H5N1 Transmission
Recombinomics Commentary
December 10, 2007
This rare case of two family members struck by the disease has drawn urgent concern from health authorities, because humans almost always contract H5N1 from infected birds.
Hans Troedsson, the World Health Organisation's China country representative, said that even if the man was infected by his son, he had no epidemic fears at present following the new case.
"We know that this strain that we have seen here in China, Vietnam, Indonesia and Thailand has a very limited capability for human-to-human transmission," he told a news conference.
The above comments are a reminder of the WHO efforts to discount human-to-human transmission in the many clusters in 2004 and 2005 and attempts to reconcile those commissions by omission with more science based acknowledgement that there have always been H5N1 clusters and none involved reassortment with human genes.
In 2004 and 2005 the WHO mantra was “no evidence of human genes”. The mantra is repeated again today regarding the latest example of human-to-human transmission in China.
In earlier examples, especially when the cluster was limited to two family members, human-to-human transmission was not eliminated, but it was not proved. Of course proving human-to-human transmission is virtually impossible when the cluster is small. As the cluster grows, or transmission chains are extended, the explanation of human-to-human transmission becomes more convincing,
Sequence analysis really doesn’t distinguish between a common source and human-to-human transmission, because both mechanisms involve similar sequences. The strongest data is the time between disease onset dates. A common source would generate dates that are within 1-2 days, while human-to-human would generate dates that are 5-10 days. The current cluster is clearly in the human-to-human category.
Human-to-human has been widely accepted in larger clusters, such as the cluster in Thailand in 2004, Vietnam in 2005, or Indonesia in 2005 and 2006. All of these clusters involved three of more contacts which were H5N1 confirmed and had gaps in disease onset dates. However, familial clusters alone signal a more efficient transmission, even if due to a common source, and in most cases, the first confirmed case in a country was part of a cluster (although in some cases there was no sample from the index case, even though the patient died with bird flu symptoms in a area with poultry of wild bird confirmations).
Countries with index cases as part of a cluster include Cambodia in 2005, China in 2005 (also Hong Kong family in 2003), Indonesia in 2005, Turkey in 2006, Iraq in 2006, Azerbaijan in 2006, Nigeria in 2007. None of the above index clusters or subsequent clusters within those countries or other countries involved reassortment with a human gene.
Thus, the current human to human transmission in China is not an indication that there has been a major change in H5N1 and announcement of no reassortment is not unexpected.
However, several of the clusters above have been associated with changes in or near the receptor binding domain. which is not addressed in the current announcement on sequence data out of China.
Elimination of the “because humans almost always contract H5N1 from infected birds” nonsense from media reports would be useful. The statement has never been true, and only adds confusion to a public that is already quite confused on the evolution of H5N1 and its pandemic potential.
Media Links
Recombinomics Presentations
Recombinomics Publications
Recombinomics Paper at Nature Precedings