Lightning storm over BostonAvian influenza has a terribly high mortality rate in the humans it infects — perhaps as many as 80% of infected people die. Why is avian flu so lethal, while other strains of influenza rarely cause serious damage in young, healthy people?

One explanation has been cytokine storms. According to this hypothesis,1 avian influenza causes a massive innate immune response, leading to the release of large amounts of cytokines. It’s the resulting inflammation that is lethal, not the damage that the virus causes directly.

The problem with this hypothesis was that it apparently didn’t help with treatment. If cytokine storms are responsible for mortality, then suppressing cytokine responses should reduce the death rate; and that didn’t happen, according to a paper2 published last year. What’s more, although cytokine levels are associated with increased mortality, they’re also associated with virus levels. That is, cytokine levels may be an indicator of the amount of virus present, rather than a direct risk factor.

Because cytokine inhibition does not protect against death, therapies that target the virus rather than cytokines may be preferable. 2

Flu cure - WellcomeFor this reason (and others) I said at the time that “I think the evidence for this is pretty weak – and for avian flu in particular, it’s been shown that cytokines are probably not the culprits at all.”

But I’m always happy to have my mind changed, and a new paper that came out a couple of weeks ago3 has made me reconsider (even though it’s mouse work, not tested yet in humans). This is from Kwok-Yung Yuen’s group in Hong Kong. (Yuen is probably best known for his work on SARS — another disease that’s been claimed to act via cytokine storms — but he has worked on on avian influenza for a long time as well.) Their trick was to use triple therapy: a dual blockade of cytokines, plus antiviral treatment. None of these treatments worked alone. If you suppress viral replication while allowing the cytokines to persist, the mice die; if you shut down the cytokines while allowing the virus to continue replicating, the mice die. But if you block cytokines and virus — then most of the mice survive.

Actually, the antiviral alone does work pretty well if you start treatment almost immediately after infection. Of course, that’s not practical in humans; humans arrive at the hospital days after they’ve been infected, and by then viral replication and cytokine levels are already roaring ahead. The exciting part about Yuen’s triple therapy is that it gives decent survival — again, in mice, not yet humans — even if you don’t start treatment for 2 days after infection. This makes it a practical treatment for humans, and since the anti-inflammatory drugs they used are not particularly expensive or obscure, I’m sure we will see this treatment tried out in humans sooner rather than later.

  1. Chan, M. C., Cheung, C. Y., Chui, W. H., Tsao, S. W., Nicholls, J. M., Chan, Y. O., Chan, R. W., Long, H. T., Poon, L. L., Guan, Y., and Peiris, J. S. (2005). Proinflammatory cytokine responses induced by influenza A (H5N1) viruses in primary human alveolar and bronchial epithelial cells. Respir Res 6, 135.
    de Jong MD, Simmons CP, Thanh TT, Hien VM, Smith GJ, Chau TN, Hoang DM, Chau NV, Khanh TH, Dong VC et al. (2006) Fatal outcome of human influenza A (H5N1) is associated with high viral load and hypercytokinemia. Nat Med 12:1203-1207. doi:10.1038/nm1477[]
  2. Salomon, R., Hoffmann, E., and Webster, R. G. (2007). Inhibition of the cytokine response does not protect against lethal H5N1 influenza infection. Proc Natl Acad Sci U S A 104, 12479-12481. [][]
  3. Zheng, B., Chan, K., Lin, Y., Zhao, G., Chan, C., Zhang, H., Chen, H., Wong, S.S., Lau, S.K., Woo, P.C., Chan, K., Jin, D., Yuen, K. (2008). Delayed antiviral plus immunomodulator treatment still reduces mortality in mice infected by high inoculum of influenza A/H5N1 virus. Proceedings of the National Academy of Sciences, 105(23), 8091-8096. DOI: 10.1073/pnas.0711942105[]