By Turna Ray
23andMe's new report on Alzheimer's disease risk has raised some eyebrows at Duke University regarding the fact that the direct-to-consumer genomics firm is testing customers for APOE variants without a sublicense for patents on testing methods held by the university and licensed to Athena Diagnostics.
"We are always looking at entities, particularly DTC companies, who may be marketing APOE tests without a sublicense from Athena, are making misleading representations of those tests and their clinical interpretations, and are generally violating patent claims we own and have licensed, accordingly," a spokesperson from Duke's Office of Corporate & Venture Development, told PGx Reporter. "I'm sure we will be examining with Athena to ensure compliances."
Earlier this month, when 23andMe joined other DTC genomics firms in offering customers the option to learn their APOE genotypes, several prominent leaders in the genomics field criticized the move. Given that Alzheimer's is an irreversible disease lacking viable medical interventions and APOE genotypes don't definitively predict who will get the disease, these critics felt it was irresponsible for DTC firms to be selling this genetic risk data to the general population (PGx Reporter 04/20/2011).
But intellectual property could pose another obstacle to DTC firms who are reporting this data as part of a commercial service.
"There is an active patent held by Duke University with an exclusive license to Athena — and well-defined rules to perform the test on a doctor's signature certifying cognitive impairment," Allen Roses, director of Duke University's Deane Drug Discovery Institute, told PGx Reporter.
"Charging as a commercial test for Alzheimer's disease risk without a valid sub-license from Duke and Athena would be in violation of the patent on several counts. This is not a qualified research use but has a purely commercial intent," Roses said.
Athena garnered exclusive, worldwide patent rights to the diagnostic uses of APOE from Duke in 1995. The association between the APOE4 variant and increased risk of late-onset Alzheimer's disease was first discovered in Roses' lab at Duke.
Whether Duke will take any legal action is not known. Going on comments from Duke's Office of Corporate & Venture Development spokesperson, it's clear the university is considering its options as to how to deal with DTC firms it believes are infringing its APOE patents.
23andMe is not the only company reporting APOE genotypes to its customers. Decode Genetics' DecodeMe service reports on variants in the APOE and CLU genes to gauge the risk of developing the illness in customers of European descent. Navigenics, which previously sold genomic risk information under a DTC model but now markets its service through doctors and employer wellness plans, also reports information on APOE status and the risk of Alzheimer's disease.
23andMe refused to discuss this issue, stating that the company "does not comment publicly on the status of specific patents or licensing agreements." Navigenics and Decode did not respond to queries about their licensing practices regarding APOE testing before press time.
It's difficult to gauge what "compliance" action Athena or Duke will demand of DTC firms from a legal standpoint, if any, since the status of gene patents is currently in flux due to the ongoing case, Association for Molecular Pathology et al. vs. the United States Patent and Trademark Office et al. The case, in which the American Civil Liberties Union and the Public Patent Foundation are challenging a number of Myriad Genetics' patents on BRCA 1 and BRCA 2 genes, was recently heard by the Federal Circuit Court of Appeals, and will likely go to the Supreme Court (PGx Reporter 04/06/2011).
In the opinion of Robert Cook-Deegan, director of Duke's Center for Genome Ethics, Law, and Policy, there are currently more reasons for Athena not to take any legal action against DTC firms than there are in support.
According to Cook-Deegan, it's not clear if the courts would interpret Duke's method claims for gauging APOE variants to be infringed by a microarray test, which is the technology underlying most DTC genomics' firms gene scans. Additionally, Athena's "patents are very broad method patents that merely claim an association," he noted. "These are just the sorts of claims that the Supreme Court signaled as questionable in the Metabolite dissent, so Athena would be taking a real risk of losing" any patent challenge.
[ pagebreak ]
Cook-Deegan was referring to the case LabCorp v. Metabolite, in which Metabolite Laboratories sued the Laboratory Corporation of America for infringing its license to a patented method for measuring the level of the amino acid homocysteine in the body and using that data to determine whether a person has vitamin deficiency. Although the Supreme Court dismissed the case as "improvidently granted" in 2006, Justice Stephen Breyer wrote in a dissenting opinion at the time that Metabolite's claimed association between homocysteine levels and vitamin deficiency was an unpatentable natural phenomena. Specifically, he stated that the company's claimed processes for determining vitamin deficiency also wasn't patentable, since the processes described in the patent were "no more than an instruction to read some numbers in light of medical knowledge."
An example of a patent claim that might be at issue when DTC genomics firms gauge and report Alzheimer's risk to customers based on APOE variants is US Patent 5,508,167, which was awarded to Duke University in 1994. The patent claims "a method of detecting if a subject is at increased risk of developing late-onset Alzheimer's disease" through the direct or indirect detection of "the presence or absence of an APOE4 isoform in the subject; and observing whether or not the subject is at increased risk of developing late onset Alzheimer's disease."
According to Cook-Deegan, several claims in this patent are "poorly worded and confusing," and could be hard to defend in light of Justice Breyer's Metabolite dissent. "They seem to claim any way of measuring APOE genotype and using it to diagnose or predict risk of Alzheimer's disease," he said.
Additionally, the ongoing AMP v. USPTO sheds little light on how the courts would view Duke's patents. In that case so far, the focus has largely been on claims on isolated gene sequences, and not so much on method claims. "There may be some clues about jurisprudence in such cases from the current Myriad case, if the Court of Appeals rules on the method claims," Cook-Deegan said. Although method claims haven't gotten much attention in AMP v. USPTO, "they are actually more important because method claims are both more numerous and harder to work around in gene patents."
Another deterring factor for Athena is that it was recently acquired by Quest Diagnostics, so its new parent company "may not want to alienate potential customers by aggressive patent litigation," Cook-Deegan observed. "The Duke patents will begin to expire in two years, and Athena will need to shift business models anyway."
Finally, if Athena or any diagnostic company wanted to take a consumer genomics firm to court, there is no precedent for how damages would be awarded given the unique business model of these companies. "23andMe is testing thousands of SNPs and only a few bear on APOE," Cook-Deegan said. "It might not be worth suing over."
Driving Medical Testing
Consumer genomics firms test their customers for hundreds of gene-disease associations, which are often protected by patents that belong to other diagnostics firms and academic institutions. As such, some consumer genomics firms have come up with unique workarounds when dealing with patent and licensing issues in their service models (PGx Reporter 12/12/2008).
For example, Navigenics has developed a universal royalty model for licensing gene patents for consumer genomics services like itself. In this model, Navigenics pays a percentage of its net sales revenue from a licensed SNP factoring in the overall value of the service provided. This system, according to Navigenics, keeps the licensing fees it would have to pay to patent holders to a more manageable level than if it had to obtain licenses SNP by SNP.
According to Navigenics' website this is a payment model the company has developed with input from third parties. It's not clear the extent to which this system is used in Navigenics' licensing deals with third parties.
Still, when the Duke spokesperson indicated that the university was interested in ensuring "compliance" from DTC genomics firms, it could very well involve ensuring that a creative payment model, like the one described above, exists between DTC firms and the patent holders.
Since 23andMe won't discuss its gene patent licensing policies, it is difficult to ascertain whether the company is in any real danger of facing patent litigation. To date, no DTC genomics firm has faced patent litigation for SNPs included in their services. But when 23andMe began reporting BRCA mutations linked to hereditary breast and ovarian cancer risk in early 2009, many industry observers were baffled as to why Myriad hadn't taken action against the company.
According to industry insiders, 23andMe began reporting on BRCA SNPs without Myriad's permission. Although the move reportedly surprised Myriad, the company has not pursued legal action to date.
This may just be a matter of timing. Myriad's zeal to protect its patents is currently tied up in AMP v. USPTO, and the company may not be interested in getting involved in any more patent challenges until that case is resolved.
Another reason Myriad may be turning a blind eye to 23andMe's BRCA testing is that it stands to benefit from the service. If a 23andMe customer finds that she has a BRCA mutation, she will not be able to act on this information without seeing a doctor, who will likely recommend retesting using Myriad's BRACAnalysis. Also, insurance companies will reimburse for Myriad's test if it is deemed medically necessary, but not for 23andMe's service.
It's not clear whether DTC firms' APOE testing will drive more business Athena's way, but it may, if customers of these services see their doctors after receiving their genotype data and the physicians then find a need for more detailed genetic testing in a medical setting.
Athena markets a number of APOE tests on its ADMark platform for late-onset Alzheimer's disease. These tests are sold as a tool to help doctors determine the likelihood that Alzheimer's is the cause of patients' dementia, based on the presence of APOE variants and other biochemical markers, such as Tau and beta-amyloid, family history, and other medical information.
Have topics you'd like to see covered in Pharmacogenomics Reporter? Contact the editor at tray [at] genomeweb [.] com.