While the market for whole-genome SNP genotyping arrays has maintained an intense pace over the past two years, beginning with the launch of the Affymetrix Human Mapping 500K Array Set in September 2005 and culminating in the launch of Illumina’s Human 1M BeadChip this June, the two major players in the sector have been hesitant to lay out new milestones for the future.
In August, Illumina CEO Jay Flatley told investors that the firm has the capability to launch a 2-million SNP chip now, but he anticipated that as of next summer, 40 percent of users will still be using the company’s Human 1M chip, while another 40 percent will still be using the HumanHap550 chip launched in April 2006 (see BAN 8/14/2007).
Meantime, Affymetrix’s genotyping plans lie in selling the idea of using its SNP 6.0 Array for replication in whole-genome association studies and launching a pharmacogenomics panel of SNPs for drug metabolism, called D-MET, as a service by the end of this year (see related story, this issue, and BAN 9/11/2007).
Yet despite forecasts of a 2008 spent selling the products of 2007 and 2006, many users agree that SNP chip prices will continue to decline as the intense rivalry between Illumina and Affymetrix continues and as the whole-genome genotyping market reacts to price reductions for competitive technologies, such as next-generation sequencing.
Flatley told investors at the UBS Global Life Sciences Conference in New York last week as much when describing the state of the genotyping market. “Genotyping is a highly elastic market,” he said. “Over the last five years we have seen prices fall from a dollar per genotype down to one tenth of a cent per genotype.”
According to Flatley, the decline in pricing is due to “continued innovation of the chip technology and the assay performance and increased density in these chips.” He said that the combination of the two “has resulted in a dramatic uptake in the number of samples that are being run in the genotyping market.”
Another factor that has played a role in the decline of pricing is Affymetrix’s July 2006 decision to slash prices for its 500K set from $500 to $250 a chip. As of July, the 1.8-million marker SNP 6.0 Array is selling at $375.
At the same time that Affy reduced its prices, Illumina also began to cut the price of its chips, which now run around $290 for a HumanHap550 array and $580 for the Human 1M chip. Illumina does not deny it commands a higher premium for its chips than Affy, but argues that its HumanHap550 contains content that is as useful as Affy’s 6.0 arrays, therefore making them roughly equal in price.
“I think a lot of people do not understand the value that we provide over our competition. We can keep the premium pricing over the competition because our products are just so much better,” said Carsten Rosenow, senior marketing manager of DNA analysis at Illumina.
“Even if they say they have 1.8-million probes on their array, in essence they have 900,000 SNPs only,” he told BioArray News last week. “The genomic coverage they have is essentially at the same level as our [HumanHap550],” he said. “This is how we see it. It is not just a one-to-one match like some of the analysts think.”
Affymetrix has maintained at investor’s conferences and in conference calls that its SNP 6.0 Array offers the greatest coverage of any genotyping product in the market. Regardless, both Affy and Illumina have for the moment said that their prices will remain stable in light of the recent reductions.
“The pricing question is something that we always look at and see how the market responds,” said Rosenow. “We have just reduced our prices to what we think is a very attractive price point. We certainly hope that we will keep that stable,” he said. At the Bear Sterns Healthcare Conference held last month in New York, Doug Farrell, head of investor relations at Affy, similarly said he expects pricing to remain stable for some time.
Certain to Drop?
Still, while manufacturers have declared a pricing plateau, users seem to have reached a consensus that prices will continue to fall over the next two years, perhaps in response to the launch of new products, but mostly due to competition in the marketplace.
“I am sure prices will drop further in the next two years, especially when microarray companies feel the increasing competition not only among themselves but also of whole-genome sequencing technology,” said Joris Veltman, an investigator at the Nijmegen Center for Molecular Life Sciences in The Netherlands. “If not, they will face tough times,” he told BioArray News this week.
“I expect strong competitive pressures between the vendors and with new technologies to continue genotyping’s deflationary pricing trend.”
Steve Scherer, senior scientist at the Hospital for Sick Children in Toronto, said he is certain that prices will come down, but perhaps only in conjunction with the launch of new whole-genome genotyping products from Affymetrix or Illumina.
“If these products remain as their market product I suspect they might drop a little bit in price, but this would likely only happen if there was a new version coming,” Scherer told BioArray News in an e-mail this week. “Of course there [are] always price reductions based on volumes and joint projects,” he noted.
Another factor cited by users is the perpetual competition between Affy and Illumina for market share. Yik Teo, an investigator at the Wellcome Trust Center for Human Genetics, told BioArray News this week in an e-mail that “competition is stiff in this arena with few key players, and I will expect that with further developments of genotyping arrays, especially for arrays capable of assaying and detecting copy number variants, prices of the current products will drop.”
He cited the “fairly dramatic decrease” in prices across the sector following the launch of Affy’s SNP Array 6.0 as an example of the market’s rapid response to price fluctuations.
Andres Metspalu, a professor of biology at the University of Tartu and a user of both Illumina and Affy arrays, told BioArray News in an e-mail this week that the pressure that cost reductions have already created in the market could continue to factor in future pricing strategy. Specifically, Metspalu wrote that because of lower prices, users are planning larger studies, which will in turn enable manufacturers to charge even less for the same array. “Because the volumes are increasing rapidly, the profit margin can be lower,” he wrote.
“The cost per genotype has continued its historical trend of constantly declining significantly. However, overall costs have stayed high as the field expects results from more and more SNPs and now copy number variations per individual,” wrote Michael Smith, a principal investigator at the National Cancer Institute’s Laboratory of Genomic Diversity in Frederick, Md.
“Historically the genotyping vendors have been very competitive on price. I expect strong competitive pressures between the vendors and with new technologies to continue genotyping’s deflationary pricing trend,” he wrote.
The big question remaining, though, is how low Affy and Illumina will go. If scientists can dream of a “$1,000 genome” on next-generation sequencing platforms, then what is the ideal price point for a SNP chip?
“Every scientist at these conferences has been clamoring for a $50 chip,” said Alexandre Stewart, a principal investigator at the University of Ottawa Heart Institute and an Affy user. Stewart said that his institute actually limited its most recent order of GeneChips in expectation that pricing will again decrease by the time it runs through its current stock of arrays.
“We limited our order to $1 million rather than $2 million, because if the price were to come down we’d feel like fools,” he told BioArray News last month. “I wouldn’t put it past them to lower the price again,” he added.
Illumina’s Rosenow warned against waiting for pricing to come down before ordering arrays, however, arguing that it would delay scientific discovery and the benefits that come with being first to publish.
“If people want to wait, they will miss out on the hype,” he said. “At a certain point you just can’t publish in Nature Genetics any more because your disease area has already been published on,” Rosenow added.