The $1K Genome? So What? Illumina Is On a Quest for World Domination
People can argue all day about whether Illumina has, at last, given us the $1,000 genome. The answer does matter, because the cheaper it gets to sequence a whole human genome, the easier it will be to gather lots of them. That will help us understand what makes people different, and shed light on what causes disease. Or so scientists hope.
It’s heady stuff. But that’s not the biggest story at Illumina right now.
The big story is that San Diego-based Illumina (NASDAQ: ILMN) has cemented its position as the dominant player in genomics, at what you could call the beginning of the age of genomic medicine. Through a series of in-house technology advances and savvy business strategy, the company has put itself in an enviable spot. It has a broad and loyal customer base, faces marginal competitive threats, and its market is growing beyond research into the broader world of medical diagnostics.
Illumina is still not that huge, with $1.42 billion in revenue a year ago, but it recently told investors it sees a total addressable market ahead of $20 billion. Illumina should capture 75 percent of the market through 2020, Goldman Sachs analyst Isaac Ro recently said in a note to clients.
“There is, as far out as we can see, an insatiable demand for whole genome sequencing,” CEO Jay Flatley told investors last week on a conference call.
The stock climbed 37 percent in January alone, to close at $152. Not bad for a company that was the subject of a hostile takeover bid, at $44.50 a share, less than two years ago.
“It’s a terrific company, and they are in a very dominant position,” says Jay Shendure, an associate professor of genome sciences at the University of Washington. “It’s hard to see that changing in the near future.”
For those who missed the news a few weeks ago, scientists were buzzing over two announcements Illumina made at the JP Morgan Healthcare Conference in San Francisco. One was that it had developed a new top-of-the-line sequencing system, called HiSeq X 10. It’s supposed provide whole genome sequencing at full-factory scale, with 10 sequencers (total price: $10 million) running together. At that scale, those machines can crank out 18,000 genomes per year, bringing the price down to an estimated $1,000 apiece. It’s a big deal, not only because scientists have been anticipating the $1k genome for a decade, but because it will enable scientists to run all kinds of ambitious new experiments, involving tens of thousands of patients, that weren’t possible before. Early orders have come in from Macrogen, a genetic sequencing service based in Seoul, South Korea; the Broad Institute in Cambridge, MA; the Garvan Institute of Medical Research in Sydney, Australia; the New York Genome Center; and Amgen’s (NASDAQ: AMGN) deCode Genetics unit.
The other Illumina story got less attention, but is probably even more important for the company. The NextSeq 500 was unveiled as the company’s new middle-of-the-lineup offering. It’s priced at $250,000 per instrument (about one-third the list price of its current workhorse HiSeq 2500). Unlike the large HiSeq, it’s small enough to sit on a desktop, and easy enough to use that anybody in the lab can run one, not just a specialized technician. That little desktop machine can deliver almost 10 times the output of Illumina’s current desktop sequencer, the MiSeq.
Through a series of technical advances in automated sample prep, advanced optics, and surface chemistries, Illumina’s new NextSeq can do an entire human genome sequencing run in about 30 hours, says Christian Henry, the company’s chief commercial officer.
That whole package, with the combination of lower price, high performance, and the ease of use, is exactly what someone needed to do to come up with a product that could be sold to hospital pathology labs, says Todd Smith, a Seattle-based bioinformatics entrepreneur. Hospitals don’t like sending DNA samples away to some distant lab—many prefer to sequence patients on site, get an answer fast about what’s wrong with a patient, and use the info to help guide doctors. A fast, powerful desktop machine also appeals to scientists, who want their data tomorrow, not next week, especially when they are competing to get their next big paper in Nature.
Ro, the Goldman Sachs analyst, predicts the NextSeq will almost immediately be the company’s second-best selling sequencing instrument, and will rival sales of the cheap and versatile MiSeq desktop sequencer by 2016.
People in genomics tend to be a skeptical bunch, as some commercial players tend to make grand claims that they can’t back up over time. After allowing some of the news to sink in, and reviewing notes from a recent in-depth conversation I had with Henry, I’ve been trying to put my finger on what has enabled Illumina to be so successful where others have failed. Here are some thoughts:
—It has a highly productive internal R&D team. A common view in the industry is that Illumina is a savvy acquirer of innovation, and good at integrating things, but not really a fountain of innovation itself. Many point to the company’s 2007 acquisition of Hayward, CA-based Solexa is a key piece of supportive evidence. But these latest announcements should put that notion to rest. Illumina spent a whopping $507 million on R&D in the past two years combined, and didn’t waste it. The company developed the super-versatile NextSeq and the boundary-pushing HiSeq X 10 through internal R&D the past two to three years that was all hush-hush to the outside world until the big announcements last month, Henry says.
For the geeks out there, here’s what’s new. The Illumina team, inspired by advanced cameras in smartphones, developed better cameras for taking pictures of the fluorescent tags that tell its machine when it’s looking at an A, C, G, or T base unit of DNA. It also developed a new surface chemistry that makes it possible to take a picture of just two of the base molecules, not all four, in a sequencing run. That reduces the number of pictures taken by 50 percent, which improves speed, lowers cost, and doesn’t sacrifice accuracy, Henry says. Not content to stop there, Illumina created a new fluidics system to handle the mixture of chemical reagents for a sequencing run. The new fluidics reduce the liquid volume going through the system, which reduces the volume of reagents, and reduces the chance of the machine breaking down with a clogged or disconnected tube. The company has even come up with a new kind of flow cell for sequencing runs that can be shipped and stored in dry form instead of in liquid. When I asked Shendure about this, he shrugged about its importance, but Henry said this move addressed a top customer complaint.
While no one of those technologies is glamorous enough to grab a headline, all those efficiency and cost improvements count when taken together, Shendure says. Customers love it when you come out with a small, fast, cheap, reliable box with that much oomph. “For many groups, it’s reasonable to think about going and getting [a NextSeq]. It’s much harder to drum up resources for a $600,000 to $700,000 instrument,” he says.
—The company has segmented its markets carefully. Just like BMW has a 3-series, a 5-series, and a 7-series for different types of buyers, Illumina has thought carefully about how to come up with new products that will expand its markets, not cannibalize its existing products. The MiSeq, now priced at $99,000 (marked down from $125,000), is aimed at entry-level, occasional sequencing users, such as scientists who might want to look at the occasional exome (the 1-2 percent of the genome that codes for proteins in the body), or lower organisms like bacteria. The NextSeq is a step up at a list price of $250,000, with enough power to sequence whole human genomes on a desktop, in a simple enough format for everyday users like Shendure and newer folks like hospital pathology labs. The HiSeq 2500 ($740,000 list price) is still the high-end workhorse that crank through the high volumes that the NextSeq can’t quite handle. And the Hiseq X 10 ($10 million for a group of 10 instruments), on the top end, is for the relatively few customers that can spend big bucks on factory-scale genome sequencing. Importantly, Illumina always works to extend the innovations it makes in one product to the rest of its lineup.
—Management talent, management focus, management depth. Illumina, as I’ve said before, often strikes people as having an exceptionally focused and disciplined management team. Flatley has been around since 1999, almost the beginning, which gives him a ton of institutional knowledge of the company, the sequencing industry, and the needs of his customer base. But there’s more than just one visionary leader at work. Henry, who joined in 2005, switched from chief financial officer a couple years ago to general manager of the life sciences business, and recently was trusted with the important role of chief commercial officer, a move not many finance people could make. Rick Klausner, the former director of the National Cancer Institute, recently joined as chief medical officer, giving the company a lot of insight and connections into the vast oncology market. And Illumina recently recruited Francis deSouza, the former president of computer security giant Symantec (NASDAQ: SYMC), as president. These are the kinds of people that can provide fresh thinking for the challenges ahead, like how to best integrate bioinformatics into Illumina’s machines, and capture the vast opportunity in cancer diagnostics.
—Illumina is playing the long game. Illumina doesn’t resort to gimmicks to boost its numbers for a quarter or two. If you listen to its quarterly earnings calls, you hear things about how the company is constantly working to diversify its customer mix, so that it’s not overly dependent on any one set of buyers—like, say, NIH-funded researchers who live at the mercy of Tea Party wingnuts in Congress.
For example, Illumina showed its long-range thinking last year when it started work on a new San Francisco satellite headquarters in the heart of the action in the Mission Bay district. Not only can Illumina bring some recent acquisitions all under one roof here, but it’s also setting up an attractive recruiting beachhead in the middle of one of the top two biotech neighborhoods in the world. Not every Illumina recruit wants to move to San Diego, and getting this setup will make it easier to lure in some top Bay Area talent.
—It’s too disciplined to fall for the shiny new thing. When Ion Torrent came along with a hot DNA sequencing platform a couple years ago (semiconductor based technology on a desktop, at a low price), Life Technologies acquired it and essentially gave up on its traditional product. Instead of losing faith in its own product line and rushing out to buy its own shiny new thing, Illumina instead raced through a 15-month development cycle to churn out the MiSeq, Henry says. That low-cost, desktop offering built on Illumina’s strength with the high-end HiSeq. Partly because Illumina already had such strong customer relationships, it ended up successfully fending off the Ion Torrent challenge, especially when the new entrant failed to hit some key product improvement deadlines.
—It will buy things when it sees a commercial advantage. Illumina made three acquisitions in the past year in the San Francisco Bay Area, all of which provide some insight into its thinking. It bought Moleculo to gain technology for reading longer stretches of DNA at a time, as a way to offset one of the technical advantages of Pacific Biosciences’ (NASDAQ: PACB) instrument. It bought Verinata Health to make its way into the noninvasive prenatal testing business, one of the important early markets emerging for genomic diagnostics. And it bought NextBio to help it gain strength in the software/interpretation side of genomics.
—A customer-centered culture. Illumina sees itself being successful if it can make its customers successful, Shendure says. It listens to customers, incorporates their ideas, and then comes back to the customers with suggestions of its own in a dynamic feedback loop. “You can imagine a more passive company that may just want to be a provider of instruments and let the markets develop as they would,” Shendure says. “Their growth requires [that] they find new markets beyond research. They’re not going to be passive about catalyzing these new markets. They do [listen]. They do workshops, and bring in academics. I’ve never gotten the feeling they are just trying to pitch us on something.”
So there are some of the key elements I see to Illumina’s success. Technology certainly moves fast in DNA sequencing, and markets can be quite unpredictable. At any point, Illumina could get arrogant or complacent or lose its edge some other way. But I don’t see it happening. The age of the $1,000 genome is the age of Illumina.