By Monica Heger
Spurred by a recent study that used sequencing to determine that GlaxoSmithKline's Rotarix vaccine for rotavirus contained fragments of a porcine circovirus, the Food and Drug Administration's Vaccine's and Related Biological Products Advisory Committee met last week to discuss whether the FDA should advise clinicians to discontinue the use of the vaccine, and how next-generation sequencing could be used as a quality-control measure for vaccines and the cell substrates used in their production.
The committee issued no formal recommendations, but concluded that the benefits of the vaccine outweighed its theoretical risks, since the circovirus has not been shown to be infectious in humans and the vaccine passed all its safety tests despite containing pieces of the virus.
The committee also acknowledged that next-generation sequencing is a powerful tool for ensuring vaccines are free of harmful contaminants, but expressed concerns about data interpretation as well as the challenges of implementing a regulatory framework for sequencing-based QC.
Eric Delwart, an investigator at the Blood Systems Research Institute in San Francisco, found the porcine circovirus earlier this year when he sequenced eight different vaccines using a metagenomics approach on Roche's 454 GS FLX (IS 4/6/2010).
The Rotarix vaccine was the only one contaminated, and the FDA recommended that physicians temporarily suspend use of the vaccine until further tests were completed, although the World Health Organization and the European Medicines Agency did not recommend any changes to the administration of the vaccine.
Additional testing by Merck of its own rotavirus vaccine RotaTeq found that it, too, had fragments of two porcine circoviruses — PCV1 and PCV2. The source of the viral fragments in both vaccines is unknown but it is believed to have come from the protease trypsin, which is derived from pigs and used in the production of the vaccines.
At last week's meeting, which was webcast, the committee and representatives from the FDA debated the benefits and drawbacks of using sequencing technology for vaccine quality control. Meanwhile, Beckman Coulter Genomics and the biologics safety testing company BioReliance said they are already offering sequencing services to pharmaceutical companies and vaccine developers to screen for contaminants.
"There's no question that sequencing cell substrates that are going into humans is going to be the standard of the future," said Jack Stapleton, a virologist at the University of Iowa and chair of the committee.
"These technologies are here and will have to be incorporated into vaccine development," agreed committee member Pablo Sanchez, professor of pediatrics at the University of Texas Southwestern Medical Center.
Currently, methods for testing vaccine safety include non-specific in vivo animal tests and in vitro tests in cell lines, as well as specific molecular tests, including PCR-based tests to screen for known viruses. Each of these methods has its own limitations. For example, there are some pathogens that are infectious in humans, but don't replicate in the rodents or eggs that are used for in vivo animal tests. And PCR-based tests will only detect known viruses that researchers are looking for.
Eric Delwart, who discovered the circovirus in Rotarix, said during the meeting that while the metagenomics sequencing strategy he used still needs to be optimized and that data interpretation is challenging, it is a good method for detecting many types of known and unknown viruses.
A massively parallel deep sequencing approach still requires an amplification step, which could introduce bias, Delwart said, "and it's very important that this approach be optimized, and its sensitivity determined in comparison to PCR." But, he added, "we've detected viruses of all types," including both known and unknown viruses. "The bigger step is in determining which viruses are pathogenic."
Some companies have already begun offering sequencing services as a quality control measure for vaccine developers. John Kolman, senior director of research and development at BioReliance, said that the company has begun offering sequencing services to its clients, which include pharmaceutical companies and vaccine manufacturers.
BioReliance uses the 454 platform for sequencing, but rather than performing whole-genome sequencing, it typically sequences just the transcriptome. The company, which has been offering sequencing services for about one year, previously used arrays for its QC services, and Kolman said it made the decision to add sequencing because with arrays, "you're going to limit yourself to what you ask to see, rather than everything you can see."
Kolman said that the company also has its own customized bioinformatics and is building a database of new adventitious agents — contaminating DNA that can appear by chance in vaccines.
Beckman Coulter Genomics has also begun offering sequencing services to pharmaceutical companies, said Elena Jouravleva, a Beckman Coulter scientist who advocated for FDA guidelines on the use of sequencing to screen vaccines. "Sequencing allows us to look for the unknown instead of targeting the presence of the known," she said.
While sequencing offers clear advantages, one challenge will be interpreting the plethora of data it will inevitably reveal. "Novel viruses are being discovered and will continue to be discovered. If anything is definite, this is it," said Keith Peden, a senior investigator at the FDA. However, he said, discovering a novel virus doesn't provide researchers with any information about its effects in humans, and pathogenicity studies can take years to complete.
Peden noted that while sequencing's advantage over PCR and other specific techniques is that it can identify agents that researchers aren't looking for, one disadvantage is in interpreting the information. "The data is extremely complex to analyze" and "interpretation can be unclear," he said. "New technologies can be a potentially powerful means to support a safe product development. However, there are regulatory challenges."
The committee members acknowledged that it would be important to figure out an approach for dealing with new findings of contaminants in vaccines. "As you point your telescope in a different place, you'll see more and more of these things," said Bruce Gellin, director of the National Vaccine Program Office. "It's going to be important that we have some ability to evaluate these findings."
The committee suggested that if agents are discovered for which there are already safety data, then vaccine testing and production shouldn't be halted, but the agent should still be studied further. New agents for which there is no data will have to be tested before they go into humans, the committee said.
Going forward, the committee said it would be important for the FDA to be adequately funded so that it is able to test and validate sequencing technology for the study of vaccines because there is currently no regulatory framework for using sequencing to determine vaccine quality.
"It's important that FDA scientists have the access and expertise to apply these new technologies to vaccine evaluation," said Melinda Wharton, the deputy director of the national center for immunization and respiratory diseases at the Centers for Disease Control, and temporary member of the committee.
Stapleton agreed: "The FDA should have those abilities so that you have a regulatory mechanism to validate sequence data."