Image source: The Motley Fool.
NanoString Technologies Inc (NSTG 5.33%)
Q4Â 2018 Earnings Conference Call
March 07, 2019, 4:30 p.m. ET
Contents:
- Prepared Remarks
- Questions and Answers
- Call Participants
See all our earnings call transcripts.
Prepared Remarks:
Operator
Good day, ladies and gentlemen, and thank you for standing by. Welcome to the NanoString 2018 Fourth Quarter Financial Results Conference Call. At this time, all participants are in a listen-only mode. Later, we will conduct a question-and-answer session and instructions will follow at that time. (Operator Instructions) As a reminder, this conference call is being recorded.
I would now like to introduce your host for today's presentation, Mr. Doug Farrell. Sir, please begin.
Doug Farrell -- Vice President of Investor Relations & Corporate Communications
Thank you, operator. On the call with me today is Brad Grey, our President and CEO; as well as Tom Bailey, our CFO.
Earlier today, we released our financial results for the fourth quarter and fiscal year 2018.
During this call, we will make a number of statements that are forward-looking, including statements about financial projections; existing and future collaborations; future business growth trends and related factors; prospects for expanding and penetrating our addressable markets; our strategic focus and objectives; and the development status and anticipated success of recent and planned product offerings. Forward-looking statements are subject to risks and uncertainties, many of which are beyond our control. Those risks and uncertainties are described from time to time in our SEC filings. Our results may differ materially from those projected on the call today, and we undertake no obligation to update those statements publicly.
Now I'd like to return the call to Brad.
Brad Gray -- President & Chief Executive Officer
Thanks, Doug. Good afternoon and thank you for joining us today. We've entered an extremely dynamic period for the Company. Following a strong 2018 performance, we began the new year with a renewed sense of confidence in energy. We have returned our core business to double-digit growth and generated substantial interest in our GeoMx Digital Spatial Profiler. We are currently defining a new open-ended market in the field of Spatial Genomics that shares many characteristics and customers with the rapidly growing field of single-cell genomics. When we commercially launch our GeoMx DSP less than a month from today, we will kick-off a product cycle that we expect to both accelerate our revenue growth and transition us into a multi-platform company. 2019 is set up to be a very exciting year.
During my prepared marks, I'll provide some highlights on our overall performance in 2018, before outlining our strategic objectives and key milestones for the current year. I'll then turn the call over to Tom to review our operating results for the fourth quarter and our financial outlook for 2019.
During 2018, our team did a fantastic job of accelerating the growth of our core nCounter business, and I am extremely proud of their performance. We generated growth of 16% in our product and service revenue, a roughly fourfold improvement over our growth in 2017. The commercial team's focused and execution ensured that the investments we made in our sales channel really paid off, and we intend to sustain this momentum in 2019.
During 2018, we grew our installed base of nCounter Systems to about 730 systems at an increase of more than 20% year-on-year. We continued to broaden nCounter sales channel, -- sales into smaller labs with our lower cost SPRINT instruments accounting for 50% of new nCounter placements. In addition, we successfully expanded into new markets such as immunology and neurology, and non-oncology applications accounted for an impressive 40% of system sales, up from only 15% in 2017.
We generated total consumable sales of more than $53 million last year, an increase of 18% over 2017. This growth was the combined result of our strength in commercial channel and the 11 new panel products that we introduced last year. The largest market remains oncology, where our cancer-research panels grew by about 30%, led by our immuno-oncology portfolio, including our number one selling PanCancer Immune Profiling Panel and our fastest growing IO 360 panel. 2018 also saw meaningful consumable growth outside oncology, including immunology and neurology panels, which grew 65% year-on-year.
Finally, Prosigna delivered another successful year with 40% growth. In parallel, we successfully advanced our pre-launch activities for GeoMx DSP without distracting attention from our core nCounter business. We expanded our technology access program and have completed more than 70 projects for more than 50 customers. We've designated three leading research centers as GeoMx Centers of Excellence and have installed seven beta GeoMx instruments at customer sites.
Perhaps the most impressive achievement was securing pre-orders for 30-plus GeoMx instruments, and the clearer sign yet that we have seen of the market's hunger for spatial profiling and the compelling nature of the GeoMx product profile. A full 70% of these pre-orders came from customers outside of our TAP program, who purchased GeoMx site unseen.
Overall, we bring great momentum into 2019, and our objective is to build on this momentum by focusing in three areas. Our first objective is to sustain double-digit growth in our core nCounter business by continuing to execute on the strategies that made us successful in 2018. This starts by selling a similar number of new nCounter instruments to what we delivered in 2018.
We expect that SPRINT instruments will once again account for about half of new instrument placements, as we continue to diversify in the smaller labs and research areas beyond oncology. The pace of nCounter instrument sales is expected to strengthen over the course of the year with some GeoMx system placements in the second half of 2019 being bundled with new nCounter systems.
We expect that the most important driver of nCounter growth in 2019 will be consumable sales. We began the year with an nCounter installed base that is 20% larger than this time last year and believe that the combination of a focused consumable sales channel and compelling products can maintain the annualized consumable sales at $75,000 to $80,000 per system rate that we delivered in 2018.
We planned a modest expansion of our highly effective consumable sales channel, adding key account managers and inside sales reps who can service this expanded customer base. In addition, we plan to continue launching new panel products, expanding into markets outside of oncology by introducing panels developed in collaboration with leading experts in fields such as metabolism and fibrosis.
On the diagnostics side, we expect to continue to grow Prosigna revenues, driven primarily by recent positive health technology assessment such as the NICE decision in the United Kingdom. Both Prosigna and LymphMark continue to garner attention from researchers looking to identify biomarkers that select patients for therapy, as illustrated in December by the presentation of a total of over 80 nCounter-based abstracts at the American Society of Hematology or ASH meeting and at San Antonio Breast Cancer Symposium.
While we are not expecting to generate revenue from our LymphMark assay during 2019, we do expect to see the results from the ongoing Phase III robust study, which could be followed by regulatory submissions.
Our second strategic objective is to launch our GeoMx Digital Spatial Profiler on a trajectory for long-term success. Over the last two years, we've executed a comprehensive market development plan for GeoMx, and during that time, we've become convinced that Spatial Profiling will evolve large market with diverse applications. The interest in GeoMx is coming from two primary markets. The first is translational research, with a search for biomarkers is driving the need for spatial analysis of protein and RNA expression. The second target market is more basic discovery research where scientists have embraced single-cell RNA sequencing to discover previously unknown cell populations, and are now hungry for technologies that will show these -- where these cells are located within tissue.
We have engineered GeoMx with the capabilities required to serve both of these markets. Our 2019 launch is focused on translational researchers who value the GeoMx that GeoMx can provide both RNA and protein profiling and has the high throughput required to efficiently process hundreds of FFPE clinical trial samples, as read out on an nCounter system.
In 2020, we intend to expand into basic discovery research by introducing ultra-high-plex RNA expression assays that can be analyzed using ubiquitous, next-generation sequencers, following the strategy that we believe will replicate the success of the most popular single-cell technologies.
While GeoMx will not be the only technology used for spatial profiling, we believe that our early entry into the market and unique product profile will leave us well positioned to capitalize on this growing field and to help shape how the market develops.
Last week, we seized on the opportunity to present -- presented by the Advances in Genome Biology and Technology or AGBT meeting to hold the Inaugural Spatial Genomics Summit and to introduce the power of GeoMx to researchers at the bleeding edge of Discovery Research. Our Spatial Genomics Summit drew a crowd of nearly 150 attendees, many of whom flew in a day early for the event, and it set the tone for a busy week.
During the plenary session on the first morning of the conference, our collaborators from the Broad Institute highlighted GeoMx as an important new spatial capability, sharing data generated in their labs during the first few weeks after receiving their Beta instrument, which profiled 1,400 RNAs across six different tumor types.
Over the days that followed, spatial genomics emerged as a major theme of the meeting, as presentations by researchers using GeoMx and other emerging technologies created substantial buzz. This included a total of three oral presentations and six posters containing GeoMx DSP data as well as a workshop featuring two of our technology access program customers who presented their experience in front of an audience of nearly 300 researchers.
In our annual survey of AGBT attendees, performed by the DeciBio group and published online earlier this week, spatial profiling was identified by AGBT attendees as the most exciting theme of the meeting with GeoMx as the top of mind associated technology. Overall, we stimulated substantial interest in GeoMx and generated hundreds of new leads that we look forward to beginning to convert into orders over the next 24 months.
Our official commercial launch of GeoMx is now just a few weeks away and will take place during the annual AACR Meeting in Atlanta. The launch will target the cancer-focused translational researchers, who form the core of our nCounter customer base, and will unveil details of the system's specifications, consumable panels and pricing.
Our launch event will include presentations by TAP customers from Stanford University and our early access customers at the Mayo Clinic. In addition, at least 11 studies will be presented by biopharma companies and academic researchers that include digital spatial profiling data generated under our Technology Access Program. These studies will build upon the three peer-reviewed publications and more than a dozen abstracts previously presented and should provide compelling illustrations of the unique capabilities of the GeoMx platform.
The April launch will be followed by the first installations of commercial GeoMx systems beginning during the third quarter. While we look forward to spending the second half of 2019 getting GeoMx into the hands of the leading research institutions, it will be even more important to position GeoMx for success in 2020 and beyond.
Therefore, our top priority will be to ensure that customers have an exceptional user experience. This will require that we manage the pace of installation to ensure that customers enjoy the level of service and support that they have come to expect from us.
During the beta installations over the past several months, we found that approximately three weeks are required to install a GeoMx system, train its users and guide them through their first experiments. While, by 2020, we plan to streamline this process to bring the timeline down to just two weeks per system, we expect to operate under this high-touch model through the balance of 2019.
As a result, during the second half of 2019, we expect to install approximately 25 to 30 of the GeoMx pre-orders received last year. The incremental orders that we booked during 2019 are likely to go into a wait list and be fulfilled in 2020. We believe that the GeoMx launch will accelerate our revenue growth beginning in the second half of the year and will keep us busy through 2020. Given the magnitude of the addressable market opportunity, which we estimate will grow to about $4 billion in 2020, it is critical that GeoMx receive the attention and resources it deserves.
In parallel to commercializing GeoMx, we will continue to invest in our third strategic objective, which is to advance our Hyb & Seq platform toward the commercial launch in 2021. In 2018, the Hyb & Seq program made substantial progress. We scaled up our chemistry to contain the entire 4,096 sequencing barcodes and incorporated the use of ordered arrays, which have allowed us to increase the throughput of our platform about 25-fold. Our contract manufacturer has now delivered 10 prototype instruments, each of which is roughly 100 times faster than the modified SPRINT instruments previously used in our proof-of-concept studies.
During the AGBT conference last week, we presented three studies that illustrate the progress we're making and highlighted key clinical applications. In the field of oncology, our collaborators at Cambridge University presented the first data from our fully scaled sequencing chemistry, where they used all 4,096 Hyb & Seq barcodes to sequence the entire coding region of the TP53 gene using a sample-to-answer workflow.
In the field of infectious disease, our collaborators at the Broad Institute presented a unique -- technique for Hyb & Seq-enabled culture independent pathogen identification and rapid phenotypic antibiotic susceptibility testing.
Finally, NanoString's own scientists presented a novel end-to-end sequencing analysis pipeline called HEX (ph) similar design to assemble Hyb & Seq readout and accurately detect genomic variants. Hyb & Seq a healthy amount of attention at AGBT and potential customers seem to understand and appreciate the value proposition of providing simplicity and rapid turnaround time. We believe that the NGS market is hungry for alternative platforms and that Hyb & Seq is well differentiated in the competitive landscape.
During 2019, we're focused on reducing the sample input requirements, increasing the number of targets simultaneously profiled and refining the use of the platform for gene expression. The next major technical update on Hyb & Seq is slated for the Association of Molecular Pathology Meeting during the fourth quarter, when we plan to present data from a solid tumor panel. In the meantime, we look forward to updating you on our progress on this program during future calls.
Now I'll turn the call over to Tom to review our financial results for Q4 and outline our financial targets for 2019.
Thomas Bailey -- Chief Financial Officer
Thanks, Brad. I'll begin by reviewing our fourth quarter and fiscal 2018 results, and then I'll conclude with our outlook for 2019. For the fourth quarter of 2018, product and service revenue was $23.6 million, which was above the high-end of our Q4 guidance and represents a year-over-year growth of 12%. Our product and service revenue included $5.9 million in instrument sales and $15.3 million in consumable sales.
Life science consumables revenue, excluding Prosigna, was $13.1 million, reflecting 14% year-over-year growth, driven by strong panel adoption.
Prosigna sales were $2.2 million, an increase of 24% over the prior year. Service revenue was $2.5 million with a 36% growth posted over the prior-year, driven by our Technology Access Program for GeoMx DSP. Revenue from collaborations was $6.4 million, and we received cash from collaborators of $3.8 million. The majority of cash received and collaboration revenue recognized was derived from our partnership with Lam Research, with the balance derived from our Celgene collaboration.
Gross margin on product and service revenue was 54%. This was primarily driven by instrument sales mix as well as year-end cleanup of some non-cash items.
R&D expense was $16.5 million, an increase of 21% over the prior year. This increase was driven by our investment in Hyb & Seq, which is fully funded by Lam Research. Our SG&A expense of $20.3 million increased by 3% over the prior year. This was the result of an increase in professional fees incurred from our Sarbanes Oxley compliance and audit activities.
For the fourth quarter, stock-based compensation expense was $2.8 million and depreciation and amortization expense was $1.1 million. We ended the quarter with $94 million of cash, cash equivalents and short-term investments. For the full year 2018, our product and service revenue was $83.5 million, above the high-end of our previously raised guidance, representing 16% year-over-year growth. Gross margin was 57%, in line with our guided range.
Our total 2018 R&D expense was about $1.6 million above our guidance, primarily driven by investments made to prepare for and support the launch of GepMx DSP. Our total SG&A expense was in line with our guidance and cash used in operating activities with capital expenditures was $59.6 million, also in line with our guidance.
I'll conclude my comments now with our financial outlook for 2019. For 2019, we expect product and service revenue of $98 million to $103 million, representing total annual growth of 17% to 23%. We expect revenue growth in the first half to be driven primarily by nCounter consumable products and for instrument revenue to strengthen in the second half of 2019 following the commercial launch of GeoMx.
We expect $92 million to $95 million of that total product and service revenue would come from our base business. We expect our core nCounter business trajectory to be consistent with last year with instrument revenue in line with 2018 and total consumable pull-through of $75,000 to $80,000 per installed system, including Prosigna.
We expect to see a similar seasonal pattern of roughly 45% of our base nCounter revenue recorded in the first half of the year and 55% in the second half. We expect to recognize from $6 million to $8 million of GeoMx revenue in the second half of the year, reflecting the installation of approximately 25 to 30 instrument pre-orders that we received under our GeoMX Priority Site program in 2018. We expect total revenue of $118 million to $123 million, which includes approximately $20 million of revenue recognition from collaborations. It's worth noting that in the absence of any new partnering activity, we expect 2019 will be the last year for any material recognition of deferred collaboration revenue.
We expect gross margin to be in the range of 57% to 59%, consistent with 2018. This outlook reflects a balance of the positive gross margin impact of increasing consumable sales in our core nCounter business and the offsetting impact of GeoMx DSP sales, which will consist almost entirely of instruments in 2019.
In 2019, we expect to realize significant operating leverage by growing operating expenses at about one-quarter of the rate of the growth in our products and service revenue. For research and development expenses, we expect to record $62 million to $64 million or just slightly above the amount we recorded in 2018. We expect about one-third of our total planned expenditures to be offset by support payments from Lam.
For selling, general and administrative expenses, we expect to record $82 million to $84 million, an increase of 5% to 7% as compared to 2018 with our overall growth in these expenses being substantially less than our expected rate of revenue growth. As for other items, net interest and other expense is expected to be approximately $7 million with approximately $5 million representing the cash portion of interest expense.
Stock-based compensation expenses are expected to be approximately $13 million. Planned capital expenditures for 2019 are expected to be between $9 million and $10 million, and we expect our general working capital needs may increase in 2019 depending upon the trajectory of our GeoMx DSP orders and installations, which could impact our levels of necessary inventory and receivables.
Lastly, cash used in operating activities and for capital expenditures is expected to total between $55 million and $60 million in 2019, which is approximately the same as 2018. For the first quarter, we expect total revenue of approximately $25 million to 26 million, including $20 million to $21 million in product and service revenue and about $5 million in collaboration revenue.
Now I'll turn the call over to Brad for our closing comments.
Brad Gray -- President & Chief Executive Officer
Thanks, Tom. During 2018, NanoString got its groove back, returning nCounter to double-digit growth. In 2019, we expect to launch the GeoMX Digital Spatial Profiler to accelerate our growth as we transition to a multi-platform company. We believe that the next several years will bring a wave of research enabled by spatial profiling, and we have positioned the Company accordingly. We expect it to be an exciting multi-year ride.
With that, I'll open up the lines for your questions.
Questions and Answers:
Operator
(Operator Instructions) Our first question or comment comes from the line of Daniel Brennan from UBS. Your line is open.
Daniel Brennan -- UBS Investment Bank, Research Division -- Analyst
Great, guys. Thanks for the questions. I guess I wanted to start off maybe with AGBT, maybe some of the feedback there. Brad, I think you mentioned a significant number of leads. But I think the market that you're addressing there is, as you mentioned, more of a 2020 versus 2019 events. Maybe you can just expand a little bit in terms of kind of what you learn there and kind of how you think about that market developing, if anything, this year or if not, then in 2020?
Brad Gray -- President & Chief Executive Officer
Yes. Thanks for the question, Dan. I mean, I think DSP is a product where, at every possible juncture, the demand for spatial profiling has exceeded our previous expectations. This happened with our Technology Access Program, which largely address the needs of our translational research market. It happened with the GeoMx Priority Site program where we pre-sold 30 instruments againthe translational market. And really, it happened again last week where our expectations for the enthusiasm from the basic research community for spatial genomics or spatial profiling was overwhelming. A year ago, we first debut -- at AGBT, we first debuted the opportunity for spatial profiling, and we were surprised at that time by level of interest from classic sequencing jocks in the idea of looking at whole tissues and where cells and genomics vary across those tissues. But this year, we really weren't the only ones talking about that. There were several other companies and their collaborators and customers presenting on the idea of spatial genomics, and it would position very much in the -- as the next logical phase of the Human Cell Atlas effort that's been going on for several years. And so it was really rewarding to hear bright voices from other leading companies as well as institutions like the Broad, who were saying the spatial is really a next big wave. 2019 is going to be a year where we focus primarily on the traditional translation researcher, who really was not represented by the groups who were at AGBT last week. But we will open up the product for next-generation sequencing readout early next year, and that's really when we expect to unlock the demand that's characterized by the enthusiasm at AGBT. I mean, that being said, we are following up on literally hundreds of leads that we captured at our symposiums, at our workshops and on posters in our booth last week, and we'll spend this year really developing those leads into the actual purchase orders with the hope of having spatial genomics enabled at leading genome centers in 2020. But overwhelm -- I mean, I think overwhelmingly, we felt positively surprised about the degree to which the community seems to be embracing spatial as the next frontier in genomics.
Daniel Brennan -- UBS Investment Bank, Research Division -- Analyst
Great. And then maybe just as a follow-up, in terms of the placement number for this year, in terms of the focus on making sure this goes as high quality launch possible, can you walk through like the thought process there in terms of possibly given the large nascent or growing market with a lot of interesting new competitors coming and kind of how you're balancing the focus to please customers at the same time we're trying to be a first mover and kind of place -- kind of capture the share that's in front of you, not make you customers wait too long in terms of maybe additional hiring that you might do or how you kind of plan to ramp the kind of the installation process? Thanks.
Brad Gray -- President & Chief Executive Officer
Yes. So the first thing that we're focused on of course is capturing demand where it exists, even well ahead of the actual fulfillment of that demand. And you saw us, I think, execute that very well in the second half of 2018 with the GeoMx Priority Site program, which captured committed orders from people well in advance of six months of instrument delivery. We will continue doing that throughout 2019, we are not standing still. We do expect to continue to take orders, even though the fulfillment of those orders may take some time. People will get in line for exciting new technology knowing that they'll get access as fast as they can. So we don't expect that our moderation of install pace puts us is any sort of competitive disadvantage or any risk of missing the market opportunity.
When you ask sort of how do we arrive at an estimate and the guidance to expect installations of 25 to 30 systems in the second half, let me kind of walk you through that logic. Remember, so far, we've installed seven systems. So our experience in terms of installation and customer support is still relatively early in its development. Those systems each took a team three weeks to go ahead from all the way from installing the system to training the customer to actually hold in their hands their first experiments. Now that means that if we start early in the second half of the year, the two teams I have up and running right now will be able to get 16 installations done. And if I scale that up to three teams, I'll be able to get 25 installations done. So the bottom end of our range basically assumes that I have three teams continually supporting customers in a high quality way through the balance of the second half. In the top end of my range assumes I scale that up to four teams. So that's really how we arrived at that. I think we will, of course, revisit as we gain more experience in installations, what the most aggressive responsible pace of installation might be, and we'll provide you updates on that over the course of the year. But based on the seven installations we've done and our desire to provide absolute white-glove service, we think this is the right place for us to initiate guidance for the year.
Daniel Brennan -- UBS Investment Bank, Research Division -- Analyst
And maybe one final follow-up. Just will you provide what should we expect in terms of color on the pipeline. I know you gave the 30-plus orders at JPMorgan. And I'm sure you don't want to give too much clarity on this, but since this is such an exciting part of the story, what should we expect on that front in terms of clarity going forward in terms of the pipeline of DSP orders? Thank you.
Brad Gray -- President & Chief Executive Officer
I think the next earnings call that will likely have will be our Q1 call, which traditionally is held early in May. By that point in time, we'll have finished the first quarter and we'll have been through our AACR our launch event. I think we'll be able to provide some color on the nature and volume of customer dialogs at that time. That being said, I mean, we will only be less than a month after the official commercial launch of the product. So I don't expect a miraculous wave of orders in the first few days after our full commercial launch. So I think we'll be -- we'll provide some degree of color, but as you said, it will probably be reticent about absolute transparency on the order flow and funnel. But we'll do our best -- we'll do the best we can to provide commentary over the course of the year on how demand is evolving in this very important part of our business. We understand the hunger for that.
Daniel Brennan -- UBS Investment Bank, Research Division -- Analyst
Great. Thank you.
Operator
Thank you. Our next question or comment comes from the line of Tycho Peterson from JPMorgan. Your line is open.
Tejas Savant -- JP Morgan Chase & Co, Research Division -- Analyst
Hey, Brad. It's Tejas on for Tycho. Just wanted to get some color on the supply chain for DSP. I know in the past you've talked about the box being sort of fairly optimized at this point in terms of the supply chain and there's some work to be done on the consumables front. Can you just give us an update on where that process stands today?
Brad Gray -- President & Chief Executive Officer
Sure. So our GeoMx instrument will be -- as all of our instruments are here at NanoString, will be contract manufactured for us. We placed an order for commercial instruments that should more than fulfill the 2019 demand for installations. We'll expect to receive those instruments late in the first half of the year and sometime in the second quarter, we'll want to bring those in and put those through internal QC measurements, load them with software and make sure that they're as robust as expected before we begin to place those in the hands of customers in the third quarter. I don't see a lot of risk in the timeline of receipt of instrument inventory from our contract manufacturer.
In terms of the development timelines, as we've discussed in the past and people who already at AGBT saw firsthand, the software that allows customers to interact with the imaging data, the region of interest selection and then the co-visualization of the image in the GeoMx data is a hugely important part of the customer experience, and we've invested very heavily in that, getting our software right ahead of launch. In terms of the development timelines, that's the most important set of work items that remains to optimize and debug. People who were at AGBT saw a fully functional set of software capabilities. So you know that we are very far along the line there. Again, I don't see a large risk in terms of completion of that software activity ahead of the second quarter.
So I'd say overall, it's an interesting launch situation to be in, and those launch situations are in a sense demand-moderated in terms of their pace. I don't think that's the situation we're actually in here. Some are supply moderated. I don't actually think we're in that pace either. We're in a situation where we're modulating the pace of revenue recognition on launch by customer experience. What's the fastest we can bring customers while continuing to delight them? And that's good because that's something that we have a reasonable degree of control over, and we will of course be making the right balance of investments between scaling up our capacity there, meeting the potential for revenue growth and thinking long-term about the customer experience.
Tejas Savant -- JP Morgan Chase & Co, Research Division -- Analyst
Got it. That's helpful. And then just turning to the base business, you've talked about nCounter and SPRINT revenue expected to be largely flat year-over-year. Can you just walk us through your underlying sort of assumption in terms of what's going on in the market that is making you call for a flat year-over-year growth profile there?
Thomas Bailey -- Chief Financial Officer
Sure. Flat instrument revenue is what we've delivered for approximately the last several years. 2017 was actually down from 2016, 2018 was just 3% growth over 2017. So for the last several years, we found ourselves in a situation where we're selling into oncology market that we've done a pretty good job of penetrating already, we are mostly selling into new smaller accounts with our SPRINT instruments, maybe further down in the market than we have penetrated in the past. That's well of small biopharma companies that are just in the company formation stage. And then we are diversifying beyond oncology into areas like immunology and neurology, which are new markets for us. So I think that trend will very much continue in 2019, it will be the same set of dynamics penetrating those parts of the oncology market that we haven't yet reached and then continuing to move beyond oncology. Remember that we're asking our same incident rep at the same time to take on the burden of selling GeoMx instrument as well in coordination with our technical sales specialists that we've added. So what we're actually asking of each individual in our channel is quite a bit more this year in terms of order capture. So it's not like we're standing still, but I think from our nCounter business, by continuing to place the same number of instruments every year with our very healthy consumable pull-through stream, we can still generate double-digit growth for the next several years on that platform while of course layering on top the GeoMx launch.
Tejas Savant -- JP Morgan Chase & Co, Research Division -- Analyst
Got it. And then one final one here on collaboration revenue. It seems -- looks like it came in a little bit light versus at least our models. So was it all related to Lam revenue recognition there? And also can you walk us through drivers behind the slight sort of push out on Hyb & Seq timing?
Thomas Bailey -- Chief Financial Officer
Yes. I can comment specifically with respect to the revenue (inaudible) Brad (inaudible) comment on Hyb & Seq. The collaboration revenue is -- the revenue recognition is determinant upon the pacing of the program, and so depending upon what activities get done and when exactly they get done that revenue recognition can fluctuate just a little bit from quarter-to-quarter. So it's really an activity based thing as opposed to something that we can always dial-in to perfection, and so some of the fluctuation in the actual revenue recognition is related to that, and that's true to a much lesser extent for the Celgene as well. There is some collaboration revenue -- reported collaboration revenue for Celgene as well. That's the reason for it. It's really a timing thing and then in subsequent periods, we're likely to catch up and you saw some of that fluctuation from quarter-to-quarter happening throughout the course of 2018 and probably be a bit of that this year as well.
Brad Gray -- President & Chief Executive Officer
Yes. Tejas, I'll like the question about the Hyb & Seq timeline. Really there's two things at work here. I mean, first, sequencers are challenging to build and sometimes the development efforts take a little longer than you think, and I think our reframing of the launch timeline from 2020 to 2021 is not really a full year slip, there's a several quarter slip, based at least in part on technical timelines changing. I'd say the other and maybe more strategic answer is a company can only do so many things very, very well. We think with the GeoMx opportunity, we have a real once in a lifetime opportunity to really define a very exciting market, be an early player in a very exciting market, and we want management bandwidth and commercial bandwidth focused on capturing that opportunity over the course of 2019 and 2020, and are reluctant to throw a launch of the new product on top of GeoMx in its very early phases now that we've actually come to market. So that's an important factor as well. By moving it to 2020, we allow GeoMx to fully mature, we allow the Hyb & Seq program to be developed at a natural pace in a high quality way, and we can add another wave of growth beginning in 2021.
Tejas Savant -- JP Morgan Chase & Co, Research Division -- Analyst
Got it. Thanks so much, guys.
Operator
Thank you. Our next question or comment comes from the line of Catherine Schulte from Baird. Your line is open.
Catherine Schulte -- Robert W. Baird & Co. Incorporated, Research Division -- Analyst
Hey, guys. Thanks for the questions. Given your GeoMx placement rate is resources-constrained at this point, would you be open to a potential commercial partnership to ensure a smooth launch or ramp either now or after you open it up to the sequencing market?
Brad Gray -- President & Chief Executive Officer
Catherine, I think our focus is on the customer experience in these first few quarters of launch. And I think now more than ever, we want to control that experience. So in the very near-term, while we are focused in 2019 on selling into our nCounter installed base, we want to make sure to delight our customers. We are not looking to partner with anybody. I don't think we need the partnership and frankly, I'd be a little worried to hand the destiny of this very exciting product to anyone else at this point in time. That could change in the future. I mean, I think the two events that could cause us to revisit partnership would be, of course, opening the platform up for next generation sequencing read-out in 2020. That could bring you a much larger market opportunity to us, and of course, in the longer-term, clinical applications that maybe weren't having a clinical pathology channel for GeoMx sales. But neither of those is an urgent priority for us. So expect to watch us goad alone and hopefully executing a very high quality way, this commercial launch in 2019.
Catherine Schulte -- Robert W. Baird & Co. Incorporated, Research Division -- Analyst
Okay. And then about a year ago, you mentioned that roughly 20% of nCounter users we're interested in GeoMx. Where does that interest level stand today, I assume much higher?
Brad Gray -- President & Chief Executive Officer
Yes. I think modestly higher, it's probably 25% to 30% today on an installed base that's 20% bigger. So if you do the math, that's about a 50% increase in the total number of people who own nCounter today and are engaged. It would have been basically 20% of a 600-instrument installed base a year ago, now it's 25% to 30% of a 730-instrument installed base. So that's a substantial growth and what I'll call the tech translational researcher interest level. Remember, we're still coming up on the official launch. So I expect that number to increase when we go to the major meeting for cancer researchers, the number one meeting of the year, which takes place in just a month's time, and really began to (inaudible) what this technology can do.
Catherine Schulte -- Robert W. Baird & Co. Incorporated, Research Division -- Analyst
Okay. Then last one from me. You mentioned the potential Celgene robust trial read out this year. Any thoughts on timing there and then what are the next steps following readout if that trial is successfil?
Brad Gray -- President & Chief Executive Officer
Yes. I think if you go to clinicaltrials.gov today, which is sort of the official non-confidential resource for the timing of clinical trial readouts, you'll see that the primary analysis is scheduled for a month or two from now, so sometime in the spring. And I would expect that it will be up to Celgene about when and how they make those results public, but following a positive result, we would expect it to begin final preparations for a submission for PMA sometime later in 2019.
Catherine Schulte -- Robert W. Baird & Co. Incorporated, Research Division -- Analyst
Great. Thank you.
Operator
Thank you. (Operator Instructions) Our next question or comment comes from the line of Doug Schenkel from Cowen & Company. Your line is open.
Adam Wieschhaus -- Cowen and Company, LLC, Research Division -- Analyst
Hi there. This is Adam Wieschhaus on for Doug, thanks for taking my questions. I believe you guided 2019 nCounter pull-through to be at the same levels that it was in 2018. Do you believe there could be upside to this range considering your augmented consumable sales force is still ramping in 2018? It seems like the maximum pull through is still quite a bit higher than that.
Brad Gray -- President & Chief Executive Officer
Yes. Thanks for the question, Adam. No, I don't see a tremendous amount of upside to the pull-through at this time. $75,000 to $80,000 was the range we gave last year, the full year range ended up near the top end around $80,000 for the full year. But a couple of things have been going on in our installed base over the last year. One is it's 20% bigger now. And so in many ways, the additional consumable sales channel that we're in the process of adding is addressing that expanded customer base more than kind of increased penetration of our existing customer base. And two is our installed base continues to shift toward smaller laboratories with the spread system accounted for half the new instrument placements. And so that is a very natural drag on consumable pull through per system as smaller labs continue to become a larger fraction of our customer base. So heading into 2019, we think the $75,000 to $80,000 per system ranges is still where we expect to be in this current year.
Adam Wieschhaus -- Cowen and Company, LLC, Research Division -- Analyst
Thanks, Brad. And how would you characterize your initial GeoMx demand relative to level you've seen with your earlier roll-out, such as the SPRINT instrument. And maybe I missed it, but how does that three-week installation period compared to previous roll-out and does that short at all of the customer already have in nCounter?
Brad Gray -- President & Chief Executive Officer
So I mean, first, the demand for GeoMx is unlike anything I've ever seen in my career, and certainly, it's unlike anything that we've seen here at NanoString on past roll-outs. The nCounter system, as distinctive as it was when it was launched in 2009, it was entering a gene expression market that had a lot of different alternative technologies at that time, such as microarrays or quantitative PCR, and it was finding a unique place in the mid plex with a very simple workflow. In contrast, GeoMx is entering a market that's just developing, where there has been tremendous latent demand for spatial profiling at high plex kind of sitting on one hand on the translational research side and on the other kind of sitting alongside single-cell RNA-seq, but there have been no real solutions, and so we're fortunate to have a very compelling product profile in a market with a lot of latent demand, where there's plenty of room for growth without bumping into competitors. And that's a very different situation than the nCounter system was. So I'd say, In addition, we have of course been executing, for 12 to 24 months, a very careful pre-launch market development effort that has used major meetings, our Technology Access Program, et cetera, to educate the market about this opportunity, and so we're coming to market with a certain amount of demand as evidenced by the pre-orders. So it's a very different situation, it's a very exciting one.
In terms of our capacity to install, the three weeks is quite a bit longer than it takes to install an nCounter system, and that's really for good reason. An nCounter system is a very simple to use genomics system with a reasonably common set of procedures that you do from a simple liquid (ph) paneling, et cetera, workflow. In contrast, GeoMx is a totally novel system that asks researchers to interact through software in the selection of regions of interest, the analysis of images and then the merging of those images with data that is genomic and high plex in nature. It's unlike anything that I've ever actually seen before. And so both in terms of the workflow and in terms of the experimental design and analysis, it requires some handhold in the first time to introduce them to these new operations that they haven't done. And right now, that's why it's taking three weeks. Now that number should come down over time as our team gets better at -- and more efficient installation and of course as we learn how to train as efficiently and effectively as possible, we think that number will come down to two weeks, but it'll never quite reach the simple out of the box kind of experience of nCounter. Today, you asked does it help accelerate training meaningfully if you already have an nCounter system, I'd say probably at the margin. Of course, we don't have to teach you how to use an nCounter system if you already own one and know how to use it, but honestly that's a relatively small part of the overall experience of GeoMx, the vast majority is focused on these other operations. So it only makes a marginal increase in the speed with which a site can be brought up.
Adam Wieschhaus -- Cowen and Company, LLC, Research Division -- Analyst
Okay. I appreciate all the color.
Doug Farrell -- Vice President of Investor Relations & Corporate Communications
Thank you. I'm showing no additional questions in the queue at this time. I'd like to turn the conference back over to management for any closing remarks.
Brad Gray -- President & Chief Executive Officer
Thank you very much. If you did miss any portion of the call, there will be a replay available in the next 30 minutes or so. Callers can access that by dialing 888-793-942 for domestic callers. International callers, please dial 734-385-2643, the passcode for the conference call is the same 4080336. Thank you very much for your time, and thanks for joining us today.
Operator
Ladies and gentlemen, thank you for participating in today's conference. This concludes the program. You may now disconnect. Everyone, have a wonderful day.
Duration: 49 minutes
Call participants:
Doug Farrell -- Vice President of Investor Relations & Corporate Communications
Brad Gray -- President & Chief Executive Officer
Thomas Bailey -- Chief Financial Officer
Daniel Brennan -- UBS Investment Bank, Research Division -- Analyst
Tejas Savant -- JP Morgan Chase & Co, Research Division -- Analyst
Catherine Schulte -- Robert W. Baird & Co. Incorporated, Research Division -- Analyst
Adam Wieschhaus -- Cowen and Company, LLC, Research Division -- Analyst
Transcript powered by AlphaStreet
This article is a transcript of this conference call produced for The Motley Fool. While we strive for our Foolish Best, there may be errors, omissions, or inaccuracies in this transcript. As with all our articles, The Motley Fool does not assume any responsibility for your use of this content, and we strongly encourage you to do your own research, including listening to the call yourself and reading the company's SEC filings. Please see our Terms and Conditions for additional details, including our Obligatory Capitalized Disclaimers of Liability.
