When drug-metabolism enzyme data is actually used to guide prescription decisions, Canadian company Seryx would like to be the clinician's middleman.
In a common vision of personalized medicine, a clinician discovers the genotype of a patient or a pathogen, and uses it along with more traditional clinical factors, such as weight and gender, to adjust the dose or the choice of drugs prescribed to that patient.
But the middle part is complicated. Gene variants can suggest adverse drug responses, the rate of drug metabolism, or the susceptibility of a pathogen to treatment, among other things. In the case of drug metabolism, for example, there are several metabolizing enzymes and genes, and drugs can compete with each other for one or more of them simultaneously. It would seem that there is a market for tools to help clinicians — who are rumored to be wary and uninformed about pharmacogenomics — sift through the data for decent dosage estimations, but so far, surprisingly few companies have stepped forward.
Seryx is planning to offer a database service in mid-September that uses drug-metabolizing-enzyme genotypes and other clinical information to point clinicians to research in the clinical literature that should inform their drug and dosage choices. The service, called Signature Genetics, is an early version of the company's Drug Regimen Optimizer, which will accept genotyping data and clinical data to suggest actual drugs and dosages for individual patients. Seryx filed a patent for the DRO last week.
"I can tell you for sure that HIV is further ahead than any subspecialty in medicine in this area, there's no question about it. All the other specialties will go down this path, but they're going to have to learn from what's been done in HIV."
The Drug Regimen Optimizer "is going to be the most comprehensive IT tool that exists," taking into account multiple gene-variant, drug-drug interaction, and other clinical information, "and actually pumping out recommendations," Reggie Downey, Seryx vice president of sales and marketing, told Pharmacogenomics Reporter.
Because the DRO will produce clinical recommendations rather than simply directing clinicians to relevant research, the product must be regulated by the US Food and Drug Administration. The company plans to submit the full DRO for FDA clearance in 18 months, and it plans to have the final product on the market in the second quarter of 2008, said Downey.
It seems that Seryx is the only one tackling drug-metabolism enzymes. The company announced that it had filed a patent for the second version of the DRO "so that if anybody is working on anything that's similar, then they'll have to disclose," said Downey.
Like Roche Diagnostics, with its CYP450 AmpliChip, Seryx is planning to market the first version of its product to psychiatrists, whose commonly prescribed selective serotonin reuptake inhibitors and antipsychotics are very often metabolized by the CYP450 family of enzymes. "The psychiatrists are already educated on it, they're already aware of the 2D6 variants," Downey said. "It's a crowd that's got a will and an interest to incorporate pharmacogenetic testing into their practice."
By the time the second version of the DRO is ready, CYP450 testing should be reimbursed and mainstreamed, allowing Seryx to sell its wares to family practices, Downey said. "In three years, we anticipate that family doctors will be doing this, and in three to five years, it will be routine practice," he said. Downey also anticipates markets in respiratory medicine and oncology.
But so far, Seryx doesn't know for sure what size the DRO's market will be. The market for CYP450 testing, which Seryx plans to indirectly exploit through its genotyping-data interpretation, should reach "about $100 million" in three or four years, Downey said, attributing the estimate to Roche Diagnostics. "It's evolving, so those are numbers that we're trying to crunch out right now," he said. "Until we have a definite answer on that, we're working with the same numbers that they have."
Signature Genetics, the initial product due in September, costs $290 for one year of access to its database for single-gene analysis. Analysis involving two or three genes will cost $390 or $590, respectively. The database contains information about CYP450 genes 2D6, 2C19, 2C9, 1A2, as well as the NAT2 gene. All analyses involve clinical factors that can impact the rate of drug metabolism, such as a patient's current drug regimen or a patient's gender.
Looking for a Parallel Application
The only technology currently on the market that is comparable to Seryx's proposed service is HIV drug-resistance prediction. This kind of service, offered mostly by academic laboratories and clinical laboratories, attempts to determine which drugs are most appropriate for the treatment of a particular patient based on the genotypes of HIV infecting that patient.
As a result of years of HIV genotyping and sequencing research, there are a number of academic laboratories focused on understanding the relationship between certain mutations and drug resistance, said Mike Bates, vice president of clinical research at ViroLogic. "If you're in Atlanta, [for example], there is a big sequencing lab at Emory [University] that you can send your patient's virus to. They can generate a sequence for you, and there are lots of algorithms floating around that will give a prediction about whether a strain is drug resistant or sensitive," he said.
ViroLogic offers a similar system, called GeneSeq, that is offered by large clinical labs, Bates said. The algorithms offered for determining HIV drug resistance from sequence data "are not all equal," said Bates. Of a database of approximately 175,000 HIV-strain drug-response phenotypes, the company has correlated about 75,000 to particular genotypes, he said.
If CYP450 testing and data-interpretation go the way of the HIV field, there is still much complexity to add to the host of clinical factors and genetic data Seryx is collecting. Because mutations in distant parts of HIV genes can attenuate the effect of well-known mutations in HIV, the virus' drug response can vary wildly with little actual genetic variation. "I can tell you for sure that HIV is further ahead than any subspecialty in medicine in this area, there's no question about it," said Bates. "All the other specialties will go down this path, but they're going to have to learn from what's been done in HIV."
Bates said he didn't have the market size of HIV genotype-interpretation on hand, and ViroLogic was unable to supply the figures before this article went to press.
— Chris Womack ([email protected])