POSTED 1 DECEMBER 2005
Influenza vaccine: Could good enough be better than perfect?
The bird flu has killed millions of birds and more than 60 people in Asia. So far, the virus does not seem to infect one person directly from another. But if the virus does start spreading that way, a worldwide epidemic is likely. According to the World Health Organization, "Experts agree that another influenza pandemic is inevitable and possibly imminent."
In a typical year, the WHO estimates that flu infects 10 to 20 percent of the world population, and 250,000 to 500,000 die. But the current avian influenza virus, called H5N1, has killed half of the people known to have caught it.
So a global outbreak could be a rerun of the 1918 "Spanish flu," which killed 40 to 50 million.
Three strategies have been discussed to fight an avian influenza once it starts to spread between people: Drugs, quarantine, and vaccines.
Each strategy has major pitfalls.
Drugs like Tamiflu reduce symptoms and infectivity, but stockpiles are pitifully low, and the drugs must be taken within 48 hours after symptoms appear, preferably much sooner.
Isolating the sick is a time-tested infection control, but according to Graeme Laver, a flu expert at the Australian National University, quarantine won't work because people without symptoms can spread influenza. "It's absolutely impossible. So many people will be infected and shedding virus, but they won't know it, they will have no fever but they will spread the virus like crazy." Even the island nation of Australia can expect to be infected through airports, he cautions. Immigration officers aren't doctors, he suggests, and even doctors can't spot symptoms before they exist.
Vaccines are harmless virus particles that alert the immune system to create antibodies before infection. Development of the annual flu vaccine starts in early spring, when virologists guesstimate which virus strains will appear next flu season, and then manufacturers race to inactivate the virus and grow it in millions of chicken eggs. The best flu vaccines are based on the exact structure of the threatening virus, and they are fairly effective -- but not perfect -- against common flu. But influenza is so contagious, and H5N1 so deadly, that millions of people could die waiting months for a well-matched vaccine.
And that brings us to our subject: poorly matched vaccine, defined as any vaccine that can be made from existing strains of virus, before the virus changes and starts to move from person to person. This discussion acknowledges a tradeoff: A vaccine produced before an epidemic is unlikely to work as well as a well-matched vaccine produced after the epidemic starts. But a stockpile of poorly matched vaccine could slow the epidemic and save lives while a better vaccine is produced.
Conventional wisdom has little time for poorly matched vaccines for influenza. The World Health Organization, for example, dismisses them entirely, stating in its fact sheet that "at least four months would be needed to produce a new vaccine, in significant quantities, capable of conferring protection against a new virus subtype."
Still, in 2005, researchers reported three separate animal tests showing that three different vaccines, based on different strains of existing virus, could protect mice against deadly H5N1 influenza -- bird flu. Details will follow, but briefly:
Suzanne Epstein of the U.S. Food and Drug Administration used a DNA vaccine and a viral booster.
Iain Stephenson of University Hospital, Leicester (U.K.) added an immune booster (an adjuvant) to a triple dose of vaccine.
Robert Webster of St. Jude's Children's Hospital (Memphis, Tenn.) tested a vaccine that has now moved into early human tests at the National Institutes of Health.
Stuck with a poorly matched vaccine?
Vaccines have conquered many viral infections: Smallpox is extinct in the wild, and polio nearly so. But these viruses are stable, and vaccines are a lot less useful against shifty foes like HIV or influenza. In its continual quest to evade drugs and immune-system attack, flu virus mutates at random and also recombines, swapping genes with related viruses.
Illustration from the National Institute of Allergy and Infectious Diseases/National Institutes of Health
Flu virus can undergo antigenic drift, the small change that explains why last year's flu vaccine doesn't work so well this year. Or it can go through antigenic shift, a major change in virus subtype that makes the virus look new to the immune system. After antigenic shift, the virus gets new H and N numbers: H stands for hemagglutinin, and N for neuraminidase; both are surface proteins that can be attacked by the immune system. Hemagglutinin comes in 16 flavors, and neuraminidase in nine.
H1N1 caused the 1918 epidemic; H3N2 caused the 1968 epidemic, and H5N1 is the bird flu that's causing a globeful of fretting this year. H1N1 and H3N2 remain widespread, but because most people have gained some immunity after infection by these subtypes, the common flu is seldom deadly. But existing immunity is little or no help against the new subtypes that appear after an antigenic shift. If the dreaded H5N1 virus does start attacking humans in a big way, existing vaccines probably won't protect us because they contain viral components not found on H5N1.
Or so goes the conventional wisdom. Many experts are even skeptical that a vaccine prepared against the strain of H5N1 now moving among birds would protect against a future human-to-human strain of H5N1.
But given the research showing that influenza immunity could be broader than previously thought, would it make sense to intensify research into poorly matched vaccines? Such a vaccine could spawn an immune attack against unchanging viral components, able to act on various subtypes, or it could be based on a good guess about the nature of a future pandemic virus.
"There is a concept in textbooks, that human immunity to influenza is subtype specific," says Suzanne Epstein, head of the Immunity to Influenza Virus section of the U.S. Food and Drug Administration. "There is a conservatism in the field based on that, but there is some published data to suggest that humans do have" a broader immunity across subtypes.
In the face of a fast-spreading, deadly epidemic, epidemiologist Ira Longini of Emory University says a poorly matched vaccine could slow the virus and make it less deadly. "The idea would be to raise population level immunity... where instead of spreading like pandemic, it spreads more like inter-pandemic flu, the kind we see every year."
Megan Anderson, project assistant; Terry Devitt, editor; S.V. Medaris, designer/illustrator; David Tenenbaum, feature writer; Amy Toburen, content development executive