Nucleic acid test for use on the cobas 6800/8800 Systems

Background and objectives: HIV viral load testing is recommended for all patients on antiretroviral treatment. As viral load testing continues to scale up worldwide, highly sensitive and specific HIV viral load tests that support high throughput testing are needed to provide timely and accurate results. We evaluated the performance characteristics of the new cobas® HIV-1 test for use with the cobas® 6800/8800 system (cobas® HIV-1), and also assessed clinical correlation with the reference standard cobas®AmpliPrep/cobas®TaqMan® HIV-1 Test, version 2.0 (TaqMan®HIV-1 v2.0). Methods: The limit of detection (LoD), linearity, precision, analytical specificity, diagnostic sensitivity, and specificity of the cobas® HIV-1 test were evaluated. The method comparison to the TaqMan®HIV-1 v2.0 was assessed using Deming regression and Bland-Altman plot analysis, and the mean of the paired viral load difference between the two tests was also calculated using a paired t-test. Results: The cobas® HIV-1 test demonstrated an LoD of 13.2 copies/mL for HIV-1 M subtype B and a similar LoD for all other HIV-1 subtypes with a linear range of 13 to 1.73 × 107 copies/mL. The assay precision was high with a pooled SD of 0.05 to 0.12 log10 across the linear range. Diagnostic sensitivity using clinical samples was 99.25%, with a specificity of 100%. cobas® HIV-1 demonstrated excellent correlation to TaqMan®HIV-1 v2.0 (R2=0.98). The overall percentage agreement between the two tests was also high (>95%) with a cutoff of 50 and 200 copies/mL. Conclusions: The cobas® HIV-1 test is a highly sensitive and specific method for viral load monitoring that has an improved limit of detection, lower sample volume, and increased throughput capability compared to TaqMan®HIV-1 v2.0. The novel cobas® HIV test has excellent correlation with the FDA-approved TaqMan® HIV-1 v2.0 test.


Intended use
cobas® WNV for use on cobas® 6800 and cobas® 8800 Systems is a qualitative in vitro nucleic acid screening test for the direct detection of West Nile virus (WNV) RNA in human plasma.
This test is intended for use to screen donor samples for WNV RNA in plasma samples from individual human donors, including donors of whole blood and blood components, as well as other living donors. This test is also intended for use to screen organ and tissue donors when donor samples are obtained while the donor's heart is still beating or from cadaveric (non-heart beating) donors.
Plasma from all donors may be screened as individual samples. For donations of whole blood and blood components, plasma samples may be tested individually or in pools comprised of not more than six individual samples. For all other donors, samples may only be screened as individual samples.
This test is not intended for use as an aid in diagnosis of WNV infection.
This test is not intended for use on samples of cord blood.

Summary and explanation of the test Background: Screening of blood for transfusion-transmitted viral infections
West Nile virus (WNV) is a single-stranded, positive-sense, arthropod-borne (arbovirus) RNA virus that belongs to the Flaviviridae family, genus Flavivirus, and the Japanese encephalitis virus serocomplex. 1,2 The Japanese encephalitis serocomplex also includes Japanese encephalitis virus, St. Louis encephalitis virus, Murray Valley encephalitis virus, and Kunjin virus (now known to be a WNV variant). [3][4][5] Phylogenetic studies have identified 2 main lineages of WNV: lineage 1 and lineage 2. Strains from lineage 1 are found in Africa, India, Australia, and the Western Hemisphere and have been responsible for recent epidemics in Europe, the Mediterranean basin, and the Americas. Strains from lineage 2 have been reported in sub-Saharan Africa 6 and more recently Southern Europe. 7,8 Like other arboviruses, WNV is maintained in an enzootic cycle between blood-feeding mosquitoes and susceptible vertebrate hosts (birds). 9 Birds serve as the natural reservoir vertebrate host and mosquitoes of the genus Culex are the principal enzootic vectors for WNV, while humans and mammals (e.g., horses) are incidental and, usually, dead-end hosts because they rarely develop viremia of sufficient titer to infect efficiently arthropod vectors. 2,9,10 WNV is distributed widely throughout Africa, the Middle East, southern Europe, western Russia, southwestern Asia, and Australia (Kunjin subtype of WNV), because of WNV's ability to infect numerous mosquito and bird species. 1 Human outbreaks, mainly associated with mild febrile illnesses, were reported infrequently in Israel and Africa until the mid-1990s. 1 Since the mid-1990s, new viral strains, likely with African origins, have resulted in increased numbers of infections in parts of Russia and southern and Eastern Europe, with large outbreaks of increased clinical severity occurring in Romania, Russia, Israel, and Greece. 1,8,11 WNV now circulates in many countries in the Western Hemisphere, but only the United States and Canada have experienced substantial human disease incidence. 1,12 WNV first emerged in the United States in 1999 in New York City and spread rapidly across the entire United States in subsequent years. 13 WNV is now endemic in all 48 contiguous United States, as well as all Canadian provinces. 1 WNV has produced the three largest arboviral neuroinvasive disease (encephalitis, meningitis, or acute flaccid paralysis) outbreaks ever recorded in the United States, with nearly 3,000 cases of neuroinvasive disease recorded each year in 2002, 2003, and 2012. 1 High viral activity occurs during the warm months of the year. 1 94% of patients with WNV infection develop symptom onset in the summer months. 1 WNV is estimated to have infected more than 4 million people in the United States between 1999 and 2012, 1 with a reported total of 16,196 patients with WNV neuroinvasive disease, including 1,549 related deaths. 1

Rationale for NAT testing
WNV was first shown to be transmissible by transfusion and organ transplantation during investigations of an epidemic in the United States in 2002. 14,15 WNV can be transmitted via transfused red blood cells, platelets, fresh frozen plasma, and heart, kidney, liver, and lung transplants, although mosquito bites cause most WNV infections in humans. 1,2,14,16 WNV can also potentially be transmitted through hematopoietic progenitor cell transplantation. Transplacental and perinatal transmission of WNV has been reported. 1 Breast milk transmission, patients undergoing kidney dialysis, and occupational exposure (e.g., laboratory workers [percutaneous or conjunctival exposure]; poultry farm workers) are other rare modes of WNV transmission. 1,12 Infection usually produces lifelong immunity. 9 Transfusion-transmitted WNV usually occurs during the acute phase of infection, when infected individuals are viremic and asymptomatic but have not yet seroconverted. 17 Since few infected donors develop clinically-significant disease, questioning blood donors for recent illness suggestive of WNV infection is ineffective at identifying infected/seropositive donors. 18,19 Data gathered from blood donor screening shows that extremely low-titer WNV viremia from very recently infected donors who have not yet developed WNV antibodies efficiently transmit WNV infection. 9,20 Donations with very low viral loads have been implicated in cases of transfusion-related transmission of WNV, 21 which poses particular danger for immunocompromised patients, who are the recipients of the majority of blood transfusions. 22 Nationwide nucleic acid testing (NAT) for WNV RNA was implemented in 2003 to insure transfusion safety. 9 During the first 2 years of WNV NAT screening of blood donations in the United States, 1,039 positive donors were identified among 27.2 million donations (0.4 per 10,000 donations), but the numbers ranged as high as 1 in 150 donors in some areas during epidemics. 18 NAT screening of blood donations in the United States and Canada has nearly eliminated the risk of transfusion-transmitted West Nile virus infection. 1 Between 2003 and 2013, approximately 3,000 WNV infections were interdicted. 23 Among persons who become infected with WNV, approximately 80% are asymptomatic, 20% to 25% develop West Nile fever, 1, 24 and 1 in 150 to 250 develop neuroinvasive disease. 1,25 West Nile fever consists of sudden onset headaches, malaise, fever (usually low grade), myalgia, chills, vomiting and other gastrointestinal symptoms, rash, fatigue, and eye pain, which can last a few days to a few weeks or even months. 1,24 West Nile neuroinvasive disease can manifest as meningitis, encephalitis, meningoencephalitis, or acute flaccid paralysis, which can lead to irreversible neurological damage, coma, and death. 1,26-31 WNV infection is also associated with myocarditis, pancreatitis, fulminant hepatitis, rhabdomyolysis, multifocal choroiditis, vitritis, and autonomic instability. 1 The sequelae of neuroinvasive disease can persist for months to years after recovery from acute infection. After discharge from the hospital, individuals with West Nile encephalitis often require assistance with activities of daily living. 1,31,32 Neuropsychiatric symptoms, including depression and anxiety, as well as neurocognitive deficits, may persist for months to a year or longer. 1,20,33 About 10% of individuals who develop neuroinvasive West Nile disease die as a result; advanced age is the most important risk factor. 2 The risk of fatality is 17% for patients age 70 years or older, compared to a 0.8% risk of death for patients younger than 40 years of age. 1,33 Other risk factors for death include encephalitis with severe muscle weakness, altered level of consciousness, diabetes, cardiovascular disease, hepatitis C virus infection, and immunosuppression. 1,12,33 Explanation of the test cobas® WNV is a qualitative test that is run on the cobas® 6800 System and cobas® 8800 System. cobas® WNV enables the simultaneous detection of WNV RNA and the internal control in a single test of an infected, individual donation or pooled plasma from individual donations.

Principles of the procedure
cobas® WNV is based on real time PCR technology on a fully automated sample preparation (nucleic acid extraction and purification) followed by PCR amplification and detection system. The cobas® 6800/8800 Systems consist of the sample supply module, the transfer module, the processing module, and the analytic module. Automated data management is performed by the cobas® 6800/8800 software which assigns test results for all tests as non-reactive, reactive, or invalid. Results can be reviewed directly on the system screen, and printed as a report or sent to a Laboratory Information Management System (LIMS) or other result management system. Samples can either be tested individually or, optionally, can be tested in pools consisting of multiple samples. The cobas p 680 instrument or cobas® Synergy software with the Hamilton MICROLAB® STAR IVD (cobas® Synergy Core), may optionally be used in a pre-analytical step if pooling is to be performed.
Nucleic acid from the sample and added armored RNA internal control (IC) (which serves as the sample preparation and amplification/detection process control) are simultaneously extracted. In addition the test utilizes two kit controls: a positive and a negative control. Viral nucleic acid is released by addition of proteinase and lysis reagent to the sample. The released nucleic acid binds to the silica surface of the added magnetic glass particles. Unbound substances and impurities, such as denatured protein, cellular debris, and potential PCR inhibitors (such as hemoglobin) are removed with subsequent wash reagent steps and purified nucleic acid is eluted from the magnetic glass particles with elution buffer at elevated temperature.
Selective amplification of target nucleic acid from the donor sample is achieved by the use of virus-specific forward and reverse primers which are selected from highly conserved regions of the viral nucleic acid. A thermostable DNA polymerase enzyme is used for both reverse-transcription and amplification. The master mix includes deoxyuridine triphosphate (dUTP), instead of deoxythimidine triphosphate (dTTP), which is incorporated into the newly synthesized DNA (amplicon). [34][35][36] Any contaminating amplicon from previous PCR runs are destroyed by the AmpErase enzyme [uracil-N-glycosylase], which is included in the PCR mix, when heated in the first thermal cycling step. However, newly formed amplicon are not destroyed since the AmpErase enzyme is inactivated once exposed to temperatures above 55°C.
cobas® WNV master mix contains detection probes which are specific for WNV and IC nucleic acid. The specific WNV and IC detection probes are each labeled with one of two unique fluorescent dyes which act as a reporter. Each probe also has a second dye which acts as a quencher. The two reporter dyes are measured at defined wavelengths, thus permitting simultaneous detection and discrimination of the amplified WNV target and the IC. 37,38 When not bound to the target sequence, the fluorescent signal of the intact probes is suppressed by the quencher dye. During the PCR amplification step, hybridization of the probes to the specific single-stranded DNA template results in cleavage by the 5' to 3' nuclease activity of the DNA polymerase resulting in separation of the reporter and quencher dyes and the generation of a fluorescent signal. With each PCR cycle, increasing amounts of cleaved probes are generated and the cumulative signal of the reporter dye is concomitantly increased. Since the two specific reporter dyes are measured at defined wavelengths, simultaneous detection and discrimination of the amplified WNV target and the IC are possible.

Reagent storage and handling requirements
Opened reagents shall be stored and will be handled as specified in Table 5 and Table 6.
When reagents are not loaded on the cobas® 6800/8800 Systems, store them at the corresponding temperature specified in Table 5. Reagents loaded onto the cobas® 6800/8800 Systems are stored at appropriate temperatures and their expiration is monitored by the system. The system allows reagents to be used only if all of the conditions shown in Table 6 are met.
The system automatically prevents use of expired reagents. Table 6 allows the user to understand the reagent handling conditions enforced by the cobas® 6800/8800 Systems. Additional materials required

Instrumentation and software required
The cobas® 6800/8800 software and cobas® WNV analysis package shall be installed on the instrument(s). The Instrument Gateway (IG) server will be provided with the system. The cobas® Synergy software shall be installed, if applicable.

Precautions and handling requirements Warnings and precautions
As with any test procedure, good laboratory practice is essential to the proper performance of this assay. Due to the high sensitivity of this test, care should be taken to keep reagents and amplification mixtures free of contamination.
 For in vitro diagnostic use only.  All samples should be handled as if infectious, using good laboratory procedures as outlined in Biosafety in Microbiological and Biomedical Laboratories and in the CLSI Document M29-A4. 39,40 Only personnel proficient in handling infectious materials and the use of cobas® WNV, cobas® 6800/8800 Systems and optionally cobas p 680 instrument or the Hamilton MICROLAB® STAR IVD with cobas® Synergy Core should perform this procedure.  All human-sourced materials should be considered potentially infectious and should be handled with universal precautions. If spillage occurs, immediately disinfect with a freshly prepared solution of 0.5% sodium hypochlorite in distilled or deionized water (dilute household bleach 1:10) or follow appropriate site procedures.  cobas® WNV Control Kit and cobas® NHP Negative Control Kit contain plasma derived from human blood.
The source material has been tested by licensed antibody tests and found non-reactive for the presence of antibody to HCV, antibody to HIV-1/2, HBsAg, and antibody to HBc. Testing of normal human plasma by PCR methods also showed no detectable HIV-1 (Groups M and O) RNA, HIV-2 RNA, HCV RNA, HBV DNA, HEV RNA, WNV RNA, or CMV DNA. No known test method can offer complete assurance that products derived from human blood will not transmit infectious agents.  Do not freeze whole blood.  The use of sterile disposable pipettes and nuclease-free pipette tips is recommended. Use only supplied or specified required consumables to ensure optimal test performance.  Closely follow procedures and guidelines provided to ensure that the test is performed correctly. Any deviation from the procedures and guidelines may affect optimal test performance.  False positive results may occur if carryover of samples is not adequately controlled during sample handling and processing.

Reagent handling
 Handle all reagents, controls, and samples according to good laboratory practice in order to prevent carryover of samples or controls.  Before use, visually inspect each reagent cassette, diluent, lysis reagent, and wash reagent to ensure that there are no signs of leakage. If there is any evidence of leakage, do not use that material for testing.  cobas omni Lysis Reagent contains guanidine thiocyanate, a potentially hazardous chemical. Avoid contact of reagents with the skin, eyes, or mucous membranes. If contact does occur, immediately wash with generous amounts of water; otherwise, burns can occur.  cobas® WNV kits, cobas omni MGP Reagent, and cobas omni Specimen Diluent contain sodium azide as a preservative. Avoid contact of reagents with the skin, eyes, or mucous membranes. If contact does occur, immediately wash with generous amounts of water; otherwise, burns can occur. If these reagents are spilled, dilute with water before wiping dry.  Do not allow cobas omni Lysis Reagent, which contains guanidine thiocyanate, to contact sodium hypochlorite (bleach) solution. This mixture can produce a highly toxic gas.  Safety Data Sheets (SDS) are available on request from your local Roche representative.  Dispose of all materials that have come in contact with samples and reagents in accordance with country, state, and local regulations.

Good laboratory practice
 Do not pipette by mouth.  Do not eat, drink, or smoke in designated work areas.  Wear laboratory gloves, laboratory coats, and eye protection when handling samples and reagents. Gloves must be changed between handling samples and cobas® WNV kits and cobas omni reagents to prevent Other than noted above, samples are stored at 2-8°C. In addition, plasma separated from the cells may be stored for up to 30 days at <-18°C with three freeze/thaw cycles. Refer to Figure 1.

Cadaveric blood samples
 Cadaveric blood samples collected in EDTA anticoagulant tubes may be used with cobas® WNV. Follow the sample collection tube/bag manufacturer instructions for handling and centrifugation.  Cadaveric blood collected in EDTA anticoagulant may be stored for up to 8 days at 2-8°C with the following conditions: o Samples must be centrifuged and plasma must be separated from cells within 24 hours of draw. o For storage above 8°C, samples may be stored at up to 25°C, for 24 hours. Other than noted above, cadaveric EDTA plasma separated from the cells may be stored for up to 30 days at <-18°C with up to three freeze/thaw cycles. Refer to Figure 3.

Instructions for use Automated sample pipetting and pooling (optional)
Either the cobas p 680 instrument, or cobas® Synergy Core can be used as an optional component of the cobas® 6800/8800 Systems used for automated pipetting and pooling of aliquots of multiple primary samples into one pooled sample. Refer to the cobas p 680 instrument Operator's Manual or to the cobas® Synergy software User Assistance for more information.

Procedural notes
 Do not use cobas® WNV reagents, cobas® WNV Control Kit, cobas® NHP Negative Control Kit, or cobas omni reagents after their expiry dates.  Do not reuse consumables. They are for one-time use only.  Refer to the cobas ® 6800/8800 Systems Operator's Manual or to the cobas® Synergy software User Assistance as applicable for details on optional pooling procedures for proper maintenance of instruments.

Running the cobas ® WNV
The test procedure is described in detail in the cobas® 6800/8800 Systems Operator's Manual and the cobas p 680 instrument Operator's Manual or to the cobas® Synergy software User Assistance as applicable for details on optional pooling procedures. Figure 4 below summarizes the procedure.

Results
The cobas® 6800/8800 Systems Software automatically detects WNV RNA simultaneously for the samples and controls. Invalidation of results is performed automatically by the cobas® 6800/8800 software based on negative and positive control failures. Table 9 Control flags for negative and positive controls

Negative Control
Flag Result Interpretation The entire batch is assigned invalid if the result for the (-) C is invalid.

WNV (+) C Q02 Invalid The entire batch is assigned invalid if the result for the WNV (+) C is invalid.
If the batch is invalid, repeat testing of the entire batch including samples and controls. Two parameters are measured simultaneously for each sample: WNV and the internal control. Final sample results for cobas® WNV are reported by the software. In addition to the overall results, individual target results will be displayed in the cobas® 6800/8800 software and should be interpreted as follows: For WNV lineage 1 and 2 viruses, PROBIT analysis on the data combined across dilution series and reagent lots was used to estimate the LoD, along with the lower and upper limit of the 95% confidence interval (Table 11). The reactivity rates observed in the LoD studies for lineage 1 and 2 are summarized in Table 12 and Table 13, respectively.

Inclusivity
The performance of cobas® WNV to detect flavivirus variants of WNV was determined by testing unique cultured isolates for each variant. A total of 10 individual WNV Lineage 1 positive cultured isolates were tested after dilution with normal, virus-negative (WNV) human EDTA-plasma at the concentration of approximately 36 copies/mL. All 10 cultured samples were detected (Table 14).
For the flavivirus variants of WNV, a total of two positive cultured isolates of Japanese encephalitis virus (JEV) were tested with four replicates after dilution with normal, virus-negative (WNV) human EDTA-plasma. A total of one positive cultured isolate of Saint Louis encephalitis virus (SLEV), Murray Valley encephalitis virus (MVEV) and Kunjin virus (KUNV) was tested using four replicates of each isolate after log dilutions were prepared with normal, virus-negative (WNV) citrate plasma. All cultured isolates were detected (Table 15).

Analytical specificity
The analytical specificity of cobas® WNV was evaluated for cross-reactivity with 27 microorganisms at 10 6 particles, copies, or PFU/mL, which included 20 viral isolates, six bacterial strains and one yeast isolate (Table 16). The microorganisms were added to normal, virus-negative (WNV) human EDTA-plasma and tested without WNV and with WNV added to a concentration of approximately 3 x LoD of cobas® WNV. The tested microorganisms do not cross-react or interfere with cobas® WNV. Plasma samples from each of the disease states (Table 17) were tested without WNV and with WNV added to a concentration of approximately 3 x LoD of cobas® WNV. These disease states do not cross-react or interfere with cobas® WNV.

Analytical specificity -interfering substances Endogenous interference substances
Plasma samples with abnormally high levels of triglycerides (up to 35.3 g/L), hemoglobin (up to 4.7 g/L), unconjugated bilirubin (up to 0.21 g/L), albumin (up to 61.3 g/L), and human DNA (up to 0.004g/L) were tested without WNV and with WNV added to a concentration of approximately 3 x LoD of cobas® WNV. Samples containing these endogenous substances did not interfere with the sensitivity or specificity of cobas® WNV.

Exogenous interference substances
Normal, virus-negative (WNV) human EDTA-plasma samples containing abnormally high concentrations of drugs (Table 18) were tested without WNV and with WNV added to a concentration of 3 x LoD of cobas® WNV. These exogenous substances did not interfere with the sensitivity or specificity of cobas® WNV.

Whole system failure
The whole system failure rate for cobas® WNV was determined by testing 100 replicates of EDTA plasma spiked with WNV. These samples were tested at a target concentration of approximately 3 x LoD and were run in pools of one (undiluted). The study was performed using the cobas® 8800 System with cobas p 680 instrument (pipetting and pooling).
The results of this study determined that all replicates were reactive for WNV, resulting in a whole system failure rate of 0%. The two-sided 95% exact confidence interval was 0% for the lower bound and 3.62% for the upper bound [0%: 3.62%].

FDA/CBER panel evaluation
The sensitivity of cobas® WNV was determined by testing the FDA/CBER WNV Lot Release panel (Table 19).
In total 10 panel members were tested undiluted in each three replicates across three reagent lots (one replicate per panel member per reagent lot).
The results of this study determined that all FDA/CBER panel members with WNV titers of 5 cp/mL to 100 cp/mL were detected by cobas® WNV and that all panel members of 0 cp/mL were non-reactive for WNV, irrespective of the reagent lot. A total of two (2) independent dilution series of 6 concentrations and a blank were prepared by diluting the WNV secondary standard in negative pooled moderately hemolyzed cadaveric EDTA samples and pooled highly hemolyzed cadaveric EDTA plasma samples. Each dilution series was tested using one of two unique reagent lots of the cobas ® WNV test.
The results are summarized in Table 20 and Table 21.

Sensitivity using clinical samples
The clinical sensitivity of cobas® WNV for WNV RNA was evaluated by testing a total of 60 individual virusnegative cadaveric samples, of those 35 individual samples were classified as moderately hemolyzed (straw to pink colored) and 25 individual samples were classified as highly hemolyzed (red to brown colored). In addition a total of 60 individual virus-negative living donor samples were tested. All cadaveric and living donor samples were divided evenly across three reagent lots, five clinical samples spiking groups (for WNV) with 12 samples per group. Each cadaveric and living donor sample was spiked with a unique clinical samples (WNV) at approximately 5 x LoD of the respective sample. Each cadaveric sample was diluted 1:5.6 with cobas omni Specimen Diluent on the instrument and tested using the cadaveric sample testing procedure.
All of the cadaveric and the living-donor samples had a reactive rate of 100% (95% confidence interval: 94.0 -100%).
The clinical sensitivity observed in cadaveric sample was equivalent to the sensitivity observed in living donor samples as determined by Fisher's Exact Test and summarized in Table 22.

Specificity
The specificity of cobas® WNV in cadaveric EDTA plasma samples was evaluated and compared with the specificity in living donor sample by testing single replicates of 64 individual cadaveric EDTA plasma samples, of those 40 individual donor samples were classified as moderately hemolyzed (straw to pink colored) and 24 individual samples were classified as highly hemolyzed (red to brown colored), and 60 individual seronegative living-donor plasma samples. The study was performed with three independent cobas® WNV reagent lots. Each cadaveric sample was diluted 1:5.6 with cobas omni Specimen Diluent on the instrument and tested using the cadaveric sample testing procedure. All the cadaveric and living donor EDTA plasma samples were non-reactive for 100% specificity. The specificity observed for cadaveric samples was equal to the specificity observed for living-donor samples as determined by the Fisher's Exact Test (α = 0.05) as summarized in Table 23.

Reproducibility
The reproducibility of cobas® WNV on the cobas® 6800/8800 Systems was determined using 20 cadaveric EDTA plasma samples (moderately and highly hemolyzed) spiked with Roche Secondary Standard for WNV RNA to approximately 5 x LoD of cobas® WNV. The results were compared to the reproducibility obtained with 20 living-donor EDTA plasma samples spiked with the Roche Secondary Standard to approximately 5 x LoD of cobas® WNV.
Testing was performed for the following variable components:  day-to-day variability over 6 days  lot-to-lot variability using three different reagent lots of cobas® WNV One replicate was tested with each of the three reagent lots over six days for up to 18 replicates per cadaveric and living donor sample. Each cadaveric sample was diluted 1:5.6 with cobas omni Specimen Diluent on the instrument and tested using the cadaveric sample testing procedure. All valid reproducibility data were evaluated by comparing the reactive rates of living donors and cadaveric samples (two-sided 95% Confidence Intervals) across all variable components. The Fisher's exact p value was calculated for the test of statistical significance of the difference between proportions of reactives observed with cadaveric and living donor samples. No significant differences were observed. cobas® WNV is reproducible over multiple days and reagent lots for cadaveric and living donor EDTA plasma samples. The results from reagent lot-to-lot variability are summarized in Table 24.

Clinical specificity
The clinical specificity of cobas® WNV was evaluated by testing randomly-selected blood donations at four external laboratory sites. Individual samples and samples in pools of six were tested. Three different cobas® WNV reagent lots were used in the study. Clinical specificity of the cobas® WNV was calculated as the percentage (95% two-sided CI) of WNV donor status-negative donors who had cobas® WNV non-reactive results. There were 63,243 evaluable donations from pooled testing and 10,823 evaluable donations from individual testing.  The cobas® WNV pool specificity for index donations was 100% (10,573/10,573; 95% CI: 99.96% to 100%). None of the 10,573 pools of six were cobas® WNV reactive. An invalid rate of 1.6% due to internal control or instrument failures was observed for pooled sample results.

Reproducibility
The reproducibility of cobas® WNV for use on the cobas® 6800/8800 Systems was established by testing an eight member panel composed of two negative plasma samples and two samples positive for WNV at three different concentrations (approximately 0.5 x, 1.0 x, and 3.0 x the LoD of cobas® WNV).
Operators at each of three sites with the cobas® 8800 performed five days of testing with each of three lots of cobas® WNV reagents and two valid panel runs (i.e., two batches, each batch composed of one panel and two independent controls) per day were completed to yield up to 180 tests per panel member virus type at each of the three concentrations.
All valid batches and test results were analyzed by calculating the percentage of reactive test results for each panel member (Table 28). This study demonstrated that cobas® WNV for use on the cobas® 6800/8800 Systems shows reproducible performance across the variables assessed (lot, site/instrument, day, batch, and within batch) for detecting WNV.

Symbols
The following symbols are used in labeling for Roche PCR diagnostic products.