- "The H5N1 virus continues to change. It will continue
to change," Paul Gully, a senior WHO adviser, told the conference.
- "Influenza A viruses replicate but they don't replicate
very well. They are bad at replicating themselves so ... mutations continue
to occur," he said.
- Gully said the mutation of the virus in Egypt "probably
occurred by chance".
- The above comments show that WHO advisors continue to
cling to random mutations to explain H5N1 evolution and Tamiflu (oseltamivir)
resistance. However as additional sequences become public, it becomes
increasingly clear that the changes in H5N1 are not due to recent mutations,
but are linked to transmission and transport by wild birds leading to
dual infections and recombination, resulting in novel gene sequences in
H5N1, which are evolving at an increasing rate.
- The N294S polymorphism in the Gharbiya cluster is an
example of such evolution, that is not directly linked to Tamiflu treatment
of the fatally infected patient, but may very well be linked to the increased
use of a Tamiflu blanket in H5N1 infected areas.
- N294S was identified in collections from two related
patients in Gharbiya. Samples were collected prior to Tamiflu treatment
as well as two days after the start of treatment. All four samples had
H5N1 with the Tamiflu resistance marker, N294S. However, it is unlikely
that the change was due to chance, as indicated above.
- Potential problems with Tamiflu resistance were evident
in initial in vitro data linked to analysis of IC50 values for the nine
N serotypes. N2 was among the serotypes most sensitive to oseltamivir,
while N1 was among the serotypes least sensitive to oseltamivir. The
IC50 for N1 was 20-40 fold higher than N2.
- The potential problems identified in the in vitro tests
were again signaled by in vitro studies. Mice pre-treated with Tamiflu
that was 10-20 fold higher than the FDA approved dose of Tamiflu were
only partially protected from challenge with H5n1. Although the elevated
concentration was linked to bioavailability differences between mice and
humans, the failure of the Tamiflu to protect the mice raised additional
concerns that the dose approved for seasonal flu might lead to Tamiflu
resistance in humans, in part because H5N1 was lethal and virulent in
- Tamiflu treatment in humans began in Southeast Asia in
2004, in part because the Clade 1 H5N1 was already resistant to amantadines,
which were ion channel blockers that targeted the M2 protein.
- Tamiflu resistance was reported in early 2005 in a patient
who had been initially treated with the lower prophylactic dose for Tamiflu.
- Two Tamiflu resistance markers, H274Y and N294S were
identified in separate clones from the patient. Although these changes
were assumed to have arisen via de novo mutations in the patient, because
wild type H5N1 was also found, an alternative explanation involving isolates
with changes was supported by the prior identification of both markers
in H5N1 infected birds in Hong Kong.
- H274Y had been previously identified in A/chicken/Hong
Kong/ 3123.1/2002. Similarly, N294S had been previously identified in
A/ Duck/Hong Kong/380.5/2001. Thus, both Tamiflu resistance markers were
present in H5N1 infected birds in Hong Kong. N294S was also detected
in China, A/duck/Zhejiang/bj/2002.
- The presence of Tamiflu resistance markers in H5N1 birds
in the region offer opportunities for acquisitions of these markers via
recombination instead of de novo mutations.
- These linkages have now extended outside of Asia. In
May, 2005 H5N1 was identified in long range migratory birds at Qinghai
- Although the H5N1 was readily distinguished from prior
H5N1 isolates, it did share polymorphisms with H5N1 infected wild birds
in Hong Kong and Japan.
- The long range migratory birds carried the Qinghai strain
of H5N1 to nature reserves in Mongolia and southern Siberia, which link
back to migration paths in eastern Asia. The presence of the Tamiflu
resistance markers soon followed. H274Y was identified in Qinghai H5N1
swan isolates from Astrakhan, A/swan/Astrakhan/1/2005 and A/swan/Astrakhan/Russia/Nov-2/2005
while N294S has been identified in A/Egypt/14724-NAMRU3/2006 and A/Egypt/14725-
NAMRU-3/2006 as well as corresponding isolates from samples collected
prior to Tamiflu treatment.
- Thus, the Tamiflu resistant marker, N294S, described
as a chance occurrence is unlikely to be due to chance or recent copy
errors as suggested by the WHO consultant above. Tamiflu blankets hae
been applied to the 2006 H5N1 in Turkey. This widespread useage can
lead to an increased incidence of Tamiflu resistance markers, which may
have lead to teh presence of N294S in the Gharbiya cluster.
- WHO consultants have uniformly indicated that H5N1 evolves
via random mutations due to copy errors. However, as the sequence database
grows the tracking of these changes follow well defined patterns in H5N1
isolates that are easily explained by homologous recombination.
- The failure of WHO consultants to understand the underlying
mechanism of H5N1 evolution and drug resistance remains a cause for concern.