External quality assessment (EQA) scheme for antiviral susceptibility detection in influenza viruses for the Community Network of Reference Laboratories for Human Influenza in Europe (CNRL) 2010/2011

The first European external quality assessment (EQA) exercise for antiviral susceptibility detection in influenza viruses was carried out during winter 2010/11. The objectives of the exercise were to a) offer participants an independent mechanism to check performance, and b) provide information on performance of antiviral susceptibility testing at the network level. Twenty different laboratories from 16 European countries (see Annex 1) participated in the exercise. Each participant received a panel of ten coded samples, including recent influenza A and B viruses containing substitutions known to confer resistance to antiviral drugs. Participants tested the viruses using the antiviral susceptibility testing methodology currently used in their laboratories. Although there was good participation from CNRL member laboratories (~60%), many laboratories chose not to participate, suggesting there is further potential for development within the network. All twenty participating laboratories returned good results from the genotypic detection of the H275Y mutation in influenza A(H1N1)pdm09 virus (90% correctly reported consensus result), demonstrating that these laboratories have rapidly and successfully implemented the new assays since the emergence of the pandemic virus in April 2009. The proficiency for detection of a mixture of resistant and sensitive A(H1N1)pdm09 viruses was lower, although the majority of laboratories (79%) did identify resistance in this sample. Analysis suggested that sequencing methods may have some limitations regarding mixture identification. Fewer laboratories (55–75%) performed genotypic tests for other influenza (sub)types. The priority for expansion of these assays within the network would depend on the risk of resistance emergence. Fifteen laboratories returned good results for amantadine resistance detection. The widespread resistance and limited use of these drugs suggests this is a lower priority for capacity development. Twelve laboratories reported phenotypic testing results, while eight laboratories did not, possibly due to limited resources or lack of technical support for this test. Areas of difficulty included influenza A(H1N1)pdm09 mixture analysis, analysis of NAI-resistant H3N2 and influenza B virus. Variation between genotypic and phenotypic susceptibility testing was observed, suggesting that results achieved with different assays are not easily compared. Participants were asked to provide an interpretation of variations in the results for genotypic and phenotypic testing in terms of virus susceptibility to antiviral drugs. This was a challenging aspect of the exercise, leading to different interpretations for similar results. The interpretations of genotypic test results varied widely (20–95% correct match to consensus), compared with the interpretations of phenotypic results, which were more consistent but still showed some variation (67–100% correct match to consensus). As there is no widely accepted definition of resistance or reduced susceptibility, results are often subject to individual interpretation. Furthermore, different countries use locally defined baseline values for phenotypic testing, which could lead to varying interpretations. One of the goals of the EQA was to gain insights into these variations, which could inform work towards harmonisation of interpretation of antiviral data and reporting of results. Provision of guidelines on the interpretation of resistance mutations and phenotyping results would help to improve consistency. Other strategies towards achieving consistency between laboratories include the harmonisation of baseline values and the provision of a reference set of viruses. A reference panel of influenza A and B viruses for assessment of resistance to neuraminidase inhibitors has been developed and is available from the isirv Antiviral Group. The technical challenges of implementing phenotype testing should be taken into account for future activities towards the development of the network, which may be best served by a supranational phenotype testing service. Future capacity building should focus on the development of widespread capability for genotypic and phenotypic detection of oseltamivir resistance in A(H1N1)pdm09 and other human influenza viruses. Although global oseltamivir resistance is currently at a low level, it is acknowledged that there is real potential for the emergence of oseltamivir resistance in A(H1N1)pdm09 viruses. The results of this exercise will be used to guide the development of the network and promote harmonisation of antiviral susceptibility testing, data interpretation and reporting through the implementation of recommendations for further improvement and additional training activities