Commentary
Vaccine Mismatch Drives H1N1 Tamiflu Resistance Recombinomics Commentary 23:59
July 17, 2008
A total of 200 (69%) of 290 influenza A (H1N1) viruses were characterized as A/Solomon Islands/3/2006-like, the influenza A (H1N1) component of the 2007--08 influenza vaccine for the Northern Hemisphere, and 70 (24%) were characterized as A/Brisbane/59/2007-like, the recommended H1N1 component of the 2008--09 Northern Hemisphere vaccine.
All the oseltamivir-resistant viruses have been influenza A (H1N1) viruses and have been determined to share the same genetic mutation that confers oseltamivir resistance. These 84 viruses represent 10.2% of the 824 influenza A (H1N1) viruses that have been tested, an increase from four (0.7%) of 588 influenza A (H1N1) viruses tested during the 2006--07 season.
The majority of influenza A (H1N1) viruses were characterized as A/Solomon Islands/3/2006, the influenza A (H1N1) component of the 2007--08 influenza vaccine for the Northern Hemisphere.
The above comments from the CDC MMWR report on the 2007/2008 seasonal influenza in the United States is not supported the by the HA and NA sequences from approximately 150 influenza isolates from the 2007/2008 season released by the CDC. Similarly, the data from the phylogenetic analysis also fails to support the above comments.
The public data from the CDC indicates all of the reported H1N1 isolates in the 2007/2008 season were either Brisbane/59/2007-like (clade 2B) or Hong Kong/2652/2006-like (clade 2C). Although Solomon Island/3/2006 (clade 2A) was in circulation in the 2006/2007 season, by the time the vaccine was created and distributed, clade A was replaced by clade B and clade C.
The differences between the clades can be seen in slide 9 from a CDC phylogram of NA. Clade 2 was distinguished from New Caldeonia by V223M and D385N. However, each sub-clade had a large number of addition non-synonymous changes. Clade 2A had M23I, H68N, G81E, K172R, S265T, D452G, while clade 2B had H49N, K82E, G248K, T286I, K329E, G357D, while clade 2C had S85P, R130K, M187I, I266M, L230I, V397I, T4531. Thus the NA sequences placed these isolates into three easily distinguished subclades.
The differences can also be seen in slide 8 which has an HA phylogram. However, there are a larger number of polymorphisms that distinguish all three sub-clades from New Caledonia (T90K, Y101H, K144E, R149K, R212K, T269N). Consequently, although each clade 2 sub-clade had specific distinguishing changes which were similar to the list above (K81R and V132T for clade A, D45N, K149R, R192K, E276K for clade B, and S46N, R192M, A193T, T197K for clade C.
Thus, even though the three sub-clades are easily distinguished by phylogenetic analysis, the 6 HA polymorphism that distinguish clade 2 from clade 1 led to the classification of the clade 2B and clade 2C circulating in the US as clade 2A, even though there were no clade 2A isolates in circulation in the US, or anywhere else in the 2007/2008 season (as indicated by the various 2007/2008 phylograms, including another from the CDC, which were recently released as well as additional 2007/2008 isolates from Europe, Asia, Africa, and South America. In all cases only clade 2B and 2C were circulation in the 2007/2008 season.
Thus, the H1N1 vaccine that targeted Solomon Island/3/2006 (clade 2A) which was circulating with clade 2B and 2C in 2006/2007 had reduced efficacy in 2007/2008 because clade 2A had been completely replaced by clade 2B and clade 2C.
Attempts to remedy this mis-match led to the changing of the H1N1 target for the 2008/2009 season from Solomon Island/3/2006 (clade 2A) to Brisbane/59/2007 (clade B).
However, while the mis-matched H1N1 was used to vaccinate worldwide in 2007/2008, clade 2B was spreading resistance to both neuraminidase inhibitors, Tamiflu (oseltamivir) and Relenza (zanamivir), while clade 2C was spreading resistance to the M2 channel blockers, amantadine and rimantadine.
The resistance to Tamiflu has captured the most attention. In Norway, the majority of H1N1 isolates have H274Y. Percentages above 50% have been reported in Norway and double digit frequencies have been reported in many northern countries, including France, Russia, and Canada. The H274Y has appeared on multiple Brisbane clade 2A sub-clades, supporting independent introductions via recombination. This pattern has also been see in Relenza resistance which was on clade 2A in 2006/2007 and in clade 2B in 2007/2008. The vaccine mismatch has allowed clade 2B to spread worldwide, facilitating the spread of Tamiflu resistance.
The mismatch for H1N1 was in addition to mismatches with H3N2 and influenza B, which has led to a 2008/2009 vaccine that has replaced all three viral targets.
Media Links
Recombinomics Presentations
Recombinomics Publications
Recombinomics Paper at Nature Precedings