Paradigm Shift Intervention Monitoring
twitter
Live feed of underlying pandemic map data here
Commentary
CDC Reliance on
Negative D225G/N Data Raises H1N1 Concerns
Recombinomics
Commentary 19:32
March 7, 2010
The above CDC comments, in response to the Norway study showing strong linkage between D225G and severe (3) and fatal (11) cases compared to mild (0) cases, mentions a subset of CDC sequences from US cases. The acknowledgment of only 8 cases of D225G in sequences generated by the CDC, when at least 15 D225G positive US sequences have published indicates the CDC is relying on negative data from direct sequencing to exclude positive sequences from isolates grown in eggs. Reliance on negative data is dangerous when the negative data is based on an assay that lacks sensitivity, and is refuted by positive data from virus grown in eggs as well as mammalian cells.
Others have identified D225G in direct sequencing. The most challenging was the generation of sequences from samples collected from fatalities in the 1918/1919 pandemic. Since the virus was degraded and fragmented, it was not possible to grow the virus. Instead the samples were sequenced directly and five HA sequences were generated, all of which covered the receptor binding domain. Two sequences were from London (1918 and 1919) while 1918 sequences were generated from New York, South Carolina, and Alaska. This direct sequencing approach identified D225G in two of the five sequences, including the only sequence from 1919. The D225G in 1918 New York sequence demonstrated that D225G was widespread and present in both waves tested. Since H1N1 in 1918/1919 was readily transmitted and resulted in an estimated 50 million casualties, the finding of D225G in 40% of the 1918/1919 samples tested is significant.
However, the presence of D225G and or D225N in direct sequences from autopsy lung samples in Ukraine is also striking. 27 of 37 samples had D225G, D225N, or both. Moreover, like Norway, virtually all samples positive for D225G or D225N were from fatal cases. Thus, the significant association of D225G with fatal cases has been demonstrated by direct sequencing, as well as from sequences from virus grown in mammalian cells, as was done in the study from Norway.
However, the CDC has also generated D225G and D225N sequences in mammalian cells even though the direct sequencing was negative. This dataset was generated in the fatal cluster at Duke Medical Center. Four patients infected with the Tamiflu resistant H1N1 that also had a rare HA marker, Y233H. Thus, it was clear that all four patients were infected by the same H1N1. The direct sequencing produced partial HA sequences with a wild type receptor binding domain. Three of the four patients died and three of the four also had samples collected prior to Tamiflu treatment, which were likely the samples collected from the first three patients which were collected within 1 day of each other. Four samples were collected from these three patients and three of the four samples had D225G or D225N casting serious doubt on the true lack of D225G or D225N in the original samples as indicated in the sequences released by the CDC. Moreover, both samples collected from the same patient had increasing levels of D225G. The first collection had D225G as a mixture with wild type, while the second collection had D225G in the absence of D225. The sequence from the original sample from this second collection was not released.
The detection of D225G and D225N in isolates from mammalian samples raises serious concerns about the reliance on negative data from direct sequences, as was seen in the Duke samples, as well as April samples which were positive for D225G in virus grown in eggs.
The assumptions that the D225G and D225N in lab cultures are artifacts have serious consequences for the selection of vaccine targets as well as the monitoring of clinically relevant polymorphisms, such as D225G and D225N.
Independent evaluation of the direct sequence from the second collection from the Duke patient, as well as the tracings of the underlying data for all sequences from original samples which were positive for D225G/N in viral cultures would be useful. The reliance on negative data for public comments and policy continues to be hazardous to the world’s health.
Media Links
Recombinomics
Presentations
Recombinomics
Publications
Recombinomics
Paper
at Nature Precedings