It’s curtains for CodeLink.
GE Healthcare, abandoning hope of a profitable future for the product line, will stop manufacturing its CodeLink printed bioarrays by April 2007 and will stop accepting requests from existing customers in February, the company told BioArray News this week.
GE has not indicated that it will sell its assets related to the CodeLink business, but said it will help clients switch to rival platforms as CodeLink is phased out, leaving the future of the 7-year-old microarray technology in doubt.
The decision to scuttle CodeLink comes at a time when the array unit was branching out into content-specific and customizable arrays and building its service infrastructure. Now many of GE Healthcare’s existing customers believe they will be forced to adopt rival platforms (see related story, this issue).
According to a GE spokesperson, in the nearly three years since GE Healthcare obtained CodeLink through its acquisition of Amersham BioSciences, “GE has invested in the CodeLink business with the goal to bring a differentiated product to the research segment and establish a significant position in the microarray research market.” However, the company said it was not confident in the future viability of the product line.
“Unfortunately, while we believe we have delivered a high-quality product, our prospects to profitably grow this product long-term have fallen short of expectations, and as a result GE has decided to discontinue the CodeLink printed microarray product line during early 2007,” the spokesperson said.
She added that GE will continue to manufacture and sell CodeLink activated blank slides for homebrew microarrays indefinitely.
GE does not break out revenue for the CodeLink bioarray unit in its quarterly or annual earning’s reports.
The spokesperson said there are several reasons GE decided to stop producing the printed array slides, “one of which is that the majority of future molecular diagnostics products will use technologies other than microarrays primarily due to the small number of genes that will need to be measured for most clinical molecular diagnostics applications."
CodeLink is “designed to measure thousands of genes, and thus cannot compete with the cost and ease-of-use attributes required to succeed in future clinical molecular diagnostics,” she said.
The company will support all of its current customers and will serve researchers who are using CodeLink until they finish their projects, the spokesperson said. At that point CodeLink sales representatives will contact customers to help them transition to alternative platforms from competitors.
GE Healthcare has not announced its intentions for CodeLink once it is wound down. The spokesperson said that it is possible that GE could put the CodeLink portfolio on the block, but so far the company has only decided to discontinue making and selling pre-printed arrays.
Birth, Growth, Death
CodeLink was originally developed by scientists at Motorola Life Sciences in 1999, and was sold to Amersham Biosciences in August 2002 (see BAN 8/2/2002). GE Healthcare gained the CodeLink unit through its $9.5 billion acquisition of Amersham Bioscience two years later (see BAN 4/14/2004).
The decision to leave the printed array business comes after GE helped grow the CodeLink line from a basic expression business offering whole-genome human, mouse, and rat arrays to more tailored content for clinical research.
As recently as two months ago the firm launched a series of mammalian inflammation bioarrays for human, rat, and mouse and announced plans to introduce metabolism and women’s-health-themed arrays next year (see BAN 10/10/2006). Earlier this year it also rolled out a new reagents kit and introduced an online custom array design tool called iCenter (see BAN 3/28/2006).
In addition, scientists at GE Healthcare played a prominent role in the first phase of the Microarray Quality Control project. In fact, CodeLink scientist Richard Shippy told BioArray News that the MAQC results — which the authors claim show concordance across all commercially available microarray platforms — demonstrated CodeLink’s “high level of detection sensitivity” that enabled CodeLink to detect “30 percent more genes relative to the other commercial microarray platforms” in the project (see BAN 10/3/2006).
Ironically, one CodeLink user who helped develop the platform suggested that the MAQC project could have helped to finish off CodeLink. Scott Magnuson, a former developer of the CodeLink technology at Motorola Life Sciences, told BioArray News this week that because the MAQC results showed concordance across platforms, it made CodeLink a less-compelling sell. Magnuson currently offers CodeLink analysis services as CEO of GenUs Biosystems, a microarray services firm.
“The majority of future molecular diagnostics products will use technologies other than microarrays primarily due to the small number of genes that will need to be measured for most clinical molecular diagnostics applications.”
“I think that MAQC showed that all platforms are reasonably close in consistency,” Magnuson said. “That may have contributed to GE’s decision. Maybe they felt that there wasn’t enough to separate them from their competitors.”
He also said that GE perhaps did not do enough to sell the technology after it acquired it with Amersham. “I am not confident that GE did a great job marketing it and selling it,” said Magnuson. “From what I understand CodeLink was really turning the corner with regards to cost and data.”
In response, a GE Healthcare spokesperson said that the MAQC results “illustrated that CodeLink is a great platform” and that the company is “proud of the CodeLink technology and the technical performance of the product.”
GE reiterated that its decision to jettison CodeLink was based on its view that the platform could not be competitive in clinical molecular diagnostics.
This week, GenUs’ Magnuson also said it was not too late for another company to step in and save the technology, offering Invitrogen, Qiagen, and Eppendorf as companies that could show interest.
“There are still plenty of options out there,” Magnuson said. “Who knows where [CodeLink] will wind up?”