By Allison Proffitt
January 10, 2024 | CEO Jacob Thaysen presented Illumina’s updates yesterday at the JP Morgan Healthcare Conference. Thaysen took over as CEO at Illumina in September after serving as senior vice president of Agilent Technologies and president of its Life Sciences and Applied Markets Group.
2023 has been an eventful year for the sequencing pioneer with ongoing legal debate and December’s eventual announcement that Illumina would divest GRAIL; Francis deSouza, Illumina’s former CEO, stepping down in June; a new CTO, Dr. Steven Barnard, in August; and $175M in cost cuts over the course of 2023 which included headcount reductions and major real estate changes.
Thaysen presented Illumina at a turning point.
“I’m incredibly excited about the future for Illumina,” Thaysen said. “The mission of Illumina is to improve human health by unlocking the power of the genome. This mission… truly also summarizes why I joined Illumina here a few months ago. I’ve been impressed [over] the past few months when I’ve talked to many of our employees about their commitment to support our customers and drive genomics forward. I really share the passion that the Illumina team has for sequencing. I see that we’re really the center of the global transformation of genomics in the future.”
The core Illumina business is operating from a position of strength, Thaysen reported. Q4 results were ahead of expectations, he said, primarily driven by fleet expansion for NovaSeq X instruments and increased consumable sales. The company had expected to ship 330 to 340 NovaSeq X instruments in 2023, but shipped 352 instead. They enter 2024 with a 38-instrument backlog.
Low throughput instruments were the largest sellers in the Illumina portfolio with MiSeq, MiniSeq, and iSeq 100 selling about 920 instruments together in 2023. The NovaSeq 6000, which the company would like to transition, sold about 60 instruments. In all, the company placed 2,200 new instruments in 2023, and the entire Illumina portfolio now represents more than 25,000 globally installed platforms.
Interestingly, Thaysen said the focus for 2024 is not primarily expanding that install base. “While we will continue to place new instruments, our focus in ’24 will be mostly about ensuring that our customers are successful with instruments they’ve already purchased, and drive, of course, a lot of sequencing on those instruments.”
Increased sequencing means increased consumables sales, which was already an area of growth in 2023. Workflow tools will also receive greater attention in the coming year. He called Illumina’s informatics offerings—including DRAGEN, Illumina Connected Analytics, and BaseSpace—a “well-kept secret, and something that really differentiates us going forward.”
Illumina offers an ecosystem of tools and solutions that position the company as a full, end-to-end provider, Thaysen said, from sample prep to insights. There are library prep kits and Clarity LIMS for sample prep, analyzation comes from the informatics tools, and interpretation tools include Emedgene, Illumina Connected Insights, and PrimateAI.
“When I’ve spoken to customers, this is something they’re really excited to work with us on,” Thaysen said. “They want to focus on, of course, optimizing applications and driving the research, and not focus on building workflows, productivity, ease-of-use, and so on. This is something we do very well, and we can really focus on innovation. This will drive, of course, more sequencing and allow our customers to be very successful also.”
Thaysen listed a few areas of innovation to watch in 2024: new flow cells for the NovaSeq X (25B and 1.5B); Illumina Complete Long Reads enrichment panel; and XLEAP-SBS chemistry for NextSeq. Illumina’s recent Partek acquisition is expected to expanded Illumina’s tertiary analysis capabilities.
Last week, Illumina announced a collaboration with Janssen to develop Illumina's novel molecular residual disease (MRD) assay, a whole-genome sequencing (WGS) multi-cancer research solution that detects circulating tumor DNA (ctDNA) from solid tumors to better understand the persistence or recurrence of disease following clinical intervention. In contrast to existing MRD solutions with complex workflows, Illumina plans to develop a research solution that will provide a cost-effective, highly sensitive, and automated workflow, with the potential to achieve a turnaround time of five to seven days.
Illumina, in collaboration with Nashville Biosciences, named five founding members of the Alliance for Genomic Discovery (AGD)—AbbVie, Amgen, AstraZeneca, Bayer, and Merck—in July. Last week Bristol Myers Squibb, GSK, and Novo Nordisk signed onto the AGDs’ effort to whole-genome sequence (WGS) 250,000 DNA samples and will have access to the resulting data for use in drug discovery and therapeutic development.
2024 Priorities
In 2024, Illumina’s top three priorities are to drive commercial excellence and grow the NovaSeq X install base, Thaysen said, to focus on operational excellence, and to resolve GRAIL as quickly as possible.
The growth of the install base feels a bit contradictory to earlier statements in the presentation, and Thaysen continued to caveat it. He pointed out that NovaSeq X adoption is early in its lifecycle and many Illumina high throughput customers have not yet transitioned from NovaSeq 6000 to X, but will in the future. He also expects 2024 to be a year of fleet expansion, where early adopters may add additional instruments as they scale up, and he predicts new-to-high-throughput customers will begin with X. But he doubled down on customer success.
“Even though we’ll continue to place instruments out there, as mentioned before, our key focus is to make our customers successful and, of course, ensuring that they are successful in driving sequencing activities,” Thaysen said. He highlighted that not only is NovaSeq X lower cost, but it has the capability of running different workflows on different lanes in the flow cells, allow much more complex analysis to be done. He mentioned a customer running bulk RNA on one lane of the flow cell and running single cell RNA on another lane, combining the data for deeper insights.
The operational excellence focus includes the expense savings of 2023 as well as “programmatic efforts” in 2024 to optimize spending, including portfolio optimization and prioritized investments, focusing on “specific innovation that had the highest impact on our customers. We’ll continue to be very committed to innovation. But we’ll be very deliberate about where we make our investments and where we don’t,” he said. “You will see that will have an impact in ’24, as we continue to focus on areas where we can improve our operations, both to improve productivity, but also to be more predictable in how we run the business.”
Finally, Thaysen expressed hope that Illumina would be able to divest GRAIL, as required, by the second quarter of 2024. “Potential outcomes might impact our balance sheets,” he conceded, While it’s too early to provide guidance on the specific details, we’re certainly working to minimize any impacts.”
Long vs Short
Finally, in Q&A, Thaysen weighed in on the long read vs. short read outlook from Illumina’s perspective. Long read sequencing has its uses, Thaysen said, but he recommended considering not what the technology can do, but what the customer wants.
“Let’s talk about the samples here. I think that the most important growth and where we’ve seen most samples coming from is FFPE for solid tumors, which is fragmented DNA, and also for liquid biopsies, which is also fragmented DNA. Hence, it’s really more aligned with short read technologies, and thereby short read will be the vast majority of sequencing right now, but also in the future and where most growth will come from.”
Joydeep Goswami, Illumina’s chief strategy and corporate development officer and chief financial officer, who joined in February 2023, joined Thaysen for Q&A. He also reported market demand for mostly short read sequencing, even in applications such as population genomics. “Long reads are used for specific partial cohorts within the larger cohorts. For larger programs, we’ve got hundreds of thousands of samples. The long read piece of it is only limited to 10,000—so less than 10% of the overall population analyses. It’s a complementary approach; it’s not one replacing the other,” Goswami said. “Then there’s the cost component. If you’re getting most of the answers with the technology that’s several-fold cheaper, there’s no need to try to get into long reads, other than specific answers that you cannot get with short reads.”