Operationalization of COVID-19 Rapid Diagnosis Using Xpert® Xpress SARS CoV-2 Assay in Resource-Limited Settings: Early Implementation Lessons From Uganda

Introduction: The novel Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) that causes COVID-19 disease is a global challenge. Several countries have adopted testing, isolation, and tracing strategy towards the control of the COVID-19 pandemic, but access to rapid and accurate testing is still a global challenge. The conventional PCR – based assay is the most commonly used test yet it has huge costs, infrastructural, and procurement logistical challenges. The Xpert® Xpress SARS-CoV-2 test is an automated in – vitro diagnostic test for the qualitative detection of nucleic acid from SARS-CoV-2 within a turnaround time of 60 minutes on the widely used GeneXpert Dx Instrument Systems. Here we document the best practices and challenges encountered with the operationalization of Xpert® Xpress SARS-CoV-2 testing in a resource-limited setting. Materials and Methods: The Xpert® Xpress SARS-CoV-2 implementation followed an operational work plan that included; Laboratory COVID-19 policy and planning, situational analysis of the Laboratory network, country Xpert® Xpress SARS-CoV-2 assay verication, and rollout at Mutukula Port Health Laboratory. The Laboratory strategy was based on a set of six objectives; conducting infrastructural modications, building a strong COVID-19 testing capacity, developing robust Laboratory Quality and Information Management Systems, establishing a Bio-risk management and Bio-banking capacity. Results: The Xpert® Xpress SARS-CoV-2 testing implementation team that was appointed by the Ministry of Health (Uganda) successfully established the Xpert® Xpress SARS-CoV-2 testing Laboratory at Mutukula border in Uganda. As of 9 th July 2020, this Laboratory had tested a total of 10,990 samples with a median turnaround time of 75 (IQR: 60 – 75) minutes for samples of persons entering through Mutukula Land Point of Entry as compared to the median TAT 1980 minutes before it was established. The laboratory had only one discordant result out of 20 panels in the inter-laboratory comparison retesting program. testing for rapid diagnosis of COVID-19 is feasible and signicantly reduces the long TAT observed with conventional RT-PCR based testing. The operationalization of the Xpert® Xpress SARS-CoV-2 testing is largely dependent on the initial planning, adequacy of resources, and preparedness within the laboratory network. Challenges include; the difference in approaches to COVID-19 response, the attitude of truck-drivers/persons on Infection Prevention and Control measures, language barrier, and waste management issues.


Introduction
In December 2019, numerous cases of mysterious pneumonia with unknown etiology occurred in Wuhan, Hubei Province, China [1,2,3]. This pneumonia quickly spread to other provinces of China and abroad.
On the 7 th of January 2020, a new coronavirus was identi ed in the human throat swab sample by the Chinese Center for Disease Control and Prevention (CDC) and was consequently named as 2019 Novel CoronaVirus (2019nCoV) by the World Health Organization (WHO) [1,2]. Upon worsening of the situation, the WHO declared the outbreak as the public health emergency of international concern (PHEIC) on January 31, 2020 [4]. On 11 th February 2020, the virus was rechristened as Severe Acute Respiratory Syndrome Corona Virus-2 (SARS-CoV-2) by the International Committee on Taxonomy of Viruses [5], and WHO subsequently declared the epidemic disease caused by SARS-CoV-2 as Coronavirus Disease 2019   [6]. The clinical features of COVID-19 are diverse, ranging from an asymptomatic state to acute respiratory distress syndrome and multi-organ dysfunction [7]. The frequent clinical features comprise fever, cough, sore throat, headache, fatigue, headache, myalgia and breathlessness [7], and conjunctivitis has also been described. Thus, they are indistinguishable from other respiratory infections [7]. As of June 21 st, 2020, COVID-19 had continued to spread worldwide, causing over 8.914 million cases and over 466 thousand deaths [6]. The European region is the most affected with more than 50% of cases and 60% of deaths reported [8]. The United States of America has the highest number of cases globally (695,350 cases) and the highest number of con rmed deaths (32,427 deaths) [9]. The African region is the least affected with 13,892 cases and 628 deaths, but the numbers are increasing [9]. Uganda has so far con rmed >340 cases of COVID-19 as of 30 th May 2020, and no related deaths [9,10].
Generally, the mortality rates for cases globally remain between 1% and 2% [11]. From the onset of this public health crisis, the need for rapid and accurate laboratory testing was highlighted, and researchers responded by developing the rst diagnostic tests for COVID-19 within days of the release of the viral genome sequence (12).
Strategies to contain the disease There are advances towards the development of effective vaccines and drugs [13]. However, these may delay to reach the mass population, and so till then adopting appropriate prevention and control is the best option to contain the SARS-CoV-2 pandemic [13]. Perhaps the most important insight from the international COVID-19 response to date has been that to successfully slow transmission and safeguard health systems, it is essential to accurately diagnose and effectively isolate and care for all cases of COVID-19 including those with the mild or moderate disease either in health or home settings, depending on the context and degree of illness [14]. To enhance the above strategy, the primary focus of most countries has been the rapid identi cation, testing, and treatment of patients with serious and severe COVID-19, and the sheltering of individuals at the highest risk of unfortunate consequences through tracing, quarantining, and supporting contacts [14].
Diagnosis of COVID-19 and the Xpert® Xpress SARS-CoV-2 PCR assay COVID-19 diagnosis relies on a combination of clinical symptoms with or without radiological imaging, and con rmed by molecular techniques such as SARS-CoV-2 PCR [6]. However, it is worth noting that the major diagnostic tool employed globally as observed from various published literature for the diagnosis of the SARS-CoV-2 has been the Real-Time Polymerase Chain Reaction (PCR) Nucleic Acid-based test [7,15]. Currently, either a Real-Time Polymerase Chain Reaction (RT-PCR) targeting the envelope (E gene) in combination with polymerase (RdRP-gene) is being used as described [16] or targeting the nucleoprotein (N-gene) as recommended by the US Centers for Disease Control and Prevention [17]. However, numerous bottlenecks have been stated since its implementation. From a technical point of view, the process is time-consuming with sample RNA extraction and PCR run-time of approximately 6 hours and a turnaround time of 12-24 hours. Since providing fast results is of critical importance to stop the pandemic through ensuring timely and adequate treatment of patients, developing high-quality rapid point of care diagnostics is essential. Among the most known Point of Care (POC) tests include the antigen-based and the molecular-based tests. The antigen-based tests have the potential in providing quick diagnosis however; these have demonstrated low sensitivity with the implication of high rates of false negatives thus unreliable to use [6].
The novel molecular point of care test, Xpert® Xpress SARS-CoV-2 FDA-approved assay, can signi cantly reduce the time taken to get a result. This helps to facilitate faster clinical decision-making [18,19]. The assay employed on the GeneXpert analyzer (Cepheid, Sunnyvale, CA) was launched in March upon receipt of Emergency Use Authorization from the United States Food and Drug Administration on 20 th March 2020 [8]. The assay's cartridge contains primers and probes and internal controls used in RT-PCR for the in vitro qualitative detection of SARS-CoV-2 RNA in upper respiratory specimens collected in viral transportation medium/normal saline [20]. The assay is currently among the rst in class for several reasons: (i) improved sensitivity over prior attempts at using Nucleic Acid Ampli cation Testing (NAAT) strategies, (ii) short turnaround time (iii) a modular format allowing testing across a spectrum of volume needs, (iv) the possibility of automation (In nity group of analyzers), (v) simplicity, (vi) speed, and (vii) Less human involvement hence less human errors. Additional advances of the technology include the cartridge designed with all relevant components ("lab-in-a-cartridge" is a term frequently used to describe the platform format) and the analyzer with its modular format. This gives random access in nature and has connectivity capabilities to various laboratory information systems and a remote connectivity platform for the monitoring of instrument performance at a central level.

Adoption of the Xpert® Xpress Sars-CoV-2 Assay in Resource-Limited Settings
Several countries haven't yet incorporated or are planning to incorporate the Xpert® Xpress SARS CoV-2 assay in their national COVID-19 control and management operations as of July 2020. This evident reluctance to implement the assay even amidst FDA approval may largely be attributed to the lack of implementation models to guide the integration of the assay in the national COVID-19 control and management operations among many other limitations. Besides, there are uncertainties around eld performance, overall program cost, and its impact, with no guidance on how to operationalize the assay given the dynamic nature of clinical and diagnostic algorithms that are subjected to relatively frequent reviews regularly due to the many uncertainties around this novel disease hence the purpose of this guidance.
Like many countries, SARS-CoV-2 testing in Uganda was previously limited to the centralized National Reference Laboratory using in-house PCR at the Uganda Virus Research Institute (UVRI). This has been associated with long sample transit turnaround time (TAT), low thorough-put, and arguably long results TAT. Earlier in the COVID-19 outbreak in Uganda, most of the cases were among truck drivers through the national transit routes in the East and Southern parts of the country, hence as a preventive measure, the National COVID-19 Task Force recommended testing for all truck drivers into the country. However, this led to long results turnaround time and general frustration among truck drivers given that these drivers were not allowed to enter the country except with negative COVID-19 test results. Testing for COVID-19 at the border was the appropriate biosecurity tool, however, the long results TAT associated with border testing meant frustration of border tra c ow and hence regional trade. Xpert® Xpress SARS-CoV-2 with a TAT of 2 hours with minimal infrastructural and biosafety requirements was one of the solutions to such a challenge. Tanzania registered her rst COVID-19 case on the 16 th March; by April 30 th , she had a total of 480 cases inclusive of community cases, and 16 COVID-19 related deaths which were so far the highest among East African states [22]. By that same time, Uganda had registered 83 cases with zero community cases and deaths [23].
We provide our experience on the implementation of the Xpert® Xpress SARS-CoV-2 assay for the diagnosis of COVID-19 at a Port Health Laboratory (PHL) in rural Uganda. This may be useful for countries with similar challenges that may be in the early stages of the establishment of a PHL for diagnosis of COVID-19.

Health policy and planning
This involved de ning the composition, roles, and responsibilities of an Xpert® Xpress SARS CoV-2 implementation team. This team was tasked to develop the Xpert Xpress National diagnostic algorithm and guidelines and update the National Guidelines for COVID-19 testing.

National Xpert® Xpress SARS-CoV-2 Assay Veri cation
This was done by the Xpert® Xpress SARS CoV-2 implementation team to provide evidence that the Xpert® Xpress SARS-CoV-2 assay could achieve the performance characteristics obtained during the manufacturer's method validation and whether the method was suitable for its intended use in the population of patients being targeted using Conventional PCR as a reference standard. Ten known positives (n=10) and 20 known negatives (n=20) samples were tested at the Department of Medical Microbiology, Mycobacteriology (BSL-3) laboratory, Makerere University. The agreement between the results of the Xpert® Xpress SARS-CoV-2 test and Conventional PCR was established as well as the Inter-Operator Reproducibility of the Xpert® Xpress SARS-CoV-2 assay. The veri cation activity was also used as a training exercise for the laboratory personnel. The trained laboratory personnel had two years' experience of the use of GeneXpert systems at the National Tuberculosis Reference Laboratory and some of them had been trained as "Super Users" by Cepheid.

Situational analysis of the National Lab Network to identify potential Xpert® Xpress SARS CoV-2 Laboratories
The Xpert® Xpress SARS CoV-2 implementation team conducted a situational analysis to establish testing sites that were strategically located from a public health perspective to impactfully yield the best results. Key elements to be assessed included testing site location, infrastructure (design, adequacy of space), accommodation for staff, availability of engineering biosafety requirements, GeneXpert platforms, software versions (Dx systems) of existing GeneXpert platforms and database back-ups at the sites, water, electricity, internet, and power back-up systems and security for both personnel and GeneXpert platforms. Mutukula (Uganda) is located in the extreme southern Kyotera district and it is the major crossing point, for both human and commercial tra c between Uganda and Tanzania [21].
A detailed operational plan was developed to implement recommendations resulting from the situational analysis. The plan entailed improvement on the existing infrastructure to accommodate the Xpert® Xpress SARS-CoV-2 assay; risk assessment; building a strong COVID-19 sample collection and testing capacity; establishing Laboratory Quality and Information Management systems; and building of Bio-risk management capacity. a) Risk assessment for the Mutukula Xpert® Xpress SARS CoV-2 Testing Laboratory WHO requires that all Public Health Laboratories conduct a local and institutional risk assessment to ensure competence to safely perform the intended testing with appropriate risk control measures in place [24]. We, therefore, conducted a Laboratory risk assessment according to the template provided in the February 2020 WHO Laboratory biosafety guidance related to coronavirus disease 2019 (COVID-19). This was done in a step by step manner from gathering information (hazard identi cation), evaluation of the risks, risk characterization, development of a risk control strategy, selection to development of a strategy for implementation of the identi ed risk control measures b) Lab design and construction of a Biosafety Level II laboratory WHO interim guidelines for COVID-19 laboratory biosafety, requires that non-propagative diagnostic laboratory work (for example, sequencing, nucleic acid ampli cation test [NAAT]) on clinical specimens from patients who are suspected or con rmed to be infected with COVID-19 should be conducted at a facility using procedures equivalent to Biosafety Level 2 (BSL-2) [24]. The pre-existing space at the Mutukula port health building was found to be ideal for the BSL2 lab. This was modi ed in terms of design to accommodate the necessities for handling the infectious SARS-CoV-2.

c) Laboratory equipment and materials
The WHO interim guidelines for lab COVID-19 testing required the manipulation of samples required to be undertaken in an appropriately maintained and validated Biosafety cabinet (BSC) class II. The Gene Xpert® system is the platform required for the execution of the Xpert® Xpress SARS-CoV-2 assay. Refrigerators (2 -8°C) or cooler boxes are required for short term storage of samples while long term storage requires freezers (-80 and -20 °C). Appropriate disinfectants with proven activity against enveloped viruses must be availed and used for the recommended contact time, at the correct dilution and within the expiry date after the working solution is prepared.

d) Human resource
Mobilization of human resources that were acquainted with the Cepheid GeneXpert Systems. These then had to be trained and deemed competent before performing Xpert® Xpress SARS-CoV-2 testing. To ensure that the newly established laboratory and its facilities meet the safety standards and provide a safe environment for laboratory personnel to conduct their work during the pandemic, an onsite inspection by the Ministry of Health and its partners had to be conducted.

g) Implementation of Quality Assurance activities
The Laboratory established a comprehensive and systematic Quality Assurance program to enable the achievement and maintenance of high levels of accuracy and pro ciency in COVID-19 testing, to ensure the reliability and reproducibility of results and therefore inspire con dence in its customers. Activities that were implemented included assay veri cation, training and competence assessment, equipment maintenance, installation and validation of the Laboratory Information Management System (LIMS), lot testing, establishment and monitoring of quality indicators, and Continuous Quality Improvement (CQI) using feedback from quality audits, Customer Satisfaction Survey (CSS), and External Quality Assessment (EQA e.g. inter-laboratory comparison) responses.
Mutukula Port Health Laboratory participated in an Inter-Laboratory comparison re-testing program with Tororo COVID-19 Mobile Field-Testing Lab, which works in almost a similar environment with MPHL.
Tororo Mobile Laboratory uses 200µl primary sample, manual extraction by DAAN kit, and Altona protocol duo target (E and S) detection on Biorad CXF97 whereas MPHL loads 300µl of a primary sample for Xpert® Xpress SARS-CoV-2 assay that targets both E and N2 genes with automated extraction and detection.
h) The assessment for capacity and capability to conduct COVID-19 testing This was conducted by WHO consultants using the WHO 2019-nCoV Laboratory Assessment tool. This tool was designed to assess ten (10) core capacities of laboratories that have implemented testing for SARS-CoV-2. These included: Organization and management; Documents; Specimen collection; Handling and transport; Data and information management; Consumables and reagents management; Equipment management; Facilities; Human resources; Bio-risk management; and Public health functions.

Results
Health policy and planning The sensitivity, speci city, positive predictive value, and negative predictive value were 100% for the Xpert® Xpress SARS-CoV-2 test against the gold standard Conventional PCR as stated in Table 1. The Inter-Operator Reproducibility of testing using Xpert® Xpress SARS-CoV-2 was 100% as stated in Table 2.

Results of the situational analysis
This was based on a set of criteria as summarized in Table 3. Upon evaluation of the situational analysis results, the Port Health Laboratory at the Mutukula border met all the requirements that were included on the checklist. Risk assessment of the laboratory Upon risk evaluation, the overall initial high risk was mainly due to homogenization of the sample, loading the sample into the cartridge during the Xpert® Xpress SARS-CoV-2 assay, and waste handling. The laboratory developed and implemented a risk control strategy with effective and sustainable measures including the use of a certi ed BSC Class II, appropriate personal protective equipment (PPE), development of administrative controls, appropriate use of disinfectants, safe housekeeping practices, timely collection and disposal of generated waste. The administrative controls included biosafety training, development of policies, procedures, bench aids, and use of signages. The residual risk was very low hence a safer working environment for personnel.

Laboratory design and biosafety
The laboratory was modi ed by the installation of two workbenches, an air conditioner, a hand-washing sink, and a sample reception window in consideration of the risk control strategy. The laboratory met the structural and biosafety requirements as stipulated by the WHO Laboratory Biosafety Manual, 3rd edition, for BSL-2 laboratories and WHO Laboratory biosafety guidance related to coronavirus disease 2019 (COVID-19) as elaborated in Tables 1 and 2.   The disinfectants selected had proven activity against enveloped viruses, including sodium hypochlorite (bleach; for example, 1000 parts per million [ppm] (0.1%) for general surface disinfection and 10 000 ppm (1%) for disinfection of spills; 70% ethanol or 80% isopropanol for hand, surfaces, and GeneXpert platform decontamination. The contact time for the bleach was set to 10 minutes.
Onsite assay veri cation results The sensitivity, speci city, positive predictive value and negative predictive value were 100% for the  (Fig 2). The assessment for capacity and capability to conduct COVID-19 testing The overall score for the capacity and capability to conduct COVID-19 testing by Mutukula Port Health Laboratory was reported to be 90%. This was above the limit of 80% that is required by any laboratory that implements or intends to conduct COVID-19 testing (Figure 7).

Discussion
This Ugandan experience indicates that, with collective efforts and technical support from locally available expertise, it is achievable to set-up and optimally operate a safe BSL-2 Xpert® Xpress SARS-CoV-2 testing facility with satisfactory quality performance in a resource-limited setting. Setting up an Xpert® Xpress SARS CoV-2 implementation team that is acquainted with the international and national requirements for construction, equipment installation, Bio-risk management, Quality Assurance, COVID-19 diagnostics in addition to recruitment of experienced personnel in the use of GeneXpert Systems and outbreak rapid response before initial laboratory set-up is key to achieving the set start-up targets. Initial and continuous support from the Ministry of Health, its partners, multi-agencies at the port as well as open discussions on the possible limitations and nding out the alternative solutions ahead of time was also crucial.
Risk assessment was very crucial as it ensured a safe environment to perform the Xpert® Xpress SARS CoV-2 testing with appropriate mitigation measures in place thereby reducing the residual risk to an acceptable very low level.
The performance in the national veri cation of Xpert® Xpress SARS-CoV-2 assay (for the sensitivity, speci city, positive predictive value, and negative predictive value against the gold standard Conventional PCR) that was above 90% proved the suitability of the assay for its intended use in the population of truck drivers and returnees.
Understanding the satisfactory quality performance standards as well as planning the routes to achieving them upfront, including early monitoring of performance with corrective and preventative action projects and management commitment, have enabled MPHL to realize continuous quality improvement. This is evident with the exceptional growth of MPHL from a Zero to three (3) stars ranking within 60 days of operation. The baseline laboratory performance was assessed by internal auditors, employing standard assessment tools accessible at (26). From our ndings, due to the initiation of onsite testing, the results turnaround time was reduced by 17-fold which quickened on tra c ow at the border point as well as case detection and management. Even though the proportion of samples released on time is within the targeted ranges by July, root cause analyses should be done for those the few results that fall out of TAT.
The Inter-Laboratory comparison re-testing program agreement of 95% was within the acceptable range (our target of 80% and above). This was possible due to the strict pro ciency procedures in the laboratory which require a mixture of senior and junior technologists at all times working on the bench to ensure effective competency transfer, training and competence assessments, and the effective use of standard operating procedures.
Generally, given the short time of operation the lab has been in existence, that performance as per the WHO assessment is very su cient. The lowest score was from the core capacity; organization and management (67%). From our understanding, this is a big challenge to many government-owned facilities for which the top management is always afar.

Conclusion
Rapid diagnosis of COVID-19 using the Xpert® Xpress SARS-CoV-2 assay is feasible in resource-limited settings. Implementation of Xpert® Xpress SARS-CoV-2 testing for rapid diagnosis of COVID-19 at border points is critical in reducing the long TAT observed with Conventional Molecular based testing. The operationalization of the testing is largely dependent on the initial planning, human resource adequacy, and preparedness within the laboratory network. The key to the successful implementation of the testing is to involve an excellent network of agencies at the port. Challenges realized included the difference in approaches to COVID-19 response by Tanzania and Uganda, the attitude of truck-drivers on Infection Prevention and Control (IPC) measures, language barrier, and waste management issues.