Mobile Point of Care Diagnostic Testing for Ebola Virus Disease in DRC

In the Ebola outbreak in Guinea up to 90% of patients tested negative but were rarely diagnosed for differentials and Malaria and other diseases went under-diagnosed, which most likely lead to more deaths from these infectious agents. Another important aspect is that, the fear of the patients of having Ebola, while they are suffering from other disease must be overcome by providing the correct diagnosis of the type of the pathogen timely. [Dunbar NK et al.].

The RPA assay is six times faster than conventional real-time PCR while yielding the same analytical sensitivity and specificity. In addition, the RPA format (equipment size, cold-chain independent reagents) makes rapid on-site testing feasible and affordable. We developed a mobile laboratory, which consists of a hard plastic glovebox a Diagnostics-in-a-Suitcase (DiaS), and a solar panel and power pack set. The disassembled glovebox is kept in a metal box (80 × 60 × 41 cm) with other necessary materials (disinfectant solution, extraction kits, filter tips, racks, vortex, heat block, autoclavable plastic bags and personal protective equipment (PPE). The total weight is 28 kg for the box and 16 kg for the DiaS. Sample inactivation and RNA extraction using the SE kit so far were done in the glovebox. This allowed handling of hazard group 4 samples. The RT-RPA assay was performed in the DiaS containing the ESEQuant TS2 device with integrated touchscreen to operate the device and display the results without the need of a laptop, which is not easy to disinfect (Figure 1).The mobile lab was successfully deployed in Conakry when an increasing number of cases, due mainly to non-reporting, reticence, and infection transmission chains during burials developed towards the end of 2014. To provide quick results for the safe and dignified burial (SDB) programme two mobile laboratories were set up in the Matoto Gbessia Port II district. Altogether 928 post-mortem swab samples were tested. Overall, 120 samples (12.9%) scored positive. Positive results were obtained within 30-40 min and reported directly to the SDB teams [Faye O et al.].

APPROACH

Objective:

To assess the performances and operational characteristics of the differential RPA we will use i) a panel of well characterized EBOV sera from samples collected in Guinea and DRC ii) sera from EVD suspect cases, and iii) sera from routine diagnostics at IPD and NIBR.

Study population:

The study will be carried out on material from EVD suspected persons, archived and prospective EBOV negative samples and selected routine diagnostic samples.

Inclusion and exclusion criteria: All subjects (male, female, adults and children) meeting the WHO case definition of "suspected" and "probable" EVD case eligible for real time RT-PCR assay, according to the currently used case definition in DRC. For the prospective evaluation, oral informed consent from the patient/legal guardian will be requested for eligibility to participate in the study and exclusion criteria will be inability to provide consent, samples or condition that could jeopardize the subject's safety.

Sample size:

A panel of 50-100 RT-PCR positive samples for each differential pathogen, and 50-100 RT-PCR negative samples will be assembled and used. With an expected RPA assay sensitivity and specificity of 98% (confidence interval of 95%), a minimum of 30 positive and negative cases for each parameter shall be included in the study. Based on our experience in Guinea, 10% of RT-PCR tests on samples from suspect cases are EBOV positive. Consequently 300 suspected EVD cases need to be screened to obtain at least 30 EVD positive samples. A prevalence gradient of P. falciparum > S.typhi > DENV > YFV would be expected in the remaining 270 samples. The minimum number of positives and negatives to be tested for each parameter will be calculated according to their prevalence.

Samples collection and processing:

For the prospective study in febrile patients and potential new EVD outbreaks, saliva and serum will be collected via oral swabbing and venepuncture as per SOP performed in the EVD outbreak in Guinea harmonized with SOPs used in DRC for patients with EVD symptoms complying to the protocol inclusion criteria. EBOV-RPA and real time RT-PCR will be performed and results will be classified as concordant negative or positive or discordant. Patients with EBOV negative and positive results will respectively be discharged or hospitalized according to criteria established locally. Discordant results will be retested on the same sample and a supplemental one collected 24 hours later. If discrepancy cannot be resolved, results will be discussed with the clinician to take the best decision for the patient and communities regarding the context. Archived samples will be tested directly from aliquots stored in the freezer.

Sample extraction protocol:

Initially all collected sample will be subjected to inactivation step using lysis buffer in BSL-3 portable Glove box (Figure 2). Two rapid extraction protocols will be validated and compared for their suitability for field use, (i) a NaoH based alkaline lysis protocol provided by TwistDX which can be used directly on whole blood and serum [Osmundson TW et al.], (ii) a filter paper based nucleic acid extraction method allowing total nucleic acid extraction from crude samples [Zou Y et al.].

Both protocols require a maximum 10 minutes. Extracts will be sent to the BSL-4 laboratory in Winnipeg, Canada (NML), to confirm inactivation of Ebola virus in the sample extract by inoculation on cell culture. The RPA will be operated from the suitcase laboratory, which contains all equipment and reagents needed (Figure 1)

.

Data collection and storage:

Clinical data and results will be recorded via an existing questionnaire used in Guinea, harmonized to procedures in DRC and entered into a database. Data include but are not limited to age, sex, date of admission, date of onset, date of sampling, detailed symptoms, results of testing, final classification of the patient, disease outcome.

Appropriate quality control measures will be implemented to monitor the reliability of the process for data collection, entry general management compliance, and for the quality of data generated. At the end of the study all data collected, will be controlled for quality and locked for data analysis.

Standard operating procedure:

SOPs of all applicable research processes will be made available.

Data analysis methods:

Clinical data will be analysed to give an accurate description of the study population. Each collected variable will be described using statistical indicators such as mean, median, standard deviation for quantitative variables and frequencies for qualitative variables. The 95% confidence intervals will be provided as well as significance tests (such as Xi-square) for comparison between analytical and clinical performance of the EVD diagnostic tests.

Test accuracy:

Postive ref Negative ref tests result tests result Positive test result TP FP

Negative test result FN TN

Sensitivity: Proportion of true positives samples. (TP/TP+FN) Specificity: Proportion of true negative samples. (TN/TN+FP) The sensitivity and the specificity will be combined in a kappa value to estimate the overall performance of the tests.

Test usefulness:

A scoring system will be developed to assess the user friendliness