Illumina Shows its Stuff to Wall Street, Stock Still Slides
San Diego’s Illumina has a pretty amazing story to tell about exponential growth. The maker of genetic analysis tools has beaten or matched Wall Street earnings forecasts for a dozen quarters in a row. It has grown from 239 employees in 2004 to an estimated 1,604 by the end of this year. But when it invited investment analysts to its headquarters yesterday for an in-depth rundown on the state of the company, Illumina’s stock fell 5 percent on another lousy day for the markets.
The Wall Street crowd apparently thinks it’s a tough economy for selling super-sophisticated genome analysis machines, which cost more than $450,000 fully-equipped, plus as much as $200,000 a year in consumable supplies to keep them running. Since Illumina (NASDAQ: ILMN) did a 2-for-1 stock split on Sept. 23, its value has sunk 33 percent, to $27.71 at the close Thursday.
CEO Jay Flatley painted a broad picture of what Illumina does in his opening remarks at analyst day, which I watched on a webcast. It makes tools for high-speed DNA sequencing, genotyping of a segment of an individual’s genetic code, and tests that examine to what extent an individual’s genes are dialed on or off. By bringing down the cost of full genome sequencing to less than $100,000, Illumina is one of the companies that has been ushering in the era personalized medicine, along with its competitors, Foster, City, CA-based Applied Biosystems, Roche, and Cambridge, MA-based Helicos Biosciences. Illumina’s new goal is to bring that price down to $10,000, Flatley said.
“It’s within our sights,” he told analysts.
Getting there isn’t easy. Illumina has had struggles to manufacture its arrays to keep up with demand, mostly from government, academic, and hospital labs, Flatley says. It placed special emphasis—unusual for a Wall Street presentation—on its initiatives to recruit and develop enough talented workers to keep up with its rapid-fire growth.
To give a sense of where that growth is coming from, Illumina brought in Peter Gregersen of the Feinstein Institute for Medical Research on New York’s Long Island. Gregersen, who does genetic research for autoimmune diseases like rheumatoid arthritis, gave a fascinating talk about how the new tools are enabling him to ask new questions.
Autoimmune diseases are a bunch of conditions in which the immune system goes haywire and starts attacking healthy tissue instead of foreign invaders like viruses and bacteria. These diseases affect millions of people—about one in 20 people in the country—with rheumatoid arthritis, lupus, multiple sclerosis, Type 1 diabetes, and inflammatory bowel disease, to name a few, Gregersen said. Their causes aren’t really known, although researchers suspect multiple genes play a role. For rheumatoid arthritis, about 10 genes are thought to be responsible, and for lupus it’s 20, he said.
By looking broadly at the genetic profiles of patients with these diseases—and with costs coming down to the point where large numbers of patients can feasibly be sequenced—new sequencing tools will lead researchers to much deeper understanding of these diseases. “This is an engine of hypothesis generation,” Gregersen said.
The day is also coming when the Illumina tools are going to be used for selecting patients whose genes make them more or less likely to benefit from expensive biotech drugs. Drugs for rheumatoid arthritis, like Amgen’s etanercept (Enbrel), and Abbott Laboratories’ adalimumab (Humira) typically are very helpful for about 30 percent of patients, and don’t really work at all for another 30 percent, Gregersen said. Companies haven’t shown much interest in genetic screening of patients to pick responders, but that’s starting to change, and it’s likely the FDA will someday require these kinds of studies to avoid another disaster with rare side effects like those seen with Merck’s Vioxx, he said.
“There is an unstoppable tsunami of applications,” of genetic analyses coming, Gregersen said.
One analyst, Doug Schenkel of Cowen & Co., asked whether it’s too difficult to get access to patient samples to do the kind of wide-ranging studies of large numbers of people that Gregersen was referring to. Patients in clinical trials not only need to give informed consent before taking an experimental drug, but need to give separate consent to allow their DNA to be analyzed. That usually takes 15 to 30 minutes for a nurse to do, and can take extra effort to get cleared by an institutional review board that monitors the ethics of clinical studies. “It isn’t cheap,” to do these kinds of genetics studies, Gregersen says. But he added that recent efforts by people like Harvard geneticist George Church, who has put his DNA profile on the Internet, is taking away some of the stigma that might chill this line of research.
Illumina insists the challenges can be overcome, and big dollar opportunities lie ahead. The market for its instruments with life sciences researchers is worth $2.6 billion a year, followed by another $1 billion a year opportunity in agriculture applications like crops and livestock, said Tristan Orpin, Illumina’s senior vice president of commercial operations. The company is also making a play in the $3 billion molecular diagnostics market, including more refined early detection of cancer. One more market segment—albeit one that’s younger and harder to define, Orpin said—is consumer genetics. That’s the field where companies like Mountain View, CA-based 23andMe (led by Anne Wojcicki, the wife of Google co-founder Sergey Brin) sells tests to help patients gauge their risk of developing certain diseases later in life.
Illumina’s chief financial officer, Christian Henry, suggested the company might use its $700 million cash horde to buy other companies if the opportunity is right. So far the company hasn’t needed it for operations. Henry forecasts the company will increase its revenue by 20 percent to 30 percent a year over the next three years, and that it will close 2008 with a profit of 84 to 87 cents a share. Then again, if the markets keep turning down, and customers get gun-shy about buying $450,000 machines for cool new experiments, $700 million looks like a pretty decent rainy-day fund.