defective, with mutations that impede their ability to spread and reproduce. But a different kind of accidental change, instead of undermining survival, can take the virus in a new, more perilous direction.
The flu virus reproduces so vigorously that there are enough viable copies to propel it forward. The other copies, those with damaging mutations, are cast aside, as the virus is swept along a path of ever-shifting forms and threats. Ultimately, the virus hits upon precisely the set of mutations needed to infect people and spread like a common cold, the recipe for pandemic. It’s just a matter of chance. Each time the virus replicates, it’s a roll of the dice. Each new bird that’s infected, each person who’s exposed, each one who’s sickened is another toss. Throw the dice enough times and they’re sure to come up snake eyes.
As if the threat of mutation wasn’t enough, the virus can also take a shortcut through what’s called genetic reassortment. Flu viruses are notoriously promiscuous because of a rare gift for swapping genes with other flu viruses. Though most other viruses can’t do this, flu can go out and acquire entirely new attributes. This is because the genetic material in a flu virus—unlike nearly all other RNA viruses—is composed of separate segments that can each be individually replaced. If two different flu strains infect a person or even a pig at the same time, a new hybrid strain could emerge that is both lethal and has the tools to spread with ease. The poster child for gene swapping is swine flu. It was produced by the recent, seemingly improbable encounter of two different flu viruses: one known to circulate among pigs in the eastern hemisphere and another among pigs in the western hemisphere. The latter strain was a so-called triple reassortant, born from even earlier flu strains originating in humans, birds, and swine.
With the wholesale genetic changes that reassortment allows, it doesn’t require too many rolls of the dice to splice together a pandemic this way. “You can move a whole lot of characteristics in one go,” explained Robert Webster, a veteran virologist at St. Jude Children’s Research Hospital in Memphis, Tennessee, and the dean of avian flu researchers. “Flu is an RNA virus and it’s also a segmented RNA virus. That gives it a double whammy.”
Among flu strains, none unnerves disease specialists as much as H5N1 bird flu. In the decade after it surfaced, the virus spread over a swath of Earth unprecedented for a highly lethal avian virus. It extended its reach among animals, even infecting mammals like tigers and leopards. It grew more tenacious. The disease persisted longer in birds and spread more easily among them than only a few years earlier. The dice were being rolled faster and faster.
Researchers have concluded that the continuing outbreaks “appear out of control and represent a serious risk for animal and public health worldwide.” No matter how many times governments claim they’ve expunged the virus, it returns. In some countries, like Indonesia and China, the disease has become deeply entrenched in poultry, posing a permanent threat of contagion to their neighbors. Global eradication, according to senior animal-health experts at the UN Food and Agriculture Organization, “remains a distant and unlikely prospect.”
Yet this strain is not the only avian virus menacing humanity. A little-noticed but equally novel avian strain called H9N2 has also proven it can infect people, including several in Hong Kong and mainland China since 1999. This pathogen, like its better-known cousin, has quietly spread across the birds of Asia and the Middle East and on to Europe and Africa. Studies have suggested that human cases of H9N2 are more common than generally acknowledged, and human transmission may have already occurred. Most worrisome, scientists say the H9N2 virus is actually a better fit for receptors in the human airway, giving it perhaps an edge in the pandemic sweepstakes. “The establishment and prevalence of H9N2 viruses in poultry pose a significant threat for humans,” an international team of researchers reported.
A separate family of novel strains, the H7s, has meantime been circulating in both North America and Europe. Several of these pathogens have also shown an increased affinity for receptors in the human airway. Researchers have urged “continued surveillance and study of these viruses as they continue to resemble viruses with pandemic potential.”
But the H7s, like H9N2, so far remain fairly benign, far less lethal than H5N1. The latter, with a recorded human mortality rate of about 60 percent, is so savage that most flu specialists agree it is the one to be most feared.
Some medical scholars dissent. Although another flu pandemic is inescapable, they doubt that H5N1 will be the source. They note that years have passed, tens of millions of birds have been infected, and countless people exposed without the virus crossing the pandemic threshold. “If it was going to happen, it would have happened already,” said Dr. Peter Palese, chairman of microbiology at the Mount Sinai School of Medicine in New York. Moreover, he suggested that H5N1 wasn’t nearly as virulent as many of his colleagues claim. “I feel the virus is awful for chickens. But this is not a virus that has been shown to really cause disease in humans except in unusual circumstances when the dosage has been extraordinarily high,” he told me, adding that a person has had to practically sleep with a sick chicken to catch a bad case. Perhaps there is some hidden, immutable attribute of the virus that precludes it from ever spreading easily among people. Maybe the dice are loaded, never to come up snake eyes no matter how many times they’re tossed.
This line of reasoning is comforting but, unfortunately, unconvincing to many other virologists. “Such complacency is akin to living on a geological fault line and failing to take precautions against earthquakes and tsunamis,” wrote a leading team of flu specialists. How much time does a virus need to become a pandemic strain? Scientists don’t know. There’s scant information about the virological events that preceded previous pandemics. Had the 1918 strain been smoldering in animals for many years before it crossed to people? Had the 1957 and 1968 strains been circulating for a long time, bouncing between birds and people, but gone unnoticed because these pathogens did not cause mass poultry die-offs like H5N1? Is a decade a long time for a virus to evolve into an epidemic strain? With severe acute respiratory syndrome (SARS), for example, it wasn’t too long. Malik Peiris, the Hong Kong microbiologist who identified the virus behind the 2003 outbreak of SARS, cites evidence that people were exposed to that microbe for quite a number of years before it finally acquired the ability to be transmitted among humans. Once it did, it spread like fire. Flu could do the same.
“The virus has evolved in alarming ways in domestic poultry, migratory birds, and humans in just the last four years,” Margaret Chan told a conference of American business leaders in early 2007. “Global spread is inevitable.”
Chan’s remarks came a month after she’d become director general of WHO. She herself had traveled far since the scare of 1997. In 2003, after staring down outbreaks of both bird flu and SARS, she had left Hong Kong for Geneva. Before long, she was the agency’s assistant director for communicable diseases and its special envoy for pandemic influenza, which she identified as the most serious health threat facing humanity. By the time she ascended to the world’s top health post in January 2007, she was well schooled in flu and convinced that a pandemic was coming.
“If you put a burglar in front of a locked door with a sack of keys and give him enough time, he will get in,” she later warned at a summit of international health policy makers in Seattle. “Influenza viruses have a sack of keys and a bag full of tricks. They are constantly mutating, constantly delivering surprises.” She cautioned that a pandemic strain would be unstoppable once it became fully transmissible. No corner of the world would be spared. So no country could count on outside relief as with earthquakes or tsunamis. “This will almost certainly be the greatest health crisis experienced for almost a century,” she said.
But back in the fall of 1997, as the mystery of Hoi-ka’s death had faded with Hong Kong’s steamy summer, flu had all but vanished from Margaret Chan’s mind. She was facing a new crisis. A public health clinic in Hong Kong had been mistakenly dispensing toxic mouthwash to sick babies instead of syrup for their fever. Many of the children had developed diarrhea and vomiting. The public was clamoring for an explanation. The scandal captured the city’s grim mood as an historic year approached its end. Months earlier, with the world watching on television, Britain had ended more than 150 years of colonial rule by relinquishing sovereignty over Hong Kong to China. But the sheen quickly came off the handover. The Asian financial crisis that autumn rocked Hong Kong. The stock market crashed. The property market tanked. Tourism dried up. Even the weather was rainier than usual.
In late November, Wilina Lim’s lab received a sample from a two-year-old boy who had been briefly hospitalized in another building at Queen Mary with a fever, cough, and sore throat. The lab staff tested the specimen for seasonal flu. They drew a blank. But now they had the chemical reagents required to check for H5N1. When they ran this test, it came back positive.
Lim called over to the health department headquarters. Dr. Thomas Tsang, a senior medical officer
