Tamiflu resistance of influenza H1N1 strains
Yesterday the New York Times ran an article on the resistance to Tamiflu of current influenza H1N1 strains circulating in the US. I wrote a post about this issue on 22 December 2008, so I'm happy to see the Times following my lead. But there is an issue with the Times article that I'd like to address here. According to the article:
The single mutation that creates Tamiflu resistance appears to be spontaneous, and not a reaction to overuse of the drug.
Drug-resistant viruses are not 'reactions' to overuse of the drug. The drug selects, from the diverse viral population in an individual, those viruses that can multiply in its presence. Usually the drug-resistant mutants are already in the host, and outpace other drug-sensitive viruses. Is that what the writer means by 'spontaneous'? Not in this case. What apparently happened is that the mutation that causes drug-resistance, a change from histidine to tyrosine at position 274 of the viral NA protein, emerged in parts of the world were little Tamiflu is used. There was some other reason why this change was selected for in those populations. The article implies that the his->tyr change accompanied a second amino acid change at position 193 of the HA protein which improved the ability of the virus to infect people. This change did not affect resistance to Tamiflu, but apparently it only persisted when the change at 274 was also present. It so happened that the 274 change also conferred resistance to Tamiflu. Thus, when this virus arrived in parts of the world where Tamiflu is used, the resistance was noted. None of this is made particularly clear from the article.
I also have an issue with the author describing the amino acid changes in the 'N' and 'H' genes. The correct nomenclature is NA and HA. The author might have been mislead by the strain designation which uses only 'H', e.g. H1N1. It's a small point but I believe that the devil is in the details.
What about the two other anti-influenza drugs? And the other strain currently circulating, H3N2?
Most of the flu in the US now is caused by H1N1 strains. So although the H3N2 strains are sensitive to Tamiflu, it's not much help.
The Tamiflu-resistant H1N1 strains are sensitive to another drug, Relenza (zanamivir). But that drug must be inhaled and is not appropriate for everyone. However, these H1N1 strains are sensitive to Rimantadine, so its use is a good alternative. Most H3N2 strains are resistant to Rimantadine, which is why it has not been used much in recent years.
Nevertheless, our anti-influenza drug arsenal is much too small. It's worth recalling the following information from Principles of Virology (ASM Press):
With about 1016 human immunodeficiency virus (HIV) genomes on the planet today, it is highly probable that somewhere there exist HIV genomes that are resistant to every one of the antiviral drugs that we have now or are likely to have in the future.
AIDS is no longer a death sentence because we have a deep arsenal of antiviral drugs that can control the infection. Patients are treated with a combination of three anti-HIV-1 drugs at a time. When resistance inevitably emerges, the patient is switched to another combination of three. The high levels of HIV-1 replication in many hosts, coupled with the large numbers of viral mutants that are produced, ensure that resistance will emerge.
Influenza virus shares similar features as HIV-1: high replication rates in many hosts, and the generation of large numbers of viral mutants. Therefore any antiviral strategy that employs only three drugs is bound to fail. The difference with influenza, of course, is that an excellent vaccine is available, and should be used whenever possible. The antiviral compounds should only be used in the face of an outbreak when immunization has not been sufficiently comprehensive. However, I suspect that the use of Tamiflu and Relenza is far more prevalent than desired. How many people rush for a prescription at the first signs of a respiratory infection? And how many of those have already been immunized? This was not the intended use for these antiviral compounds.
If we want to seriously use antiviral to treat influenza (which I don't think is a good idea except in certain cases), we need to have a far deeper arsenal of antiviral drugs.
