For nearly two years now, German startup Febit has tantalized visitors to conference exhibit halls with displays of its ‘magic’ microarray machine — a silver box with a sleek, curved surface that promises to synthesize custom oligo arrays, perform experiments, and scan the arrays all within the confines of its machinery.
Now, after promising initially to launch the product at the end of 2002, the Mannheim, Germany-based company said this week it has finally begun shipping a version of this array machine, Geniom One, to customers.
“This is a major step in the life of the company,” said Cord Stahler, Febit’s CEO. “We founded Febit exactly five years ago, with the concept of an instrument which can analyze genes and do all the process steps” in microarray synthesis and analysis. “It took us five years to the date to develop that instrument, and five generations” of prototypes.
This fifth-generation instrument is similar to the prototypes that the company has previously showcased, said Stahler, but allows users to synthesize eight arrays in parallel, and includes major improvements in the accompanying software.
The company’s decision to wait to launch its instrument until it had apparently worked the bugs out contrasts to that employed by genomics instrument companies, who have often hurried to release sequencing machines or other instruments in what is really a beta format, in order to grab a piece of the market pie before it is devoured by others.
But in this case, Febit does not report suffering from its strategy. “We really started the true marketing campaign in October of this year,” said Stahler, and “we sold more than we can ship this year.” The company has begun shipping of a half a dozen of its Geniom One instruments to customers, including one big phrama, as well as others in biotech and academia. (Stahler did not disclose the names of the customers because the company had not been given permission to do so.)
The company is shipping these first instruments to customers in Europe. Febit plans to make the Geniom One available to US customers and is preparing to set up a US facility on the East Coast, in the coming year, according to Stahler, so that it can service its instruments without forcing customers to ship them to Europe.
Febit’s decision to focus initially on the European market first can be seen as partially a marketing one, due to the fact that Affymetrix has dominated this late-blooming market to a lesser extent than it does the US market. But Stahler offered the now-common refrain that his technology is “complementary” to, not competitive with, Affymetrix arrays. He pointed out that the Geniom One not only allows users to conduct second-pass analysis on a chosen subset of genes (a niche many companies are trying to fill), but also enables the quick synthesis of arrays for organisms for which no catalog arrays are available from the major platform companies. “If you look at the number of organisms and viruses sequenced, [there are] more than 2,000 in GenBank,” said Stahler, “and you can address them all with our product.” Instead of designing the experiments around an organism-specific chip, researchers using Febit’s technology can “design their experiments around a specific disease, then choose which organisms are involved in the process,” he said.
The current generation of Geniom One instruments is de-signed to produce chips that each have 48,000 features, and which are subdivided into eight separate arrays that can be used in parallel experiments. The arrays each consist of 34-micron pixels that use 30-mer oligos. As little as 30 micrograms of input sample is necessary for analysis by the system, which then produces the chips, performs the experiment, and provides a result in the form of an array image, information on the company’s website indicates.
Febit has routinely obtained raw CVs of below 0.20 for the arrays, and generally obtains CVs between 0.10 and 0.20, said Peer Stahler, the company’s chief scientific officer and co-founder (and Cord Stahler’s brother). On the company’s website, bar graphs from comparisons of Geniom One to Affymetrix Y98 chips — an old GeneChip—and cDNA arrays indicate that the Geniom One produces an overall lower CV than these chips when the same sample is hybridized and analyzed.
Febit did not provide pricing information on the instrument. However, the company told BioArray News in 2002 it was available for early access customers at a fee of $300,000 per machine, as well as $1,000 per chip cartridge.
In the coming year, Febit plans to “probably take a little more money into the company,” said Stahler, to follow the €30 million the company raised in November 2002 from a team led by the EMBL Technology Fund and Infineon Ventures.
Additionally, Peer Stahler revealed that Febit is developing applications for the system including 3’ end oligos, which “opens the arrays for molecular biology-like assays such as primer extension and mini-sequencing,” and is developing assays for the instrument for specific organisms. “We validated a bunch of human genes so that the complete human genome will be available for Geniom One users,” he said.
The company is also continuing on its quest for a perfect instrument. “The next major development will be a modular system where you will have a synthesis unit and a specific hybridization reader unit,” Peer Stahler said. “The synthesis unit will have higher throughput than the current instrument has.” The idea, he said, is for core labs to churn out arrays using the synthesis unit, then satellite labs with the hybridization reader unit to use the arrays.
But the company’s main goal this year will not, unlike in previous years, be instrument development, according to Stahler. “There’s only one goal in 2004 for the company, [and that] is reasonable sales,” he said. For sales to be reasonable in Stahler’s mind, they do not have to be in the double digits of millions, but must be significantly more than a million euro. With these significant sales, he said, “we can really show that people are interested in our product and buy it.”