Commentary
Fatal H1N1 Tamiflu Resistance
Recombinomics Commentary 17:58
September 5, 2008
Oseltamivir was administered for the influenza virus infection, beginning on the sixth hospital day, but it was discontinued on day 13 because sequence analysis revealed the H274Y mutation, and no decrease in the viral load was observed. In retrospect, the H274Y mutation was present in the specimen obtained before oseltamivir therapy was initiated. The patient's hospital record and his family indicated that he had had no contact with patients who had received oseltamivir. On day 15, amantadine was added to the patient's treatment regimen. Four days later, the neutrophil count increased, indicating bone marrow recovery. Mechanical ventilation was discontinued on day 20, and zanamivir by inhalation was initiated. However, respiratory failure occurred on day 22, mechanical ventilation was reinstituted, and therapy with zanamivir was discontinued. On day 26, the influenza virus was no longer detectable. Because sequence analyses showed an amantadine-resistance mutation in the viral M2 protein (L26F) and zanamivir therapy had been limited to three doses, clearance of the virus was probably due to recovery of the immune system. A second CT scan, obtained on day 28, revealed progression of the pulmonary infiltrates. Because of the poor prognosis, mechanical ventilation was discontinued on day 34. The patient died 3 days later.
The above comments are from a correspondence to the New England Journal of Medicine on a patient who was sequentially treated with three antivirals, oseltamivir (Tamiflu), amantadine, and zanamivir (Relenza). The death of the patient was directly linked to his underlying cancer, but the monitoring of viral load and sequencing of sequential isolates provided additional insight into antiviral resistance.
The oseltamivir treatment did not decrease viral load, but when H274Y was identified in an isolate, treatment shifted from oseltamivir to amantadine, which cause a brief decline in viral load, but the emergence of L26F in the M2 target. This led to a shift to zanamivir, and an associated re-emergence of a wild type M2, indicting the L26F variant was not evolutionarily fit. Zanamivir treatment was only for three days, and the viral load subsequently declined, but the H274Y was present in all isolates, including an isolate from a sample collected prior to treatment.
The above case demonstrates the evolutionarily fitness of H1N1 with H274Y, and the failure of oseltamivir treatment to reduce the viral load. Thus, in countries with high levels of H274Y in H1N1, initial treatment with zanamivir is preferred in the absence of sequence data demonstrating a wild type N1.
H274Y was reported widespread in northern Europe, but recent reports suggest resistance may now be at 100% in South Africa, Australia, Guatemala, and Honduras. The vaccine targets for 2007/2008 in the northern hemisphere, and 2008 in the southern hemisphere have Solomon Island/3 as the H1N1 target. The new vaccine for 2008/2009 has replaced Solomon Island/3 (clade 2A) with Brisbane/59 (clade 2B). The isolates described above are clade 2B, but the dominant sub-clade in South Africa has already acquire a cluster of polymorphisms, which may limit the utility of the new vaccine.
Thus, H274Y levels in H1N1 may increase in the upcoming season, which would lead to increased zanamivir usage. However, zanamivir resistance, Q136K or Q136R, has been reported on a clade 2B background in 2008 isolates in Thailand and the United States as well as 2007 isolates in South Africa and Australia, raising concerns of an increase in evolutionarily fit H1N1 with Q136K.
Media Links
Recombinomics Presentations
Recombinomics Publications
Recombinomics Paper at Nature Precedings