This story has been updated from a previous version. Illumina has submitted an undisclosed genetic test with its BeadXpress system to the FDA, not a PGx panel as previously reported.
Marking a shift in strategy, Illumina plans to submit its iScan system and accompanying arrays and software to the US Food and Drug Administration for clearance, according to a company official.
Greg Heath, Illumina's senior vice president and general manager of diagnostics, said the company will submit a cytogenetics package to the FDA as part of its preliminary investigational device exception, or pre-IDE, process.
As part of that process, test makers can send analytical or clinical protocols to the FDA for review and comment before proceeding with studies and discuss possible regulatory pathways.
The decision marks a shift in strategy for Illumina, which had previously positioned its digital microbead-based BeadXpress reader as its diagnostics platform of choice while marketing its high-density arrays for research use. Now it will seek FDA clearance for its arrays, too.
"We will be migrating the iScan system into the clinical space and we will bring this through the FDA," Heath said during Illumina's annual R&D day for analysts, which was webcast earlier this month. "The first area we want to attack is in the space of cytogenetics. We decided to submit a pre-IDE to the FDA to start the dialogue about using our cyto arrays for this clinical indication."
Illumina launched its automated, high-throughput array scanner, the iScan system, in April 2008, and debuted an offering for cytogenetics the following November that included its HumanCytoSNP-12 DNA Analysis BeadChip and KaryoStudio cytogenetic software (see BAN 11/18/2008). The cyto array market includes a myriad of competitors, including Affymetrix, Agilent Technologies, BlueGnome, Oxford Gene Technology, Roche NimbleGen, and others.
Each of these firms is looking to cash in on the increasing popularity of arrays in cytogenetics when used either on their own or in tandem with existing technologies like fluorescent in situ hybridization. During his talk, Heath said the firm anticipates that by 2015, arrays will support half of the assays used in cytogenetic testing.
Illumina's submission to the FDA includes the iScan, plus the HumanCytoSNP-12 chip and HumanOmni1-Quad, launched last May (see BAN 5/19/2009). The HumanCytoSNP-12 contains nearly 300,000 genetic markers per sample that target cytogenetic abnormalities found in genes and disease pathways linked to mental retardation, autism, and other common chromosome anomalies.
The 4-million-feature HumanOmni1-Quad includes rare variation SNP content 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. The package also includes Illumina's KaryoStudio software.
Heath said that Illumina's pre-IDE discussion will allow the firm to "start the discussion with the FDA about intended use, structure of clinical trials, and get some guidance from them about how we are going to approach this." The company has already submitted its BeadXpress reader to the FDA paired with an undisclosed genetic test.
Heath said the firm submitted the assay with BeadXpress to bring the instrument under the FDA's discretion. The firm intends to offer tests via its Clinical Laboratory Improvement Acts-compliant lab in San Diego.
"We've gone back and forth in Q&A sessions with [the FDA] and we don't have clearance yet but we feel that it's imminent," Heath said of the BeadXpress system.
He said the firm has also been offering a research-use-only 185-marker pharmacogenomics panel for drug metabolism to some pharmaceutical partners, which use it in their drug-development programs. Out of the 185 markers in the assay, he said there is "huge potential" to eventually bring each one through the FDA's 510(k) approval process.
Heath said that Illumina has other continuing diagnostic programs, including one each on gastric and ovarian cancers. The company is also developing a multi-drug-resistance panel, a herpes panel, and a respiratory viral panel, all of which will likely be deployed on the BeadXpress.
Since it acquired the platform's foundational VeraCode technology with its buy of CyVera in 2004, Illumina has positioned BeadXpress as the vanguard of its molecular diagnostics lineup.
[ pagebreak ]
However, Illumina's views on how the platform may be used have evolved. While a rival like Affy has staked out the diagnostics market with the help of its Powered by Affymetrix program, Illumina at first seemed to rule out following suit.
"The sweet spot for diagnostic applications will be at multiplex levels between 10 and 150 markers," CEO Jay Flatley told BioArray News in March 2007 (see BAN 3/27/2007). "Microarrays are overkill at this level of complexity and do not offer the flexibility or cost dynamics that this market demands." By November of 2008, however, Flatley envisioned the future use of arrays in routine diagnostics, particularly in personal genetic testing.
"We believe that whole-genome genotyping will be done as a standard of care in about five years," Flatley said at the time. "If it doesn’t happen, we should be disappointed because it won’t be the result of a limitation of our technology, it will be because of a limitation of our imaginations or our infrastructure" (see BAN 11/18/2008).
Flatley revived the idea of a newborn screening panel during this month's R&D day. Eventually, "all pharmacogenomic testing will be done on one microarray, and ultimately it makes sense to do [that] on every baby that is born," he said.
Illumina's interest in molecular diagnostics is backed up by sizeable market estimates. Tristan Orpin, the company's vice president of commercial operations, said that the molecular diagnostics market is worth about $3 billion per year.
Arrays Not Dead Yet
Illumina's decision to initiate pre-IDE talks for its array platform comes at a time when the firm has felt compelled to address the impact that next-generation sequencing platforms, including its own, are having on the array market.
Earlier this month, for example, Flatley told analysts at the JP Morgan Healthcare Conference in San Francisco that sequencing is still "not an applicable technology for the next round of [genome-wide association studies] that will happen over the next few years" and that "arrays are a hundred times cheaper, a hundred times faster, and materially more accurate than sequencing will be over the next couple of years" (see BAN 1/19/2010).
That hasn't stopped some in the industry from forecasting the decline of the technology. Following the launch of Illumina's high-throughput HiSeq 2000 sequencer earlier this month, for instance, an analyst from Barclay's Capital said the new sequencing products made products from pure-play array firm Affymetrix "obsolete," particularly in the gene expression analysis market.
Likewise, Morningstar Research said it was placing Affy's stock "under review," and said that there is a risk that "customers may choose to forego upgrades to Affymetrix's devices to buy next-generation sequencing devices."
Based on statements from Flatley and other Illumina officials during the company's annual R&D day, however, Illumina isn't worried about sequencing replacing its genotyping array business, though it has positioned its sequencing instruments as an alternative to the expression array market, where Affy continues to be the dominant player.
"We have heard a lot in the last few quarters about the death of microarrays, and we hope we can convince you that the death of microarrays has been exaggerated," Flatley said during the R&D day. "We think that microarrays are the optimal solution where the content you want to analyze is known, the content doesn't change with time, as it does with cancer, and you are looking at a part of the genome that is well assayable."
Flatley also said the steady decline in sequencing costs should embolden researchers to adopt Illumina's arrays in the applied markets, notably ag-bio.
"Driving the price down [of next-gen sequencing] opens up the next layer of the ag market," Flatley said. "If it only takes $10,000 to sequence a genome, then a whole new family of organisms will be sequenced, and those markers will wind up on microarrays."
Illumina has already launched focused arrays for bovine, ovine, equine, canine, and porcine research. A maize-focused chip is expected to launch soon (see BAN 1/19/2010).
Still, Illumina officials said they are eager to see researchers transition their expression projects from arrays to its sequencers.
[ pagebreak ]
"We have now driven the price per human genome down below $10,000, and the price per transcriptome down below $200, and this is very important because it begins to enable the transition from array-based expression to sequencing-based expression," he said.
Orpin estimated that the market for sequencing will grow from $1.4 billion to $1.9 billion by 2012, and predicted that eventually next-generation sequencers will "be as ubiquitous as a PCR machine."
He noted that there are "certain array-based apps like chromatin immunoprecipitation-on-chip that have largely disappeared as a consequence" of Illumina's ChIP-seq application. In the expression market, the move of gene expression projects from arrays to sequencers "might not be quite as dramatic," Orpin said.
Still, he said that "if sequencing continues to reduce in cost, and the throughput of those systems goes up, with the richness of data that is supported by sequencing, people will transition to sequencing."
Orpin added, though, that "sequencing to us does not compete with genotyping" and called it the "driver to do high complexity high-throughput genotyping on arrays."