The Crux of Ebola Diagnostics

(See the brief report by Kainulainen et al, on pages 1380–5.)

Although there are no approved therapies for Ebola virus disease (EVD), early and reliable diagnosis is paramount to managing cases so that isolation, contact tracing, and infection control procedures can be implemented in a timely fashion. Because of the nonpathognomonic clinical presentation of EVD, any diagnostic uncertainty or delay may expose individuals who are not infected, with devastating consequences [1]. In addition, delays in initial recognition of the outbreak may delay appropriate local, national, and international responses.

Obviously, nonlaboratory parameters such as epidemiological risk as well as the presence or absence of EVD symptoms are key factors in the evaluation and management of suspect cases [1–3]. To this end, individuals who have had possible exposure to ebolaviruses but manifest no symptoms may not warrant testing, as none of the currently available diagnostic tests have demonstrated the sensitivity to reliably detect ebolaviruses prior to the onset of symptoms. Furthermore, such asymptomatic patients may not need to be strictly isolated as they pose a minimal risk of transmission. However, such exposed patients should be closely monitored, and isolated and tested if they were to become symptomatic. These risk stratification measures allow for more rapid case identification, help prioritize diagnostic testing, and lessen further transmission. We recognize that risk stratifications of EVD patient management may also be applied differently in distinct countries and under certain circumstances according to specific national contingency plans.

Although the techniques utilized for laboratory testing are not vastly different for other pathogens, diagnostic testing for ebolaviruses and other such high-consequence pathogens has been limited to laboratories with highly trained staff and enhanced biosafety and biosecurity requirements [1, 2, 4]. These issues are still pertinent during outbreaks, as typically initial diagnosis is performed in international high-containment reference laboratories, as exemplified by the recent West African outbreak [5]. Once the outbreak has been identified and established, diagnostics can be handled by mobile laboratories deployed into remote regions, such as in sub-Saharan Africa, with limited but rapid and targeted diagnostic approaches such as real-time reverse-transcription polymerase chain reaction (PCR) or other, largely molecular detection methods [6, 7].

Currently, ebolavirus-specific testing can be achieved in several ways: (1) virus isolation and characterization; (2) antigen detection for viral proteins; (3) serologic testing for virus-specific antibodies (immunoglobulin M, immunoglobulin G, and rarely neutralizing antibodies); and (4) molecular detection for viral genome sequences [2, 4, 6] (Table 1). Virus isolation and subsequent characterization (ie, by immunofluorescence or electron microscopy) has historically been the gold standard of testing, but is rapidly being challenged by PCR assays, other nucleic amplification strategies (eg, loop-mediated isothermal amplification), and next-generation sequencing. Today these molecular detection methods may be considered the new gold standard, especially after the West African Ebola outbreak, because of their wider application, higher sensitivity and specificity, rapid performance, and ability for field deployment [6, 7].

Table 1.

Established Diagnostic Approaches

Approach	Platform	Advantages	Disadvantages	Comments
Antigen detection	Antigen capture ELISA	Rapid; fieldable; potential for bedside testing	Lower sensitivity	Confirmatory test
Serology	Direct (IgG) or antibody-capture (IgM) ELISA	Convalescence; helpful in retrospective and ecologic analyses	Lower sensitivity; limitations in specificity; delayed response	Confirmatory test; serosurveys
Nucleic acid detection	Real time RT-PCR, loop-mediated isothermal amplification	Rapid; sensitive; fieldable	Potential for contamination; potential for false-negatives due to genome mutations	New gold standard
Isolation	Cell culture, animal	Virus isolate for characterization	Slow; containment required	Old gold standard
Particle structure	Electron microscopy	Structure	Slow; expensive; potential for false-positives	Rarely used

Abbreviations: ELISA, enzyme-linked immunosorbent assay; IgG, immunoglobulin G; IgM, immunoglobulin M.

Equally important is a standardized laboratory algorithm that should be applied [4, 7]. The algorithm likely differs in a high-containment reference laboratory from that employed in a field laboratory or any interim diagnostic arrangement [4, 6, 7]. Testing individual specimens with a single approach and the right algorithm is probably acceptable in an outbreak setting with an established causative agent, even though not optimal, but certainly not adequate for initial diagnosis, which should include confirmatory testing by an independent approach. General diagnostic principles should be considered such as multiple target molecular detection to avoid false-negatives due to genetic changes or follow-up sampling for serological assays to demonstrate antibody class switch and titer rises. Reliable diagnosis is likely more important than rapid diagnosis because of the tremendous implications on public health measures. A good compromise is often necessary.

A major drawback to the current mainstream approaches to ebolavirus diagnostics (PCR, serology, antigen detection) is that they fail to determine virus infectivity. In this issue of The Journal of Infectious Diseases, Kainulainen and colleagues [8] have partially addressed this dilemma with a novel reporter cell line capable of detecting live ebolavirus replication. A minigenome expression cassette was introduced into Vero E6 cells, the standard cell line for ebolavirus isolation. Upon infection, the ebolavirus replication complex recognizes the promoter on the minigenome leading to expression of a reporter gene, here zsGreen (ZSG, Clontech), that can be easily detected by its fluorescence. Interestingly, despite being based on Ebola virus Makona, the causative agent of the West African Ebola outbreak, the minigenome promoter is detected by members of all other known ebolavirus species (Zaire, Sudan, Bundibugyo, Reston, and Tai Forest ebolaviruses), but not by related members of the genus Marburgvirus. Even though this assay has not eliminated the necessity of high biocontainment, it is simpler, faster, and more easily scalable than usual isolation in tissue culture. Therefore, this is a very welcome improvement for diagnostics, with important public health implications.

The recent West African Ebola outbreak has established molecular diagnostics as a key component of outbreak control [6, 7, 9, 10]. Bedside testing based on antigen detection is being developed for faster diagnosis and higher safety, as these tests would eliminate or at least limit specimen transport [11, 12]. These efforts address the need for immediate outbreak response and patient triage, but do not answer the question of virus infectivity. This is particularly pertinent for patient discharge decisions, in convalescing patients or with nontraditional specimens (including semen, urine, breast milk), as virus components, such as viral proteins or nucleic acid, may be detectable by sensitive methods and yet the patients may not be infectious. For example, immune-complexed virus from convalescing patients may not be viable or infectious. Semen, urine, and breast milk may also contain immunologic or other, yet unidentified factors that lower or abrogate infectivity. Currently, only culture-based techniques will definitively demonstrate virus viability and help to define the transmission risk, an important public health control parameter. This is particularly important for convalescent patients who have shown to be persistently infected with ebolavirus for several months with potential for sexual transmission [13, 14].

Unfortunately, current assays to determine infectivity prevent rapid diagnosis and screening of larger sample numbers, as they require high containment conditions and will not lend themselves to outbreak situations such as those in sub-Saharan Africa. The novel assay developed by Kainulainen et al [8] addresses some of those drawbacks, but for biosafety reasons cannot complement the molecular assays in field conditions. One could perhaps foresee that in-country or regional reference laboratories may one day allow for limited virus replication under safe conditions and, thus, those laboratories may be able to safely accommodate this or other new assays in the future to address issues of infectivity for certain outbreak response activities.

In general, initial diagnosis of ebolavirus infections is still made in reference laboratories largely located in developed countries [4]. This was not any different with the recent West African epidemic [5]. Normally, the algorithm includes confirmation by independent assays based on distinct principles, one of which being virus infectivity testing [4, 7]. Therefore, the new assay by Kainulainen et al [8] would be helpful in replacing or complementing virus isolation. It likely could also be adapted to serve as a new and faster platform to test for ebolavirus neutralization and serve for better testing of non-blood samples.

A key problem remains early detection, which starts with clinical awareness, recognition of suspected cases, and the subsequent confirmation through laboratory diagnosis—a sequence of events that has caused the largest delays in the past. Once an outbreak is established, early mobilization of on-site laboratory response is paramount but has been slow, even though effective. Therefore, more effective strategies are needed for clinical education, assay development, and effective deployment [15]. Despite having powerful diagnostics and algorithms in place, the high-containment diagnostic laboratory world is still in search for the ideal ebolavirus test: one that in the field can safely, rapidly, and inexpensively detect minute quantities of only viable virus at the bedside.

Notes

Financial support. Work on filoviruses for J. E. S is supported by the Canadian Institutes of Health Research (CIHR) and the Public Health Agency of Canada. Work on filoviruses for H. F. is supported by the Intramural Research Program, National Institute of Allergy and Infectious Diseases, National Institutes of Health.

Potential conflicts of interest. All authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.