A paper claims to have solved on of the most important puzzles in infectious disease research: what made the 1918 influenza pandemic so lethal?
“Ever since the great flu pandemic of 1918 it has been a mystery where that virus came from and why it was so severe,” says Professor Michael Worobey of the University of Arizona. Between 50 and 100 million people died from the outbreak dubbed “the Spanish flu”, several times more than World War I, yet history books generally relegate it to a footnote.
While most of those who die from normal flu outbreaks are the very young or the elderly the 1918 outbreak was different – the heaviest death toll was among young adults, including those that had been in good health beforehand.
Interest is more than historical. There is little reason to doubt that a similar strain could return, and while modern medical technology would limit the death toll, no healthcare system in the world could cope with the volume of patients struck down in such a short period. Moreover, airtravel would likely see the disease cross the globe far faster than in an era when circumnavigating the world in 80 days was still the benchmark for speed.
The reason we still lack a universal flu vaccine is that the virus is always scrambling the hemagglutinin and neuraminidase proteins on its surface. "Imagine a soccer ball studded with lollipops," said Worobey. "The candy part of the lollipop is the globular part of the HA protein, and that is by far the most potent part of the flu virus against which our immune system can make antibodies. If antibodies cover all the lollipop heads, the virus can't even infect you." A different sort of lollipop, but with the same type of stick, will still attract antibodies that won't prevent infection but may slow the development of the virus “which protects you from severe disease and death," Worobey said.
Once infected or inoculated our body produces defenses against a particular combination, but these are of little or no use against a different form. The 1918 version has been labeled H1N1. Other versions of H1N1 have appeared since – the so called “swine flu” epidemic of 2009 was treated with such alarm because it was also H1N1 - but none have been so lethal.
Worobey compared samples from different H1N1 outbreaks to try to find the origins of the 1918 catastrophe. He thought this might settle debate between the two competing theories – a jump from birds or the reshuffling of genes between existing human and swine varieties. Instead in the Proceedings of the National Academy of Sciences he proposes a third option in which a human flu virus with the H1 protein in circulation for 10-15 years gained the N1 genes from a version infecting birds. He believes previous researchers have missed this because after 1922 a slightly different form of H1 displaced the one responsible for the 1918 outbreak, and scientists have been examining the wrong one until samples taken during the original outbreak were found.
"It sounds like a modest little detail, but it may be the missing piece of the puzzle," Worobey said. "Once you have that clue, many other lines of evidence that have been around since 1918 fall into place."
It has previously been suggested that the reason people over 35 were less likely to die in the 1918 epidemic might be because they had been exposed to a form of flu with some similarity to the epidemic of that year; the same lollipop stick in Worobey's analogy. While the common features might not have been enough to stop them getting sick the immune system would have had just enough recognition to save their lives. Worobey endorses this, suggesting widespread exposure to either H1 or N1 in childhood would explain the relatively low death rate among the elderly from the pandemic.
However, Worobey also turns the idea on its head by proposing exposure to a very different version of the flu virus primed the immune system to misread epidemic version. The finger of suspicion is pointed at an H3N8 virus common between 1889 and 1900. Tens of millions of people would have been exposed to this virus in childhood and recovered, but Worobey said, “A person with an antibody arsenal directed against the H3 protein would not have fared well when faced with flu viruses studded with H1 protein. And we believe that that mismatch may have resulted in the heightened mortality in the age group that happened to be in their late 20s during the 1918 pandemic."
Of course people could have been infected by both the H3 and the H1 versions at different times, but essential to Worobey's thesis is the idea that the first infection has the most impact in terms of the antibodies produced, what some researchers have called “original antigenic sin”. Even if they were infected by an H1 version in the early years of the century, those whose first exposure was to H3N8 were vulnerable.
Whether prior infection by the H3N8 virus was better than nothing remains unclear. On the one hand, island populations that may never have encountered flu at all experienced death rates higher than anything seen in urban areas. On the other,in places with good health records there is an extraordinary match in the death rates between those people who would have been exposed to H3N8 as their first dose of flu and the proportion who died (see chart below),
Worobey found support from recent outbreaks to back his theory up. In recent years The H5N1 virus has been much more likely to kill people born after 1968 than before. On the other hand, H7N9 is lethal to those born earlier, particularly in the late 1940s. He attributes this to each group having been exposed to a particular virus when young that offered some protection against one of the recent strains, but not against the other.
The good news here is that much of the population has now been immunized against numerous strains of flu. While these might not be enough to stop people getting sick from a novel version, it should keep the death rates down if we experience something as potentially devastating as the 1918 outbreak again. Further research on the way specific hemagglutinin and neuraminidase match up might help us design vaccines that would offer widespread partial protection to cover all options for future pandemics.