Scientists have already created a live, attenuated version of the H9N2 bird flu virus, which first appeared in Hong Kong in 1999. Animal results for the H9N2 vaccine have been "encouraging," Coelingh says. With the H5N1 strain, researchers at MedImmune and NIAID are using recombinant-DNA technology to pluck the genetic code for part of the bird flu and graft it to the backbone of the MDV. The company hopes to start clinical tests on H5N1 sometime early next year.

As MedImmune and NIAID scientists test more strains of influenza with the MDV, they can expect to offer more protection for the population. "We're very hopeful because our seasonal FluMist data indicate that this live, attenuated vaccine does indeed protect against drift variants," Coelingh says, referring to new, unexpected mutations.

GSK also has invested in new vaccine technology. In September the company announced the acquisition of a facility in Marietta, Penn., which is trying to replace chicken eggs, the substrate used for manufacturing vaccines since the 1950s, with a cell culture. The new process could cut production time by two months. GSK also bought Seattle-based Corixa over the summer and announced plans in September to purchase Vancouver-based ID Biomedical, which has a vaccine production contract with the Canadian government that runs through 2011. The moves increase the vaccine-production clout of a company whose third-quarter vaccine sales have risen by 20 percent, to nearly $700 million.

When H9N2 emerged in 1999, Chiron was again on the scene to make a vaccine for the new strain. "The key message is we've worked on a number of potential pandemic-like strains," says Rob Budge, a Chiron spokesman. "And the point of that work is you can take learnings from that as to how you would construct an optimal vaccine against a pandemic."

Chiron delivered its H9N2 vaccine to NIAID in March and expects to deliver an H5N1 vaccine either this month or early next year.

Pharma's Comeback

Fifty years ago, there were six flu-vaccine manufacturers in the United States. Today there is one, Maryland-based MedImmune. Companies around the world currently manufacture enough vaccine to protect about 900 million people, or less than 15 percent of the global population. The United States receives about 80 million to 100 million doses of flu vaccine each year, which leaves roughly two-thirds of the country unprotected.

"The infrastructure has crumbled," says Paul Offit, MD, a vaccine expert at the Children's Hospital of Philadelphia. "There were protections [against lawsuits] put in place for pharmaceutical companies in the mid-80s, but it was too little too late. I mean, there was just a big exodus from vaccines."

Offit notes that about 36,000 people die of flu each year, "and we don't give enough flu vaccine to prevent epidemics."

Vaccine makers steadily left the market in the latter half of the 20th century due to dwindling demand and rising litigation costs, but that trend is about to change. On November 1, President Bush announced at NIH in Bethesda, Md., that is he asking Congress to "remove one of the greatest obstacles to domestic vaccine production: the growing burden of litigation."

Bush also outlined a $7.1 billion plan calling for better monitoring of the virus, stockpiling antivirals, manufacturing vaccines, and preparing governments and agencies for an outbreak.

"I'm very encouraged by the enthusiasm and all the hard work," says Coelingh, who worked 10 years at NIAID before joining MedImmune in 1994. "As far as I can tell, the expressed intention on the part of policymakers is that we will be moving toward a universal vaccination recommendation in the United States within the next few years."

The bird flu scare, at least for the moment, is forcing pharma and policymakers to work together. After all, a little fear can go a long way.

Looking Ahead

COMBATING A DISEASE as nimble and adaptive as influenza requires scientists to mine for alternative methods of production and revolutionary new products. Here are some ways in which scientists have ventured beyond the conventional in constructing a bird flu defense:

DNA vaccines: UK-based PowderMed is working on DNA vaccine technology that injects the flu virus's nucleotides—usually the ones that code for the protein hemagluttin—directly into a person's Langhans cells in the skin. The person's cells then read that DNA and transcribe it into viral proteins, which allows the body to develop an immune response. The biggest advantage of a DNA vaccine is its short production cycle—about two weeks, compared with the six to eight month cycle required for conventional vaccines.

PowderMed is expected to start Phase II trials early next year (the company is currently lobbying for funding for its vaccine delivery device, shown above). San Diego-based Vical is also working on a DNA vaccine.

Cell-culture vaccines: Several companies, including Chiron, ID Biomedical, and Crucell, a Dutch biotech company, are developing methods for vaccine production using cell cultures instead of embryonic chicken eggs. Mammalian cell substrates can shave one to two months off the egg production time.

Protein vaccines: Novavax—in partnership with Wave Biotech—is working on a Virus-Like Particle (VLP) vaccine. The VLP is comprised of three co-expressed virus proteins—hemagluttin, neuraminidase, and matrix M1—that are injected into the body. The VLP lacks genetic material so it cannot replicate, but animal studies have shown that it triggers an immune response.

Novavax has worked on this product since 2000, and it hopes to get clinical data next year. VLP vaccines have a six-week production cycle, according to Rahul Singhvi, the company's president and CEO.

Viral-vectored vaccine: Researchers at Purdue University have been working on an adenoviral-vector vaccine for H5N1 for the past three years. The production process involves attaching part of the bird flu gene, extracted through standard molecular techniques, to a relatively benign, widely circulating virus called the adenovirus. The vaccine is then injected into the host, where it replicates influenza proteins without harming the host. Lead researcher Suresh Mittal says his team has been tweaking the delivery system (the "adenoviral vector") for the past 10 years. Preclinical results with H5N1 are due later this month.

Nanoviricides: Also called "chemical viruses," these are chemically synthesized agents that destroy specific targets while remaining invisible in their hydrophilic shells to the host's immune system. They are the "cloak and dagger" operatives of medicine, slipping into viruses for the kill before sacrificing themselves in a tidy "post-op cleanup" that leaves no residue or side effects.

That's the vision of Anil Diwan, president of Nanoviricides, a seven-month-old company based in West Haven, Conn. Diwan, who has been developing nanoviricide technology for the past 13 years, acknowledges that his peers in the scientific community ridicule him for believing he can synthesize a drug akin to a "guided missile." But even so, he remains confident that nanoviricides will be commercially available within the next five years (Diwan says animal studies last month showed nanoviricides were safe to use). "There is a real need to find a different approach," Diwan says. "To find a different way of attacking the virus itself."