key: cord-0722684-iawkxck1 authors: Brunelle, Sharon L; Bird, Patrick M; Boone, Jeremy; Nelson, Maria; Johnson, Zerlinde; Coates, Scott title: Comparison of the Modified Centers for Disease Control and Prevention 2019-Novel Coronavirus Real-Time RT-PCR Method for Detection of Infectious and Heat-Inactivated Virus on Stainless Steel date: 2021-04-02 journal: J AOAC Int DOI: 10.1093/jaoacint/qsab046 sha: aba06ed70905e34a99bac60f8ab25b9c339f0347 doc_id: 722684 cord_uid: iawkxck1 BACKGROUND: Infectious Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2) was used in the validation of methods for detection of SARS-CoV-2 on stainless steel surfaces in the AOAC Research Institute Emergency Response Validation project. Handling infectious virus requires Biosafety Level (BSL)-3 facilities. OBJECTIVE: To compare the recovery and detection of infectious and heat-inactivated (65 °C for 30 min) SARS-CoV-2 from stainless steel by the modified US Centers for Disease Control and Prevention (CDC) 2019-Novel Coronavirus Real Time Reverse Transcription Polymerase Chain Reaction (RT-PCR) Diagnostic Panel. METHODS: Viral stocks were diluted in viral transport medium and deposited onto stainless steel test areas at 2 x 10(3) and 2 x 10(4) genomic copies for low and high, respectively. Test areas were sampled, and aliquots of the resulting test solutions analyzed by RT-qPCR according to the CDC method. Results were analyzed by Probability of Detection (POD) statistics. RESULTS: The low level, where fractional positive results (25–75%) are expected, yielded POD(I) = 0.80 (0.58, 0.92) for the infectious virus and POD(HI) = 0.15 (0.05, 0.36) for the heat-inactivated virus. The bias, dPOD(HI) = -0.65 (-0.80, -0.35), demonstrated a statistical difference between infectious and heat-inactivated virus detection. No difference was observed at the high inoculation level. CONCLUSION: Despite the statistical difference observed, the use of the heat-inactivated virus is a viable alternative for matrix extension studies using a method comparison study design. HIGHLIGHTS: The use of heat-inactivated SARS-CoV-2 can mitigate the need for a BSL-3 facility for matrix extension validation of alternative methods in SARS-CoV-2 studies. The method originally included 3 target regions within the nucleocapsid gene, N1, N2 and N3, but the N3 target was later dropped, and the current method includes the N1 and N2 targets only (4) . In addition, the CDC method includes a primer/probe set for detection of the human RNase P gene (RP), which serves as an internal control for human clinical specimens. In May of 2020, the AOAC Research Institute initiated an Emergency Response Validation (ERV) process to validate multiple candidate methods for detection of SARS-CoV-2 on food-grade stainless steel surfaces through the Performance Tested Methods SM (PTM) program. Several published studies have indicated the virus is able to survive on nonporous surfaces for up to 4 days depending on conditions (5, 6, 7) . In order to use the CDC method as a reference method in the ERV-PTM project, the 8) . Briefly, swabs were moistened by dipping into a 15 mL conical tube containing 2.0 mL viral transport medium (VTM; 9). A 2" 2" test area was sampled by swiping in two directions while applying × pressure to the surface and rotating the swab. The swab was snapped at the break point and placed back in the tube containing VTM. Swab tubes were vortex mixed briefly and placed in a refrigerator at 2-8°C within 15 min of surface sampling until nucleic acid extraction was performed. The swab tubes were vortex mixed briefly after removal from refrigeration and RNA was extracted Infectious SARS-CoV-2 (isolate USA_WA1/2020) was sourced from the World Reference Center for Heat-inactivated (HI) SARS-CoV2 (isolate 2019-nCoV/USA-WA1/2020) was sourced from the American Type Culture Collection (ATCC VR-1986HK; Manassas, VA) and was received as a frozen stock. The HI virus was prepared at ATCC from cultured virus (1.6 10 5 TCID 50 /mL) treated at 65°C for 30 min × according to product documentation. The certificate of analysis reported no visible cytopathic effect when ≥10% of the heat-treated seed was incubated with Vero E6 cells at 37°C with 5% CO 2 for 7 days. Upon receipt, the HI viral suspension was thawed once to remove an aliquot for determination of genomic copy (GC) concentration, then refrozen and stored at -70°C until use. RT-qPCR using the CDC N1 primer/probe set and a standard curve of synthetic SARS-CoV-2 RNA (ATCC VR-3276SD; Manassas, VA) was performed as described above to determine the GC concentration of the infectious and HI viral suspensions. The synthetic RNA standard curve consisted of 1 10 1 , 1 × × 10 2 , 1 10 3 , 1 10 4 , and 1 10 5 GC/µL. Viral stocks were serially decimally diluted in nuclease-free × × × water to a 10 -5 dilution. RNA standards and viral stock dilutions were analyzed in triplicate by RT-qPCR using the N1 primer/probe set as described above. A standard curve was generated by plotting Cq values against the log 10 RNA concentration. Linear regression analysis was performed and used to determine the RNA concentrations of the viral stocks based on the mean Cq values of triplicate analyses of all dilutions that were within the standard curve range. The presence of infectious virus in the WRCEVA stock was verified using cell culture. Approximately ScholarOne Support phone: 434-964-4100 email: ts.mcsupport@thomson.com Page 6 of 13 3 10 6 Vero E6 cells (ATCC, Manassas, VA) were plated into a T75 flask with 15 mL DMEM/F12 and × incubated overnight at 37 ± 1°C in a humidified incubator with 5% CO 2 . The following day, the DMEM/F12 was replaced with fresh medium and inoculated with an aliquot of virus stock. The inoculated cells were incubated for 5 days at 37 ± 1°C in a humidified incubator with 5% CO 2 . Determination of cytopathic effect (cpe) was made by microscopic examination. Dilutions of infectious viral stock were prepared in VTM from two pooled frozen aliquots. Dilutions of HI viral stock were made in VTM from an aliquot of a single frozen vial from ATCC. Infectious and HI virus stocks were diluted to 1. Stainless steel plates (grade 304, 12" 12") were selected to mimic food preparation surfaces. The × plates were cleaned prior to use by first treating with Goo-Gone to remove tape residue (if needed) followed by washing with a dish soap solution and rinsing with water. The plates were then disinfected by wiping with a 10% bleach wipe, thoroughly rinsed with water, then wiped with a 70% isopropanol solution, and autoclaved (121°C for 30 min) prior to use. Test areas ( 21 h) at room temperature. During the inoculation and drying process, the room temperature was in the range 21.5 to 24.6°C and relative humidity varied from 36 to 56%. To compare the infectious and heat-inactivated SARS-CoV-2 virus, twenty test areas were inoculated at the low level and five test areas were inoculated at the high level for each virus preparation. In addition, 5 test areas received VTM alone as a negative control. The fifty-five test areas were sampled with swabs as described above, the resulting swab tubes were labeled with random ID numbers, and the swab tubes were held at 2-8°C for 24 h to simulate overnight shipment of swabs to a testing laboratory. A second analyst not familiar with the ID key code performed the nucleic acid extraction and RT-qPCR according to the modified CDC method as described above. Statistical analysis of results was performed according to the probability of detection (POD) model (11) . Infectious SARS-CoV-2 was confirmed to be infective by observance of cpe after microscopic examination of inoculated Vero E6 cells incubated for 5 days as described. Genomic content was calculated from RT-qPCR with the CDC method N1 primer/probe set. Results of triplicate analyses of viral stocks at 1:100, 1:1000, 1:10,000, and 1:100,000 dilutions were used in the calculations. Triplicate determinations were subjected to statistical analysis by the Grubbs' test and one outlier in the 1:100,000 dilution of the heat-inactivated virus was removed. Viral stocks were thus determined to contain 1.6 10 9 GC/mL (SD r = 2.2 10 8 GC/mL, RSD r = 13.8%) and 9.8 10 7 GC/mL (SD r = 2.4 × × × × 10 7 GC/mL, RSD r = 24.5%) for the infectious and heat-inactivated viral stocks, respectively. The latter value is slightly lower (within 2-fold) than the reported value of 1.9 10 8 GC/mL provided by ATCC as × determined by Droplet Digital PCR for the heat-inactivated viral stock. For the comparison study, viral stocks were diluted to 1.3 10 4 GC/mL for the low level and 1.3 × × 10 5 GC/mL for the high level and 150 µL aliquots were applied to the test areas, resulting in 2.0 10 3 × GC/test area for the low level and 2.0 10 4 GC/test area for the high level. Table 1 presents the results × of the comparison study. Since the internal control, RP, was not applicable for surface testing (human DNA is not expected to be present), there was no determination of invalid results due to PCR inhibition. The external positive and negative controls produced expected results and all negative control test areas yielded negative results for both N1 and N2. The smallest bias is observed for the N1/N2 results with dPOD HI = -0.3 (-0.52, -0.05) and the largest bias is observed for the N1+N2 results with dPOD HI = -0.65 (-0.80, -0.35). When the 95% confidence interval of the dPOD does not include zero, the bias is considered significant at the 5% level. Therefore, for the low level of inoculum, the bias observed between the infectious and heat-inactivated virus as measured by the modified CDC method is statistically significant for all targets and combinations of targets. Since the quantification of the viral stocks was based on the N1 primer/probe set and the detection of virus from the inoculated stainless-steel surfaces was based on the N1 and N2 primer/probe sets, the expectation is that the N1 POD results would therefore be equivalent, but this is not the case. Even for the N1 target alone, the dPOD HI is statistically significant at the low inoculation level. It is important to note that no difference in detection rate was observed at the high inoculation level for N1, N2, N1+N2, or N1/N2. There are several possible factors contributing to the observed significant difference between detection of heat-inactivated and infectious virus at the low level. Unlike a method comparison study, this study compares the detection of two distinct preparations of virus by a single method. When working at fractionally positive concentrations, small changes or errors in concentration could have large effects on the probability of detection, depending on the slope of the POD curve. Since we don't know the shape of the POD curve for either of the virus preparations, we cannot estimate the effect of the error in the estimation of genomic copy concentration by RT-qPCR (RSD r 13.8% and 24.5% for the infectious and heat-inactivated preparations, respectively) on the POD results. Other potential contributing factors come from the difference between the determination of GC concentration of viral stocks and detection of virus on dried surface areas. These factors include differential recovery of the viral preparations from the stainless-steel surface, differential release from the foam-tipped swab, Disease outbreak news: Pneumonia of unknown cause -China US Food and Drug Administration US Centers for Disease Control and Prevention (2020) 2019-Novel Coronavirus (2019-nCoV) Real-Time RT-PCR Diagnostic Panel: Instructions for Use Surface sampling of coronavirus disease (COVID-19): A practical "how to" protocol for health care and public health professionals US Centers for disease Control and Prevention (2020) SOP# DSR-052-05, Preparation of Viral Transport Medium ScholarOne Support The authors thank Laura Rose, US Centers for Disease Control and Prevention, and Sanjiv Shah, US Environmental Protection Agency, for advice on modifying the CDC method for environmental use. The authors thank Laura Rose, Efi Papafragkou, US Food and Drug Administration, and Jacquelina Woods, US Food and Drug Administration, for helpful comments on the manuscript All authors declare no conflict of interest.