John Bishop, Cepheid's chief executive with Cepheid's GeneXpert machine (Photo by Steve Castillo)
Getting DNA Results--Fast
October / November 2004
Volume 2 Number 5
When people need lab test results these days, they need them fast. Postal workers want to know if the mail they're about to touch is contaminated with anthrax. An obstetrician monitoring a young woman about to give birth needs to know if Group B strep threatens her baby. A surgeon has to know if a patient's breast cancer has spread.
All of these problems can be solved by analysis of genetic material. But until Cepheid came along, it often took a few days to get lab test results. The Sunnyvale, Calif.-based instruments company sells machines that automated the process of DNA analysis and delivers results within 30 minutes.
Born at Lawrence Livermore National Laboratory, the technology is finding new applications in biohazard detection, life sciences research, and hospital laboratory tests. It is the fruit of a government-invented technology that has been commercialized and is now helping government: by alerting postal workers in hundreds of distribution centers to the presence of killer anthrax microbes.
"It is a very successful license," said Norma Dunipace, manager for partnership development at the Livermore lab. "It has come full circle, and we have bought some of their machines for use in our biological research."
Founded in 1996, Cepheid now has 200 employees and in its most recent quarter ended June 30 it generated $10.5 million in revenues. The company isn't profitable yet, but it expects to become profitable by 2006 as its postal contract and other businesses ramp up. During the contract's first phase, Cepheid expects to generate at least $26 million in revenue from the machines and the disposable cartridges that go with them, says Cepheid CEO John Bishop.
The postal contract also helped Cepheid raise $58 million in cash in a secondary public offering in February of this year.
Thomas Weisel Partners estimates that the gene-based testing market is $1 billion and growing at 30 percent to 50 percent a year. The competition includes bigger companies like Roche, Applied Biosystems, Qiagen and Caliper. Cepheid's automated technology, which has the slogan "sample in, answer out," was a long time in the making as it combined advances in both micro fluidics and microelectronics.
Its seeds began at the Livermore lab, where a research scientist named Allen Northrup pioneered a high-speed process that looks at a sample of blood, urine or feces for a specific section of DNA that is unique to a microbe such as anthrax. In the test, the sample from a patient is mixed with known strands of anthrax DNA. If the microbe is present in the sample, the DNA will bind to the similar genes. Then, using a process called a polymerase chain reaction that heats and cools the sample, the machine reproduces the bound DNA a billion-fold.
The sample then becomes big enough to be flagged with colorful particles. Once the particles are detected using optical technology, the machine spits out a positive result. Northrup figured out how to reduce the laboratory instrument to the tiniest size possible using silicon and mechanical technology.
In March 1996, Northrup and two-time Silicon Valley entrepreneur Kurt Petersen, an electrical engineer by training, got together and decided to start Cepheid. Petersen also recruited Tom Gutshall, a medical diagnostics veteran who became the company's first CEO. It took Petersen and Gutshall nearly a year and dozens of trips to the lab to negotiate a license to the technology, but they didn't want to go to investors without the license in hand.
"That actually is a short period of time in terms of how long it takes to get a license," Dunipace at Livermore says. "But when someone with the track record of Kurt Petersen walks in, you pay attention."
In March 1997, with license in hand, Gutshall raised $3.2 million from friends, family and the Band of Angels, a Silicon Valley entrepreneur group that pooled money for new startups. That was enough to get the company off the ground and assemble the team of electronics and life science experts. But the product was nowhere near complete, says Petersen, who recently stepped down as president and chief technical officer. (He left to start another electronics firm, but remains on the board of the company.)
"You can't expect to just take it out of the national lab and start selling it," Petersen says.
The company staffed up its research team. Meanwhile, it raised $8 million in 1998 and another $19 million from venture capitalists in 2000. Then it went public in June 2000, raising $35 million at $6 a share. That enabled it to refine its manufacturing process for making the detection machines and disposable plastic cartridges that are used to hold samples in the machines.
Cepheid started shipping its first product, Smart Cycler®, for life sciences researchers in May 2000. Scientists use the system to devise their own tests for detecting what they want. This market was key because Cepheid could license the tests the scientists created, such as a method for finding cancer cells, that had commercial value. The product, which can test 16 different cartridges simultaneously, earned Cepheid an R&D 100 technology innovation award from R&D magazine.
Cepheid was still a long way from making money, but investors were hot for biotechnology stocks in 2000. The human genome project had been completed and scores of companies were being created to capitalize on it. Cepheid's promoters could point to hospitals that were stuck with the costs of labs that had very skilled technicians who had to be available 24 hours a day to diagnose diseases from test samples.
The urgency for the equipment exploded in the fall of 2001 when postal workers and others were contaminated by deadly anthrax spores sent through the mail. The company worked closely with government contractor Northrup Grumman to win the contract for the biohazard detection systems. Cepheid tested its second product, GeneXpert®, with the postal service for about two years and then began making it in 2004. That product is a system of cartridges and an analysis machine automates the preparation of a test sample and delivers DNA test results in as little as 30 minutes.
In 2002, Bishop replaced Gutshall as CEO and Gutshall remained chairman. Bishop says that the first phase of the contract calls for deploying the machines in 283 postal distribution centers across the country. The second phase of the contract could generate revenues equal to or higher than the first phase.
Because the second phase won't be awarded until 2005, Bishop doesn't anticipate becoming profitable until 2006. That was why he decided to raise more money in early 2004.
Bishop's mission is to make the company profitable and expand it to the next level. The company is doing so by extending its research and by licensing new tests for different kinds of disease detection. It is focusing on tests for infectious diseases and cancer. Among the tests it is trying to create is the detection of smallpox.
Cepheid might be one of those unpredictable companies that lives or dies on contract wins with very irregular financial payments, but it has a business model that makes sense in the long run. The company built two generic platforms, GeneXpert and Smart Cycler, that serve as platforms for any kind of test. Those platforms are tailored to run specific tests through the test cartridges, which are disposable and have to be repurchased.
Sales of the cartridges are more predictable and therefore help Cepheid deliver more predictable financial results. Sales of actual instruments account for 75 percent of sales, while cartridges are 25 percent. Over time, Bishop expects those proportions will reverse.
To date, Cepheid has sold more than 1,600 of its Smart Cycler machines, which go for about $34,000 each. Among the customers are university labs, hospitals, and government institutions such as the Food and Drug Administration. It makes the machines and cartridges in its 70,000 sq.ft. manufacturing facility at its Sunnyvale headquarters.
Cepheid lost $17.3 million in 2003 but it expects to reduce that loss this year as the postal revenues accelerate. By raising the money earlier this year, Cepheid was able to give itself a hefty cash cushion and also pay for a license from rival Roche, which owned some fundamental patents on DNA testing that Cepheid needed, Bishop says.
Petersen and Bishop look back on the spinoff from Livermore and see a lot of lessons. Among them: they maintained good communication with the lab long after the licensing contract was negotiated, meeting two or three times a year. They also say that patience paid off in dealing with the long negotiating period and the long time it took to commercialize the technology. The company won some Army contracts early on, but Petersen made sure that Cepheid didn't bog down in low-volume government business and kept focused on the commercial market. And though it licensed important technologies, the company has been filing for and continues to collect patents on technologies that it has invented itself.
Adds Dunipace, "You have to be really passionate about the idea, and they were."
One day, some hope that the company can come up with universal tests to detect any kind of disease and even deliver test results immediately while a patient is in a doctor's office. While Bishop says those are laudable goals, "I don't see it, at least not in the next five years. If you could do tests much faster, you could do much better patient management."
Dean Takahashi is a staff writer at the San Jose Mercury News. He can be reached at dtakahashi@mercurynews.com
Copyright © 2012 | Innovation America