key: cord-0773849-qxivrhj8
authors: Gniazdowski, V.; Morris, C. P.; Wohl, S.; Mehoke, T.; Ramakrishnan, S.; Thielen, P.; Powell, H.; Smith, B. D.; Armstrong, D. T.; Herrera, M.; Reifsnyder, C.; Sevdali, M.; Carroll, K. C.; Pekosz, A.; Mostafa, H. H.
title: Repeat COVID-19 Molecular Testing: Correlation with Recovery of Infectious Virus, Molecular Assay Cycle Thresholds, and Analytical Sensitivity
date: 2020-08-06
journal: nan
DOI: 10.1101/2020.08.05.20168963
sha: 70eed0b9c04b5b1b5b0950696a7fd066654a73ae
doc_id: 773849
cord_uid: qxivrhj8

Repeat molecular testing for SARS-CoV-2 may result in scenarios including multiple positive results, positive test results after negative tests, and repeated false negative results in symptomatic individuals. Consecutively collected specimens from a retrospective cohort of COVID-19 patients at the Johns Hopkins Hospital were assessed for RNA and infectious virus shedding. Whole genome sequencing confirmed the virus genotype in patients with prolonged viral RNA shedding and droplet digital PCR (ddPCR) was used to assess the rate of false negative standard of care PCR results. Recovery of infectious virus was associated with Ct values of 18.8 {+/-} 3.4. Prolonged viral RNA shedding was associated with recovery of infectious virus in specimens collected up to 20 days after the first positive result in patients who were symptomatic at the time of specimen collection. The use of Ct values and clinical symptoms provides a more accurate assessment of the potential for infectious virus shedding.

Repeat molecular testing for SARS-CoV-2 may result in scenarios including multiple positive 24 results, positive test results after negative tests, and repeated false negative results in 25 symptomatic individuals. Consecutively collected specimens from a retrospective cohort of 26 COVID-19 patients at the Johns Hopkins Hospital were assessed for RNA and infectious virus 27

shedding. Whole genome sequencing confirmed the virus genotype in patients with prolonged 28 viral RNA shedding and droplet digital PCR (ddPCR) was used to assess the rate of false 29 negative standard of care PCR results. Recovery of infectious virus was associated with Ct 30 values of 18.8 ± 3.4. Prolonged viral RNA shedding was associated with recovery of infectious 31 virus in specimens collected up to 20 days after the first positive result in patients who were 32 symptomatic at the time of specimen collection. The use of Ct values and clinical symptoms 33 provides a more accurate assessment of the potential for infectious virus shedding. 34

Molecular methods for SARS-CoV-2 nucleic acid detection from nasopharyngeal swabs have 37 been the gold standard for COVID-19 diagnosis. Although diagnostic approaches target 38 different genes within the SARS-CoV-2 genome, they have shown comparable analytical 39 sensitivity and high specificity (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) . Sensitivity of the assay is associated with the shedding 40 pattern of SARS-CoV-2 RNA, which can vary based on the source of respiratory specimen and 41 based on the course of illness (18) (19) (20) (21) . 42

Infection control personnel and physicians managing COVID-19 patients and patients under 43 investigation (PUI) continue to face several diagnostic dilemmas related to a lack of 44 understanding of the clinical sensitivities of SARS-CoV-2 molecular diagnostics and the 45 correlation between viral RNA detection and shedding of infectious virus. Retesting of patients 46 has become a common practice especially when there is a strong clinical suspicion or exposure 47 history and there is an initial negative result (22) . A single positive molecular result should be 48 sufficient for confirming COVID-19 diagnosis, however, repeated testing of hospitalized patients 49 for determining isolation needs and infection control measures has become a part of managing 50 this patient population. Two negative molecular assay results from two consecutively collected 51 respiratory specimens more than 24 hours apart has been the strategy used for discontinuation 52 of transmission precautions and returning to work (23). Repeat testing on patients has revealed 53 that SARS-CoV-2 RNA can be detectable for weeks after the onset of symptoms (24). In 54 addition, there have been reports of patients who had initial negative molecular tests that tested 55 positive on subsequent tests. In general, molecular detection of SARS-CoV-2 RNA does not 56 necessarily denote the presence of recoverable infectious virus. A few studies, as well as data 57 from the CDC, showed that higher viral loads are associated with recovery of infectious virus 58 and that virus recovery is generally not reported after 9 days from symptom onset (20, 25, 26) . 59

A case study, in which severe infection was associated with recovery of infectious SARS-CoV-2 60 France) were used using software version 2.1.0.1. The input specimens' volumes were 500 µL 108 and the final elution volume was 50 µL. Specimens for automated systems were processed 109 following each assay's FDA-EUA package insert. 110

VeroE6 cells (ATCC CRL-1586) were cultured at 37°C with 5% carbon dioxide in a humidified 112 chamber using complete medium (CM) consisting of Dulbecco's modified Eagle Medium 113 supplemented with 10% fetal bovine serum (Gibco), 1mM glutamine (Invitrogen), 1mM sodium 114 pyruvate (Invitrogen), 100µg/mL penicillin (Invitrogen) and 100 µg/mL streptomycin (Invitrogen). 115

Cells were plated in 24 well dishes and grown to 75% confluence. The CM was removed and 116

replaced with 150 µL of infection media (IM) which is identical to CM but with the fetal bovine 117 serum reduced to 2.5%. Fifty µL of the clinical specimen was added to one well and the cells 118 incubated at 37°C for one hour. The inoculum was aspirated and replaced with 0.5 ml IM and 119 the cells cultured at 37°C for 4 days. When cytopathic effect was visible in most of the cells, the 120 IM was harvested and stored at -70°C. The presence of SARS-CoV-2 was verified by one of two 121 ways. SARS-CoV-2 viral RNA was extracted using the Qiagen Viral RNA extraction kit (Qiagen) 122 and viral RNA detected using quantitative, reverse transcriptase PCR (qPCR) as described (47). 123 SARS-CoV-2 viral antigen was detected by infecting VeroE6 cells grown on 4 chamber LabTek 124 slides (Sigma Aldrich) with 50 µL of virus isolate diluted in 150 µL of IM for 1 hour at 37°C. The 125 inoculum was replaced with IM and the culture incubated at 37°C for 12-18 hours. The cultures 126 were fixed with 4% paraformaldehyde for 20 minutes at room temperature and processed for 127 indirect immunofluorescence microscopy as described (48). The humanized monoclonal 128 antibody D-006 (Sino Biological) was used as the primary antibody to detect Spike or S protein, 129 followed by Alexa Fluor 488-conjugated goat anti-human IgG. The cells were mounted on 130

Prolong antifade and imaged at 40X on a Zeiss Axio Imager M2 wide-field fluorescence 131 microscope (49). 132 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted August 6, 2020 . . https://doi.org/10.1101 Oxford Nanopore whole genome sequencing 133

Whole genome sequencing was conducted using the Oxford Nanopore platform following the 134 ARTIC protocol for SARS-CoV-2 sequencing with the V3 primer set (50). Eleven indexed 135 samples (and one negative control) were pooled for each sequencing run and 20 ng of the final 136 pooled library was run on the Oxford Nanopore GridION instrument with R9.4.1 flowcells. 137

Basecalling and demultiplexing was performed with Guppy v3.5.2 and reads were assembled 138 using a custom pipeline modified from the ARTIC network bioinformatics pipeline 139 (https://artic.network/ncov-2019). As part of this custom pipeline, reads were mapped to a 140 SARS-CoV-2 reference genome (GenBank MN908947.3) using minimap2 (51). Coverage was 141 normalized across the genome and variant calling was performed with Nanopolish v0.13.2 (52). 142

Sites with low coverage (based on the negative control coverage) were masked as 'N'. Variant 143 calls were also independently validated with two other variant callers-medaka 144 (https://nanoporetech.github.io/medaka/snp.html) and samtools( 145 https://wikis.utexas.edu/display/bioiteam/Variant+calling+using+SAMtools)-and all sites with 146 disagreements or allele frequency <75% were manually inspected in Integrated Genome Viewer 147 (53). Sites with minor allele frequency 25-75% were replaced with IUPAC ambiguity codes. 148

The ddPCR procedure followed the assay's EUA package insert (54). Briefly, RNA isolated 150 from NP specimens (5.5 µL) were added to the mastermix comprised of 1.1 µL of 2019-nCoV 151 CDC ddPCR triplex assay, 2.2 µL of reverse transcriptase, 5.5 µL of supermix, 1.1 µL of 152 Dithiothreitol (DTT) and 6.6 µL of nuclease-free water. Twenty-two microliters from these 153 samples and mastermix RT-ddPCR mixtures were loaded into the wells of a 96-well PCR plate 154 (Bio-Rad, Pleasanton, CA). The mixtures were then fractionated in up to 20,000 nanoliter-sized 155 droplets in the form of a water-in-oil emulsion in the Automated Droplet Generator (Bio-Rad, 156

Pleasanton, CA). The 96-well RT-ddPCR ready plate containing droplets was sealed with foil 157 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 6, 2020. . https://doi.org/10.1101/2020.08.05.20168963 doi: medRxiv preprint using a plate sealer (Bio-Rad, Pleasanton, CA) and thermocycled to achieve reverse 158 transcription of RNA followed by PCR amplification of cDNA in a C1000 Touch thermocycler 159 (Bio-Rad, Pleasanton, CA). Following PCR, the plate was loaded into the QX200 Droplet 160

Reader (Bio-Rad, Pleasanton, CA); the droplets in each well were singulated and flowed past a patients had an initial positive result that was followed by a negative test ( figure 1A and B). Our 175 data indicates that of all the patients that had repeat testing, 81.5% continued to have negative 176 results, 5.7% had an initial negative followed by a repeat positive test, and 6.8% had a final 177 negative test result after an initial positive (figure 1B). 178

Infectious virus isolation and viral RNA load. To understand the correlation between a positive 180 molecular result and virus recovery, 161 patients' specimens that were positive by molecular 181 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 6, 2020. median Ct values associated with recoverable virus were 18.8 ± 3.4 and 18.17 respectively, 185 which was significantly lower than the mean and median Ct values that did not correlate with 186 infectious virus recovery (27.1 ± 5.7 and 27.5 respectively) (paired t test, P<0.0001) ( Figure 2) . 187

Samples with a Ct value below 23 yielded 91.5% of virus isolates. However, 28.6% of 188 specimens that were negative for viral growth on VeroE6 cells were in that same Ct value range 189 ( Figure 2 ) and 11.9% were below a Ct value of 20. 190

with longitudinal positive results were tested within a time frame that ranged from less than one 192 day to more than 45 days. To assess the correlation between the repeated positivity, viral loads, 193 and recovery of infectious virus, we evaluated a randomly selected subset of 29 patients. We 194 examined the Ct values of all test results, days between testing, as well as viral growth on cell 195 culture (if performed) (Table 1) . Except for two patients (#24 and 25) (and the first three whose shedding was not associated with a specific outcome as one patient was never admitted (# 24), 202 one was hospitalized with no oxygen requirements (# 10), and two had more severe disease (# 203 8 and #29). Recovery of infectious virus was associated with persistence of symptoms in all but 204 one patient (# 24). Longitudinal specimens of patients were sequenced to assess any changes 205

in the viral genome that could have resulted in prolonged shedding or could possibly suggest a 206 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 6, 2020. . https://doi.org/10.1101/2020.08.05.20168963 doi: medRxiv preprint reinfection. The successful recovery of complete viral genome sequences at multiple time points 207 from 7 patients provided evidence that these patients were carrying the same virus over time, 208 however in one case, the second time-point sample had additional variants, and in two cases 209 minor variants appear in the later time point sample (denoted as IUPAC ambiguity codes, since 210 two alleles are present in the sequencing reads) ( Table 2 ). Of note, two different isolates 211 collected from patient #14 in the same day were included in this analysis for validating our 212 sequencing reproducibility. (Table 3) . Attempted recovery of infectious virus from these specimens 219 was negative. 220

Repeat negative testing of patients with clinical disease or exposure history with 1,788 patients were tested more than once between March 11 th and May 11 th 2020 without any 222 positive result. To examine the possibility of false negative results of the standard of care 223 molecular SARS-CoV-2 diagnostic assay due to limitations in the analytical sensitivity, we used 224 the SARS-CoV-2 droplet digital PCR (ddPCR). We selected 198 negative from 185 patients that 225 received repeated testing over time, of which 163 patients had from 2 to up to 5 negative 226 results. We selected 15 that had positive SARS-CoV-2 serology and multiple negative RT-PCR 227 results. A few included 22 specimens from patients who had an initial positive result but turned 228 negative on a repeat test or the reverse. Of the total 198 tested, 11 specimens were positive by 229 ddPCR (Table 4 ). Only one patient who had a positive serology test (patient # 51) had a positive 230 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 6, 2020. . https://doi.org/10.1101/2020.08.05.20168963 doi: medRxiv preprint ddPCR result and 4 of the 11 patients had positive specimens by RT-PCR collected at other 231 days (54-57). 232

The molecular detection of SARS-CoV-2 genome has been valuable not only in diagnosis, but 234 also in making decisions related to infection control measures and return to work. Several 235 outcomes were observed with repeat molecular testing including: i) prolonged, consistent viral 236 RNA shedding, ii) alternating negative results and positive RNA shedding, and iii) false negative 237 results. Our data shows that prolonged positivity could be associated with recovery of infectious 238 virus especially when symptoms persist. Our data also shows that RNA positive specimens after 239 a negative result are not associated with recovery of infectious virus. 240

The ddPCR assay detected a few positives that were missed by our standard of care testing in 241 the subset of patients who were highly suspected of infection based on clinical symptoms. 242

Overall, our data confirms that SARS-CoV-2 RNA is detectable for a prolonged time, and 243 recovery of infectious virus is associated with persistent symptoms. Importantly, our data also 244

shows that the standard of care molecular diagnostics' analytical sensitivities are affected by the 245 shedding pattern of the viral RNA rather than the assay's performance. 246

The use of a diagnostic test's Ct values as an indicator of the presence of infectious virus has 247 been proposed. One report suggested that a Ct value above 33-34 is not associated with 248 recovery of infectious virus (55) and another report concluded that cell culture infectivity is 249 observed when the Ct values were below 24 and within 8 days from symptoms onset (25). Our 250 data shows that the average Ct value that was associated with cell culture growth is 18.8. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 6, 2020. DdPCR showed a slightly higher sensitivity in detecting SARS-CoV-2 RNA in a subset of 277 specimens from patients with high suspicion of COVID-19 and negative standard RT-PCR. Our 278 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 6, 2020 . . https://doi.org/10.1101 data is consistent with published reports that compared ddPCR with real-time PCR (33). It is 279 important to note that the analytical sensitivity of the ddPCR assay as reported by the EUA 280 package insert (645 copies/ mL) is comparable to standard of care real-time PCR methods we 281 use in our diagnostic laboratories that include the CDC panel assay among others (3) and all of 282 the positives detected by the ddPCR assay in this study were below the ddPCR assay's 283 analytical limit of detection (Table 4 ). The Bio-Rad ddPCR assay uses primers and probes that 284 are same as reported by the CDC assay and also includes the human RNase P gene as an 285

internal control. Including this control is very valuable to exclude insufficient sampling as a 286 cause of false negative results (64). Only a few samples that tested negative by the standard 287 PCR methodologies were later positive by ddPCR (5.7%), even in a cohort with a high suspicion 288 of COVID-19. A few samples showed conflicting results when repeated (Table 4) , likely 289 because of viral loads below the lower limit of detection of the ddPCR assay. Overall, this 290 suggests that false negative results in some cases are secondary to low viral loads likely 291 associated with temporal aspects of viral shedding. 292

Our study indicates that prolonged viral RNA shedding is associated with recovery of infectious 293 virus in a subset of patients and seems to correlate with persistence of symptoms. Higher Ct 294 values and positive RNA tests detected after viral RNA clearance were not associated with 295 recovery of infectious virus in our tested cohort. DdPCR can add an increased sensitivity in 296 detecting viral RNA. Our data support the recently updated CDC guidelines for the duration of 297 isolation after a positive COVID-19 test (23). Additional studies are required to inform using Ct 298 values and cell culture results in making clinical decisions and developing diagnostic strategies 299 that can differentiate shedding versus active replication will be very valuable for infection 300

We thank the entire clinical microbiology laboratory for the rapid response to the pandemic and 303 for offering a unique testing capacity. This work was funded by the department of pathology, 304

Research and Surveillance HHSN272201400007C (A.P., H.H.M.) and T32A1007417 Molecular 306

and Cellular Basis of Infectious Diseases (H.P. and B.S.). DdPCR was performed in 307 collaboration with Bio-Rad Laboratories. We thank Winston Temp, PhD and Stuart C. Ray, MD 308 for their valuable contribution to the SARS-CoV-2 genomic analysis pipeline. 309 310 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 6, 2020. . https://doi.org/10.1101/2020.08.05.20168963 doi: medRxiv preprint CoV-2 detection from nasopharyngeal swab samples by the Roche cobas 6800 SARS-CoV-2 test 353 and a laboratory-developed real-time RT-PCR test. J Med Virol. 2020. 354 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 6, 2020. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted August 6, 2020. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted August 6, 2020. . https://doi.org/10.1101/2020.08.05.20168963 doi: medRxiv preprint Table 1 . Patients with multiple positive molecular results overtime and correlation between the time of testing, isolation of infectious virus on cell culture, and the cycle threshold (Ct) value of the diagnostic assay. *symptomatic at the time of specimen collection. N/A: Not Available 

Comparison of Four Molecular In Vitro Diagnostic Assays for 313 the Detection of SARS-CoV-2 in Nasopharyngeal Specimens

Comparison of Abbott ID 315 Now, Diasorin Simplexa, and CDC FDA EUA methods for the detection of SARS-CoV-2 from 316 nasopharyngeal and nasal swabs from individuals diagnosed with COVID-19

Comparing the analytical 319 performance of three SARS-CoV-2 molecular diagnostic assays

Rapid and sensitive detection 321 of SARS-CoV-2 RNA using the Simplexa COVID-19 direct assay

Commercially Available and Laboratory Developed Assays for in vitro Detection of SARS-CoV-2 in 324 Clinical Laboratories

Clinical Evaluation of Three Sample-To-Answer 326 Platforms for the Detection of SARS-CoV-2

Reverse Transcription-Polymerase Chain Reaction Systems for the Detection of 329 Severe Acute Respiratory Syndrome Coronavirus 2

Panther Fusion and a laboratory-developed test targeting the envelope gene for detection of 332 SARS-CoV-2

Comparison of Abbott ID Now 334 and Abbott m2000 methods for the detection of SARS-CoV-2 from nasopharyngeal and nasal 335 swabs from symptomatic patients

Performance of the rapid Nucleic Acid Amplification by Abbott ID NOW COVID-19 in 337 nasopharyngeal swabs transported in viral media and dry nasal swabs

Comparison of two commercial molecular 340 tests and a laboratory-developed modification of the CDC 2019-nCoV RT-PCR assay for the 341 detection of SARS-CoV-2

Comparison of Cepheid Xpert Xpress and 343

Now to Roche cobas for the Rapid Detection of SARS-CoV-2

Comparison of the Accula 346 SARS-CoV-2 Test with a Laboratory-Developed Assay for Detection of SARS-CoV-2 RNA in Clinical 347 Nasopharyngeal Specimens

The Detection of SARS-349

CoV-2 using the Cepheid Xpert Xpress SARS-CoV-2 and Roche cobas SARS-CoV-2 Assays

Viral Load Kinetics of MERS Coronavirus 449 Infection

Factors Associated With Prolonged Viral 451 Shedding in Patients With Avian Influenza A(H7N9) Virus Infection

Prolonged persistence of measles virus 454 RNA is characteristic of primary infection dynamics

Suboptimal biological 457 sampling as a probable cause of false-negative COVID-19 diagnostic test results

 Table 3 . Patients with positive molecular results after one or more negatives and correlation with the time of testing, isolation of infectious virus on cell culture, and the cycle threshold (Ct) value of the diagnostic assay. ND, target not detected .   1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  29  30  31  32  34  35  36  39  43  44  45  46  47  50  51  55 Table 4 . ddPCR sensitivity of detection in patients with consecutive negative results (47-53) and negative specimens collected from known positive patients (54-57). ddPCR copies shown for the N1 target. -ve: negative result by the standard of care RT-PCR. +ve: positive results by the standard of care RT-PCR with no available Ct value.