| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Review |
1 Health Protection Agency, East of England, Clinical Microbiology and Public Health Laboratory, Addenbrooke's Hospital, Hills Road, Cambridge CB2 2QW, UK
2 Respiratory Virus Unit, Health Protection Agency, Colindale, London NW9 5HT, UK
Correspondence
Martin D. Curran
martin.curran{at}addenbrookes.nhs.uk
Joanna S. Ellis
joanna.ellis{at}hpa.org.uk
Received 10 April 2007
Accepted 29 June 2007
Introduction
Given the serious threat to public health, early and accurate diagnosis of H5N1 infection is critical for effective clinical management, disease containment and infection control measures. Since re-emerging in 2003, avian H5N1 has maintained a steady global advance from the epicentre in South-East Asia, infecting wild birds and poultry, spreading into Europe and parts of Africa. Viral evolution has led to significant genetic variation among H5N1 viruses, associated with increased virulence and an expanded mammalian host-range. Numerous lineages, HA clades and subclades have been described, with recent isolates clustering into two distinct clades: clade 1 and the more diverse clade 2. Moreover, the ability of H5N1 to directly transmit from infected birds to humans following close contact, although a rare event, has been repeatedly demonstrated in more than 270 cases with an approximate case fatality of 50–60 % (http://www.who.int/csr/disease/avian_influenza/en/). Although H5N1 remains an avian virus not yet adapted to transmission between humans, there is concern that small genetic changes may significantly alter the pandemic potential of this virus, allowing it to emerge as the next influenza pandemic strain. Highly sensitive, specific and rapid tests for H5N1 detection are therefore essential to control the threat to human health posed by this zoonotic infection. In view of the genetic diversity, molecular assays capable of detecting a wide range of circulating H5N1 variants are essential.
Pandemic planning and H5N1 threats in the UK
The detection of H5N1 in an imported finch in a quarantine centre in October 2005, and in a dead Whooper swan washed up in Scotland in March 2006, avian influenza (H7N3) outbreaks in two poultry farms in Norfolk in April 2006, and more recently the H5N1 outbreak in a turkey farm in Suffolk in February 2007, emphasize the clear and present danger of H5N1 to animals and wildlife in western Europe, despite the distance from South-East Asia. Human exposure to H5N1 may occur not only in returning travellers potentially exposed in distant regions, but also as a consequence of exposure to virus in wildlife or poultry in the UK. The ability to detect and confirm rare zoonotic infections is usually the preserve of a reference laboratory. However, the threat posed by H5N1 and consideration of strategies to mitigate a pandemic (Ferguson et al., 2006) suggest that at least one counter measure that can be considered is the use of antivirals to contain an epidemic at source. This requires the ability to detect and accurately diagnose infection at or close to the source/outbreak with minimum delay, a tactic consistent with global experience during the SARS epidemic in 2003, underlining the desirability of decentralised specialist testing, at least to a regional level.
UK regional diagnostic capability
Development of robust and rapid regional diagnostic capability for detection of H5N1 is part of national pandemic planning activity. Delivery of a seamless, integrated national diagnostic service, capable of delivering a reliable result for either a returning traveller with an influenza-like illness or exposed poultry worker, requires coordination of scientific, public health and logistical activities through a functional laboratory network. Diagnosis of H5N1 influenza can be achieved by a variety of techniques, the most reliable and sensitive of which are nucleic acid amplification techniques applied to clinical respiratory samples (de Jong et al., 2006). Such tests target the most highly variable gene of the H5N1 influenza virus genome, the viral HA gene. However, as with any test used in a very low incidence setting (such as with rare zoonotic infections), the predictive value of a single positive test is <70 %. It is essential to ensure that there is a confirmatory strategy to improve the certainty of diagnosis. In view of the frequency of returning-traveller queries (over 100 000 UK nationals return every week from South-East Asia), it was recognized that H5 testing needed to be based in regional laboratories, close to the patients, with the diagnostic service available 24/7 to deliver a confirmed result in a timely fashion, 3–4 h following receipt of a sample.
H5 assay development
The last few years has witnessed some exciting new developments relating to rapid H5 diagnosis. A number of real-time RT-PCR assays exist for the detection of avian influenza viruses, including assays that can detect all influenza A subtypes with an internal control (Di Trani et al., 2006; Das et al., 2006), a triplex assay that can specifically detect H5, N1 and all influenza A subtypes (Payungporn et al., 2006) and a monoplex assay that specifically detects H5 only (Ng et al., 2005; Chen et al., 2007). Evaluating which of the assays are suitable for primary screening and which might be better used for confirmation, ensuring that such assays are both sensitive and specific for the detection of H5 in relevant clinical material and are fully validated for clinical use, is a tall order. Within the UK, prior to the establishment of the National Influenza H5 Laboratory Network in 2005, significant progress had been made in some regional laboratories towards implementing multiplex real-time PCR methods for respiratory virus screening (Coyle et al., 2004; Moore et al., 2004; Templeton et al., 2004; Gunson et al., 2005), but H5 diagnosis was provided centrally at the Influenza Reference Laboratory in London and the Regional Virology Laboratory in Glasgow. Access to panels of influenza virus subtypes and different strains of H5 influenza to ensure subtype specificity and sensitivity of detection for relevant circulating strains is essential for the validation of assays that will be used to decide important/high profile public health measures. Clinical material from human H5 cases is globally highly restricted, since such material is in short supply and generally located outside western Europe. Assay validation involves comprehensive testing of reference influenza H1–H16 isolates, recently circulating H5 viruses from all H5 HA clades, related respiratory pathogens and clinical material containing human seasonal influenza. Preparation and analysis of simulated clinical material spiked with relevant concentrations of different influenza H5 (clade 1, 2 and 3) and other virus strains is an important part of assay validation, but is generally the preserve of specialist reference laboratories. However, measuring assay performance for different platforms (TaqMan, LightCycler, Rotor-Gene, Smart Cycler etc.) and fine-tuning technical/confirmatory strategies suitable for use in diagnostic settings require input from laboratories providing front-line diagnostic services in order to develop a coherent national strategy.
H5 assay validation
Creation of laboratory partnerships to overcome difficulties in developing tests suitable for a wide variety of settings is essential to ensure that robust and reliable assays are implemented. In the UK, collaborations between national reference and regional virology laboratories has led to the development of several national standard operating procedures (SOPs) for molecular detection of H5 (Ellis et al., 2007; M. D. Curran, J. S. Ellis and M. C. Zambon, unpublished results), and a suite of confirmatory assays targeting different regions of the H5 HA using different technical approaches and platforms. Pyrosequencing, melting point analysis (fluorescence resonance energy transfer), restriction digestion of the screening assay RT-PCR amplicon (RFLP), an agarose gel-based RT-PCR with a post-amplification restriction enzyme digestion step (PCR-RFLP) and sequencing have been comprehensively validated on appropriate platforms to distinguish amplification of a circulating H5 strain from the assay control material (Ellis et al., 2007). This allows regional laboratories some choice in assay and confirmation strategies to provide a more streamlined work flow alongside other molecular diagnostic work performed in the same laboratories. Efficient diagnostic testing requires multiplex assays that can detect all influenza A subtypes (H1–H16) and influenza B, as well as specifically identify the presence of H5, with an internal control in a single reaction. A quadriplex real-time assay, which was developed for the network, fulfils these criteria, and allows simultaneous differential diagnosis (flu A or flu B) and the whole process of nucleic acid extraction, reverse transcription and amplification to be monitored, providing quality assurance and thereby minimizing the possibility of issuing a false-negative result (M. D. Curran, J. S. Ellis and M. C. Zambon, unpublished results).
Establishing a network
Whilst the requirement for a national/regional H5 laboratory network was evident, the practicalities of establishing it required substantial coordination. Choice of laboratories for inclusion in a laboratory network was based on:
Initial choice of laboratories and assays was followed by a comprehensive training course demonstrating selected assay protocols, to ensure that key laboratory personnel would be able to implement appropriate assays in their own laboratories. Participants returned to their laboratories with knowledge of the diagnostic and confirmatory assays. Regional laboratories aimed to provide a molecular-based diagnostic service, as well as a surveillance service, for influenza A, B and the avian H5N1 virus. A single assay was initially adopted as a front-line H5 screening test (Ellis et al., 2007), providing the necessary national standardization and adaptability to other real-time platforms. While consensus existed for a confirmatory strategy using different assays, the development of a second real-time confirmatory assay, capable of distinguishing a control from wild-type H5N1 and run in parallel with the diagnostic assay, streamlined testing and made a 24/7 ‘test and confirm’ service an achievable objective. To facilitate low-level surveillance and maintain proficiency, weekly runs using both assay formats performed on 10–20 routine clinical respiratory specimens by all laboratories providing an H5 service were undertaken.
Network support activities
Supporting activities for the UK National Influenza H5 Laboratory Network include:
A secure H5 laboratory network website has proved extremely useful for facilitating information deposition and exchange, identifying problems with methodologies, providing technical and information updates, clinical guidelines and newly released SOPs, and allowing laboratories to see and provide supporting evidence and data to substantiate issues raised during regular monthly teleconferences. Monthly teleconferences have been pivotal to the success, and ongoing development and testing strategy of the network. As part of the network support by the reference laboratory, regular sequence review of newly deposited H5 sequences is undertaken, and new H5N1 isolates from around the world are analysed to ensure that no modifications are necessary as a consequence of sequence diversity occurring in the target sites, and that all approved diagnostic assays used by the network are effective. Centrally provided CE-marked inactivated H5N3 virus control material, which is suitably easy to distinguish from currently circulating H5 strains, is used as positive control material throughout the network.
Annual proficiency testing
Annual proficiency testing has been introduced to ensure that participating laboratories are accredited for H5 diagnostic testing, and demonstrate that regional laboratories in the National Influenza H5 Laboratory Network can provide a quality-assured diagnostic service for H5 detection. The first H5 proficiency-testing panel of simulated clinical samples containing high and low titre Vietnam H5N1 viruses (inactivated), and influenza A and B viruses, was distributed in October 2005 simultaneously to 19 laboratories in the UK and 1 laboratory in the Republic of Ireland, timed to arrive at each laboratory at the same time. The results of proficiency testing demonstrated an overall excellent performance by the laboratories. All participating laboratories had succeeded in establishing H5 assays. The high completion rate (18/19) and overall performance showed that laboratories had been able to implement H5 specific assays in a 3–4 month period, irrespective of local problems. The following conclusions were possible following the first proficiency exercise:
Proficiency testing has continued, with a second proficiency panel in November 2006. This was designed to test whether laboratories could reliably detect virus strains representative of all circulating H5 lineages at clinically relevant concentrations. The panel contained 12 samples, including all 3 major H5 HA clades, as well as a number of other influenza A subtypes (H1, H3 and H7) and influenza B. All UK regional laboratories included in this exercise demonstrated the capability to detect and confirm diverse H5 strains, and deliver H5-screening results in a very timely way, which was substantially improved from the previous exercise, and suitable for dissemination in the event of a suspect case of avian influenza. Timed proficiency-panel exercises are an important and useful way of assessing laboratory service provision, testing technical decision making in complex algorithms and clinical interpretation, as is required in high-profile outbreak situations where the time to result delivery is calculated in hours, not days. Results of proficiency panels provide objective evidence of improving proficiency and familiarity with assays, and performance in these assays provides a means of deciding on service delivery capability for the National Influenza H5 Laboratory Network.
Network performance
The number of suspect H5N1cases in the UK fulfilling the clinical algorithm for exposure and illness presentation (http://www.hpa.org.uk/infections/topics_az/influenza/avian/algorithm.htm, http://www.hpa.org.uk/infections/topics_az/influenza/avian/documents/J3poultrymanagement080207.pdf) between 2005–2006 has been limited (approximately 1–2 per month), although presentation of possible cases has occurred throughout the UK. The outbreak of avian H7N3 influenza (low pathogenicity) in a poultry flock in Norfolk in April 2006, and its subsequent identification in two other free-range poultry farms close by, provided an opportunity to examine the national strategy for avian influenza. Suspect H7 human cases arising during this episode were screened at the local regional laboratory and specimens fast-tracked to the National Reference Laboratory for confirmation and further analysis. On the 27th April 2006, a poultry worker presented with conjunctivitis, and testing within 4 h identified influenza A in all specimens (eyes, nose and throat). This was subsequently confirmed as H7N3. Samples from a further 17 suspect human cases, all displaying conjunctivitis or other respiratory symptoms, were screened using the quadriplex assay and found to be negative for influenza A. Despite only one human case in this outbreak, the outbreak proved extremely informative, and highlighted strengths and minor weaknesses in the regional testing strategy, suggesting attention to certain aspects of logistics planning.
More recently, an outbreak of H5N1 in turkeys in a farm in Suffolk, England, at the beginning of February 2007 again triggered outbreak-response testing in two different regions of the UK. Of six cases presenting with influenza-like symptoms, none were found to contain the highly pathogenic H5N1 subtype, but three did test positive for influenza A, confirmed to be the circulating influenza A H3N2 virus. Fifty to sixty per cent of suspect H5N1 cases in travellers returning to the UK have been cases of circulating seasonal influenza A H1 or H3 or influenza B viruses, emphasizing the need to test for seasonal influenza in parallel with H5 testing (J. S. Ellis & M. C. Zambon, unpublished results), in the event of returning-traveller illness or outbreak investigations.
Network development and role of commercial assays
Commercial assays for diagnosis of H5 have not played a significant role in the last few years in the provision of H5 diagnostic capability in the UK. This has partly been due to the difficulty for companies that prepare kits for human diagnosis in accessing strains and sequences from newly emerging H5 viruses to achieve CE marking required on kits for human clinical use. Several companies have begun to find ways around these difficulties, often by working in partnerships with laboratories in countries in South-East Asia affected by H5N1. It is likely that the next few years will see the availability of different commercial kits for use in diverse laboratory settings. One role of the National Influenza H5 Laboratory Network will be to evaluate such kits and determine their suitability for different laboratory environments. Several kits are available that are based on the technique of isothermal amplification of viral RNA. For example, a rapid and sensitive CE-marked test, based on a novel amplification method (Imai et al., 2006, 2007), is now available (http://loopamp.eiken.co.jp/e). While nucleic acid extraction is needed for this test, there is no necessity for a real-time PCR instrument or thermal cycler since amplification is isothermal (63 °C) and amplification can be visualized by eye, or using a UV lamp. Such a test would be extremely useful in a simple front-line clinical laboratory lacking complex equipment close to an outbreak/incident site. Alternatively, an example of a more complex methodology is a recent CE-marked multiplex diagnostic test that permits the detection of 16 different respiratory viruses in a single test, including avian H5 influenza, seasonal influenza H1 or H3 viruses and SARS virus (http://www.idtag.tmbioscience.com), and requires moderately complex equipment. Such tests have the potential to significantly streamline molecular diagnostics across microbiology disciplines, but may not be suitable for front-line laboratories with simple equipment and few staff.
Conclusion
The creation of a national laboratory network for influenza H5 diagnosis ensures a rapid testing response to suspect H5N1 clinical cases can be mounted anywhere in the UK and Republic of Ireland. Ongoing influenza A surveillance using molecular methods within the network provides enhanced capability within the UK for detection of newly emerging influenza strains in humans. Assessing the suitability of commercial tests and matching them to appropriate laboratory environments is part of national and regional pandemic preparedness, and an important role for an influenza laboratory network, which also needs to adapt and evolve in response to the formidable challenge offered by the virus it is attempting to track. The supporting structures within the network provide a generic capability and capacity to permit the network to adapt quickly to any new emerging infectious pathogen, and provide necessary national molecular diagnostic capability in the absence of commercially available reagents.
ACKNOWLEDGEMENTS
The authors would like to thank all participants in the UK National Influenza H5 Laboratory Network. National SOPs are available upon request from the authors.
REFERENCES
Chen, W., He, B., Li, C., Zhang, X., Wu, W., Yin, X., Fan, B., Fan, X. & Wang, J. (2007). Real-time RT-PCR for H5N1 avian influenza A virus detection. J Med Microbiol 56, 603–607.
Coyle, P. V., Ong, G. M., O'Neill, H. J., McCaughey, C., De Ornellas, D., Mitchell, F., Mitchell, S. J., Feeney, S. A., Wyatt, D. E. & other authors (2004). A touchdown nucleic acid amplification protocol as an alternative to culture backup for immunofluorescence in the routine diagnosis of acute viral respiratory tract infections. BMC Microbiol 4, 41[CrossRef][Medline]
Das, A., Spackman, E., Senne, D., Pedersen, J. & Suarez, D. L. (2006). Development of an internal positive control for rapid diagnosis of avian influenza virus infections by real-time reverse transcription-PCR with lyophilized reagents. J Clin Microbiol 44, 3065–3073.
de Jong, M. D., Simmons, C. P., Thanh, T. T., Hien, V. M., Smith, G. J., Chau, T. N., Hoang, D. M., Van Vinh Chau, N., Khanh, T. H. & other authors (2006). Fatal outcome of human influenza A (H5N1) is associated with high viral load and hypercytokinemia. Nat Med 12, 1203–1207.[CrossRef][Medline]
Di Trani, L., Bedini, B., Donatelli, I., Campitelli, L., Chiappini, B., De Marco, M. A., Delogu, M., Buonavoglia, C. & Vaccari, G. (2006). A sensitive one-step real-time PCR for detection of avian influenza viruses using a MGB probe and an internal positive control. BMC Infect Dis 6, 87[CrossRef][Medline]
Ellis, J. S., Smith, J. D., Braham, S., Lock, M., Barlow, K. & Zambon, M. C. (2007). Design and validation of an H5 TaqMan real-time one-step reverse transcription-PCR and confirmatory assays for detection and verification of influenza A virus H5 infections in humans. J Clin Microbiol 45, 1535–1543.
Ferguson, N. M., Cummings, D. A., Fraser, C., Cajka, J. C., Cooley, P. C. & Burke, D. J. (2006). Strategies for mitigating an influenza pandemic. Nature 442, 448–452.[CrossRef][Medline]
Gunson, R. N., Collins, T. C. & Carman, W. F. (2005). Real-time RT-PCR detection of 12 respiratory viral infections in four triplex reactions. J Clin Virol 33, 341–344.[CrossRef][Medline]
Imai, M., Ninomiya, A., Minekawa, H., Notomi, T., Ishizaki, T., Tashiro, M. & Odagiri, T. (2006). Development of H5-RT-LAMP (loop-mediated isothermal amplification) system for rapid diagnosis of H5 avian influenza virus infection. Vaccine 24, 6679–6682.[CrossRef][Medline]
Imai, M., Ninomiya, A., Minekawa, H., Notomi, T., Ishizaki, T., Van Tu, P., Tien, N. T., Tashiro, M. & Odagiri, T. (2007). Rapid diagnosis of H5N1 avian influenza virus infection by newly developed influenza H5 hemagglutinin gene-specific loop-mediated isothermal amplification method. J Virol Methods 141, 173–180.[CrossRef][Medline]
Moore, C., Hibbitts, S., Owen, N., Corden, S. A., Harrison, G., Fox, J., Gelder, C. & Westmoreland, D. (2004). Development and evaluation of a real-time nucleic acid sequence based amplification assay for rapid detection of influenza A. J Med Virol 74, 619–628.[CrossRef][Medline]
Ng, E. K., Cheng, P. K., Ng, A. Y., Hoang, T. L. & Lim, W. W. (2005). Influenza A H5N1 detection. Emerg Infect Dis 11, 1303–1305.[Medline]
Payungporn, S., Chutinimitkul, S., Chaisingh, A., Damrongwantanapokin, S., Buranathai, C., Amonsin, A., Theamboonlers, A. & Poovorawan, Y. (2006). Single step multiplex real-time RT-PCR for H5N1 influenza A virus detection. J Virol Methods 131, 143–147.[CrossRef][Medline]
Templeton, K. E., Scheltinga, S. A., Beersma, M. F., Kroes, A. C. & Claas, E. C. (2004). Rapid and sensitive method using multiplex real-time PCR for diagnosis of infections by influenza A and influenza B viruses, respiratory syncytial virus, and parainfluenza viruses 1, 2, 3, and 4. J Clin Microbiol 42, 1564–1569.
This article has been cited by other articles:
|
|
 |
|
  J. Duitama, D. M. Kumar, E. Hemphill, M. Khan, I. I. Mandoiu, and C. E. Nelson PrimerHunter: a primer design tool for PCR-based virus subtype identification Nucleic Acids Res., May 1, 2009; 37(8): 2483 - 2492. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
