ONTTF - Oxford Nanopore Technologies plc (ONTTF) Q4 2022 Earnings Call Transcript

2023-03-21 14:57:04 ET

Oxford Nanopore Technologies plc (ONTTF)

Q4 2022 Earnings Conference Call

March 21, 2023 08:00 AM ET

Company Participants

Gordon Sanghera - Chief Executive Officer

Tim Cowper - Chief Financial Officer

Conference Call Participants

Charles Weston - RBC Capital Markets

Odysseas Manesiotis - Berenberg Capital Markets LLC

Paul Cuddon - Numis Securities

Zain Ebrahim - JP Morgan

David Westenberg - Piper Sandler

Veronika Dubajova - Citigroup

Presentation

Gordon Sanghera

Good afternoon and welcome to our 2022 Full Year Results Presentation. I'm Gordon Sanghera, Chief Exec, Oxford Nanopore, and I have with me today Tim Cowper, our CFO. I will provide a short summary of our financial highlights for 2022 and key business highlights for the year. Tim will then walk you through our financials in more detail. The company's vision is to enable the analysis of anything by anyone anywhere. This slide shows a overview of the company, a summary of the company. I will leave you to read that at your leisure. I would like to move directly to the results summary.

We saw continued momentum in our business in 2022. Our total revenues for 2022 were ?198.6 million. This included a ?51.8 million one-off legacy COVID testing LamPORE revenue, DHSC contract settlement. The core life science research tools revenue was ?146.8 million. Overall growth, including EGP and COVID sequencing was 16%. Excluding EGP and COVID sequencing and currency, the underlying growth was 30%. A key metric for our business is a number of publications. In 2022, our customers published 3,000 papers highlighting the breadth, depth and diversity of Nanopore sequencing applications.

As I've just said, our core life science research tools revenue was ?148.6 million. 76% of this revenue comes from consumable sales. We've increased our customer base from approximately 6,300 to 8,200. Our gross margin increased approximately 250 base points to 56.3%, and we have a cash position of ?558 million of cash and cash equivalents. We are seeing strong momentum across the business. We are delivering on revenue and margin.

We continue to push the boundaries of uncovering the dark genome, as well as continuing to innovate on our platform. We continue to simplify end-to-end workflows as well as launch new applications. We've doubled our commercial headcounts since the IPO to meet the growth and commercial targets we set out at IPO. We continue to make progress in applied market opportunities, and I shall give you a few examples of that later.

Innovation is and remains at the heart of our strategy. On our platform, we launched P2 Solo and we have a strong order book. We also launched our short fragment mode making Nanopore sequencing the only platform that can address any read length from 20 base pairs right through to millions of bases. We are also providing NVIDIA's A100 GPUs, increasing our base calling speed and enabling full methylation analysis.

We continued to upgrade the electronics on our P24 and P48 infield fleet, providing the latest hardware to run our latest chemistries. In 2022, we transitioned our Q20+ chemistry kit from developer to broad early access. Our Simplex single molecule accuracy is 99%. With consensus, we deliver accurate small variant calling comparable to shortly sequencing, but in addition, we provide structural variation and methylation.

Our Simplex chemistry routinely delivers infield 100 gigs of sequence data per PromethION Flow Cell. Our Duplex single molecule accuracy is 99.9%, improving rare variant and de novo assembly of genomes. We are seeing infield replication of our Q20+ plus accuracy in the hands of customers. Currently, full Nanopore sequence genomes are available from 3 45 on our PromethION platform with a technology roadmap to sub $200 per genome. Remember, this includes short reads, long reads, full methylation plus copy number variation, structural variation, and all in one cost effective [indiscernible].

We continue to make good inroads into the 6.2 billion sequencing market for life science research tools. This market is growing at approximately 15% per year. And at this point I am going to hand over to Tim, who will give us a deeper dive on the financials, and I will come back at the end and talk about some use cases, particularly on the applied market side.

Tim Cowper

Thank you, Gordon. Good afternoon, everyone. During 2022, we have continued to deliver a strong financial performance whilst investing in driving future profitability. Firstly, we delivered strong revenue and margin growth in our core LSRT business. LSRT revenue was up by 30% on an underlying basis and up by 16% on a reported basis. LSRT gross margin increased by approximately 250 basis points, despite global supply chain challenges.

Total revenue was ?198.6 million, including ?51.8 million from our legacy diagnostic sales related to COVID testing. We continued to invest for the future with a focus on innovation and commercial infrastructure. Adjusted EBITDA, an alternative performance measure was a loss of ?78.6 million compared to a loss of ?57.7 million in 2021. This reflects a 48% increase in commercial headcount in the period, and a 39% increase in adjusted R&D spend.

In addition, post period end, we achieved the goal we set out to the IPO of doubling our commercial team. Despite significant investment in R&D and commercial infrastructure, we finished the year with cash, cash equivalents and liquid investments of ?558 million, representing a net cash burn of ?60.2 million during the year.

Let's look more closely at LSRT revenue. There are a number of figures on this slide, but the key message I'd like to leave you with is that the underlying revenue, excluding the EGP, COVID sequencing and foreign exchange was 30%. This growth is driven by new customer acquisition. Including the EGP and COVID sequencing, LSRT revenue growth was 16% on a reported basis and 10% on a constant currency basis. This includes a significant headwind from the EGP, down ?17.4 million year-on-year, which was partially offset by a year-on-year increase in COVID sequencing of approximately ?8.6 million. EGP revenues declined in part due to accelerated orders of consumables in the fourth quarter of '21, which reduced their demand for flow cells in the first quarter of 2022.

Let's look at LSRT revenue by category. Starter Pack revenue increased by 23%. This growth reflects increased customer acquisition in 2022. Consumable sales grew by 12% in the year to ?95.7 million. This was despite a ?15.5 million decline in consumable spend by the EGP. This decline was offset by a continued increasing consumable utilization from our broader customer base across all geographies and all customer types.

Excluding the EGP, consumable revenue increased by 43% in 2022. As a reminder, consumables consist of the flow cells and preparation kits that customers purchase separately after they've completed their starter packs. New customers acquired in 2022 will drive consumable growth in 2023.

Let's look at the revenues by customer type. We focus on our customers and segregate them by revenue size into three categories. S1, S2 and S3. We delivered strong revenue growth across all customer groups in 2022, excluding the EGP on the left of the S1 customers. Total revenues in this group grew by 29% in 2022. Growth accelerated in the second half with H2 revenue growth of 44% compared to 16% in the first half.

The relationship we announced with our channel partner Avantor 18 months ago is working well, and this is reflected in the strong S1 growth in the second half. As a reminder, Avantor are helping expand our commercial reach for entry level products such as the MinION in Europe and North America. They're driving both new business and ongoing consumable business. The direct contact has also seen the reactivation of entry level devices already in customer's hands.

Revenue from the S2 customers grew by 36% in 2022, driven by an increase in customer acquisition, whilst COVID sequencing accounted for approximately 20% of S2 revenue in 2022. S2 is a very diversified group, including human genetics, clinical research, cancer research and work on infectious diseases. Other areas include plant and animal research, environmental studies and microbiology.

Revenue from the S3 group, excluding the EGP grew by 33% in 2022, reflecting strong growth in the broader underlying customer base. Population genomics, including the EGP, represent about a third of total S3 revenue dominated by seven customers. COVID sequencing by S3 customers contributed approximately 19% to S3 revenue in 2022.

S3 revenue growth significantly slowed in the second half of 2022. H2 revenue declined by 5% in the second half compared to 94% growth in the first half. This is primarily due to the high-level of COVID sequencing by S3 customers in the second half of 2021 compared to the low level of COVID sequencing by S3 customers in the second half 2022.

Finally, we are increasingly investing in improving the level of service to more difficult to reach countries. We rely on an expanding network of regionalized distributors to help us. Revenue for these indirect customers increased by ?8.4 million or 86% in the year, which was largely due to growth in China. COVID sequencing accounted for approximately 25% of indirect revenue.

Turning to the next slide, you can see that this revenue growth was driven by a significant increase in active customer accounts. We delivered a net increase of more than 1,900 active accounts in 2022, taking total active customers to more than 8,200. S1 customer numbers increased significantly to 7,210. S1 average revenue was $5,200 per customer in this group, down from $5,800 in 2021, reflecting the significant increase in new accounts.

We also added over 200 new active customers in this S2 group, whilst maintaining average revenues of $66,000. Average revenue per account declined slightly due to the increase in new customers as well as movements between groups. We also delivered a net increase of 17 customers in the S3 group.

As you can see, average revenue per account, excluding the EGP declined by 8%, which is predominantly due to a number of customers moving up from S2 to S3. Importantly, if we look at customers that were in the S3 group in both 2021 and 2022, average revenue per account increased by 18% in the year to $898,000. As I've said, 2022 has been a year of customer acquisition, which will drive growth in 2023.

Moving to regional results for our LSRT revenue. Revenue across the Americas was ?48.3 million, up 45% or 32% on a constant currency basis. This reflects increased investment in commercial resources and growing demand for our unique technology platform. Revenue growth in this region is principally driven by research in human disease and genomic surveillance in the U.S and Canada, but also reflects the expansion in South America through an emerging network of distributors.

Revenue in Europe was ?43.3 million, up 30% or 29% on a constant currency basis. Revenue growth reflects the increased commercial headcount across the region. Rest of world revenue was ?39.8 million, up 40% or 33% on a constant currency basis. This includes very strong revenue from China, up by ?8 million year-on-year. Increased demand in China was driven by strong performance of MinION and GridION for infectious disease.

Now, let's move to gross margin. We spent a lot of time investing in our supply chain and making operational improvements in 2022, including automation, improvements in manufacturing techniques and the recycling of electronic components. As a result, we delivered a 250 basis points increase in LSRT gross margin in 2022, despite significant global supply chain challenges and inflation re-pressures.

We also benefited from our decision to maintain high inventory levels of core components, our long-term relationships with suppliers and our core internal capabilities. For example, we were able to navigate supply shortages by purchasing and adapting alternative generic components to work in our products. During the second half of 2022, we continued to make further progress on driving margin expansion.

LSRT gross margin in the second half was 57.8%, up approximately 300 basis points from the first half. We have increased investment in innovation and expanded our commercial infrastructure to drive future growth. In terms of headcount, we ended the year with more than 1,000 employees, up from 803 at the end of 2021.

Reported R&D expenses of ?64.8 million include the reduction in the provision for social security costs on pre IPO share awards. Excluding this item and adding the capitalization of development costs, the total spend on R&D increased by 39% in the year to ?93.9 million. This was principally due to a 31% increase in R&D headcount.

Reported SG&A costs were ?157.4 million, down from ?161.8 million in 2021. After adjusting for share-based payments, adjusted SG&A expenses increased by 18%. This was principally due to a 48% increase in the commercial team and a 30% increase in corporate staff in the period. You can find a reconciliation of adjusted R&D and SG&A expenses in the appendix.

Adjusted EBITDA, an alternative performance measure was a loss of ?78.6 million compared to ?57.7 million in 2021. This includes adjustments for the settlement of the DHSC contract, share base payments and the sale and lease back as you can see on this slide.

Moving now to guidance. We expect annual LSRT revenue growth of greater than 30% on an underlying basis in both 2023 and in the medium term. As a reminder, underlying revenue growth excludes the EGP, COVID sequencing and foreign exchange. Including the EGP and COVID sequencing, we expect full year 2023 revenue growth to be in the range of 16% to 30% on a constant currency basis. This includes a significant headwind from COVID sequencing.

As I have previously discussed, there was some volatility in the EGP and COVID sequencing revenues over the last 2 years. This is reflected in our overall growth in 2022 and guidance for 2023. This is now largely behind us and isn't expected to impact forecasts going forward. The 3-year contract signed with the EGP in 2021 is now expected to continue in the range of ?15 million to ?20 million per year until the contract completes in November, 2024.

We believe that COVID sequencing has now largely stopped. As such, we expect COVID sequencing revenues of approximately ?6 million in 2023, which reflects a year-on-year headwind of approximately ?20 million. We do not expect COVID sequencing revenues in 2024. Our margin targets are unchanged. We are targeting greater than 60% gross margins for the financial year 2023 and greater than 65% in the medium term. We also continue to target adjusted EBITDA breakeven by the end of 2026.

Before I turn back to Gordon, I would like to quickly summarize the four key takeaways. First, the core business is doing well. 2022 was a year of strong customer acquisition. Those new customers will drive consumable growth in 2023. Second, we delivered very strong gross margin progress despite significant global supply chain challenges in 2022. The operation improvements we have made will continue to drive margin expansion in 2023 and beyond.

Third, we have a strong balance sheet with cash, cash equivalents and liquid investments of ?558 million. And finally, we are very excited by the progress we are making in new market applications. This will be a key driver of long-term growth, which Gordon will discuss in more detail. Over to you Gordon.

Gordon Sanghera

Thanks, Tim. I want to finish off by discussing how our technology is being used in discovery and translational research and early progress towards applied market clinical and industrial applications. In discovery and translational research, we are working with the NIH who are sequencing brain samples from dementia patients with a particular focus on elucidating structural variants, and the association with neurodegenerative conditions such as Alzheimer's.

Cancer 2.0, our collaboration with Genomics England is moving into year two of the study with two genomic laboratory Hub, Pathfinder Labs starting pilot programs to deploy cancer genomic insights into the NHS. Building on the work previously reported on same day sample to answer on whole genome sequencing at Stanford, they are now expanding the program to resolve, hard to diagnose disease faster than ever before.

Here in the U.K., the Royal Devon and Exeter NHS Foundation are looking to deploy rapid whole genome sequencing to evaluate benefits in care pathways for human genetic disorders. On the infectious disease front, we are building on our placement of Nanopore sequencing in public health labs during the pandemic. We're focusing on rapid genomic surveillance in a post-COVID world, from avian flu to monkey pox. The anti-platform was the first sequence avian flu in an 11-year old Cambodian girl.

At Guys and St. Thomas's, they’ve been running a proof-of-concept trial providing daily respiratory metagenomic services to characterize pneumonia and respiratory infection in the ICU. 200 results have been returned same day versus 3 to 4 days in a traditional path lab. At Guys and St. Thomas's were also at the forefront of sequencing isolate samples from the recent U.K strip outbreak, providing novel insights into invasive cases and pediatric deaths. On the TB front, the drug resistant workflow, we are looking to extend and run further proof-of-concept clinical studies in public health laboratories where in countries where there is high prevalence of TB.

In my final section, I would like to focus on our collaborations to accelerate Nanopore sequencing into applied markets in clinical and industrial applications. As you may have seen last week, we announced our collaboration with Cyclomics. Cyclomics is developing a kit, a workflow to access liquid biopsies. Just to remind you, liquid biopsy is cell-free or circular and tumor fragments of cancer cells. These fragments provide early detection of cancer compared to traditional technologies.

Cyclomics have developed a workflow that enables highly sensitive and accurate methods for detecting circular in tumor or cell-free DNA on the MinION. They have demonstrated on head and neck cancer samples near 100% accurate sequencing of the gene TP53, which has very low abundance mutations for head and neck cancer. Our collaboration will enable Nanopore customers to access the Cyclomics workflow as an early access developer.

Longer term, we expect this approach to be adopted for early detection of cancer in clinical settings. Our 4bases collaboration on breast cancer, we've signed a strategic collaboration with 4bases, a Swiss company that develops and markets CEIVD, genetic diagnostic solutions and assays. 4bases has tested the use of oxygen Nanopore platform with its kits and will supply them together for breast cancer screening in Switzerland and Italy. The four bases kit targets the BRCA 1 and BRCA 2 genes through PCR generating short amplicons, which are compatible with Oxford Nanopore's built-in short fragment mode sequencing, enabling the identification of variants present in these genes to provide a report listing the mutations to the end users.

This is exciting as it has the potential to enable results near the point of care and with faster turnaround times. Asuragen collaboration focuses on carrier screening. Asuragen and R&D will develop assays designed to deliver more accurate and reliable options for reproductive health and carrier screening. Asuragen will leverage its gold standard PCR technology with Oxford Nanopore's DNA sequencing capabilities to develop the first sequencing system to identify the most challenging, yet highly prevalent carrier genes in a single unified work stream that today requires multiple methods.

Although traditional sequencing methods allow identification of large numbers of genetic sequence variants, they cannot effectively resolve the many problematic genes such as those with tandem repeats. Copy number variation, structural variation and pseudo genes. And these are genes that are recommended in professional practicing guidelines for carrier screening.

6 of the 10 genes with the most common pathogenic variants in carriers are difficult to genotype using traditional short resequencing methods and often require complex and complicated combination of solutions. Oxford Nanopore and Asuragen will develop a comprehensive carrier screening assay.

I'm going to switch now to talk about an emerging customer segment, which is pharma/biotech, biologics QA and QC manufacturing or vaccines, biologics and therapeutics require mandatory testing to ensure they are safe and effective. ONT has the potential to offer significant advantages within the genetic characterization, safety and contamination investigation space. This is a highly regulated market and we are very excited about our PathoQuest collaboration. PathoQuest is a spinout from the Institute of Pasteur in Paris and is a leader in using NGS based testing for quality control of biologics, enabling rapid robust analysis At HGMP level.

PathoQuest and ONT have signed an MOU for a collaboration with the objective of optimizing and commercializing a testing service for non-GMP and GMP validated integration site analysis tests for bio-manufacturing quality control. The development of regulated platforms will be a key enabler in penetrating the biopharma QA, QC testing market.

So finally, in summary, we continue to execute on innovation, our new products and upgrades to existing platforms driving adoption. Our platform performance continues to improve through accuracy, output and reliability. With full rollout of our fleet upgrades and Q20+ plus chemistry. We are simplifying our workflows and improving analysis software pipelines, which are key drivers of growth.

Operationally, we are making progress in automation and manufacturing improvements to drive reliability and margins. We are enhancing our customer experience through key technology partnerships such as the A100 upgrades with NVIDIA. And on the commercial execution front, we are delivering on our commitments of growth in the life science research tools market, and as I've just discussed in the second part of my talk progress towards applied market clinical industrial applications.

With that, I will hand over and look forward to questions.

Question-and-Answer Session

[Operator Instructions] We'll take the first question from Charles Weston from RBC. Please go ahead.

Q - Charles Weston

Hello. Thanks for taking my question. My first is around the commercial team. You've retooled it, in terms of capabilities and scale and how should we expect this to translate to growth in the various different customer groups, and how much more investment do you expect over the next year? And obviously the subtext of that question is how much we might expect SG&A to grow, next year in 2023.

Tim Cowper

Thanks, Charles. It's -- we were very pleased to have achieved our target of doubling our customer facing team, which we did in February of last month. That was a promise we made at the IPO. Obviously, that relates in to higher SG&A costs in the year. Now we've established ourselves with a good commercial footprint. These costs will continue to grow and there'll definitely be more growth this year as we experience a full year of that commercial team that we have in place.

However, the growth over time will be significantly less than the revenue growth and the margin growth that we are planning, and that's why we believe we will be tracking towards our medium term target of EBITDA breakeven. So looking at G&A, which is also related, this was our first full year of being a public company and we put the necessary increases in infrastructure in place at the time of IPO. So, that's [indiscernible]. So the G&A portion of SG&A is a fair reflection on where we expect those numbers to grow, and that should grow at a relatively modest rate as well.

Operator

We'll now take the next question from Odysseas from Berenberg. Please go ahead.

Odysseas Manesiotis

Hi, Tim. Hi, Gordon. Thanks for taking my question. So first question, I want some additional color around the full year '23 sales guidance downgrade, and the softening of orders that you mentioned in the release. I understand that this was largely driven by correction in COVID-related sequencing, perhaps from one month to another earlier in the year. But have recent product launches from competition, and in particular, PacBio's Revio and Illumina's Infinity played their role. What has been the customer or sales feedback you've been getting so far? Thank you.

Gordon Sanghera

You do the revenue and I will do the …

Tim Cowper

Okay, okay. Thanks, Odysseas. So just as a -- just a reminder of our LSRT guidance that we've just given is for 16% to 30% growth on a constant currency basis. If you look at our underlying revenue growth during 2022, we achieved 30% level of growth. And indeed, if you look over the last few years, we've been tracking to those levels. In arriving at our guidance for this year, we've gone -- we stripped it back to the underlying revenues and we've assumed we will continue to grow in line with that growth rate that we can -- we've demonstrated we can consistently deliver. As a reminder of 2022, we -- it was a year of very strong starter pack growth, and those new customers that we've acquired, we feel very confident are going to drive the consumable revenues throughout this year.

Looking at the higher end of the range that we've got, there's a number of large projects we're looking at, and when we risk adjust those, that's how we've -- that's the reason for the wide range, if that makes sense. Definitely COVID has been the major change that we've seen this year. And now we have -- now we feel confident. COVID sequencing has now finished that has caused us to really reflect on what our guidance levels should be. But order patterns continue this year apart from COVID in line with our expectations.

Gordon Sanghera

With regard to PacBio's Revio and Illumina's Infinity, my understanding is Illumina's Infinity is coming in around a $1,000 to run a long read. And the read is relatively small when you think about what we can do anything from 20 bases up to a million. So I don't know what impact that will have. I mean, I think potentially, short read sequencing customers will be able to look at some long reads and then we'll be very interested in it. So I think it might be quite useful to get the people who are very traditionalists SBS acolytes to open their mind up a little bit.

PacBio's Revio, we'll see how that plays out in the marketplace. There's a lot of noise coming out from PacBio. We shall see how customers respond and we will continue to keep our heads down and focus on our own launches. Our Q20+ chemistry and P2 in particular, we think are both game changers as well as with the Q20+ chemistry accelerate and adoption of PromethION, MinION and GridION across the fleet.

Odysseas Manesiotis

That's [indiscernible]. And on to my second one, one for Tim, I mean, impressive to see your gross margins expand here and targets unchanged [indiscernible] across the sector. When we compare this to peers, do you think it's largely related to your different approach using materials which haven't -- which weren't as impacted by supply chain pressures, or half the manufacturing and design improvements ended up offsetting supply chain pressures more than you initially expected?

Tim Cowper

Thanks, Odysseas. And just as a reminder, I think it's an important point to make. Margins in the first half of this year were 54.8% and LSRT margins in the second half of the year grew by 300 basis points to 57.8%. When we look as we've been investing in improving our margins for a number of years right back to when we built manufacturing facility in 2019, we have also been investing in our core suppliers in order to build those relationships over a number of years. We've been concentrating on increasing our in-house capabilities so that we can be more adaptable. And as an example, we've been able to switch components at a relatively short notice when some of the generic component supply chain shortages arose during the year. We've had that flexibility because of our in-house capabilities, and we also benefit from carrying high inventory levels at the beginning of the year and higher now. And that was designed as a risk mitigation against an event such as this. Although I can promise you we did not foresee everything that was going to happen in 2022.

Odysseas Manesiotis

Okay. Thank you very much.

Operator

The next question comes from Paul Cuddon from Numis Securities.

Paul Cuddon

Good afternoon, guys. I mean, just firstly for me, the new customer capture has been impressive during the year, but I wonder if you could elaborate on the transition sort of between S1 and S2 customers and S2 to S3. Whether you are seeing an accelerating rate of capturing the user and then getting them to spend more on your kind of platforms.

Gordon Sanghera

There is some of that, but that's not how this product is set up. A PhD student may well do a PhD in genetics and be a very strong S1 customer and then join [indiscernible] and become a banker. It's meant to be an ever increasing -- it's not a -- the MinION ultimately, and I'll keep saying this, like the mobile phone will be ubiquitous eventually, and that is a massively important customer segment and they come in and out. They don't necessarily have to transition to S2 and S3.

And then within cohorts, if we look at S2 for a moment, you look at public health laboratories, they're strong users of GridION. And they’re not going to move up to PromethION or down to MinION, unless of course there's a decentralized or point of care play, then they may well do. But generally public health laboratories tend to be in centralized facilities. And so it's not a continuum. And there is some migration that PhD student chose not to go into investment banking, but decided to do a PhD post-DOC become a professor and we have seen some customers, but I wouldn't say that is a strategy, that's just happenstance and it's nice when it happens and that they move all the way up. If you look at MIT [ph] and [indiscernible], Northwestern University now formally Santa Cruz, he's now a power PromethION user.

So there is a continuum, but that's the exception rather than the rule. What will be interesting here is how P2 inter sit between those gaps. So I think rather than trying to move people from one continuum to another, we try to fill in the gaps, hence [indiscernible] the other way. And the S2 will be very interesting because you can daisy chain it to be 1 or 4 or 6 or 8 PromethIONs right up to filling the gap to 24. So watch this space. It's going to be an interesting play and we're very excited about P2.

Paul Cuddon

Excellent. Thank you.

Tim Cowper

Yes, I mean, one, just to add a couple more points to that, the strong growth in S2 customers are 26% growth in the number of active customers. Actually I think it's mainly the way the [indiscernible] within that. Companies do move in and out between S2, S3, in particular. So the average revenues have stayed pretty much unchanged in the year. When you look at S3, we've -- there has -- we've added net 17 customers in there. Most of those customers have actually moved up from S2. And as they move up, they tend to pull the average revenue down because they're new to being an S3, added within the S3 group. But if you look at the customers that were S3 customers at the beginning of -- in 2021 and were still S3 customers in 2022, we saw 18% growth in average revenue per customer to $898,000. And I think that's a trend we would expect to continue as S3 will drive growth.

Gordon Sanghera

Yes, I think that's a fair point. We're focused on reuse and utilization of boxes customers have. As Tim says, we do see S2, S3 migration, but the focus is utilization and then it naturally happens rather than being a forced sale, if you like, or a forced fit.

Paul Cuddon

Understood. Thank you. And my second question was just on the kind of emerging battleground of kind of epigenetics and whether the Nanopore kind of based approach has a better kind of dynamic range ability to detect more modifications than you can get using the more fluorescent and camera based approach.

Gordon Sanghera

Sure. So, there was a landmark study last year, I think publication that showed Oxy Nanopore can map 95% to 98% of the methylome as it's becoming referred to now. The current [indiscernible] methods are complicated. Sample prep is very long and laborious, they're error prone. And all of those things add to the error budget on short read sequencing by synthesis measuring directly the methylation in our platform as part of the sequencing offering is all part of the data analysis, the A100 upgrades that really allow us to power through all methylation.

So we are now seeing ever increasing customers starting to map the whole of the methylome. And we think -- I don't really think it's a battleground. I mean, we just know everybody will tell you that if you've got native methylation and full methylation with all CPG islands marked up, you would always pick that. We are just at the beginning of enabling that on our platforms. And I think this is going to be a huge, huge market for us. It's clear from the grail studies that the key and prominent driver of cancer is methylation. Now everybody's known that for a while, but now what we provide is a comprehensive methylome and we think that's game changing and our customers will provide that in publications and use cases.

Paul Cuddon

Excellent. Thank you very much guys.

Operator

[Operator Instructions] We'll take the next question from Zain Ebrahim from JP Morgan.

Zain Ebrahim

Hello, Zain Ebrahim, JP Morgan. Thank you for taking my questions. My first question would just be on population sequencing. So I think you mentioned in the call that population sequencing is now around a third of S3 revenues in 2022. And it sounds like there may be a risk adjustment on population sequencing type contracts within the '23 revenue guidance. What proportion of '23 revenues in your guidance do you assume from population sequencing at the lower end of the guide versus the upper end? And can you provide an update on the contract when you made in 2023 and how you're thinking about the evolution of population sequencing beyond '24 when you see the expiry of the EGP contract?

Tim Cowper

Thanks, Zain. Yes, so there are a number of large contracts within guidance. While some of them may be population genomics, they're in other areas as well. And Gordon talked about several new exciting areas that we are moving into, excuse me -- but, we risk adjust those opportunities. These are the opportunities, these are the contracts we haven't won yet. And we've always risk adjusted those opportunities and that really is the point I was making there is that's the difference between the lower end and the higher end of the guidance is there's a number of large contracts that we risk adjusted down to get to the top high-end of the guidance.

But if you look at the EGP, in particular, we -- just -- again, just to remember what's happened in November, 2021, we signed a $68 million contract, which lasted for 3 years. Our guidance of ?15 million to $20 million is based on the completion of that contract and largely speaking, not significantly different from the cadence that we saw in the second half from the EGP in second half of 2022.

Zain Ebrahim

Okay, that's very clear. Thank you.

Operator

The next question comes from David Westenberg from Piper Sandler.

David Westenberg

Hi. Thank you for taking the question. So I want to continue with some of the question -- prior question asked about kind of the S2 -- S1, S2 customer dynamics, and particularly I want to ask about maybe some of the trends with the introduction of the P2 and whether or not that might facilitate some of the customers that would've started -- at S1, maybe start immediately at S2, just given the fact that they have a instrument for -- to get more output in.

Tim Cowper

That's a great question, David. What was interesting when we put out the concept for P2, we got a lot of S1 customers, MinION users very interested in it. So I think it may well -- it'll definitely fill the gap between GridION and P24. That was kind of the design idea and you can daisy chain. There's a lot of debates, should we do an 8 channel PromethION? And in the end we've done the two channel and you can buy multiples. But what's interesting is I think it is also going to catalyze and enable access for lower spend accounts to be able to do really punchy high throughput experiments. And I think that just like the MinION has completely changed the market dynamics, I think the P2 at the low end, I mean it is $10,000, so it's still a fairly low entry point. I think we'll see growth that way as well. So, yes, it'll be interesting to see.

And then in the middle there, there's going to be as you say, the median user, which probably will be in that sort of S2 spend. But it'll be an interesting one to see how it evolves. I think it might surprise us and be a bit more ubiquitous right across S3 accounts where people can do quick look, see stuff. Even though they can do that on a 24 or 48 channel, they tend to use, particularly the large scale projects. They tend to sort of have a PromethION to run their human samples. And having a 2 channel system to check kits, check assays, check sample prep may well be something that happens as well.

David Westenberg

Got you. Thank you very much. And then, can you maybe talk about how the P2 might stack up, or how it might operate kind of on a competitive way. Do you see this as more adoption from other kind of long read sequencers, maybe that we're using a service provider before for their long read needs and now want to bring it to the bench top? Or do you actually see the P2 being used in a semi production mode from some of the service providers that also might, or is it maybe going to do that, maybe with the regular PromethION? Thank you very much. And that's my second and final question.

Gordon Sanghera

So, I think two things. We announced our automation solutions with Tecan, so that's quite important for those service providers to be able to get high throughput. And yes, I do think getting -- going from sort of nothing to a P24 is a big step. That's zero to MAC1. So having a service provider being able to buy 2 or 4 or 6 P2s and enter long read native DNA sequencing service provision with our automated solutions from Tecan will enable them to establish their market presence. And then I think this is a case where we were talking earlier about an S2 becoming an S3, they could well progress to a higher throughput as they get commercial traction.

I don't know if you saw this morning, but Alex Wittenberg was talking about [indiscernible] comprehensive mapping of the [indiscernible] genome using Nanopore only. And that's phenomenally important in crop insights as we start to think about that area. And that is a nascent market that we can really go after and the service providers can really enter it with a very low entry point to test the market. And we think that's going to be an interesting customer segment for P2. And particularly …

David Westenberg

Thank you.

Gordon Sanghera

… on the Ag bio [ph] side, the middle income countries, Africa, we think this is going to be really exciting around P2.

David Westenberg

Thank you.

Operator

We'll now take the next question from Veronika Dubajova from Citi. Please go ahead.

Veronika Dubajova

Hi, guys. Good afternoon. And two questions from me, please. One is, I'm going to start with Tim, please, if that's okay. Just trying to understand the share, the shape of the EBITDA in 2023, apologies, given the investments you've made into SG&A. And I don't know if you can be a bit more specific, if we annualize out the hires that you've added in the second half of last year and obviously through to February. What should that give us as a baseline for SG&A for 2023?

Tim Cowper

Hi, Veronika. Thanks. Yes, so when do you look at EBITDA and just say we adjust out positive and negative things that we considered not to be trends from adjusted EBITDA. So for instance, this year we removed the ?51.8 million of COVID testing revenue and we removed the profit from certain leaseback. So it is designed to show a trend. With the increase in revenues that we are planning, with the increase in margins that we are planning, and there will be an increase in costs. The EBITDA should be roughly relatively flat, may increase just a little bit during 2023. But the trend after that as the revenues grow, we've -- we benefit from the investment that we've been making over the last year or two, really should come through and reduce the EBITDA over the remaining years as we go towards breakeven. That should be a fairly constant upward trend, or reduction in losses after that.

Veronika Dubajova

Excellent. That's very helpful. Thanks for that. And then my second question is for Gordon, and I'm just curious, Gordon, obviously we are already the -- towards the end of March. Just curious if you've seen any change, COVID aside, any change in conversations you're having with customers year-to-date? And so honing in particular on that, S3 segment, the larger customers, do you have any visibility into what the funding is looking like through the remainder of 2023 and your degree of confidence in that? Thanks so much.

Gordon Sanghera

Yes, I think two comments. Because of the CapEx free model, we are not as exposed to cuts in government funded spending. And then for us, some of those medium to large projects, they come after a small project, small here is 5,000 to 10,000. So if you look at the NIH neurodegenerative study that is starting to show some interesting correlations between structural variation as everybody expected and that could then lead to a larger cohort. But that is not just technologies dependent, it's also funding dependent. And we are seeing lots of opportunities for large scale projects via small pilot scale programs. And that's our focus and the weight adjusted probability of winning those to become large substantial projects is where, as Tim said, the swing is in that sort of upper end of the revenue guidance.

There's a lot of exciting things going on, but what the world looks like in 3 to 6 months as the data comes through, as the tender processes happen and whether there is still money available to do bigger programs, that's all up in the air right now. But we're fairly confident that there are some important programs that we will do this year that if they do not bear fruition into larger contracts this year, we'll be setting us up nicely for 2024.

Veronika Dubajova

Understood. Very clear. Thank you both.

Operator

As there are no further questions, I will hand back over to Gordon Sanghera for closing remarks.

Gordon Sanghera

I'd like to thank you all for your time today. We'll be seeing some of you in briefings later this week and in subsequent weeks and on our road trips to the USA. I would just say we are 1-year in from our early launch of our Q20+ chemistry. Our P2 order book is fabulous and we are really excited and well poised to really push through.

The other thing you may have seen this morning, recently, sorry, it was Karen Miga doing telomere-to-telomere full genome on Nanopore only a year ago that was using multiple sequencing technologies. And that is really -- gives a sense of how we now provide the most comprehensive and complete genome, which will be really important as a reference Atlas.

And as we talked earlier with methylation, but the real driver of revenue will be our Q20+ on Simplex, which can meet all the accuracy criteria and small variant calling, but add so much more biology and ever increasingly that is gaining huge momentum and we think that we are really, really big driver for our growth and we will continue to innovate and iterate those chemistries to push the envelope further. Thanks for your time, everybody.

For further details see:

Oxford Nanopore Technologies plc (ONTTF) Q4 2022 Earnings Call Transcript