With scientific intensity, EuroImmun launches the first specific serological test for Zika virus
Interview by Steve Halasey
It has been just about a year since the Pan American Health Organization reported that the first locally transmitted case of Zika virus had been identified in the Western Hemisphere. Since then, thousands of cases have been identified in more than 30 South American and Central American nations, leading the World Health Organization, at the beginning of February, to declare the situation a public health emergency of international concern. The declaration calls for a coordinated international response to control the mosquitos that transmit the virus, to improve surveillance, and to “expedite the development of diagnostic tests and vaccines to protect people at risk.”
One organization that didn’t wait for WHO’s starting gun is EuroImmun AG, Lübeck, Germany, a fiercely scientific €230 million company with a global leadership position in autoimmunity diagnostics, and a significant presence in infectious disease serology and allergy testing. As it happens, the company also has a specialty interest in tropical diseases, and it has been quietly working to develop a diagnostic test for Zika since 2014, when an earlier outbreak occurred in French Polynesia.
Amid some skepticism, by the middle of this February EuroImmun announced that it had developed an in vitro diagnostic for Zika-specific antibody detection in two formats, had received CE marking for both tests, and had been granted market entry by Brazilian authorities. To find out more about EuroImmun’s test for Zika virus, CLP recently spoke with Hamid Erfanian, CEO of EuroImmun US Inc, Mountain Lakes, NJ.
CLP: The Zika virus has been known since 1947. When did EuroImmun recognize there was an urgent need and begin working to develop a diagnostic test for Zika disease?
Hamid Erfanian: We have two different technologies for Zika testing. One of them uses indirect immunofluorescence analysis (IFA), which we categorize as an indirect immunofluorescence technology (IIFT). The other is an enzyme-linked immunosorbent assay (ELISA).
EuroImmun began development of the IIFT for Zika in the spring of 2014, in response to an outbreak in French Polynesia and corresponding reports from infected travelers returning to Europe from Southeast Asia. The first prototypes of the IIFT were sent out for validation studies in 2015. On February 3, 2016, Brazilian authorities approved this version of the test for use in their country.
Meanwhile, during the summer of 2015, we began production of the Zika-specific recombinant antigen used in the ELISA version of the test. In September 2015, we accelerated our efforts, because we had begun to receive requests from Brazilian authorities in respect to their growing outbreak. After the first cases in Brazil were reported, we recognized that assays to differentiate dengue and Zika infections would be urgently needed, so we responded to this demand as quickly as possible. We released the ELISA kit for evaluation by a variety of organizations in January of this year, and received CE marking for the test just before the middle of February.
CLP: How did you obtain Zika-positive specimens to begin the development process?
Erfanian: We worked with the WHO Collaborating Center for Arbovirus and Hemorrhagic Fever Reference and Research (WHOCC), in Hamburg, Germany, which already had some well-characterized samples from patients with Zika virus infection. The center had been receiving samples from Zika-infected patients for some time. Later, after the development process had been completed, the center also performed validation of the IIFT and ELISA tests using characterized patient samples.
CLP: What challenges did you encounter while developing the test?
Erfanian: First, for the IIFT version, a suitable Zika virus isolate had to be obtained and cultivated in our own biosafety level 2 laboratory. The infected cells were then used for the production of microscope slides, and we then had to find characterized patient samples in order to validate the slides.
For the ELISA version, the biggest challenge we encountered was the potential for crossreactivity. The main problem in the serological diagnosis of flavivirus infections is the high similarity of the different flaviviruses. This is especially relevant for the dengue and Zika viruses, which are transmitted by the same vector and display nearly the same clinical symptoms. It was also considered essential that there be no crossreactivity with antibodies to the yellow fever virus—also a flavivirus—which develop after a vaccination that is often given to travelers.
In this context, our aim of developing a test system with at least 90% diagnostic specificity for anti-Zika antibodies was extremely ambitious. EuroImmun has more than 20 years of experience in developing robust ELISA kits and assays, and this background played a valuable role in enabling our team to identify a highly specific recombinant viral protein from Zika virus as the basis for this automatable kit. In the end, we outperformed our own goal, reaching a specificity of more than 99% with the antigen. Isolating this antigen helped us to overcome the challenge of crossreactivity.
CLP: Can you describe how the two tests deal with the potential for crossreactivity with other flaviviruses?
Erfanian: EuroImmun’s Arboviral Fever Mosaic 2 consists of six substrates of cells infected with chikungunya virus, dengue virus serotypes 1 to 4, and Zika virus. One drop of the patient sample can be used for testing via EuroImmun’s proprietary biochip technology, as all six different virus substrates are placed in a single slide well.
The mosaic can help in differential diagnostic delimitation of infections with any of the three viruses. Because the test uses whole virus particles, crossreactivity among antibodies against different flaviviruses has to be taken into account. While there is generally no or only low-grade crossreactivity in a primary flavivirus infection, in a secondary flavivirus infection (for example, a Zika infection following a dengue infection or a yellow fever vaccination), high-grade crossreactivity is typical. Such crossreactivity is stronger for IgG than for IgM antibodies. Investigating serial dilutions of the patient sample may enable determination of a dominant end-point titer for the causative virus.
For the ELISA format test, we have proven with more than 200 serum samples that the assays are more than 99% specific for anti-Zika virus antibodies. There is no crossreactivity with other flavivirus antibodies. These findings have been confirmed by several customers that have used the EuroImmun assays to test anti-dengue virus positive samples and samples from vaccinated people. The results are always convincing: the use of our highly specific antigen avoids crossreactivity with other flaviviruses.
CLP: What are the advantages of the two formats you selected for the Zika test?
Erfanian: In the IIFT, the use of virus-infected cells enables very sensitive detection of antibodies since all viral antigens are presented in their native conformation. Our unique biochip technique enables us to combine the Zika substrate with substrates for other viruses that occur in the same geographic area and cause very similar clinical symptoms, such as chikungunya and dengue in Brazil.
Microtiter plate ELISAs are considered the most reliable, convenient, and reasonably priced technology for performing infectious serology. When an open-channel automated ELISA processing instrument is used, the ELISA format enables customers to load large numbers of patient samples at the same time and perform the tests fully automated. Such automated instruments for ELISA testing are available in nearly all routine and infectious disease labs around the world, so this technology can be used in labs in affected countries such as Brazil, as well as in the United States and Europe for diagnosing travelers coming back from endemic areas.
To perform differential diagnoses, the test can easily be combined with well-established ELISAs for dengue virus NS1, and for chikungunya and dengue virus antibodies. Another advantage of ELISA kits is that customers with small-volume samples do not have to rely on automation, but can run the assays manually at their laboratories.
CLP: EuroImmun has a great deal of experience in antigen production. Do you expect to continue producing all of the antigen necessary for the Zika test, or will you license-out some of that production?
Erfanian: We currently produce all of our own Zika antigen in-house, and our goal is to continue doing so. We’ve directed a lot of our efforts toward making sure that we can produce an ample supply.
So, the antigen is ours, we manufacture it ourselves, and we have plans that will enable us to increase production of the antigen if necessary. Are we going to license it? At this stage, we haven’t considered that option.
CLP: Is there a preference in the marketplace for either the IIFT test or the ELISA test? How do you foresee the balance of your production shaping to meet labs’ actual use of the tests?
Erfanian: We are shipping both the IIFT and the ELISA tests. Of the two, IIFT is used more frequently in reference centers for tropical diseases, where the researchers are experienced in reading IIFT based on virus-infected cells. The ELISA would seem to have the greater appeal to routine clinical labs due to its high specificity and convenient, automatable format. It can be performed using automated instruments that are already widely available in the marketplace.
The IIFT format is a EuroImmun technology that can be described as a biochip mosaic. The mosaics contain antigens for chikungunya, dengue serotypes 1 through 4, and Zika, each distinctively arrayed inside a single slide well. The majority of customers follow our published testing algorithm that recommends testing for chikungunya, dengue, and Zika virus together—that is, from the same patient specimen.
Our published algorithm provides labs with a sequence that they can follow in order to perform Zika testing. But not every laboratory has the ability to carry out the entire sequence. For that reason, it seems logical that most routine labs would choose to use the easier ELISA test, either manually or by way of an automated platform.
CLP: So far as the two test formats are concerned, is there a difference in use patterns between public health labs and hospital laboratories?
Erfanian: It’s hard to say. A number of public health labs are currently looking at the IIFT and have requested samples in order to conduct evaluations of the test in that format. But IIFT results need to be evaluated by an experienced technician, which makes it harder for labs to perform.
But a lot also depends on what technologies a lab has available to it, and how quickly and efficiently the lab can obtain test results without requiring too much expertise. Currently, the demand that we are getting is mainly for our ELISA test. This format of the test is highly sensitive and highly specific, and laboratorians have come to rely on ELISA tests. From my perspective, it seems logical that most labs would seek to adopt the ELISA test.
CLP: Payors sometimes resist the adoption of a panel test such as EuroImmun’s IIFT that combines chikungunya, dengue, and Zika. Do you expect to encounter questions along those lines as the test comes into use?
Erfanian: I’m no expert in reimbursement, so consider this merely an informed opinion. At present, tests for chikungunya and dengue in either the IIFT or ELISA format are already being reimbursed. There is no reimbursement for Zika testing in either format. But if a Zika test were to be cleared by FDA, reimbursement would probably be made available for that test.
Because we are in the midst of a Zika epidemic, however, most labs, including state labs and public health authorities, are more concerned with being able to offer an accurate, validated test than they are with how reimbursement for the test is going to work out. The best Zika test that can be provided to the marketplace is one that is both accurate and accommodating to the testing algorithm put forward by CDC.
To overcome payor resistance to paying for a mosaic test, we could provide just the Zika virus substrate in a single test. Doing so would not create any technical problems on our end. However, we are trying to be a little cautious with respect to CDC’s recommendations, and that’s why we decided to provide a mosaic for all three viruses.
But in respect to reimbursement, I just don’t think that’s something that laboratories are at all concerned about at this stage. They just want to have the ability to offer Zika testing.
CLP: Are there any special requirements or limitations on the types of immunoassay instrumentation that labs will need to perform the ELISA-format test?
Erfanian: Not at all. The ELISA can be performed manually or with any open-channel automated ELISA system. EuroImmun offers two automated analyzers suitable for labs of different sizes, and these instruments have everything onboard to perform all the necessary pipetting, incubation, and analytical steps, and to send results to a laboratory information system.
Roughly 97% of EuroImmun’s ELISA tests are standardized in such a way that programming for all assays can be easily accommodated for any of these open-channel automated ELISA systems. The ELISA kits contain interchangeable secondary reagents and buffers, which facilitates parallel automated processing of different parameters.
The IIFT version can also be automated. EuroImmun offers a variety of automation solutions for IIFT.
CLP: How was the performance of the test evaluated, and what were the performance outcomes?
Erfanian: We ran more than 1500 samples to evaluate the sensitivity and specificity of the assays. Our studies included a panel of 29 Zika patient samples from WHOCC, more than 200 samples with different anti-flavivirus antibodies, samples with different potentially interfering factors, and several prevalence studies. Overall, these studies showed that the sensitivity of the ELISA was 97%, with specificity greater than 99%.
For the IIFT version, antibody detection is based on cells infected with the corresponding virus, which provide highly sensitive diagnostics. In clinically characterized samples, the IIFT substrates yielded sensitivities between 96% and 99% and specificities of 95% to 100% for the parameters chikungunya, dengue, and Zika. However, the Zika assay showed significant crossreactivity with samples containing IgG antibodies directed against the dengue, tick-borne encephalitis, West Nile, and yellow fever viruses. The level of crossreactivity with IgM-positive samples was lower.
CLP: What was the role of WHOCC in evaluating the tests?
Erfanian: The center was involved in evaluating the sensitivity and confirming the specificity of the tests. Its researchers ran tests on a large cohort of samples containing chikungunya, dengue, and West Nile virus antibodies. Their tests confirmed that the ELISA showed no crossreactivity, especially with the dengue and West Nile viruses.
CLP: Have you also been working with Brazilian authorities to evaluate the tests?
Erfanian: The Brazilian Agência Nacional de Vigilância Sanitária (ANVISA), which is the body that approves testing across Brazil, granted permission on February 3 for our IIFT to be used in the country. It’s our understanding that they are now using the test.
Additionally, a shipment of ELISA kits was cleared through customs in the middle of February. We are working with ANVISA to help them review the ELISA test, and we hope that they will also clear this test format for use throughout Brazil.
CLP: The CE mark covers Europe and a number of other countries worldwide. What is the regulatory status of the tests in the United States and in countries that are part of the WHO-designated Zika outbreak area?
Erfanian: For the European market, both the IIFT and ELISA formats of the test have now been granted the CE mark. So both tests are available for purchase and use in diagnostic applications.
For the United States, the tests are currently available under research use only (RUO) labeling. They can be used for research purposes or subjected to a laboratory’s in-house validation to make them suitable for diagnostic purposes.
I don’t have details of the regulatory status of the tests in all the other countries affected by the Zika outbreak. We have submitted an application for emergency use assessment and listing in response to WHO’s February 10 invitation to manufacturers of in vitro diagnostics for Zika virus. And we, or our subsidiaries, are working with other authorities to obtain whatever regulatory authorization may be required.
Pretty much every lab that has seen our data is interested in receiving test kits. But we’re not circumventing any authority’s oversight to distribute the test. All we can do at this stage is to distribute the test where we have authorization to do so.
CLP: Have you heard from all of the 30-plus countries in the WHO-designated Zika outbreak area?
Erfanian: We have heard officially from every single one of them. In addition, we are receiving a lot of personal inquiries that we just can’t respond to. We have subsidiaries in 13 different countries, and we have distributors across 100 different countries. So unless an inquiry comes through them, or is from an official source or a customer with whom we’re familiar, we may not be able to respond directly.
CLP: How will the availability of these tests help public health authorities monitor and prevent Zika outbreaks?
Erfanian: There are a lot of unknowns surrounding this infection, and some of the suspected effects have already become sensationalized in the media, including birth defects, brain damage, sexual transmission, and so on. Even the possibility that the Zika virus can result in newborns with microcephaly, which we believe to be true based on what has already been published, is really an unconfirmed suspicion. At this point, we don’t know how much of what is suspected about Zika virus is absolutely true.
Such uncertainty raises alarm among residents and travelers to countries or regions where there are mosquitos that carry the Zika virus, including certain areas within the United States. Anyone who has been in those areas and is experiencing any of the symptoms that come along with Zika infection may have good reason for concern.
Testing with RT-PCR provides reliable identification of infections, but is only viable during the viremic phase of infection, which is quite short. For many patients, RT-PCR will already be negative by the time they consult a doctor.
This is the point at which EuroImmun’s serological tests will prove helpful for the diagnosis of acute Zika virus infection. Serological methods are effective from soon after clinical onset to beyond convalescence. Detection of specific IgM or a significant rise in the IgG titer in a pair of samples taken 7 to 10 days apart is evidence of an acute infection.
Our IIFT and ELISA tests can give individuals a near-term level of assurance by confirming the absence of IgM antibodies to the Zika virus. And down the road, when an individual is outside the period when seroconversion happens, the tests can still provide that assurance, by confirming the absence of IgG antibodies that would be indicative of a past infection.
In addition to their application in acute diagnostics, serological methods are also useful for studying the long-term consequences of infection. For example, serological investigations of Zika infection may help to establish whether the dramatic rise in cases of microcephaly and Guillain-Barré syndrome observed in Zika-affected regions are consequences of Zika virus infection. If the link to congenital malformations is confirmed, Zika virus serology could play an important future role in prenatal diagnostics. Pregnant women with serological evidence of an infection could be offered intense prenatal monitoring, while seronegative women may be spared unnecessary worry.
Serology is also useful for screening donated blood. Travelers returning to unaffected countries from regions where flavivirus infections are endemic are being advised by public health authorities to defer donation of blood for a few weeks or even months. After this time, serological testing can verify the safety of donated blood products.
A further, critical role for serological studies is to monitor the growing epidemiological reach of arboviruses. As the chikungunya, dengue, and Zika viruses are expected to continue spreading around the globe, knowledge about emerging endemic regions is valuable for providing effective patient diagnostics.
CLP: WHO has indicated that it knows of about 20 different companies that are working to develop a Zika test. Do you think that EuroImmun has a significant market lead?
Erfanian: I can’t speculate on what else may be coming onto the marketplace, or when, because I don’t have exact knowledge of what other companies are working on.
The specific recombinant antigen that is used in our tests is something special. Our ELISA has exceptional sensitivity and specificity, and I would be hard pressed to believe that someone else would be able to make a similarly powerful test very quickly. One doesn’t acquire expertise in making recombinant antigens and proteins overnight. Our company has more than 20 years’ experience in the field, and our scientists have created similar tests numerous times. Even CDC has published about some of our tests. So we feel pretty confident about our latest achievement.
Steve Halasey is chief editor of CLP.