How were bird flu viruses sent to unsuspecting labs?
The Canadian Press
Published Thursday, February 26, 2009 7:45AM EST
Officials are trying to get to the bottom of how vaccine manufacturer Baxter International Inc. made "experimental virus material" based on a human flu strain but contaminated with the H5N1 avian flu virus and then distributed it to an Austrian company.
That company, Avir Green Hills Biotechnology, then disseminated the supposed H3N2 virus product to subcontractors in the Czech Republic, Slovenia and Germany. Authorities in the four European countries are looking into the incident, and their efforts are being closely watched by the World Health Organization and the European Centre for Disease Control.
Though it appears none of the 36 or 37 people who were exposed to the contaminated product became infected, the incident is being described as "a serious error" on the part of Baxter, which is on the brink of securing a European licence for an H5N1 vaccine. That vaccine is made at a different facility, in the Czech Republic.
"For this particular incident ... the horse did not get out (of the barn)," Dr. Angus Nicoll of the ECDC said from Stockholm.
"But that doesn't mean that we and WHO and the European Commission and the others aren't taking it as seriously as you would any laboratory accident with dangerous pathogens - which you have here."
Accidental release of a mixture of live H5N1 and H3N2 viruses - if that indeed happened - could have resulted in dire consequences. Nicoll said officials still aren't 100 per cent sure the mixture contained live H5N1 viruses. But given that ferrets exposed to the mixture died, it likely did.
H5N1 doesn't easily infect people, but H3N2 viruses do. They are one of two types of influenza A viruses that infect people each flu season.
If someone exposed to the mixture had been co-infected with H5N1 and H3N2, the person could have served as an incubator for a hybrid virus able to transmit easily to and among people. That mixing process, called reassortment, is one of two ways pandemic viruses are created.
Research published last summer by scientists at the U.S. Centers for Disease Control found that in the laboratory, H5N1 and H3N2 viruses mated readily. While less virulent than H5N1, a number of the offspring viruses appeared to retain at least a portion of the killing power of their dangerous parent.
Baxter International, which is based in Deerfield, Ill., said the contamination was the result of an error in its research facility in Orth-Donau, Austria.
The facility had been contracted by Avir Green Hills to make what Baxter refers to as "experimental virus material" based on human H3N2 viruses.
Christopher Bona, Baxter's director of global bioscience communications, said the liquid virus product was not a vaccine and was developed for testing purposes only. He deferred questions about the purpose of the testing to Avir Green Hills, but said the batch was to be used in animals and was never intended for use in humans.
Avir Green Hills said in an email that it took possession of the material in late December. It later sent the product to the sub-contractors. The email said the material was stored and handled throughout under high biosafety conditions.
Alarm bells rang in early February when researchers at the Czech sub-contractor inoculated ferrets with the material and the animals promptly died. Baxter learned about the problem on Feb. 6, Bona said from Deerfield.
Ferrets are susceptible to human flu strains, but they don't die from those infections. Preliminary investigation found the material was contaminated with H5N1 flu virus, which is lethal to ferrets.
Nicoll said the fact the ferrets died supports the working assumption that there were live H5N1 viruses in the material Baxter produced.
Bona said Baxter has identified how the contamination happened and has taken steps to ensure it doesn't happen again. He said Austrian authorities audited Baxter's Orth-Donau research operations after the problem came to light and are satisfied with the steps taken.
Baxter is the only flu vaccine manufacturer to work with wild type flu viruses, felt to be more dangerous than the altered and attenuated (weakened) viruses other manufacturers use.
The company uses what is known as BSL3 level precautions in all its vaccine research facilities, Bona said. (Researchers at the U.S. CDC use BSL3-plus biocontainment when working with H5N1 viruses, a spokesperson for the agency said.)
People familiar with biosecurity rules are dismayed by evidence that human H3N2 and avian H5N1 viruses have somehow co-mingled in the Baxter research facility. That should not be allowed to happen, a number of experts insisted.
The company isn't shedding much light on how it did.
"It was a combination of just the process itself, (and) technical and human error in this procedure," Bona said. When asked to elaborate, he said to do so would give away proprietary information about Baxter's production process.
Bona said when Baxter realized its error, it helped the various companies destroy the contaminated material and clean up their facilities. And staff who had been exposed to the contaminated product were assessed and monitored by infectious diseases doctors. They were also offered the antiviral drug oseltamivir (Tamiflu).
Baxter's error is reminiscent of a 2005 incident where a U.S. manufacturer of kits used by laboratories to test their detection capabilities included vials of H2N2 virus in several thousand proficiency kits. H2N2, the virus that caused the 1957 pandemic, has not circulated since 1968 and is thought to be a prime candidate to cause the next pandemic.
That mistake, discovered by Canada's National Microbiology Laboratory, set authorities around the world scrambling to retrieve and destroy the vials of virus, which had been sent to labs in 18 countries.