By Turna Ray
Proposed changes to the National Cancer Institute's intellectual property policies covering inventions arising from government-funded industry/academia collaborations have some stakeholders worried that they give too many concessions to drug developers and could restrict personalized medicine research.
The changes to the guidelines, if finalized, could restrict pharmacogenomic studies conducted by academic researchers using samples from patients treated with proprietary drugs in NCI-funded research.
In an April 6 Federal Register notice, the NCI's Division of Cancer Treatment and Diagnosis announced it was seeking public comments on proposed revisions to its IP policies governing collaborations between industry and academics who receive government funding to conduct cancer drug trials under the Cancer Therapy Evaluation Program. The deadline for the public to submit comments was May 6.
CTEP obtains drugs from biotech and pharmaceutical companies for use in NCI-supported clinical trials conducted by academic researchers. Under the terms of these industry/academic collaborations, drug firms require researchers to grant certain IP rights that come about as a result of the research, in exchange for making available their proprietary treatments for research purposes.
The changes seek to essentially provide industry collaborators with first-option commercial licensing rights for any IP arising from studies that make use of pharma companies' proprietary drugs, or samples from patients treated with those drugs. Specifically, up to five years after a CTEP-funded research is completed, industry would have the right to exert licensing options for data arising research that makes use of clinical data, tissue, or samples from patients treated with drugs used in prior CTEP studies — including biomarkers that could be used to develop companion diagnostics.
Diagnostics developers and groups representing researchers have objected to the outlined changes, asserting that the policy would harm the field of pharmacogenomics.
The Coalition on 21st Century Medicine, representing diagnostic firms, clinical labs, and researchers, wrote in submitted comments that "the proposed far-reaching expansion of intellectual property rights granted to pharmaceutical and biotechnology companies would have a dramatic chilling effect on the advancement of personalized medicine."
One player in the cancer diagnostic space, Biodesix, criticized the policy proposals as "a step away from the development of a robust, independent molecular diagnostic industry."
"In this age, the drug is only as effective as the diagnostic informing the clinician. To achieve this goal, we need many trials, more trials not less, increased access and transparency and tissue and serum from all studies," Biodesix said in its submitted comments. "It is imperative that clinical data and biological samples from government-sponsored studies be made available to researchers and industry in as open and transparent a manner as possible and with few restrictions on the subsequent commercialization of important clinical discoveries."
According to NCI, it is proposing these policy changes because "the current IP option language is silent as to the disposition of intellectual property developed from data and agent-treated samples." This policy gap has threatened to limit industry participation in such collaborations, according to NCI.
In contrast, academic investigators challenge that NCI's proposed policy changes are weighed in favor of industry and severely limit their ability to financially benefit from their research.
Although NCI officials have said that many drug developers support expanded IP rights in CTEP studies, according to an October 2009 report in Cancer Letter, the policy changes were first proposed by Genentech.
Some critics of the NCI's proposed IP policy have cited E2100, a study conducted by the Eastern Cooperative Oncology Group and sponsored by the NCI, as an example of the challenges for academic researchers pursuing pharmacogenomic research on drugs previously involved in NCI-funded research.
E2100 was the pivotal study that provided the basis for FDA's accelerated approval of Genentech's Avastin in metastatic breast cancer. However, in a correlative study that utilized tumor blocks from patients in the E2100 study, published in the Journal of Clinical Oncology in 2008 by Indiana University's Bryan Schneider and colleagues, researchers identified VEGF gene markers that were associated with longer survival in Avastin-treated patients and that appeared to protect patients against drug-induced hypertension.
Schneider told Cancer Letter that Genentech provided $69,000 for evaluation of samples in the E2100 correlative study and in return had the right to be assigned a patent covering VEGF polymorphisms in anti-antiogenesis therapy (WO 2009073540) without compensation to the researchers.
"There are several problems that come up as a result of this, which is why I find the new NCI recommendations objectionable," George Sledge, a professor at Indiana University and a co-investigator in E2100 and the correlative study, told PGx Reporter. "First, these are trials where patients put their lives on the line, and where investigators spend years developing the trials, and where the drug company more or less gets a free ride to an FDA approval, based on the work of cooperative groups."
[ pagebreak ]
Schneider is currently conducting a Phase III study called E5103, which is an NCI-supported trial led by the Eastern Cooperative Oncology Group, evaluating the benefit of adding Avastin to standard chemotherapy for breast cancer patients with a high risk of relapse. Schneider has also received a $5.8 million grant from Komen for the Cure to study whether certain genetic markers will improve disease-free survival and overall survival for women being treated with Avastin and standard chemotherapy in the E5103 trial. This trial is not funded by Genentech.
"This is the type of research that would be covered by the NCI policy change if it is finalized," Sledge said.
Since the comment period for the proposed policy change closed, Genentech's fortunes with regard to Avastin have shifted. An independent panel of experts recommended last month that the FDA revoke approval of the drug for use in treating metastatic breast cancer due to its unfavorable risk/benefit ratio — a development that has led patient advocacy organizations to call for the use of PGx strategies to keep the drug on the market for those who are most likely to benefit from treatment.
Although Genentech told PGx Reporter that it is continuing to explore biomarkers identified in E2100 and other studies with collaborators, the company has not published any studies or announced any specific PGx efforts related to Avastin (see related story, this issue).
With regard to NCI's IP policy, Genentech "wants to ensure that any proposed changes to CRADAs preserve an academic institution’s ability to develop, license, or commercialize inventions resulting from correlative research ─ without preventing or slowing Genentech’s ability to bring new cancer medicines to patients and doctors," company spokesperson Edward Lang told PGx Reporter in May.
According to NCI, changes to IP rights governing CTEP collaborations are necessary since current IP option language is "silent" on how academic researchers can use data and patient samples treated with drugs proprietary to industry.
"As a result, both [drug companies] and [research] institutions have claimed an ownership interest in inventions generated from these data and materials," the notice explains. "This lack of clarity has become a major impediment in NCI CTEP's ability to obtain proprietary [drugs] from collaborators for use in CTEP-sponsored clinical studies, which has resulted in delays and threatens the continuing ability of CTEP to provide proprietary [drugs] to CTEP-funded investigators."
NCI's policy revisions would establish two categories of IP rights for inventions arising out of CTEP-funded trials. In the first category of policies covering inventions that use drugs made available by industry, institutions would have to grant companies the option of a royalty-free, worldwide, non-exclusive license to any IP arising from the study for commercial purposes. Industry would have between three to six months to decide whether it wishes to exercise this option. Furthermore, for this category of inventions, regardless of the industry collaborator's decision to seek a commercial license, the research institution would still grant commercial entities a paid-up, nonexclusive, royalty-free, worldwide license for research purposes.
The second category of policies cover inventions that arise out of previous CTEP-funded clinical trials, use specimens from patients treated with drugs involved in CTEP-funded studies, or utilize samples from CTEP-funded tissue banks. For inventions arising from these studies, researchers must give industry non-exclusive, worldwide, royalty-free rights for research purposes and the first option to negotiate exclusive commercial use terms. Industry collaborators have between 12 to 18 months to express "interest" in obtaining an exclusive license to the invention. These terms would apply for academia-led inventions that arise up to five years after the release of data on the primary endpoint on a CTEP-funded study.
[ pagebreak ]
It is this second category of IP protections that some researchers and diagnostic firms have pointed out could not only threaten pharmacogenomic research, but impact advancements in the diagnostics sector.
"The proposed IP policy revision does not identify an effective date, thus it is unclear whether all existing or currently developed diagnostics arising from relevant NCI-sponsored trials would also fall under these terms," the Coalition for 21st Century Medicine wrote in public comments to the proposed policy. "Given the comparative financial strength of the parties, pharmaceutical and biological companies could easily undercut innovative diagnostic companies, the majority of which have significantly fewer resources. This would cripple the diagnostics industry."
Similarly, Indiana University's Sledge pointed out that the proposed policy allows drug companies to reap the benefits of the work of other parties. Of course, Genentech, a subsidiary of Roche, is one of the few drug developers with the capacity to not only develop its own diagnostics in house, but it was one of the first companies to partner with diagnostics firms to develop companion tests for its personalized cancer drug Herceptin.
Genentech has said that it has discussed the CTEP IP policy with NCI officials. Beyond the IP issue, recognizing its advantageous position in this emerging space, Genentech has played an influential role in asserting the pharmaceutical industry's interests in other key policy issues spurred by personalized medicine advances.
Last year, Genentech submitted a Citizen Petition to the US Food and Drug Administration, pointing out that the agency's practice of enforcement discretion over laboratory-developed tests creates an uneven playing field for those in industry, like Genentech, that are subject to the highest regulatory requirements when introducing a genetic test that predicts response to a drug. In deciding to regulate all LDTs this year, the agency acknowledged that the absence of FDA oversight may have given an unfair advantage to LDT developers (PGx Reporter 06/06/10).
Impact on PGx Research
"Critical to the future of personalized medicine, is the ability of innovative diagnostic companies to rely on CTEP studies that utilize clinical data or specimens from patients treated with the therapeutic agents, including patient samples obtained from CTEP funded tissue banks," the Coalition for 21st Century Medicine said in its comments.
If NCI's proposed policies are finalized, researchers fear that financial incentives for research conducted by cooperative groups would be further reduced. "Patient tissue samples are essential for advanced diagnostics research, but under existing practical, legal, and regulatory conditions, the universe of suitable patient samples is limited," the Coalition for 21st Century Medicine noted. "If that universe is effectively narrowed even more by expansive collaborator IP rights [that] discourage the use of CTEP samples (because the collaborator retains commercial rights to any discovery resulting from use of such samples), the development of advanced personalized diagnostics will suffer."
Another fear is that a drug company could conceivably license a gene discovery that would limit a drug's market and then sit on it, restricting other researchers and diagnostic firms from bringing such a test to market.
"Absolutely, that kind of delay would be the major concern," Sledge said, acknowledging that NCI's proposed IP policies would make those kind of setbacks a real possibility.
This is not to suggest that Genentech intends to purposely delay PGx research that would cut into the nearly $1 billion in sales Avastin netted from the metastatic breast cancer market — even if the FDA ends up not revoking its approval. Although Genentech has not detailed any personalization plans for Avastin, the company has generally said it is continuing research in this area. Gene associations for drug response undoubtedly take a long time to validate, and FDA's growing regulatory requirements in this area only increase the time frame in which such discoveries can be translated into companion tests and changes in drug labeling.
This translational gap may be a reason why PGx data in the E2100 correlative study did not feature at all in the recent advisory committee discussion on whether to keep Avastin on the market in the metastatic breast cancer setting.
Lang explained in an e-mail that Genentech supports NCI policy changes as the company "would like to be able to disclose, in any materials about the medicine, relevant information from correlative research using 'agent-treated tissue' that helps patients and doctors determine if a Genentech medicine is right for them, and is required by a regulatory authority (like the FDA)."
In Genentech's view, NCI's proposed IP rights are fair to industry, since the company has spent more than $2.25 billion to study and advance Avastin as a treatment for various cancer indications. Lang maintained that Genentech's policy position does not prevent institutions from licensing discoveries to other companies, or prevent those other companies from researching and discovering associated diagnostics or information. Genentech is also not attempting to "force a diagnostic company to hand over their invention or pay the original collaborator," Lang added.
For the time being however, before NCI's IP policy changes are finalized, independent PGx research related to Avastin is ongoing. NCI has not yet issued its final IP policy in this regard. A NCI contact told PGx Reporter that the final policy would be published in the Federal Register, but did not provide a publication date.