YouTube Facebook LinkedIn Google+ Twitter Xinginstagram rss  

By Malorye A. Branca

November 15, 2003 |  The first wave of single microarrays capable of interrogating the entire human genome sequence are hitting the market, signaling a new phase in the chip market’s evolution.

NimbleGen Systems claimed victory in the race to produce the first whole genome chip product, with Affymetrix close to it. Agilent and Amersham Biosciences say they’ll be selling off-the-shelf, or “catalog,” whole human genome chips by year-end. 

These new arrays are important because they save time, material, and money. But they are not a quantum-step improvement, says Gene Brown, senior director and head of Wyeth’s expression profiling group. Scientists have been surveying the entire genome for some time, albeit not on a single chip.

A key priority is to develop probes for genes expressed at low levels. Amersham says that its CodeLink chips, with a unique 3-D surface chemistry, do a better job at fishing these out. But how much better? Many established expression labs already have reams of Affymetrix-generated data and labs full of Affymetrix-friendly tools.

“The biggest challenge is to convert the Big Pharma customers who have been entrenched with Affymetrix,” says Sam Raha, Amersham’s vice president of CodeLink.

Agilent also says that its chips have improved sensitivity, and points to the industry-standard size of its microarrays -- 1 inch x 3 inch (25mm x 75mm) -- compatible with most scanners. “You don’t have the same huge upfront investment with our chips,” says Agilent’s Christina Maehr.

Both Amersham and Agilent say they are gaining market share, although Affymetrix marketing director Elizabeth Kerr says, “We’re not feeling any pressure from competitors. All the pressure we feel is from customers.”

The microarray market leaders are moving aggressively into other markets. “In the gene-expression area, our largest focus is genome screening,” Maehr says. “But we are also moving into toxicogenomics and, eventually, clinical trials and diagnostics.”

The move into diagnostics has the greatest upside, opening the doors to a market that could top $1 billion, according to recent estimates. In June, Roche Diagnostics launched the AmpliChip CYP450 array (an Affymetrix chip), which measures genetic variations in drug-metabolizing enzymes. And Agilent has a contract to produce what could become the first gene-expression-based test used in the clinic -- to detect prognostic gene signatures in breast tumors.

Diagnostic chips will carry far fewer genes than the latest research arrays. With that in mind, Affymetrix has been test-manufacturing 1mm2 versions of its standard GeneChip, with much smaller probe dimensions. As chips shrink, so does the price. According to Affymetrix, the wrapping will eventually cost more to manufacture than the actual chips, which will be priced in the “double digits.” Today, microarrays range from a few hundred dollars to $1,000 or more.

Ultimately, however, it could be Amersham that is sitting in the catbird seat when it comes to diagnostics. Besides having its own diagnostic arm, the company’s recent acquisition by GE was facilitated by both companies’ mutual interest in pharmacogenomics.

To hold its market lead, Affymetrix is covering all fronts. One strategy is to expand its specialty offerings. It is also expanding the definition of “whole genome,” so more genes, including splice variants, are represented as well as regulatory sequences.

Such advances are still a few months from the market and will not guarantee continued market dominance. Nevertheless, consultant Ken Rubenstein says, “Alternative splicing and regulatory regions are important. The priority at pharma companies is dealing with what’s available now, then looking to extend the kind of information they collect.”

Wyeth’s Brown, for example, says his headaches have nothing to do with the chips themselves. “We’re happy with the chips and the data,” he says. “The biggest challenge, whatever the platform, is the concept of annotation.” Determining what genes actually do, and which are worth following up, is a software issue.

With no single, optimal, off-the-shelf solution currently available, Wyeth is using both home-grown and commercial software, as well as consultants. “The whole field could move rapidly if there was a public annotation effort,” Brown says. “That would be a major effort, but provide a fantastic payoff.”


For reprints and/or copyright permission, please contact Angela Parsons, 781.972.5467.