This article was originally published on July 23.
Illumina expects array revenues to remain flat for the next several quarters, but to begin to pick up early next year, according to the firm's CEO.
Jay Flatley attributed a "faster than expected slowdown" in the San Diego firm's array business to researchers delaying new rounds of genome-wide association studies as they await next-generation content. They are also holding off on purchases until they find out whether their projects will receive stimulus funding under the American Recovery and Reinvestment Act of 2009, he said.
"We are very optimistic about the future of our array business and believe that the availability of new content will create a second round of association studies that will require increased sample numbers for statistical significance," he said. "Our model for this business is for it to continue to be flat for the next several quarters and then begin to accelerate again in 2010."
Flatley made his remarks last week during Illumina's second-quarter earnings call. The company reported that total revenue rose 15 percent to $161.6 million from $140.2 million in the second quarter of 2008. Illumina's net income for the quarter nearly doubled to $24.7 million from $12.7 million in the second quarter of 2008.
Investors were pleased with the results and shares of the firm's stock rose 7 percent to open at $35.38 on July 22, the day after the firm announced its earnings.
Research and development spending increased 41 percent to $33.1 million from $23.5 million in the year-ago period. Selling, general, and administrative expenses rose 18 percent to $41.9 million from $35.6 million for the second quarter of 2008.
As of June 28, Illumina had $790 million in cash and investments, of which $352 million was in cash and cash equivalents.
Earlier this month, Illumina trimmed its second-quarter revenue forecast by between 4 and 7 percent to $161 million from an initial estimate of between $168 million and $173 million, citing order delays for its arrays (see BAN 7/7/2009).
The decline in the array business has largely been offset by demand for the company's next-generation sequencing products. Flatley said that since Illumina now sees "parity in profitability from sequencing and microarrays" it is "relatively indifferent to the mix of orders between these two technologies."
The Decline and Potential Rise of GWAS
While Illumina sells a broad menu of array products, including chips for custom genotyping, gene expression, and microRNA profiling, demand for its higher density, genome-wide genotyping arrays used in association studies has receded the most, Flatley said. He attributed this lull to the fact that customers are waiting for Illumina to put rare variant content generated by the 1000 Genomes Project and other sources on its chips before diving into a second round of GWAS.
"The rapid decrease in the cost of sequencing is enabling the discovery of rare variation in the genome, which, once incorporated on arrays, will provide dramatically increased power for making disease associations," Flatley said.
He specifically cited 1000 Genomes data from a May 2009 meeting at Cold Spring Harbor Laboratory. As BioArray News sister publication In Sequence reported at the time, the 1000 Genomes researchers found around 21 million SNPs after sequencing 178 HapMap individuals from four populations at low coverage as part of a pilot study. The researchers consider novel about 11 million SNPs of the 21 million SNPs identified.
By the end of this year, the 1000 Genomes consortium plans to sequence at least 1,200 individuals — both HapMap and newly collected samples — of European, East Asian, and African ancestry at 4-fold coverage. It also plans to expand the sample sets to additional populations in the future.
"There is a high degree of confidence that buried in the content that will be discovered in 1000 Genomes is much of the remaining heritability of diseases," Flatley said this week. "I think that all it will take is the discovery of a few very key disease associations that explain significant portions of heritability, and you will see tremendous amounts of money pour into whole-genome genotyping market if and when that happens."
[ pagebreak ]
Flatley added that the company will go through cycles of making the 1000 Genomes content available on arrays. The typical product cycle from design to manufacture and commercialization is between three and six months, he said.
Illumina has already launched an array with next-gen content for clients running association studies. In May, the company launched its 4-million feature HumanOmni1-Quad BeadChip. The array includes more than 100,000 new SNPs and markers from the 1000 Genomes Project and more than 11,000 copy number variants from the Wellcome Trust Sanger Institute, the Hospital for Sick Children in Toronto, Harvard Medical School/Brigham and Women's Hospital, and Decode Genetics (see BAN 5/19/2009).
Flatley said Illumina expects a "full range of behavior" from its customers with regards to the adoption of new chips. Some customers will continue using existing arrays in their studies like the Human660W-Quad and the Human1M-Duo, while others will shift to the newer arrays as they come on line.
"We are already seeing significant adoption of the Omni and so there will be quite a number of customers that believe there are opportunities to get low-hanging fruit off the Omni with the rare variant content that we have on that chip," Flatley said.
"The first bolus of content has already come from the 1000 Genomes program and we are partnered with a number of participants in the program to sift through the SNPs they are discovering and to determine which ones are the most important ones to put on a next version of a chip," he said. "It takes time to do that, but I think by the first half of 2010, you will see this array opportunity reaccelerate."
Multi-sample Indexing and the Stimulus
Earlier this month, Illumina launched GoldenGate multi-sample indexing, a high-throughput genotyping approach that enables researchers to screen up to 16 times as many samples per reaction as the standard GoldenGate Assay while decreasing total reagent consumption.
For example, if a researcher were using Illumina's 32-Universal BeadChip, "he or she could now screen up to 512 samples on the same chip using less reagents," Flatley said. "We believe MSI will be ideal for high-throughput customers running low-complexity screening, validation, and control applications in commercial agriculture and academic markets," he said.
The same customer group is also lining up for stimulus funding. Flatley said Illumina is seeing "more than double" its normal quote activity. "We are seeing significant stimulus activity around custom content arrays as customers look to conduct fine mapping and additional association work on data from existing GWA studies," he said.
Still, Flatley noted that sequencing will be "more than likely the biggest beneficiary of stimulus funding." One difference between the sequencing and microarray stimulus projects under grant consideration is that "in sequencing, there are many, sort of, 'moon shot' projects, very bold things that customers may try to do," he said, while array-related projects are typically smaller in scale.
The large sequencing-related grants are for "scientific endeavors that have never been possible before," Flatley said. "We think there is a reasonable chance that some of those will get awarded."
It is difficult for Illumina to estimate the total amount and the timing of stimulus awards, but the company expects to begin seeing some impact on its sales beginning this quarter.
For Q3, Illumina expects revenues between $162 million and $172 million, an increase of between 8 percent and 14 percent over the third quarter of 2008. During this week's earnings call, it also reiterated its most recent forecast of full-year 2009 revenues in the range of $690 million to $720 million, which represents year-over-year growth of between 20 percent and 26 percent.