Evaluation of a rabies ELISA as an alternative method to seroneutralisation tests in the context of international trade of domestic carnivores

For several years, international movements with pets have greatly increased. Most countries have relaxed their quarantine measures and adopted a scheme combining vaccination of pets against rabies followed by a serological test to check the efficacy of vaccination. This new scheme has been strongly supported by the OIE, WHO and the European Commission to facilitate the free movement of people and pets around the world. Currently, only two reference methods are recognised and prescribed (the FAVN test and the RFFIT) to measure rabies antibody levels in serum samples for international trade. They are reliable and valuable methods of assessing the efficacy of rabies vaccination but they are time-consuming and require well-trained people and specialised laboratory facilities. A few years ago, an ELISA (Platelia™ Rabies II kit ad usum Veterinarium) was developed for domestic carnivores and wildlife. To our knowledge, this ELISA is the only one certified and prescribed by the OIE. Following its marketing, one task of the EURL for rabies serology was to evaluate the performance of laboratories using this new kit. The results revealed that 26% of the participants, which were already approved laboratories for rabies serology, failed the inter-laboratory trial. Such unsatisfactory results have never been observed during any of the previous proficiency tests organised annually since 2000 by the EURL for rabies serology using reference methods. More investigations were undertaken through internal and collaborative studies to assess the performance of this newly marketed ELISA kit. The results of the internal study revealed that even with a specificity of 100%, the sensitivity evaluated on 593 samples of domestic carnivores came to 78.2%. An issue regarding the underestimation of serum titres was also revealed during the study. The results of a collaborative study involving 23 international laboratories reinforced the preliminary conclusions regarding lack of sensitivity. Indeed, only 5 laboratories out of the 23 obtained satisfactory results. We therefore suggest adopting a threshold of 0.3 EU/mL instead of 0.5 EU/mL to increase the sensitivity of the test.