think about that NovaSeq upgrade and to either write the grants that they needed to or get access to budgets that they needed to.

And so this is the year that we want to activate that part of the HiSeq user base.

And to do that, we needed some very specific offerings.

So for example, there are some customers, some HiSeq 2500 customers that really use the rapid run mode a lot.

And so what they wanted was a flow cell that gave them lower run costs, fast run time and be able to run fewer samples and, therefore, need less data output.

And that's exactly what the S Prime is targeted at.

Now, it's coming out this quarter and it's intended to catalyze that customer, that part of the customer base, right?

The HiSeq 2500, customers that love rapid run.

And then our other HiSeq 2500 customers that have 3000 and 4000, that wanted a flow cell that can cost in, in the 12- to 18-month period, and those are not necessarily very small.

They could be labs that do $0.5 million of business in with us a year in consumable revenue.

And for them, the new pricing around S1 and S2 is a really good fit.

And so this is the area we want to catalyze the smaller HiSeq customers and that's the arsenal we have, the S Prime, the S1 and S2 at their new prices.

And then the second question you had was around population sequencing.

And your question was around so what does it take to catalyze those opportunities?

And there's a number of things that takes.

And what we did a few years ago, actually starting with GeL, was we've created in Illumina a team that's dedicated to population sequencing efforts.

So that's the team that coordinates across Illumina, but represents the point of engagement to those 50-plus population sequencing opportunities that I talked about.

And depending on the initiative, there are different roles that we play.

We're deeply connected in some countries with the scientific and medical community to explain the value of a population sequencing effort and help the local medical and scientific community articulate the population benefits to the citizens.

In some cases, we put together and are able to share reference architecture around how you would actually set up a population sequencing effort.

And in other cases, we're able to convene groups from nations around the world with representatives from Genomics England, for example, where they can share their experiences around what they learned, even in things like how you set up the cohorts for sequencing or how you collect samples.

And so those are the roles that we play.

Different initiatives around the world need different catalysts and that's what we set up.

And our next question comes from Dan Arias from Citigroup.

Francis, just going back to the sequencing consumables outlook, is there any way you can add anything to the way in which you're thinking about the pace of Xs and HiSeqs coming off?

If I look at my model, the decommissioned number for several of the quarters in '18 was a lot lower than what we have, but it looked like is stepped up in 4Q.

So just wondering if you can help at all with the assumption there.

And then maybe relatedly, I'm curious if the beginning of the year is sort of broadening your visibility into some of the big projects that very obviously stayed on the Xs in 2018.

Yes.

So we think about what we're looking for in '19.

I talked about the fact that '18, we were starting to catalyze the high-throughput genome centers.

And in nearly all cases, what they did was they would ramp up a NovaSeq but keep their existing, either X going and they were validating their workflows in NovaSeq.

And so while they were using their NovaSeqs and so you see that in the high utilization numbers, a lot of them were maintaining parallel seqs.

So one of the things that we're looking for in 2019 is to see that workflow -- more of those workflows move off the Xs and from the HiSeqs onto the NovaSeqs.

And so that's the dynamic, in the very large genome centers and the high-throughput, that we expect to see play out this year.

The other dynamic I said that we're looking for this year is with that driver, if you like S Prime, S1 and S2, to really catalyze the smaller genome centers.

And so that's going to be another driver that comes online this year that should generate additional demand and activate another part of the HiSeq customer base.

And if I can add one thing, Dan, as well.

Decommissions in the past, as we've talked about, have not always been tied to necessarily the replacement cycle and how customers are actually replacing HiSeqs for NovaSeqs.

We've had some choppiness over the course of '18 with regards to decommission.

You're right, Q4 was a bit higher in terms of -- was a higher quarter in terms of decommission.

And the other thing I would remind you as well is that 75% of our HiSeq customers still have yet to take their first NovaSeq.

So we do expect this upgrade cycle to be still in its early stages and will play out over the next quite a few quarters.

Yes.

And then if I could just, Dan, add color on so what does that mean in terms of consumable pull-through?

So what we are seeing is that customers that have both, that are in that mixed mode phase where they have either Xs and HiSeqs with NovaSeqs, they spend more in terms of consumable dollars with Illumina than the ones that just have the Xs or the HiSeqs.

And so we are seeing the entry of a NovaSeq into an environment catalyze additional sequencing activity, which is obviously good news.

And our next question comes from Bill Quirk from Piper Jaffray.

Francis, how are you thinking broadly about the oncology market here in 2019?

And I guess, to be a little specific, I believe a number of your customers should be getting fairly close to some FDA approvals, which should trigger some CMS reimbursements.

So I'd love your color there.

And then secondly, if I read the press release correctly, have you actually filed for approval for the TruSight Oncology array -- or excuse me, panel?

Yes.

So we've already engaged with the FDA and we recently announced that we have gotten from the FDA a Breakthrough Device Designation for the TruSight Assay.

And so what that means is we will get from the FDA a higher priority to get that assay through the process to get IVD clearance.

What that means specifically is that they give you higher priority in the queue.

They dedicate a team to work with you for the application.

They review your application in advance to make sure that they are comfortable with the design, the data requirements.

And that, as you are going through the clearance process, that they will add additional resources if they need to, to get the application through the process in a speedy way.

In addition, while PMAs don't always require PAI inspections, this one will not because it's a breakthrough-designated product.

And so it's a real positive to be recognized by the FDA for this designation.

So we are in process with the FDA.

We will be submitting with their help to get designation.

As you pointed out, there is just a lot happening in oncology.

And so the story of oncology and NGS over the last few years was primarily a research and translational story.

And now, we're seeing a lot more of a clinical story showed up, partially because we're seeing just more therapies on the market that require a companion NGS test.

And the approval by the FDA last year of larotrectinib as the first precision oncology therapy that is solely driven by the presence of a genomic biomarker is a really big step forward.

So we're seeing more therapies, whether it's solely driven by a genomic biomarker or like Keytruda where you're getting expansion of indications based on genomic biomarkers.

So you're seeing more therapies that require NGS testing.

You're seeing a more favorable reimbursement environment, as you talked about.

We had the national coverage decision earlier in the year.

We've seen more private payers and commercial payers reimburse for NGS testing.

And so we're seeing a lot of momentum build up in terms of having oncology drive NGS testing in the clinic.

And our next question comes from Jack Meehan from Barclays.

I was hoping you could provide some additional context into the difference we're seeing on the array consumable and service line in the quarter.

You obviously have some exposure to consumer on both lines.

So what was driving the underlying acceleration on the consumable side of things?

And can you break out, between the 2, the expectations for 2019?

Yes.

So in general, maybe starting with 2019, as we said, we obviously haven't talked about specific customers, but we've talked about the fact that we've taken a more cautious view on the array growth and specifically DTC growth in 2019, as Francis mentioned earlier, as a result of our conversations, our work with our customers in the DTC space and really looking at their forecast.

With regards to 2018 and specifically regarding your question in Q4, we've talked about -- I mean, there are some dynamics within the quarter that we've talked about before.

With regards to one customer, for instance, going back to Q3, we had some ramp-up related to instruments, some iScans that they purchased from us ahead of the holiday season.

And then in Q4, we had some orders regarding consumables that helped the quarter.

And with regards to array services in Q4, there was an expected drop with regards to the timing of array services and the samples that we process.

So these can be choppy.

These can be seasonal as we've talked about in the past.

And we do expect in the future that, that same seasonality maintains.

And our next question comes from Dan Leonard from Deutsche Bank.

So first off, for Sam, just wanted to check my math.

If I add up the different pieces of your Q1 outlook, I'm coming up with something in the zip code of $820 million in revenue.

So first, is that right?

And then secondly, you mentioned that the linearity in the quarter was more back end-loaded.

Is there anything that drove -- that you could identify that drove the difference in linearity in the quarter?

So let me start with Q1, Dan, and I'll reiterate some of the comments that I made on the call.

Essentially, we're -- you should look at instruments being down roughly $50 million from Q4 to Q1.

With regards to sequencing consumables, we expect them to be flat to slightly increasing.

And keep in mind that we did mention that there was a $5 million stocking order in Q4 that was accelerated from Q1 '19 as well.

And then, we talked about also arrays slightly increasing as well in the high single-digit range from Q4 to Q1.

So you're a little bit light on the $820 million that you mentioned, but obviously you can put the pieces together to come up with what Q1 looks like.

And then we talked about sequencing services being roughly flat as well from Q4 to Q1.

With regards to -- I think you were referring to Q4 linearity when you said in the prior quarter.

So really, there was nothing specific about that.

Usually, we do have, in Q4, linearity is more tilted towards the back end of the quarter as we approach the end of the year and that's typical in prior quarters that we've had.

So I wouldn't point to anything that was unusual than that.

And our next question comes from Mark Massaro from Canaccord.

Francis, I wanted to ask about NIPT.

First, do you expect any change from ACOG as it relates to average-risk NIPT?

And then secondly, has your guidance for '19 included any change?

And then finally, can you give us an update on when the Harvard Pilgrim study would read out and when you might be able to drive or prove utility in that setting?

Sure.

So let me start with ACOG.

So as you know, ACOG withdrew its Practice Bulletin 640, which question the utility of average-risk NIPT.

And I think as an industry, everybody expects that that's a precursor to them putting in a more favorable guideline in place.

It's hard to predict when that happens.

We haven't built in an expectation around that change into our plans.

And this time line between when that guideline changes and when you actually start to see revenue flowing.

And so because of that, we frankly haven't built in big expectations based on any ACOG changing.

In terms of the Harvard Pilgrim study, we should start to see data from that come in over the back half of this year.

And so in that time frame, you should start to see some results come up.

And then, I think there was a question from Marc around the -- whether our guidance anticipates average risk.

So that wasn't...

Yes.

And it doesn't

And our next question comes from Daniel Brennan from UBS.

So Francis, with the X and the higher-throughput, HiSeq customers and the majority of which who haven't adopted NovaSeq yet, is there anything different in 2019, whether it be pricing or marketing or strategies to kind of support your 2019 kind of placement outlook?

I know obviously on the low end, you're making some price adjustments in order to stimulate demand, but I'm wondering more on the high end.

Yes.

In the high end, we're not changing the prices of S4.

Obviously, there's lots of work being done by our commercial organization, primarily working with our customers to plan the rollout, help them with any validation that they need, and that's the work.

A lot of them, as you can imagine, have either bought a NovaSeq.

So the very largest customers have a NovaSeq.

And it's really about doing the entire fleet cut over.

So they will do it at various paces depending on the customer you are talking about.

And so beyond the usual commercial engagement with those customers and then as you can imagine, we are close to all the very large high-throughput services providers out there.

Beyond that, we're doing nothing exceptional.

And our next question comes from Sung Ji Nam from BTIG.

Was wondering if, Francis or Sam, if you might be able to comment on, for microarrays, what the growth rate was for the non-consumer genomics business in 2018 and then what's embedded in your outlook for 2019?

Yes.

We haven't specifically disclosed that, Sung Ji.

But usually, that's a more mature business that grows in the low to mid-single digits.

And that's a more predictable business, unlike DTC where we've seen significant growth over the last 2 years.

So I would say that's a good proxy as well for '19, although we haven't disclosed that specifically.

And our next question comes from Catherine Schulte from Baird.

For the customer mix outlook for NovaSeq in 2019, is it now much more of a HiSeq upgrade story from here?

Or do you think you'll still get 25% to 30% of orders coming from benchtop or new-to-sequencing customers?

And then on the flow cell side, you talked about S4 being the majority of NovaSeq consumables.

How much mix shift are you expecting away from that in 2019 as you catalyze the lower-throughput users?

Sure.

One of the really exciting things about the NovaSeq story so far has been the 30% of NovaSeq orders that have come from new to high-throughput customers.

And that -- it surprised us on the upside and it's been incredibly resilient.

If you look at our results now, over almost 2 years now, taking orders for NovaSeq and shipping NovaSeq, it's stayed around that number.

And so we expect that to continue.

In fact, if anything, you can construct a story of a having S Prime and having the new pricing in S1 and S2 will make NovaSeq even more accessible to customers who are new to high-throughput.

So I believe that number has proven to be remarkably resilient and I expect it to continue to be a strong contributor in terms of NovaSeq orders going forward.

In terms of mix going forward, the S4 customers represent some of our biggest sequencing customers.

So I expect the S4 to continue to grow as we go into '19 and beyond, but you should expect to see a bigger contribution obviously from S1 and S2, but then also from S Prime, which is going to be new to the market coming out in Q1 and build its revenue base over the year.

And our last question comes from Patrick Donnelly from Goldman Sachs.

Maybe just building on that question.

Given the focus on the smaller labs with S Prime, price reduction on S1 and S2, further incentivizing HiSeq customers to convert over to NovaSeq, is there potential for utilization or pull-through of NovaSeq to mix down as some of those lower volume users come online?

I'm just trying to get a better feel for how the moving parts could impact utilization going forward on Nova.

Yes.

So first off, look, we're going to keep our focus on the very large labs and we have teams that are dedicated just to those labs.

So we're adding a focus now and looking to activate the smaller labs deliberately in '19.

That's an incremental focus on top of the focus that we have on the very largest labs, so it's an and.

The -- there will be likely a change in the overall pull-through rate for NovaSeq as these new flow cells come on the market and build out their revenue base.

At this point, frankly, it's too early to call what that impact will be.

And that's partially why we haven't given you the pull-through range yet on NovaSeq.

And so we want to see those flow cells in the market.

We want to see utilization sort of normalize on them before we're able to sort of call an overall blended pull-through rate on the NovaSeqs.

And I know that a lot of you on the call are waiting for that.

And when we have that, we will certainly bring you that, but it's too early to call what the impact will be as these flow cells come on to market.

And this concludes our question-and-answer session.

I'll now turn the call back over to Jacquie Ross for final remarks.

Thank you.

As a reminder, a replay of this call will be available as a webcast in the Investors section of our website as well as through the dial-in instructions contained in today's earnings release.

Thank you for joining us today.

This concludes our call and we look forward to our next update following the close of the first fiscal quarter.

Thank you, ladies and gentlemen.

This concludes today's conference.

Thank you for participating and you may now disconnect.