One Flu Over the Avian's Nest

Reposting because I want to be at the top:

You may have been noticed the flurry of reports about the dangers of the Avian Flu infecting the American media in this particular news cycle. You may have been alarmed, or you may be numb to the general sense of oddly-content alarm cultivated particularly by the cable news networks, to the point where you can no longer distinguish the legitimately alarming story from the story that's only alarming by virtue of it being considered news. Well where would you turn to clarify the ensuing confusion but our intermittently updated and generally subject-less blog? I'm not sure, but here you are.

Flu pandemics occur periodically on the scale of centuries, and they differ drastically from the yearly flu that waxes and wanes with the seasons. While a bad flu season can kill significant numbers of immunocompromised, elderly and infant people, flu pandemics are caused by strains that can also easily kill healthy young and middle-aged people. The yearly flu virus mutates each year, allowing it to catch our immune systems by surprise, but it can't really change enough to catch the immune system totally off guard. In other words there is some cross-tolerance to the strain this year because of exposure to the strain last year and the year before that and before that. One essential difference between the pandemic strains and the yearly strains, one that helps make them so deadly, is exactly their difference. They are similar viruses in terms of how they reproduce inside our cells, but they look different enough on a molecular level, that our body has little to no acquired immunity, and does not begin to mount a targeted immune response until about five days into an infection, when it may be too late.

They also tend to be more virulent. This means that they reproduce at a much higher rate than the yearly flu virus, which has a bit more of a symbiotic relationship with its human host. Lets take up an historical example, as scientists have recently, by actually reusurrecting the virus that caused the 1918 flu pandemic. They did this by extracting some virus genetic material from a victim of the 1918 pandemic that has been frozen in Alaskan permafrost until recently. Why this person was trekking across Alaskan permafrost with a bad case of the 1918 flu remains unclear to me. In any case, the genetic material, injected into kidney cells in the lab can incorporate into the genetic code of the kidney cells and instruct the cells to make more of the virus. Hence, from frozen Alaskan (different from baked Alaskan) you end up quickly with resurrected deadly virus. Injecting it in mice, you find interesting things like that 39,000 times more virus particles are produced in the mice lungs after four days than produced by the yearly flu strain, and all the mice die, whereas they tend not to when infected with the yearly strain. This is a telling experiment, with implications for the nature of these pandemics, which we'll discuss below, but another interesting question along the way is: should we resurrect things like this?

Though the generally pretty smart editorial staff of Nature seem to have given it much thought and decided that it was indeed a good idea, and further they decided that it's a good idea to publish the sequence of the 1918 virus genome, others disagree. As nature reports, Richard Ebright of Rutgers University suggests that "Tumpey et al. have constructed, and provided procedures for others to construct, a virus that represents perhaps the most effective bioweapons agent now available." Ebright and others suggest that the escape of the virus from the level 3 biosecurity laboratory (as opposed to level 4) where it resides is a significant risk, "verging on inevitability."

Hmmmm. But bioterror or laboratory carelessness might not be necessary to trigger the next pandemic, because the natural world is about ready to do it for us. Whatever its risks, the resurrection and sequencing of the 1918 virus taught us something valuable and disturbing about it's origin: it was originally an avian strain. This suggests that the world is currently in a similar situation to the one it was in sometime preceding 1918. The current avian flu, classified as H5N1, is epidemic among domestic birds in southeast asia, and the New York Times recently reported the first evidence of spread to Europe, along with unconfirmed reports that the virus is being carried by wild fowl, which, unlike poultry are virtually impossible to contain or slaughter. The containment and slaughter of infected birds is the key to reducing the risk of a human pandemic.
In its current form the virus rarely jumps to humans, but when it does is shows an significant virulance, having killed 60 of the 116 people known to be infected. It's hard to get a handle on the actual mortality rate, since milder cases are less likely to be diagnosed. Also, the virus has only been shown to be transmitted from human to human in a few isolated cases. Some argue that in the course of this mutational change the virus will likely become less fatal. So it's difficult to gauge how fatal the disease is now and how it might change, but what's clear is that it's currently plenty deadly. The 1918 flu had an actual mortality rate of about 2.5-5%, but was contagious enough that roughly half of the world's population contracted the disease. This added up to about 20-50 million deaths. These are rough estimates, but in any case, the H5N1 strain currently shows that it is plenty deadly to produce a massive pandemic.

20-50 million deaths is hard to get a handle on. It's (probably) more deaths than occured in WWII. It's more deaths than AIDS has caused in decades. The 1918 virus transiently lowered life expectancy in the US by about a decade.

But surely medicine has made great strides since 1918? In some ways it has. For instance, up to half of the 20-50 million deaths in 1918 were due to bacterial pneumonia superinfection, to which the virus increases susceptibility. This complication would easiliy be treated with modern antibiotics.

But science hasn't done as well at combatting viruses with drugs as it has with bacteria. There are drugs on the market that have anti-viral effects indicated for influenza. The most notable and probably most effective of these is Tamiflu (oseltamivir). There's no good way to judge how effective this would be in changing the course of bird flu. Most likely it would help a little bit, but even this small advantage over the virus could mean the difference of life and death in an infected person. It could also be taken prophylacticly to reduce the risk that a person would accquire the virus in the first place, but this would be expensive and could lead to the emergence of a resistant strain. Unfortunately, this is a recently developed drug, and as such it's still under patent to the Pharmaceutical manufacturer, Roche, which means two things: there's not enough of it, and it's expensive. Also, other countries are far ahead of the US in stockpiling the drug, so that while some European countries have enough to cover about 40% of their populations, while last I saw we had enough for about 2%. I think it would not be unreasonable to try to buy some tamiflu at this point. Of course governments and officials don't recommend it because they don't want it all to be bought up by private individuals, but let's just say that it wouldn't be alarmist at this point. Unfortunately, it's very expensive (last I saw about $70 for the standard dose which is 75 mg 2x per day for five days) and it's a prescription medication, so I don't know how it works to buy it on the internet (but I know that you can).

A vaccine is a much better hope. This would be a preventative injection, which would provide at least partial immunity, and at worst would make the course of the disease more survivable (because the immune system would be more prepared) Unfortunately, this is a pretty slow process. There's not much money in yearly flu vaccines, so they're still produced by some pretty old technology, where the vaccine is actually cultured in massive numbers of chicken eggs. Not only is this weird and slow (at least 6 months), it could be problemmatic if the avian strain continues to proliferate. It's hard to use chicken eggs to make vaccine when all the chickens already have the disease that the vaccine is designed to prevent. There are several countries and corporations working on this, but vaccine making is a risky business, with low profits and a large potential for a bust in any given year. As such, the industry works best when it's subsidized by governments with good public health planning in mind. This is not the case here. In fact, the US relies on foreign sources for most of its yearly vaccine (you may recall the large shortage last season, which resulted in many unecessary deaths, but would have resulted in more had last year not been a mild season). As such we have next to no capacity to produce such a vaccine, nor is any American company working on such a vaccine, despite the fact that it would be very lucrative already, and in the case of a pandemic, will become so valuable that the countries that do have it will likely not share any with us for any price until their populations are completely immunized, which may not ever happen in time.

In the end it's impossible to predict how likely a pandemic is to emerge from the H5N1 strain, but one thing is clear, that as the disease continues to spread around the globe in birds, the chance of it making the jump to humans increases in kind. I've heard people I trust put it at about 50-50 to occur in the next few years. Given that with our current level of preparation, this would likely kill millions in the US alone, it seems like time to be reasonably alarmed.


www.nature.com: The 1918 flu virus is resurrected
www.nejm.com: Avian Influenza A (H5N1) Infection in Humans, Neuraminidase Inhibitors for Influenza
www.nytimes.com