mardi 7 octobre 2008

OVP, Enterprise Partners See Big Opportunity in $5,000 Human Genome Sequencing

Luke Timmerman 10/7/08, xconomy.com

It’s getting cheaper by the day to sequence the entire string of 6 billion chemical units of DNA that make up an individual human being.

Yesterday, Complete Genomics of Mountain View, CA unveiled plans for what amounts to a democratization of genomics. It will offer a service to sequence full human genomes for just $5,000, beginning in the second quarter of 2009.

At Xconomy.com, we normally focus on companies based in Boston, Seattle, and San Diego, but we couldn’t resist digging into this one, because it has multiple connections to our network cities.

Complete Genomics raised its seed capital in 2006 from OVP Venture Partners in Kirkland, WA, and Enterprise Partners in San Diego. It also counts a pair of Xconomists, Leroy Hood of the Institute for Systems Biology in Seattle, and George Church of Harvard Medical School, as scientific advisers.

So we tracked down OVP managing director (and Xconomist) Chad Waite to find out why he decided to invest in this technology versus all the other sophisticated instruments made by companies like Applied Biosystems, Illumina, 454 Life Sciences, and Helicos Biosciences. (He proudly pointed out that his Harvard Business School connection to CEO Clifford Reid gave him the inside track on this investment, and he invited Drew Senyei of Enterprise in on the action, but more on that later.)

It turns out Waite was sold on Complete Genomics because it has a fundamentally different vision of the market from its rivals.

Instead of trying to sell a machine to pharmaceutical companies and top academic labs for hundreds of thousands of dollars, Complete Genomics plans to keep the work in-house on its own proprietary machines and offer sequencing as a service.

The company plans to open 10 sequencing centers around the world over the next five years, with the capacity to sequence 1 million complete human genomes.

It will have enough bandwidth to sequence an entire genome for $5,000 in about four days, compared with $100,000 and six weeks to six months on currently marketed instruments, Waite says.

We’re disruptive on technology, and on the business model,” Waite continues. “We’re not going out and trying to sell million-dollar machines. Is there really a competitive advantage for a pharmaceutical company to have the machine? The advantage for them is in the data. They want the data.”

So how might that be really useful for companies or academics?

At that high speed and low price, it’s conceivable that drug companies will want to sequence every patient who enters a clinical trial to provide clues as to why some patients respond differently than others to experimental drugs, Waite says.

Or, they might want to run big experiments that compare the genomes of 1,000 patients with diabetes to 1,000 other people as healthy controls, to look for tiny genetic variations that might offer clues.

They could look at a bunch of prostate cancer tumor samples to try to find genomic markers that explain why the disease spreads more quickly in some people than in others, Waite says.

These concepts are truly mind-boggling when you look at the recent history of gene sequencing.

Back in 1991, when many scientists were skeptical the full human genome could ever be sequenced, Congress was told it would cost $3 billion to sequence the human genome and it would be done by 2005.

It was actually done by 2003, at a cost of $2.7 billion, according to the National Institutes of Health.

Five years later, thanks to big improvements in sequencing technology, it can now generally be done for about $100,000, depending on how you account for the expense of labor to run the machines, Waite says.

Even at that high degree of efficiency, only a couple dozen human genomes have been fully sequenced, according to this report in the New York Times.

Complete Genomics hasn’t published its methods in a top peer-reviewed journal like Science or Nature, although it has double-checked its machine for accuracy against competitors internally, and plans to publish the work soon, Waite says.

The paper hasn’t come out yet, so the following comes with a grain of salt, but the company says it achieves its lower costs by using much less of chemical reagents than existing tools. It also uses ultra high-density DNA tests that can be read with commercial imaging equipment that keeps those costs down.

The company will need that validation to convince scientists and others that they are getting their money’s worth by getting sequences done at Complete Genomics.

Before the paper comes out, though, the company hopes to receive some validation already from two very prominent early partners—the Institute for Systems Biology in Seattle, and Genentech, the biotech industry’s No. 1 company by market value.

Complete Genomics’ partnership with the ISB is expected to sequence 100 individual genomes in 2009, and 2,000 more in 2010.

The new low-cost tool “will allow us to gain a more complete understanding of the genetic components and molecular processes of diseases in order to better manage, treat and prevent human disease and better understand human health,” Hood said in a statement.

Waite and Senyei invested together in Seattle-based Corixa back in the 1990s, so it was natural that they would go in together on Complete Genomics, Waite says.

Both OVP and Enterprise Partners have participated in all three rounds of funding in Complete Genomics, while Prospect Venture Partners led the second round and Highland Capital Management led the third, according to DowJones VentureSource data that my colleague in San Diego, Bruce Bigelow, found. The company has raised about $46.5 million since its founding, and has 100 employees.

People have been talking about the dream of the $1,000 genome, so I asked Waite how low can this sequencing really go. “We think it can go lower, but we’re not really going to talk about that,” Waite says. “But this is really exciting on many levels.”

Luke Timmerman is the National Biotechnology Editor for Xconomy. You can e-mail him at ltimmerman@xconomy.com, call 206-624-2374, or follow him on Twitter at http://twitter.com/ldtimmerman.