Excellent capability for molecular detection of Aedes-borne dengue, Zika, and chikungunya viruses but with a need for increased capacity for yellow fever and Japanese encephalitis viruses: an external quality assessment in 36 European laboratories

Arthropod-borne viruses exert an increasing pressure on health globally (1–7). The World Health Organization (WHO) estimates the yearly global burden for dengue virus (DENV) at 96 million clinical cases including 400,000 deaths while four billion people are at risk for infection around the world (8). It has been predicted that an additional 2.25 billion people (range 1.27–2.8) will be at risk for infection by 2080 (4).

In the past few decades, viruses transmitted by the mosquito species Aedes albopictus and Aedes aegypti have increasingly spread across the world causing significant outbreaks in naïve populations (9, 10). Such propensity for dissemination of Aedes-borne viruses was exemplified in 2013–2014 when chikungunya virus (CHIKV) and in 2015–2016 when Zika virus (ZIKV) entered and spread in the Americas (11, 12). Currently (2023–2024), a global upsurge in DENV infections with >5 million cases including ~5000 deaths across the world is being reported to WHO, including reports from the European region (13) .

Local transmission of originally (sub)tropical viruses like CHIKV, DENV, and ZIKV has been reported in Europe since 2007 from areas with established Ae. albopictus populations (14). Local transmission of DENV has been reported from Croatia, France, Italy, Portugal, and Spain, with two cases of dengue related to travel within Europe reported in 2020 (the Netherlands ex-France) (15–20). In 2023, a record number of 130 autochthonous DENV cases were reported to the European Centre for Disease Prevention and Control (ECDC), with cases as far north as Paris, France (16). Local transmission of CHIKV has been reported from France and Italy with outbreaks of hundreds of cases in Italy in 2007 and 2017 while a few local mosquito-borne ZIKV cases were reported by France in 2019 (21–24). While these three viruses have no established endemic transmission cycle in Europe, the risk of transient local transmission is directly linked to the risk of virus importation through viremic travelers returning from Ae. albopictus- and Ae. aegypti-infested areas (14, 25). With the increasing health burden of Aedes-borne virus circulation worldwide (6, 7), the risk for virus importation through returning travelers increases as well (26–29). It is predicted that ongoing climate change will make Northern European regions increasingly suitable for Ae. albopictus establishment, which parallels an expansion of at-risk areas for autochthonous transmission of human-to-human transmissible viruses vectored by this mosquito species (30). Ae. albopictus and Ae. aegypti are considered invasive mosquito species, and introductions (but no establishment yet) have been described as far north as Sweden for Ae. albopictus and the Netherlands for Ae. aegypti (14, 25).

Besides these three viruses, other Aedes-borne viruses with a potential significant health burden, like yellow fever virus (YFV), are sporadically introduced in Europe through returning travelers (29, 31, 32). While YFV epidemics elsewhere in the world can be linked to an increase of imported cases to Europe, these introductions have not yet resulted in autochthonous transmission in Ae. albopictus- or Ae. Aegypti-endemic areas.

Therefore, it is no wonder that Aedes-borne viruses that threaten human health are increasingly the focus of (inter) national preparedness, readiness, and response activities (33, 34). One of the main components of outbreak and pandemic preparedness and control is the ability to diagnose infections timely and accurately (35–37). For this and for the purpose of reliable surveillance for EU-notifiable arboviral diseases, laboratory functions in European Union and European Economic Area (EU/EEA) countries need an adequate capacity and capability to detect. Therefore, there is a need for periodic assessment of the capability of laboratories for detection of emerging arboviruses that represent a public health threat. To strengthen the molecular diagnostic capability in the EU/EEA and EU-enlargement countries for emerging arboviruses, an external quality assessment (EQA) through proficiency testing was performed among members of the ECDC-funded Emerging Viral Diseases Expert Laboratory Network (EVD-LabNet).

The EQA focused mainly on Aedes-borne viruses, i.e., CHIKV (alphavirus), DENV, ZIKV, and YFV (orthoflaviviruses). Also, Japanese encephalitis virus (JEV), an orthoflavivirus that is spread by mosquito species belonging to the genus Culex, was included in the EQA panel. Epidemic rise of JEV in regions with visitors returning to Europe or the return to Europe after extensive visits to rural areas with endemic viral presence incidentally gives rise to imported cases (38, 39). Although these cases are not expected to lead to local transmission as humans are dead-end hosts for JEV, JEV was included in the proficiency panel to assess orthoflavivirus test specificity and to improve JEV diagnostic capacity for health care and surveillance purposes. Here, we report the results of this EQA.

Thirty-six laboratories in 24 countries (20 EU/EEA, 4 EU-enlargement) participated in this EQA exercise focusing on the molecular detection of three arboviruses (DENV, CHIKV, and ZIKV) (potentially) emerging in Europe in Ae. albopictus-infested areas and two arboviruses (YFV, JEV) that are occasionally detected in travelers returning from endemic/epidemic regions worldwide.

Importantly, all laboratories reported the capacity for the three viruses that have proven potential for autochthonous transmission in specific regions in Europe; all participants correctly identified the ZIKV- and CHIKV-RNA-positive panel entries (100% score) while for DENV, only one sample of DENV4-RNA was missed by a laboratory (99.3% score). However, four laboratories identified one of the NC samples as DENV-RNA positive while one laboratory mis-identified the YFV-RNA sample as DENV-RNA positive. At this time (2024), (false) positivity in a suspected autochthonous case in Europe will lead to an extensive local public health response (40).

Molecular detection for CHIKV, DENV, and ZIKV was included in EVD-LabNet proficiency panels in the past. During the ZIKV Public Health Emergency of International Concern, only 40% (20 of 50) of the EVD-LabNet laboratories scored sufficiently, i.e., they had at least one test operational that scored all core samples correctly. Comparing the EQA results in 2016 and 2023 for panel entries with similar RNA loads, we observe an increase in percentage correct results for the ZIKV African lineage from 66% to 100% and for the Asian lineage from 79% to 100% (41).

CHIKV was also included in a 2022 EVD-LabNet proficiency panel targeting molecular detection of alphaviruses. The 2022 CHIKV Asian lineage panel entry with a similar RNA load as in the panel assessed here showed similar results with only one single false-negative result in 2022 versus the 100% score here.

A network EQA in 2003 showed 73%–100% capability to detect DENV1-4; however, only 15 laboratories participated versus 36 laboratories now, representing a significant increase in diagnostic capacity in the EU/EEA. The 2003 EQA showed a profound lack of sensitivity for DENV detection among the participants (42). As sensitivity was not assessed in the current EQA, it is recommended to include a DENV1-4 sensitivity assessment in near-future proficiency tests as part of the work of the soon-to-be-established EURL vector-borne viral diseases laboratory network (43).

For molecular detection of YFV-RNA, 76% of the participants with reported detection capacity reported having a species-specific RT-PCR. The capability among the 30 laboratories was high; only one missed the YFV-RNA-positive sample (a possible sensitivity issue, overall 97% laboratories correct) while another laboratory identified wrongly two other panel entries as YFV-RNA positive (a possible specificity issue). This success rate is a considerable capability improvement in comparison to the previous YFV-dedicated network EQA in 2018, where only 71% of the network laboratories scored a panel entry with South America lineage YFV-RNA with a similar load correct (44).

The capability for molecular detection of JEV RNA was assessed for the first time within EVD-LabNet. The capacity for molecular detection of JEV was the lowest for all five target viruses with 23 of 36 laboratories having the capacity, of which 15 had a JEV-specific RT-PCR test. The success rate was 83% with four laboratories mislabeling the JEV panel entry: 2× it was called negative for arbovirus RNA and 2× it was called as containing YFV-RNA based on very high Ct-values. As the environmental conditions in the EU/EEA are considered suitable for JEV transmission upon introduction of the virus from JEV-endemic regions, laboratory preparedness and response planning for timely and reliable detection in animal reservoirs (birds and pigs) and dead-end hosts like humans and equids are necessary to contain emergence and possible establishment of the virus in bird/pig-mosquito cycles (45) (46). Seeing the importance of reliable and sufficient capacity for JEV detection for potential future JEV risk mitigation, the need for EQA support to the establishment of such capacity is evident. Therefore, besides encouragement of laboratories to implement JEV diagnostics, proficiency testing for molecular detection of JEV is recommended for the future including the range of different JEV genotypes and a sensitivity assessment.

The EQA described here showed once again the high variety in diagnostic tests that are used in EU/EEA laboratories for diagnosis of emerging viral infections and the importance of laboratory-developed tests (LDT) for diagnostic workflows for molecular detection of emerging viruses that are not routinely part of diagnostic portfolios (Table 4). Most laboratories relied on in-house assays with scientific literature as a pivotal source to implement these LDTs. The wide range of LDTs used in this EQA shows that the implementation of LDTs is of high importance for reliable testing in the workflows of the participating laboratories. In 2024, the in vitro diagnostic regulation (IVD-R) becomes effective in the EU/EEA (47). Seeing the high reliance of European expert laboratories on LDTs that performed with high reliability in this EQA, the effects of the implementation of the IVD-R on the reliability but also the timeliness of laboratory response operations should be evaluated and assessed very carefully. The observed absence of standardization for arbovirus detection, albeit with great reliability, potentially hampers consistent case finding and reporting for the EU-notifiable diseases caused by DENV, CHIKV, ZIKV, and YFV. To assess whether the molecular capabilities are indeed sufficient for consistent case finding, future EQAs should assess sensitivity aspects of testing more in-depth as RNA loads were quite high in the current proficiency panel and may not represent viral loads seen in clinical settings. Laboratories participating in EQAs are further encouraged to provide more detailed information on the diagnostic tests used to guide putative corrective actions needed and to increase the impact that proficiency testing can have on the quality of testing.

Finally, this EQA was limited to molecular detection of a range of mosquito-borne viruses. While molecular detection of the viral RNAs is a powerful tool in the establishment and confirmation of infection, diagnosis is often based on serology alone as viremia is typically short lived in these infections. Therefore, it is recommended to perform EQAs for detection of virus-specific antibodies in the future. However, the performance of such EQA might be hampered by the availability of sufficient, well-defined serum samples for the rare viruses like JEV and YFV and by legal constraints for distribution of such clinical materials.

We thank all the EQA participants: Institute of Public Health, Tirana, Albania; Medical University of Vienna, Center for Virology, Vienna, Austria; Institute of Tropical Medicine, Antwerp, Belgium; University Hospital for Infectious Diseases "Dr. Fran Mihaljević," Zagreb, Croatia; Cyprus Institute of Neurology and Genetics, Nicosia, Cyprus; Institute of Virology, Charité—Universitätsmedizin Berlin, Berlin, Germany; Bernhard Nocht Institute for Tropical Medicine, Hamburg, Germany; Friedrich-Loeffler-Institut, Germany; Aristotle University of Thessaloniki, Thessaloniki, Greece; Molecular virology Laboratory, Microbiology and Virology department, IRCCS Fondazione Policlinico San Matteo, Pavia, Italy; Istituto Superiore di Sanità, Rome, Italy; Scientific Department, Army Medical Center, Rome, Italy; Amedeo di Savoia Hospital—ASL Città di Torino, Torino, Italy; Padova University Hospital, Padova, Italy; U.O. Microbiology (CRREM), IRCCS Azienda Ospedaliero-Universitaria di Bologna, Bologna, Italy; Laboratoire National de Santé, Dudelange, Luxembourg; Institute of Public Health, Montenegro; National Institute for Public Health and the Environment—RIVM, Bilthoven, Netherlands; Institute of Virology, Vaccines and Sera “Torlak”, Belgrade, Serbia, Institute of Microbiology and Immunology, Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia; Laboratorio de Referencia de Virus de Andalucía; Servicio de Microbiología. Hospital Universitario Virgen de las Nieves, Spain; Centre National de Reference (CNR) Arbovirus, Marseille, France; CIBU Institut Pasteur, Paris, France; HUS Diagnostic Center, Virology and Immunology, Helsinki, Finland; National Public Health Center, Budapest, Hungary; National Institute of Health/Center for Vectors and Infectious Disease Research, Lisbon, Portugal; Instituto de Salud Carlos III, Madrid, Spain; Hellenic Pasteur Institute, Athens, Greece; Defense Institute for Biomedical Sciences, Italy; “Cantacuzino” National Medico-Military Institute for Research and Development, Bucharest, Romania; U.O. Microbiologia Laboratorio Unico—Centro Servizi—AUSL della Romagna, Rome, Italy; Erasmus Medical Centre, Department of Virology, Rotterdam, Netherlands; Oslo University Hospital, Department of Microbiology, Oslo, Norway; Molecular Diagnostics, Pathology Department, Mater Dei Hospital, Malta; Institute of Public Health of Republic of North Macedonia, Skopje, North Macedonia; UCD National Virus Reference Laboratory, Dublin, Ireland; Riga East University Hospital, Latvian Infectology Center, National Reference Laboratory, Riga, Latvia.

In addition, we thank the ECDC support staff, as well as all the staff members who provided laboratory, administrative, and other technical support at AMU and RIVM, especially Camille Placidi, Johan Reimerink, and Bas van der Veer.

This work was funded by the European Centre for Disease Prevention and Control (ECDC) under specific contract No 3 ECD.14393 ID 26009 implementing Framework contract No ECDC/2020/010. For providing access to the virus strains used, we thank the EVA-GLOBAL consortium (funded by the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 871029).

Conceptualization was done by the following: C.M.G., C.B.E.M.R., and R.N.C.; methodology was prepared by the following: L.P., L.D.P., R.M., C.B.E.M.R., and R.N.C.; validation was done by the following: L.P., C.B.E.M.R., and Y.L.; data analysis was done by the following: L.D.P. and R.M.; visualization was done by the following: L.D.P.; project administration was done by the following: L.D.P., C.B.E.M.R., and R.N.C.; supervision was done by the following: C.B.E.M.R. and R.N.C.; resources were acquired by the following: C.M.G., C.B.E.M.R., and R.N.C.; writing was done by the following: L.D.P., R.M., C.B.E.M.R., and R.N.C.; all authors reviewed the manuscript.