With $150 million from the National Institutes of Health for its next five years of funding, the Pharmacogenetics Research Network is adjusting its course a bit.
Since the PGRN kicked off in 2000 with $140 million in NIH funding, researchers have produced about 380 publications, found more than one million SNPs, and established the Pharmacogenomics Knowledge Base (PharmGKB) to share data among network researchers. With any luck, the next round of funding is set to continue influencing the US Food and Drug Administration and the pharmaceutical industry, in addition to providing a proving ground for new technologies.
The network's research has had at least one ground-breaking policy effect so far. Based on the strength of data gathered by two PGRN projects — the Polymorphism Analysis in Drug Pathways group and the Pharmacogenetics of Anticancer Agents group — concerning genetic variation in response to the drug Camptosar, a US Food and Drug Administration subcommittee revised the drug's label in June.
"The conclusion came down that there was some important information that needed to be added to the existing label for this drug," said Howard McLeod, a professor of medicine at Washington University in St. Louis and a member of the PGRN coordinating committee. The FDA set a precedent by changing Camptosar's label to highlight the risk of neutropenia to patients carrying the UGT1A1*28 allele, and urged Third Wave Technologies to develop a genetic diagnostic to identify patients carrying the allele.
"One of the most important things we discovered during the first four years that we were funded was not so much a big scientific finding, but rather the lack thereof."
Asked to explain the effect of PGRN research on the FDA's thinking so far, Rochelle Long, chief of pharmacological and physiological sciences at the National Institute of General Medical Sciences, said that the FDA "sets up its own advisory committees and invites people to come speak where they have the expertise. I think it bodes well that they think several members of the [PGRN] are leading in their own areas."
The PGRN has also helped shift the power balance in pharmacogenomics research from one in which academic researchers are dependent on the cooperation of pharmaceutical companies to one with more even terms, said McLeod. "We've had a fair amount of interaction with the companies that make the drugs commonly used for colon cancer," including Sanofi-Aventis, manufacturer of oxaloplatin, Pfizer, which makes irinotecan under the name Camptosar, and a few makers of 5-fluorouracil, he said. "Because of the NIH money, we have not had to go to them in a reliant sort of way," but have been able to enlist these companies in large publicly funded studies, producing data for the public domain, he said.
Pharma continues to be a bit gun-shy of pharmacogenomics, and the measure of independence afforded by the PGRN funding can help ameliorate that tendency. "There's some phenomenal research being done by drug companies," said McLeod. But "the marketing dogma is still 'Get everyone to try the drug once,' rather than, 'Let's find the 30 percent market share that needs to stay on this drug forever,'" although economics favor the latter approach, he added.
A Few Changes
New NIH institutes have joined the network in Phase II, expanding the scope of research that PGRN supports. The National Institute on Drug Abuse, for example, "didn't choose to participate the first time around, I think they liked what they saw, and they did choose to participate the second time around," Long said. NIGMS coordinates the PGRN activities.
Likewise, the US National Institute of Mental Health declined to join the network during its first five years, but saw interesting research being conducted with drugs important to mental health, and decided to contribute to the second funding round, said Long. "When more institutes join, rather than dropping out, that shows a genuine interest in the program, and a belief that it's going to be a successful program," she said.
Eight projects funded under the first round of grants have been renewed under the current PGRN funding, while four original projects were not included in this second five-year installment, and three new projects have been added. (See below for a complete list of current PGRN projects.) "Each one that was new brought some dimension or different approach experimentally, or a subject area to the network that wasn't there without them, and so it was thought that that would broaden the effort and provide a more comprehensive Pharmacogenetics Research Network," said Long.
New projects added for the current round of funding are: Pharmacogenomic Evaluation of Antihypertensive Responses, which is receiving $11 million to study genetic variation in response to blood-pressure drugs; the Amish Pharmacogenomics of Antiplatelet Intervention Study, a $7.5 million project with a focus on common cardiovascular drugs, such as aspirin and warfarin; and the Pharmacogenetics of Nicotine Addiction, which was awarded $10 million (see the interview with this project's principal investigator, Neal Benowitz, in this issue).
The Amish Pharmacogenomics of Antiplatelet Intervention Study, added in this round, makes an interesting contrast to collections of unrelated individuals, which constitute the majority of PGRN grants, said Long. "By looking within families within pedigrees, they could rule out and rule in certain things that they couldn't do with unrelated individuals," she said.
Another newly granted group, Pharmacogenomic Evaluation of Antihypertensive Responses, has brought in new technological approaches. "You really can't tell what a drug is doing until you have really solid measures of hypertension itself, and I'd say they had some very advanced thinking in the way they're monitoring blood pressure," said Long.
The third new group, Pharmacogenetics of Nicotine Addiction, has been added because of the participation of the National Institute of Drug Abuse, which joined the PGRN for this round. "Because they were willing to support such research, we were able to include an invitation to support those kinds of applications, when the request for applications went out," said Long. "It reflects the breadth of approaches in the pharmacogenetics area, but the network also reflects the interests of the supporting institutes too."
The projects no longer funded through the PGRN are: the UCLA Pharmacogenetics and Pharmacogenomics Research Group, which studied Mexican-Americans' response to antidepressants; Design and Implementation of Pharmacogenetic Network, an Internet-based tool for incorporating pharmacogenomic data into the PharmGKB information library; the Autonomic Pharmacodynamic Pharmacogenomics center, which studied the correlation between gene variants and drug effects in different body parts; and research conducted by Mark Rothstein, a former University of Houston Law Center researcher who examined the social, ethical, and legal implications of the use of pharmacogenomic information.
Some of the individual projects modified their approaches as research progressed during the first phase of the PGRN. McLeod, who leads the Functional Polymorphism Analysis in Drug Pathways group, said that a new approach was needed to correct for a flaw common to pharmacogenomics research. "More often than not, the candidate genes that the scientific population selects are not correct," McLeod said.
In the last year, his group began using in vitro familial genetics studies, in which researchers use cells cultured from related family members to determine whether certain physiologically important traits — such as cytotoxicity or phosphorylation — are heritable. After determining heritability, researchers perform a genome-wide scan and fine mapping to find the gene and variant responsible for the trait, McLeod said. "That's a long way from the front-line clinical application stuff that we've been primarily doing, but we really felt it was required, because if you don't have good candidate genes, then you're never going to find a meaningful biomarker," he said.
Experience seems to have made many pharmacogenomics researchers extra careful. "I think all the groups are becoming more comprehensive in their experimental approaches, and more rigorous in their statistical examination, to make sure that what they find is a real finding, and not just something that looks good and won't stand the scrutiny of further testing," said Long.
"One of the most important things we discovered during the first four years that we were funded was not so much a big scientific finding, but rather the lack thereof," McLeod said. "The one thing we're doing differently is stepping back and doing a lot more less-biased gene-discovery type work."
McLeod said that the PGRN funding has so far proved to be a great help in securing new technologies. "It's a nice situation to be in, because whether it's 454's pyrosequencing or it's Illumina BeadArrays, or the new Affymetrix custom chips, the Parallele stuff, it's available here and it's being used here and it's applied here," McLeod said.
— Chris Womack ([email protected])
The following groups have been awarded grants under the current round of funding for the NIH Pharmacogenetics Research Network:
An overview of the accomplishments of the Phase I PGRN centers is available here.