Paradigm Shift Intervention Monitoring
Commentary
Fit Tamiflu Resistant
Pandemic H1N1 in the United States
Recombinomics Commentary 11:20
August 17, 2009
The above comment on the recent MMWR dispatch warning of prolong shedding of oseltamivir resistant pandemic H1N1 in immunosuppressed patients is in error. The two patients from Washington state were the first confirmed cases of H274Y in swine H1N1 in the United States, but the presentation of the resistance data helped create the above misconception that the H1N1 muted within the patients over a period of several weeks. MMWR dispatches are used to quickly alert physicians to significant medical concerns and the inclusion of August 11 data in the August 14 publication signals the rapid publication of such data as well as the urgency of such reports. However, the description of the resistance could easily be misinterpreted and lead to the false conclusions stated in the above media report.
Recent data on the detection of H274Y in pandemic H1N1 has supported the silent spread of the polymorphism which is usually detected in patients treated with osletamivir. Since the H274Y is present as a mixture, sequence analysis of untreated patients produces a "Tamiflu sensitive" genotype, with H274. Indeed, the only reported case of H274Y in a patient who was not treated with Tamiflu was a San Francisco resident who was identified in Hong Kong through routine surveillance. However, multiple examples of H274Y in patients who have been briefly treated with Tamiflu support a relatively high level of H274Y prior to treatment. Published sequences are a consensus and represent the dominant nucleotide at a given position, so samples with H274Y in 10-20% of the virions will not be represented in the published sequence. When levels approach 50% the consensus sequence will show a mixed signal at the position, but sequences published to date have either had wild type or H274Y at position 274. This is due in part by the ability of the Tamiflu, in combination with and the host's immune system, to rapidly clear the wild type sequences (and most patients can clear the H1N1 without antiviral treatment)..
This type of data was most clearly presented in the recently released sequence from an American traveling from Honolulu to Singapore. The patient developed symptoms during the flight, but successfully passed through the thermal scanner. However, her condition did not improve and she was hospitalized the next day, May 27. The H1N1 was confirmed on May 28 and the sequence indicated the virus was Tamiflu sensitive. The infection was mild and the patient was discharge on May 31. However, a sample was collected on May 30, and the associated sequence had H274Y, signaling the rapid appearance of the resistance, and the rapid clearing of the wild type sequence (the sequence had no mixed signals).
This rapid appearance of H274Y was also seen in multiple patients on prophylactic Tamiflu. These were asymptomatic contacts of confirmed cases infected with Tamiflu sensitive H1N1. However patients in Denmark and Japan developed symptoms after 5 days of treatment and a patient from Hong Kong developed symptoms on day 6. Since the incubation period of influenza is 2-4 days, the slightly longer than normal incubation period indicated the H274Y was silently spreading as a minor component, which was subsequently detected after brief Tamiflu treatment.
In the two patients in Washington, the clearance of wild type was likely delayed somewhat by their immuno-compromised state. The first patient responded to Tamiflu treatment, but then had an H1N1 recurrence which had H274Y. Testing of samples collected earlier showed that prior to treatment the virus was sensitive, as was the sample from treatment day 4. However, H274Y was identified on day 11 after the start of treatment, indicating the H274Y reached a detectable level between days 4-11 post initiation of treatment, and the H1N1 was evolutionarily fit and produced a recurrence in the patient.
Data from the second patient produced similar results, but that H1N1 with H274Y has produced a persistent infection which has not been cleared by aggressive Tamiflu treatment or treatment with inhaled and intravenous Relenza, as well as ribavirin. The patient remains hospitalized with H274Y positive H1N1, again demonstrating the virus is evolutionarily fit. Data on earlier collections indicated that the H1N1 was sensitive prior to treatment and H274Y was detected on day 18 after the start of treatment. However results on earlier collections on day 3 and 6 of treatment were not disclosed, so the H274Y reached detectable at some time between day 1-18 post treatment. Testing of samples collected on days 3 and 6 could help resolve the initial date of detectable H274Y.
However, the above data clearly show that H1N1 is evolutionarily fit and can cause significant medical problems linked to prolonged shedding, as well as the condition of infected patients. These data increase concerns that H1N1 with H274Y could be linked to the increased number of fatal and hospitalized patients, who are frequently treated with Tamiflu.
Testing of samples collected after the start of Tamiflu treatment would be useful because, as seen in recently released data, virus which is Tamiflu sensitive prior to treatment can be quickly replaced by Tamiflu resistant H1N1, leading to significant complications.
Media Links
Recombinomics
Presentations
Recombinomics
Publications
Recombinomics
Paper
at Nature Precedings