Paradigm Shift Intervention Monitoring
twitter
Live feed of underlying pandemic map data here
Commentary
CDC
Fails to Show Swine H3N2 Transmission To Humans
Recombinomics
Commentary 05:20
November 13, 2010
The above comments from the CDC update on the most recently reported H3N2 swine flu cases are carefully worded to give the impression that the cases were due to transmission of H3N2 from swine to humans, but the report itself gives no such data for the two most recent cases, and no such data has been presented for the three prior H3N2 cases. Moreover, there is no claim of contact for three of the five cases, and for the two cases with contact, there is no data indicating the contacted swine were symptomatic or infected.
Direct data has been presented for H1N1 infections at Huron County fair in Ohio in 2007. In that outbreak pigs at the fair were infected and H1N1 virus, A/swine/Ohio/24366/07 was isolated. Virus was also isolated from a presenter (10F), A/Ohio/01/2007, and her father (36M), A/Ohio/02/2007. 24 others at the fair had flu-like symptoms (in August in Ohio), and all human and swine sequences matched and were available at a public database.
However, these cases represent the bulk of evidence for transmission of H1N1 triple reassortants from swine to humans in the United States, although many of the symptomatic cases as well as one of the confirmed cases could have been infected by human to human transmission. A NEJM report described 11 cases infected with H1N1 triple reassortants, and only five (which included the two described above) had direct contact with swine. Two had no documented exposure, and four were in the vicinity of swine, but had no direct contact.
Weaker data has been presented for H3N2 cases, which occured after the 2009 H1N1 pandemic began. Only two of the five have been cited as having direct contact, and both cases are somewhat confusing.
The first case was reported by the Kansas Department of Health and Environment (KDHE) and involved a child at the Riley County fair in July, 2009. The child developed symptoms in late July and sought medical care. The KDHE press release indicated the exhibitors were being questioned about the health of the swine at the fair. There were no reports that swine were ill, and no swine sequences from Kansas have been announced or made public.
However, a full set of sequences was made public by the CDC (at GISAID in February). The characterization sheet for the isolate, A/Kansas/13/2009, indicated the sample came from KDHE and was collected July 27, 2009 and was a “human case of swine- origin H3N2 triple reassortant”, which was consistent with the published sequences. However, the patient was listed as a 22M, indicating there were at least two infected patients in July in Kansas or there was a data entry error in the KDHE press release or the GISAID characterization sheet.
The other case with reported contact also was at a state fair. It was reported by Wisconsin, although the pager alert cited a 7 month-old child from Illinois, suggesting the child lived in one of the two states and was at a fair in the other. However, the nature of the direct contact by a 7 month old baby remains unclear.
Thus, in contrast to the outbreak at the Huron County fair, the two H3N2 cases with direct links to swine have no demonstrated link to infected swine. Moreover, only one human sequence has been made public, and the age on the characterization sheet does not match the description of the patient at the Riley County fair. The other three swine H3N2 cases lack direct contact with swine, so there are no H3N2 cases which show a human contact with a confirmed infected swine, and there are no public swine or human sequences that match the described cases.
The withholding of the sequences is curious. All cases were after the start of the 2009 pandemic, but only one sequence set has been released, although sequences were required to classify the cases as H3N2 infections by swine triple reassortants. Other than the H and N, the other gene segments in H1 and H3 triple reassortants are similar, although the sequences from the 22M from Kansas is readily distinguished from the pH1N1 sequences, including MP, which is adamantine sensitive in the H3N2 isolate.
The sequences from the five cases can be subjected to phylogenetic analysis to characterize the relationship between the isolates. Although the latest CDC report cites “differences”, there are “difference” between pH1N1 sequences, although all have a common origin. Thus, the “differences” in the two most recent H3N2 sequences do not necessarily exclude a common origin due to a jump from swine to humans followed by human to human transmission.
Although the CDC report cites a failure to find community human to human transmission of swine H3N2, the investigations appear to have failed to find any swine to human transmission, because none of the H3N2 reports describe infected swine or isolated H3N2 from the contact swine.
Thus, the CDC introduction on infected swine has no demonstrated basis for any of the five reported H3N2 cases, and only one set of human sequences has been released (but the age does not match the described case).
Release of all human H3N2 swine sequences is long overdue. The Iowa case was reported in January, 2010 and the Minnesota case was reported in May, 2010. Moreover, sequence data for the two most recent cases was generated to characterize the isolates as triple reassortants with some “differences”.
In addition, the origin of A/Kansas/13/2009 should be clarified (is there a second case or is the age mismatch due to entry errors).
The appearance of swine H3N2 in humans following the pH1N1 is of considerable concern, and the reliance on negative data, withheld sequences, a sloppy reporting is appalling.
Media link
Recombinomics
Presentations
Recombinomics
Publications
Recombinomics
Paper
at Nature Precedings