The US government and a well-known academic research center are currently developing a repository of genetic data in which patient information — including names, addresses, health details, and telephone numbers — will be freely available to scores of federal and academic researchers.
The move, spearheaded by the National Institute of Environmental Health Sciences and the University of North Carolina at Chapel Hill, aims at giving scientists enhanced ability to track down leads on gene-disease links, according to the architects of the registry. The initiative may also spur pharmacogenetics research by widening the web of data being studied and increasing patient enrollment in gene-based trials.
But will wholesale disclosures of patient information — especially to the federal government — help or hinder pharmacogenetics by frightening patients with the specter of Big Brother?
The central goal of the database, called the Environmental Polymorphism Registry, is to “study the relationship between environmental exposures, genetic susceptibility, and human disease,” according to the NIEHS and UNC.
“This is a pioneering initiative,” said Paul Watkins, professor of medicine, and director of the General Clinical Research Center at UNC and an architect of the repository. “Clearly, this is the next step that’s needed to increase our understanding of how genes interact with the environment, including the body’s response to medicines.”
The EPR “is a product of our frustration of trying to do ascertainment by genotype,” added Perry Blackshear, director of clinical research at NIEHS and co-director with Watkins of the EPR. “If we identified mutations … that are prevalent in the population [and] that we think might lead to a trait or a disease, we would like to be able to search for those polymorphisms in patients and families who have those variants for possible links with environmental sensitivity.” Traditional anonymized registries prohibit this.
The EPR will “probably initially” focus on 53 genes identified by the Environmental Genome Project, an initiative within NIEHS that identified and re-sequenced “environmentally sensitive genes” involved in factors such as toxicant and drug metabolism, cell proliferation and differentiation, cell cycle, cell death, DNA repair, signal transduction, hormone receptors, and immune and inflammatory responses, Blackshear said.
Participants of the EGP identified around 520 polymorphisms in these 53 genes and estimate that around one-tenth of them are associated with “significant, non-conservative amino-acid changes or other probably significant changes in protein expression or function,” said Blackshear.
He said there are three instances in which the new registry can benefit genetic and pharmacogenetic research. First, the EPR, which the NIEHS and UNC say will ultimately contain data on 20,000 individuals, could help scientists determine the frequency of a polymorphism in a given population and relate that information to disease populations. “It’s been possible [to determine the frequency of a mutation] given typical anonymous DNA registries,” Blackshear said. “You can ascertain the frequency of a given polymorphism in a given population, but then it becomes difficult to do something about it.”
In addition, because researchers will have access to personal patient information, the registry would enable scientists to obtain family information in order to develop haplotypes. “The simplest way would be to do family studies to see how the group of polymorphisms gets inherited,” said Blackshear. Interesting results could spur a researcher to “re-contact the patient and contact their family member and figure out how a given set of polymorphisms is transmitted through an extended family.” He said this is impossible with existing anonymous registries.
Lastly, researchers armed with the EPR will be able to assemble individuals that share a mutation and launch clinical trials. This would help them determine if the mutation is linked to a phenotype.
“The data collected from these studies may be used to define environmental risk factors and develop preventative strategies to reduce disease incidence,” Blackshear said.
The data will be banked for 25 years, and patients will have the option to be removed from the database if they wish. The data will be coded, and patient information based on the code will be available to NIEHS employees who are running the study. Researchers at UNC and its various medical facilities seeking to access the data, on the other hand, will first be required to obtain approval from a steering committee that NIEHS will organize. Specifically, the researcher would describe to the committee his intent to troll the database for particular polymorphisms, and to re-contact patients with those SNPs to ask if they would like to participate in a study he is conducting on, say, endotoxin exposure. The researcher could tell the committee he would like to search for a sub-group of people in this batch based on age, gender, race, or any other factors he chooses. Once his proposal is green-lighted, the researcher would ask patients in the sub-group to participate in the study, which would test their responses to endotoxin.
Despite the overt bioethical and privacy-rights implications surrounding the registry — Blackshear said the NIEHS had enlisted the help of a bioethicist — individuals are becoming more comfortable with submitting their DNA for testing and archiving, according to formal and anecdotal research. These people are also becoming more confident that these data will not be used against them by future employers or insurance companies, he said.
For example, Stan Finkelstein, senior research scientist and co-director of MIT’s Program on the Pharmaceutical Industry, said that “patients have no problem enrolling in gene-based clinical trials.”
Still, NIEHS and UNC sought first to gauge whether patients would be willing to forego anonymity. To that end, the partners launched a pilot study that asked patients for permission to use leftover blood from sera drawn for previously scheduled tests, such as complete blood counts from check-ups. Blackshear said that all prospective patients were told that their data would be kept on file and made accessible to the federal government as well as to the academic scientists at UNC. If patients agreed, their DNA was extracted and stored.
Of 500 patients asked to participate, between 70 percent and 80 percent agreed, said Blackshear. A second pilot study is set to start next week, he said.
“In our discussions leading up to this, some people believed that virtually all patients would agree, while some people believed that virtually no patients would agree,” said Blackshear. “Genetic testing is a fairly hot-button issue. We’ve been extremely gratified that the patients that we’ve talked to in the pilot study have been willing to participate in this, as evidenced by their cheerful, informed consent,” he added. “Even with the specter of Big Brother, they were quite happy to participate.”
Privacy-rights advocates were less enthusiastic. “I’m surprised that this passes muster under classic human subject protocols,” said Lee Tien, a senior staff attorney at the Electronic Freedom Foundation, an individual-rights-advocacy group in San Francisco. “This raises many questions: Who has access, and under what circumstances? Are there adequate audit trails for security purposes?”
He mentioned the phenomenon of “mission creep” — when data collected for one purpose is used for another. “This has always been a huge problem, and it’s always very hard to enforce,” Tien said.
“We all think that doctor-patient information is confidential. But a search warrant or a subpoena to a doctor’s office is going to get those records,” he said. “It doesn't matter how private you and I think that information is … it’s going to get out there.”