Avian influenza A (H7N9) virus...
Last Updated 03-December-2013
The latest avian influenzavirus strain added to the list of agents causing "bird flu" is called avian influenza A(H7N9) virus - a colourfully exciting reference to its key surface proteins (see influenza background). H7N9 is considered a low pathogenic avian influenza virus (LPAIV)...at least in its feathered hosts. Avian influenza A(H7N9) virus (which I may shorten to H7N9 to save my fingers) was once considered a relatively rare cause of infection although how extensively and routinely it was looked for in the past is unclear. Avian influenza A(H7N9) virus is similar to its closer cousins, H7N2, H7N3 and H7N7 and its more distant cousin H5N1 in that they are all IFAVs and they usually infect birds. However, they are each quite distinct and made up of different genes.
A time line of important H7N9 events starting with the date of illness onset for the first laboratory confirmed case. It will be updated as events unfold so stay tuned.
Avian influenza A(H7N9) virus is an IFAV (see the adjacent schematic) that has evolved in birds. THe virus has 8 single-stranded RNA segments that encode 11 proteins. However, when cases of infection by it were reported for the first time in humans (China, March 2013; very general locations of confirmed cases to date are shown in the adjacent chart), it was linked to its first two deaths10 in this host(an 87 year old male [87M, sick 19.02, died 04.03; 2 family member(s) also reported ill12 with pneumonia but H7N9 negative] and a 27 year old male butcher [27M, sick 27.02 and died 10.03], both in Shanghai, China) or critical illness (35F in Chuzhou city). Since then, cases slowly but continuously climb, at an increasing rate. Because there is a delay between illness onset and laboratory confirmation of 10-13 days, the newest cases started have illness onset dates some way back making it is still hard to judge the impact of control measures like live animal market closures and animal culling, on interrupting transmission. A timeline of the early cases is shown below and more info on precise placement of the cases can be found using this excellent map.
As of 8th May - these numbers are updated only intermittently due to absence of quality onset, death and discharge dates. Last update:03-12-2013
a Some official or media reports towards the the 1st H7N9 outbreak have not provided sufficient identifying details to know which cases have died. This number is based on statements through the media/WHO. My own tally only accounts for 31 fatalities b The PFC, a VDU-created term, is just a number - it is not meant to imply that these are all the cases happening (unlikely it ever could be) and is not a constant rate of acquisition as it is a rolling average spread across the total time period being looked at. It is meant as a guide to what is happening right now with the data we can get our hands on. Sometimes that's lots of data (as it was with H7N9 - but has not been for a while now) or very limited data (as it is with MERS-CoV cases). It will undoubtedly change in the coming weeks and years. To be pedantic and use the number of discharges as the denominator for the Case Fatality Ratio (CFR) will sensationally inflate the result in the early days. This is most useful at the end of an epidemic/pandemic, but not so much when data-in-hand is poor during the early days of many outbreak. Keeping in mind that some will take this number and multiply it by the world's population as an estimate of how many would die if the virus reached pandemic levels, I don't believe this is a helpful use of the CFR metric and is poor science. cSome sex cannot be determined hence does not add up to 100% also, the CFR is currently at 27% with the use of the higher number of deaths but I am showing 22% here as it is supported by data that allow the sex and other calculations to be made; dNo longer being updated.
The various H7N9-positive municipalities (Shanghai and Beijing) and provinces (e.g. Henan, Anhui, Zhejiang, Hunan and Jiangsu) in China have had different numbers of cases reported and the reports started at different times. The first cases were confirmed from Shanghai on March 30th and included the first case who showed signs darting back to February 19th. Jiangsu's first confirmed cases occurred March 30th (although they were not announced until April 2nd) then Anhui province followed (confirmed on 31st March) then Zhejiang province (1st April), Henan province (11th April) and Beijing reported its first case 13th April. Guangdong province confirmed the most recent case August 6th
We can see from the line chart (numbers are based on date of confirmed diagnosis) that different municipalities and provinces had differing total numbers of cases and rates of case accumulation. Shanghai was the first municipality to report cases but activity in Zhejiang province was especially rapid, rising exponentially...then abruptly halting. In the last week of April, Chinese authorities changed to a per-week reporting system instead of daily reports and important data data became limited. Key live bird market closures are highlighted (green triangle)
Researchers from the Erasmus Medical Centre [NLD] have noted here and here, that avian influenza A(H7N9) virus seems to have already adapted to a mammalian host to some extent. It does not appear fully adapted to humans as H7N9 is still showing signs of genomic instability.
The denominator problem...
You may remember from your early maths classes that fractions have both a numerator (the number on on top or before a slash/line, that represents a portion of the whole) and a denominator (the non-zero number on the bottom or after the line that represents all the parts of the whole). The avian influenza A(H7N9) virus outbreak in March-May 2013 mostly manifested as severe disease - those people being admitted to hospital needing help to breathe (requiring oxygen or mechanical ventilation) - and we did not see many mild or asymptomatic cases (virus positive but no obvious feelings of signs of illness). This might be because there were no mild cases (although a couple popped up in children and adults) or because only the worst cases were tested. In favour of the former hypothesis, enhanced surveillance in China found few H7N9-positive cases of mild disease - the most notable being children.
Many milder respiratory diseases look the same based on signs and symptoms but can be caused (and I use that word with caution) by many, many different respiratory viruses. There may be cases of respiratory infection going around in China that are H7N9-positive but not severe enough to warrant a admission to hospital or even a trip to the Doctor. Until we know more about this part of the H7N9 story, we won't know the total number of avian influenza A(H7N9) virus infections that are occurring, so we can't calculate the true denominator. More info on this in a recent article.
Obviously, if the denominator is small (like it is now with so few H7N9-positive cases and those being from mostly severe illness or deaths, then we see a high proportion (percentage) as our Case Fatality Rate (x severe cases over a small number of total cases is a high proportion). If there are mild cases of avian influenza A(H7N9) virus infection out there (as we have seen today, 15.04.13) and they are in large numbers, that denominator could be much larger and the severe disease cases will be "diluted" down to a smaller proportion (x severe cases over a large number of mild and severe total cases is a low proportion).
A graph of the number of H7N9-laboratory confirmed cases by day of illness onset (left y-axis) and the cumulative case numbers (pale blue background; right y-axis). These data are compiled for general interest, from multiple public sources including interested parties and official organizations They are compiled with all care however this site is one of my own construction and the data and opinions expressed are not supported by the WHO, Chinese CDC or any other official organization. This chart may not have every relevant case because of difficulties in obtaining complete data from these sources. Use images freely but please cite VDU and Dr. Ian M Mackay.
This graph shows the age (ungrouped) of all cases (includes fatalities; green bars) and the fatalities (red bars) among cases who were confirmed positive for H7N9. These data are compiled for general interest, from multiple public sources including interested parties and official organizations They are compiled with all care however this site is one of my own construction and the data and opinions expressed are not supported by the WHO, Chinese CDC or any other official organization. This chart may not have every relevant case because of difficulties in obtaining complete data from these sources. Use images freely but please cite VDU and Dr. Ian M Mackay.
Genome, genetic sequencing and related influenzaviruses...
According to GISAID11 data, the first H7N9 full genomes (all 8 segments) were submitted as below: =date of deposition; submitting authors/submitter and date submitted listed in italics or as superscript lining to paper below; age and sex of human host shown as XXM/F)
One of the Shanghai strains (A/Shanghai/1/2013 from the 87-year old male [87M] who died 10.03) was less similar to A/Shanghai/2/2013 (27M who died 10.03) and A/Anhui/1/2013 (35F, symptom onset 09.03). There was a
An example of the HA cleavage site targeted by proteases in birds and humans. Those with a monobasic cleavage site (only 1..or 2...basic amino acids in the cleavage-determining region-an Arg, His or Lys) are considered low pathogenicity viruses because they are cleaved by a protease (Tryptase clara)from club cells (f.Clara cells) limited in distribution to the respiratory section of the human airways (a similar protease is restricted to the guts of birds). Viruses with the multibasic cleavage site, which can be cleaved by ubiquitous proteases (e.g. furin) in many additional tissues and organs, are considered "high-path". Older H7N9 HA sequences and those from the China outbreak have what looks to be a lower potential for uncontrolled HA cleavage. The cleavage site is not the only determinant of disease-causing potential, but it is a significant factor. With thanks to Prof. Ross Barnard. These data are based on a published analysis12 and use GenBank and GISAID11 nucleotide data I analysed using Geneious Pro. Use images freely but please cite VDU and Dr. Ian M Mackay.
Other H7 influenza strains (e.g. H7N2, H7N3, H7N7) have infected humans before although, as far as we know, not to a great degree. No H7N9 has been detected previously. So far, the animal host for this avian influenza A(H7N9) virus outbreak eludes discovery, but there remains no proof that human-to-human transmission has occurred. There is no proof that it is not occurring however.
Important genetic changes, adaptation or mutations in H7N9...
This figure depicts the 4 possible parental viruses that contributed to the evolution of H7N9 through donation of gene segments. Segments of interest to the final H7N9 virus are shown in colour as are their products. Grey is used to depict genes/proteins not of interest to the final H7N9 virus. The timeline is based on mathematical modeling to estimate the time to most recent common ancestor. These data are based on three published analyses5,10,13. Use images freely but please cite VDU and Dr. Ian M Mackay.
Some things we don't know...
[With thanks to Dr. Katherine Arden for contributing to the questions and thoughts below]
Some questions that arise...
This graph shows the proportion of males (blue) and females (violet) contributing to all laboratory-confirmed cases (which includes deaths) and to fatal cases alone. It contains no data on the societal roles of males and females and therefore may bias towards the males simply because they are more involved in activities that increase their risk of exposure to the influenza A(H7N9) virus host or environmental source. These data are compiled for general interest, from multiple sources including interested parties and official organizations They are compiled with all care however this site is one of my own construction and the data and opinions expressed are not supported by the WHO, Chinese CDC or any other official organization. This chart may not have every relevant case because of difficulties in obtaining complete data from these sources. Use images freely but please cite VDU and Dr. Ian M Mackay.
Signs, symptoms and clinical issues...
The current presumed incubation period is 3-8 days (mean 5.8 days).8 Upon presentation cases generally include fever, cough (which may be productive), weakness, chills and shortness of breath progressing to viral pneumonia. It is not yet clear if this is just the "pointy end" of the cases i.e. those with the most severe disease. There do seem to be milder cases; a 4Y recovered from mild illness. This case supports that there may be other patients with milder disease who continue on without presenting to a Doctor, clinic or hospital for testing or treatment and which clinically, look like illness caused by the other 200 or so diverse respiratory viruses.
Hopefully, reports of patients being deterred from presenting to hospitals because of steep treatment costs were an exception and not a rule. Delays in treating cases may be dangerous to patient health and hinder contact tracing efforts which are very important during the early phase of an emerging viral disease.
An ongoing timeline, now infrequently updated, of the avian influenza A(H7N9) virus confirmed human cases. The most important data for this graph are date of illness onset, hospitalization, death and lab confirmation but they are presented variably by the provinces and municipalities and seem to be getting harder to find. Some holes exist but not too many. Some delays will occur as I search out more info. Info on contact with animals is rare (green line) despite media reports, so this aspect is likely underrepresented above. Each label on the x-axis is a week and each horizontal line represents a case. Most cases remain in hospital and so are continuing lines although more discharges are occurring as the outbreak progresses. The y-axis numbers indicate the Case# according to the FluTrackers chronological numbering scheme. These data are compiled for general interest, from multiple sources including interested parties, research publications and official organizations. This chart may not have every relevant case because of difficulties in obtaining complete data from these sources. They are compiled with all care however this site is one of my own construction and the data and opinions expressed are not supported by the WHO, Chinese CDC or any other official organization. Use images freely but please cite VDU and Dr. Ian M Mackay.
Case definitions for avian influenza A(H7N9) investigations...
The US Centers for Disease Control and Prevention (CDC) provide US-based advice to define cases which is paraphrased below. The full detail is here.
A dead pigeon from a Shanghai market tested positive for the H7N9 strain providing the first identification of a potential animal host for this virus, in the "wild". Avian influenza A(H7N9) virus was next found in chickens (subclinical disease).9 These findings led to the culling of over 110,000 poultry. There were also H7N9-positive samples from the environment (soil, water and faecal samples) around poultry markets in Shanghai.9 How the apparently discrete H7N9 cases have been infected remains a mystery.
This infographic shows the number of positive bird samples compared to the number of bird samples tested (from China's Ministry of Agriculture). Based on the New England Journal of Medicine study7, you can see the number of human contacts of H7N9-positive cases that have been observed (including 19 who showed signs and symptoms, were tested by RT-PCR and found negative) for signs and symptoms of H7N9 infection added to 5,551 samples screened as part of sentinel surveillance. The number of humans positive for H7N9 is shown oin the adjacent image (120 cases when the bird study details came out; 82 human cases in the NEJM article that describe the 1,689 contacts). It is meant as a rough illustration of what was pointed out by Mike Coston at the Avian Flu Diary, there are few H7N9-positive bird samples according to the Chinese Ministry of Agriculture, compared to positive human cases. These data are compiled for general interest, from multiple sources including interested parties and official organizations. This chart may not have every relevant case because of difficulties in obtaining complete data from these sources. They are compiled with all care however this site is one of my own construction and the data and opinions expressed are not supported by the WHO, Chinese CDC or any other official organization. Use images freely but please cite VDU and Dr. Ian M Mackay.
This new influenzavirus emergence highlights the serious implications of a virus jumping from its natural host to another species of host to which the virus is not well adapted. When not well adapted, the virus's ability to infect, replicate and spread, preferably without killing the spreader, is disturbed, mainly due to the new host's vigorous immune defence against such a foreign invader. The ensuing cytokine storm and tissue destruction from a zoonotic virus lead to severe disease in the human. A zoonosis is when a virus jumps from another animal host to a human host; anthroponosis is the reverse route.
This figure shows major wild bird "flyways" considered to have contributed key virus gene segments to the influenza A(H7N9) virus infecting poultry and humans Southeastern China.10 Domestic ducks in Hong Kong were believed to provide a major mixing vessel between wild birds and chickens.13
No sustained human-to-human (H2H) transmission (see the figure below) has been confirmed but suspect "clusters" exist. Some are families in which multiple respiratory infections occurred over a suspicious period of time, but only some members were laboratory tested\positive for avian influenza A(H7N9) virus positive. Proving H2H transmission is very difficult because families, friends or couples share life experiences that expose them to a common source of infection.
The sudden appearance of the first cases in humans made epidemiological investigations (virus sleuthing) an urgent and essential next step to answer how and where the virus was acquired and whether its been around in people for a while. Its possible that H7N9 does not always cause the severe disease that brought it to our attention and seroepidemiology will be needed to measure specific antibodies to define such previous infection. However, many positive cases have a traceable contact with poultry of some kind - although not all. Given the genetic relationship to wild birds and poultry, it seems likely that transmission is from poultry to human however there is still a chance that humans are already acting as the main vector. More testing is needed
Current advice to protect yourself and to break the transmission cycle is to maintain good hand hygiene using soap and running water - an alcohol-based hand cleanser can be used is hands are not visibly dirty (before, during and after preparing food, before you eat, after you use the toilet, after handling animals or animal waste; when your hands are dirty; and when providing care when someone in your home is sick), respiratory hygiene (cover your mouth and nose with a mask, tissue, sleeve or flexed elbow when coughing or sneezing; throw the used tissue into a closed bin immediately after use; perform hand hygiene after contact with respiratory secretions) and to prepare and thoroughly cook (food should reach 70'C in all parts, no pink parts) only healthy birds and bird-products well.
Shown by colour are the numbers of cases in each province, municipality or national central city in southeastern China (and the imported case in Taiwan) that have confirmed cases of avian influenza A(H7N9) virus infection of humans.
Reagents and testing methods have been distributed to laboratories within China and reported to the World Health Organization (WHO). China CDC undertakes polymerase chain reaction (PCR) testing and reverse agglutination testing for serology. The WHO has released its supported reverse-transcriptase real-time PCR-based protocol which include a housekeeping/endogenous gene control. Taiwan had previously reported it had developed an in-house RT-rtPCR. Researchers from the University of Bonn in Germany, teamed with oligonucleotide experts, Tib-Molbiol, to describe 3 newly designed RT-rtPCR assays based on an alignment of 3 Chinese H7N9 sequences, targeting HA (two assays) and NA and tested using A/Mallard/Sweden/91/2002 (H7N9) RNA. A mutated (traceable to discriminate laboratory contamination) cloned positive control was made publicly available to researchers. The choice to use a variable gene target rather than the conserved M gene was to avoid false positives due to other influenza A viruses. The US CDC has now released its RT-rtPCR and protocol with support from the FDA. The assay also contains an RNaseP endogenous control as does the WHO protocol. Taiwan's CDC recommended hospital collect a lower airway sample such as sputum. His followed the imported H7N9 case returned two negative tests on throat swabs, but was RT-rtPCR-positive using a sputum sample. There is evidence of assay variation, supporting underlining the need for independently designed, validated and compared molecular assays - and sharing of results.
A chart showing the time taken for the laboratory to confirm an H7N9 case, starting with the first confirmation on March 30th, 2013. The green line shows the average of all cases confirmed on that day while the red line is a rolling average number of days. The x-axis is the date of laboratory confirmation and the y-axis is the average number of days that it took, since disease onset, for the first confirmation; sometimes a day earlier than the official Chinese CDC date. These data are compiled for general interest, from multiple sources including interested parties and official organizations. This chart may not have every relevant case because of difficulties in obtaining complete data from these sources. They are compiled with all care however this site is one of my own construction and the data and opinions expressed are not supported by the WHO, Chinese CDC or any other official organization. Use images freely but please cite VDU and Dr. Ian M Mackay.
Hopefully, screening can expand beyond poultry to the numerous other avian species with which the Chinese population community have more direct contact.
Traditional virology techniques like the adjacent electron micrograph (comes from http://www.cdc.gov/flu/images.htm; captured by C.S.Goldsmith and T.Rowe) are not likely to get much use for specimens screening. Additionally, commercial companies are producing recombinant H7N9 protein, expressed from human cells, as reagents that could be used in antibody-testing methods.
The "lab-in-chip" assay, VereFlu, run on Veredus’ VerePLEX is also reported to detect H7N9.
Prophylactic and therapeutic options for avian influenza A(H7N9) virus...
Antiviral drugs exist to treat (amantadine, rimantadine, zanamavir[Relenza] and oseltamivir[Tamiflu]) and prevent (amantadine [avian influenza A(H7N9) virus resistance exists], rimantadine, and oseltamivir) IFV disease. Oseltamavir and zanamavir are neuraminidase inhibitors. More and more however, we are noting single amino acid mutations which confer resistance to antivirals, driving new research into different viral targets and mechanisms to treat infection once it has been acquired. Those benefiting most from the use of antivirals include people 65 years or older, pregnant women, indigenous people 17 years or older, nursing home/aged care facility residents, and anyone older than 6 months of age with an underlying medical condition (cardiac or kidney disease, severe asthma, diabetes, impaired immunity or neuromuscular disease). Early reports, evidence from the Chinese CDC and a statement from the WHO indicate that the emergent H7N9 strain is sensitive to neuraminidase inhibitors. However, researchers have identified the presence of an important mutation within one of three avian influenza A(H7N9) virus strains' publicly available (Global Initiative on Sharing Avian Influenza Data, GISAID) genetic sequences; one that can confer resistance to oseltamivir [Tamiflu] in H3N2 viruses
No specific vaccines exist yet and once commenced (underway in the United States, China and Taiwan), contemporary methods will take approximately 5-7 months to produce a marketable product. However, a report suggests an "anti-flu" injection exists that can "cure H7N9 flu". It has been under development by the China Academy of Military Medical Science. It seems [IMM: I'm unable to clarify] this is the same as Peramivir (called RAPIACTA in Japan and PERAMIFLU in South Korea), a new intravenous neuraminidase inhibiting drug, designed to target both influenza A and B viruses, and has completed Phase III clinical trials. Developed by US-listed biotech BioCryst Pharmaceuticals, it has been fast-tracked by Chinese FDA and was used to treat H7N9 patients, but mutations in 2 patients rested in treatment failure and viral load rebound compared to oseltamivir. The US CDC is developing a candidate virus seed strain. There is also a report of a rapid epitope-driven vaccine design (within 36 hours) from EpiVax Inc. FluCideTM (NanoViricides, Inc.) is an injectable (severe disease) or oral (more mild disease) anti-influenza formulation comprising a molecule that mimics the sialic acid receptor or influenza virus(both the mammalian [sialyl-a2,6-gal-] and avian [sialyl-a2,3-gal-] forms), coating the virus and preventing its attachment to cells. Phase I trials are pending. Greffex claims to have completed creation of an H7N9 vaccine, with support form NIH and NIST, carrying HA and NA genes on a gutted adenovirus background.
Traditional Chinese remedies, such as ingesting woad root (banlangan in Chinese) have been suggested by government workers. At least one peer-reviewed study suggested that there was little or no effect on NA inhibition but that metabolites of the root may, to some degree, inhibit NA.
Information sources you can trust...