On Oct. 5, Chiron Corp. CEO Howard Pien announced in a conference call that British regulatory agencies "temporarily suspended Chiron's license to manufacture Fluvirin in our Liverpool facility, which is the sole Chiron facility that is authorized to supply flu vaccine to the U.S. That action will prevent Chiron from releasing any product for the 2004-2005 influenza season."

The move deprived the U.S. of about half its expected supply of flu vaccine for the season, in turn prompting the Centers for Disease Control and Prevention to issue an interim recommendation that persons who are not at high risk of influenza complications forego or defer vaccination. (See BioWorld Today, Oct. 6, 2004.)

Whatever the exact reason for the suspension, it suggested that alternative vaccine production technologies could provide technological benefits and, if reliable, public health benefits, as well. Both the production of seed strains and the manufacturing of the actual vaccines are done in eggs, but newer techniques exist for both.

The Making Of A Flu Vaccine

In terms of vaccine development, the king of the hill is hemagglutinin, a surface glycoprotein that the influenza virus uses to bind to host cells.

"All the immunity to influenza is determined by hemagglutinin," John Treanor, professor of medicine and director of the vaccine evaluation unit at the University of Rochester in New York, told BioWorld Today.

The annual cycle of vaccine production begins in late winter. After a number of labs evaluate the different types of influenza circulating at that time, the FDA's Vaccines and Related Biologicals Advisory Committee makes an educated guess both on which new viruses are likely to pose a particular problem during the next flu season, and how well last year's vaccine is likely to protect against this year's viruses. The actual vaccine is trivalent, combining proteins raised from three seed strains. Usually, one or two of the three seed strains for the vaccine are changed each year.

Each seed strain combines sequences designed to make it grow well in eggs for production, with antigenic sequences from a circulating virus deemed likely to wreak havoc during the coming flu season. Selecting the latter is more art than science, though the proof - in the form of an effective vaccine - usually is in the pudding.

The standard technique to make seed strains is called classical reassortment and basically relies on statistics to make the desired virus.

"You take one influenza virus, which is adapted to grow very well in eggs, and one virus, which has the antigenic properties you want, and co-infect embryonated chicken eggs with the two parental viruses," Christina Cassetti, program director for basic and preclinical influenza research at the National Institute of Allergy and Infectious Diseases, told BioWorld Today. The influenza virus genome has eight segments, and those segments recombine when the two viral strains replicate in the same host.

"The virus you want has the hemagglutinin and neuraminidase you want, but other gene segments confer the properties necessary to grow very well in eggs," Cassetti said. However, the segments recombine independently of each other, and selecting the desired strain from the plethora of combinations is a cumbersome endeavor. At least from the point of view of the host egg, the result of classic reassortment is a buffet with mostly toxic offerings.

Even producing the seed strain can be a technical challenge. That was the case with H5N1, the avian influenza virus now under eagle-eyed watch by the World Health Organization. In that case, the virus killed the eggs used for producing seed strain too early in the process to allow for useful harvesting.

The CDC responded by using reverse genetics to make the seed strain. In contrast to the usually successful but scattershot technique of classic reassortment, reverse genetics is a precision technique in which cDNAs for all eight segments are made and assembled in cell cultures. Here, Cassetti said, "what you put in is exactly what you get out." The other advantage is that the DNA sequences within the segments can be selectively manipulated. That is what enabled CDC researchers to produce an H5N1 seed strain that can be produced in eggs.

"H5 hemagglutinin has a sequence known to confer virulence and kill chicken eggs," Cassetti said. "Using reverse genetics, it was possible to rationally modify that sequence," to enable vaccines from the seed strain to be produced in eggs. That manipulation did not affect the effectiveness of the H5 sequence to elicit an antibody response. Fortunately, the protein retains complete antigenicity but loses virulence, she said.

Cell Culture Production: Bigger, Better, Faster, More!

Once the seed strain has been produced, the actual vaccine also is produced in eggs. That approach has several disadvantages - everything from speed to possible infections in the chick embryos from other diseases that could theoretically contaminate the vaccines - and thus needs to be carefully monitored.

"People had been considering moving to cell culture for a while, but the fact of the matter is that flu virus just doesn't grow very well in culture," Frank Cano, CEO and chairman of the Birmingham, Ala.,-based biotech company Vaxin Inc., told BioWorld Today.

Vaxin gets around that by using an adenovirus vector to express the hemagglutinin protein. "We get a seed strain from the CDC. We will immediately take the hemagglutinin gene and put it into adenovirus. And then we're done with the flu - we're producing adenovirus," Cano said. That adenovirus is produced in the human cell line PER.C6. A monovalent version of the company's influenza vaccine is in early Phase I trials. Cano hopes that Vaxin will progress to larger-scale trials with a trivalent vaccine by late 2005 or early 2006.

Another company that has taken cell culture-manufactured flu vaccines to the clinic is Meriden, Conn.-based Protein Sciences Corp. Their FluBl k vaccine is derived from recombinant hemagglutinin (rHA). That vaccine is produced in insect cells and driven by a baculovirus expression vector. Like Vaxin's adenovirus-based system, but unlike production in eggs, making FluBl k does not require live virus, which makes it safer to produce because there is no possibility of workers being infected during the production cycle.

FluBl k is in Phase II trials at the University of Rochester. "In this current winter, we're hoping to do a small-scale proof-of-principle efficacy trial," said Treanor, who, in his capacity as director of the vaccine evaluation unit at Rochester, will be overseeing that trial. "And if that looks promising, next winter we would hopefully be able to do a pivotal efficacy trial."