Evaluation and validation of selected LAMPs and other rapid tests with emphasis on Avian Influenza and testing its suitability under field conditions
Insel Riems, 11 September, 2012
Friedrich‐Loeffler‐Institut,
Dr Sasan Fereidouni & Dr Anja Globig
This project aims at improving the diagnostic capacities of developing countries in animal disease detection. Two major goals will be targeted which are (i) the scientific validation of new diagnostic tools (“robust tests”) for sensitive and specific detection of influenza A viruses, in particular a promising novel technology, called LAMP, and (ii) enhancing international laboratory networking and capacity building of basic laboratories in Asia and Africa. Furthermore, the application and training capacities of techniques for test validation shall be enhanced.
The diagnosis of Influenza A underlies complex pathways which currently have major disadvantages for laboratories in developing countries.
The overall goal of this project is to improve diagnostic capacities in labs of developing countries. Currently, those labs are facing serious problems with the diagnosis of Influenza A viruses, and in particular, highly pathogenic (HP) H5N1 viruses:
(i) The OIE-recommended gold-standard (virus isolation by propagation in embryonated chicken eggs) takes long time and requires high biosafety standards.
(ii) Usually, molecular technology requires constant -20°C freezing conditions which are rarely fulfilled as power cuts often occur.
(iii) Material and equipment of molecular biological techniques (e.g. PCR, sequencing) are expensive and not affordable for the labs which often have a very limited budget. They also require RNA-extraction step in advance, which- by use of commercial kits – lead to further high costs.
(iv) Therefore, rapid tests (immunochromatographic assays) are frequently used as the only means of detection. The rapid tests were distributed widely by different stakeholders. However, they lack in sensitivity and are often expired when used.
Assuming this scenario for a high number of countries in the world (maybe the majority of all countries) under-detection of Influenza A is evident and thus contributes to the spread and/or endemicity of certain virus strains, e.g. HP H5N1. Limitations in the diagnosis of animal diseases further reduces reporting activities to the OIE Animal disease notification system and thus results in a lack of information related to a specific disease situation in a country.
Often, limited access to updates on modern molecular diagnostic methods leaves lab staff unaware of novel technology that may suit their laboratory capabilities. Capacity building and lab networking, also to encourage sending samples to advanced labs, is therefore essential to fill this gap.
In order to support diagnostic work in labs of developing countries it would be necessary to have diagnostic technology that fulfils the following:
- Robust
- Cheap and easy to obtain independently (without the support of donors)
- Easy to handle and interpret
- No need for sophisticated equipment
- No need for freezing conditions for long-term (1-2 years) storage
- High sensitivity, high specificity
We selected three different diagnostic approaches available on the market that seem to come closest to fulfill those conditions and tested their suitability for labs in developing countries:
(i) Loop-mediated isothermal amplification (LAMP), H5-specific RT-LAMP assay or Generic AI RT-LAMP.
(ii) Immunochromatography Antigen test (so-called rapid test).
(iii) Commercial lyophilized PAN AI RTqPCR kit targeting the M-gene.
(iv) In addition, we are looking for any possibility to simplify the laboratory working or reduce the costs of sample collection or processing.
Results :
(i) The evaluated H5-specific RT-LAMP assay produces results in short time. It has sensitivity comparable to RTqPCR while producing results more rapid. Further, the IAEA designed H5-specific RT-LAMP assay has the favourable advantage of being delivered in lyophilised format, which facilitates transport and storage for several months independent of a freezing chain. In general, this H5-specific RT-LAMP assay was found to be functional. However, the following challenges arose during our evaluation:
- The cut-off time discriminating positive and negative results is critical: By choosing different assay running times either sensitivity or specificity decreases. Therefore, interpretation of the results is difficult. The assay will need further modification to solve this problem.
- Our attempts to modify the RT-LAMP assay for result interpretation by the naked eye in failed due to the composition of enzymes and buffers required for the lyophilisation process. As such usage of the kit is restricted to laboratories which have access to sophisticated optical reading equipment (qPCR cycler or for example a Qiagen ESEQuant Tube Scanner). Although many labs also in developing countries have a qPCR cycler, they may not be well maintained (and thus not usable) or lab-staff is overwhelmed in using the challenging soft-ware. The easier-to-operate Qiagen ESEQuant Tube Scanner, on the other hand, is an expensive piece of equipment (not clear why it is so expensive) which can only be used for LAMP-assays and not for any other purpose.
Therefore, the H5-specific RTqLAMP assay is a robust tool but does not yet match all prerequisites mentioned above. The assay and/or visualization system/machine would need some modification and could then serve well as a molecular, robust and cheap tool for simple laboratories.
(ii) Our results of the evaluation of the two rapid tests based on immunochromatography confirm the poor sensitivity of those kits. Sensitivity and specificity of the kits tested are not at all sufficient for reliable diagnosis of AI infections and may only be applied on flock-basis upon HPAI-suspicion (see Marek et al., 2012).
(iii) The commercial PAN-AI RTqPCR kit in a lyophilized format is highly sensitive and specific, easy to handle and can be stored at room temperature for one year, but is expensive and needs a qPCR cycler.
(iv) We studied and compared the results of using a sample pooling policy for processing high numbers of samples. We found high efficiency of analysing sample pools of five samples by RT-qPCR. This pooling policy is beneficial with respect to saving hands-on time, consumables and, thus, costs during wild bird monitoring studies or poultry surveillance programs when negative AI results are expected for the majority of samples.
Reducing the reaction volume of influenza A virus generic as well as of subtype-specific RT-qPCRs to 12.5 µl (2.5 µl template) instead of 25 µl did not adversely affect the limit of detection of these RT-qPCRs. A significant economic benefit without impeding detection efficacy can be achieved when sample pools of five samples are analyzed by RT-qPCR using a reduction of reaction mix to the half of the original volume.
v) Another aim of the project is to enhance international laboratory networking and capacity building of basic laboratories in Asia and Africa for a possible future application of the targeted diagnostic tool/method/kit.
Therefore, a two weeks hands-on workshop (from November 07nd to 18th, 2011) on "Training on modern diagnostic tools for the diagnosis of Influenza A and enhancing networking between laboratories" featuring theoretical (8 sessions) and practical training (8 sessions) was carried out at the FLI. It was planned for three veterinary researchers from Animal Health or Research Institutes in Nigeria, Iran and Kazakhstan to participate in the workshop (Nigerian participant had to cancel the trip). Besides the program comprising novel technologies including LAMP, the participants had ample opportunity to establish contacts with researchers at the FLI and were able to visit various OIE Reference Laboratories.
It is envisaged to closely collaborate with IAEA for the modification or development of further LAMP assays which is currently pending. Work within the frame of the project will focus on research and finally validation of the newly developed assays.
(vi) We studied the influence of swab types (rayon versus nylon) on the sensitivity of AIV RNA detection and we couldn’t find a meaningful difference in this study. Considering the lower costs, rayon swabs seem to be a preferred choice for sample collection especially in the laboratories with limited resources. The nylon and rayon swabs yielded similar Ct values for immediate RNA extraction after swab collection or longer storage. Shaking of swabs in VTM slightly improved recovery rates for the rayon as compared to the nylon swab. Standard operating procedures (SOPs) should include a shaking step for at least one minute to ensure optimal sensitivity.