key: cord-0843165-5hquvmxa
authors: Kim, Ji Yeun; Bae, Joon-Yong; Bae, Seongman; Cha, Hye Hee; Kwon, Ji-Soo; Suh, Mi Hyun; Lee, Hyun Jung; Jung, Jiwon; Kim, Min Jae; Cui, Chunguang; Park, Heedo; Lee, Jungmin; Park, Man-Seong; Kim, Sung-Han
title: Diagnostic usefulness of subgenomic RNA detection of viable SARS-CoV-2 in patients with COVID-19()
date: 2021-08-13
journal: Clin Microbiol Infect
DOI: 10.1016/j.cmi.2021.08.009
sha: f5053641adb17b6d5d1eeb060ef01fb0fddeed13
doc_id: 843165
cord_uid: 5hquvmxa

OBJECTIVES: The development of a rapid diagnostic test for viable SARS-CoV-2 is important for infection control. Real-time RT-PCR assays detect non-viable virus and cell culture differentiates viable virus but it takes several weeks and is labor-intensive. Subgenomic RNAs may reflect replication-competent virus. We therefore evaluated the usefulness of subgenomic RNAs for diagnosing viable SARS-CoV-2 in patients with COVID-19. METHODS: Patients with various severities of confirmed COVID-19 were enrolled at a tertiary hospital between February and December 2020. RT-PCR assay results for genomic and subgenomic RNA of SARS-CoV-2 from nasopharyngeal swab, sputum, and saliva specimens were compared with cell culture results. RESULTS: A total 189 specimens from 20 COVID-19 patients were tested in genomic and subgenomic PCR assays, and cultured on Vero cells. Of these 189 samples, 62 (33%) gave positive culture results, 93 (49%) negative results, and the remaining 34 (18%) indeterminate results. Compared with cell culture results, the sensitivities of genomic RNA and subgenomic RNA of the N and S genes were comparable as 100%, but the specificity of subgenomic RNA (N, 65% and S, 68%) was higher than that of genomic RNA (N, 23% and S, 17%, p value < 0.001). The mean durations of positive culture and subgenomic RNA were 11.39 ± 10.34 and 13.75 ± 11.22 days after symptom onset (p value = 0.437), respectively, while that of genomic RNA was 22.85 ± 11.83 days after symptom onset (p value < 0.001). CONCLUSION: Our comparison of subgenomic RNA detection with symptom duration and SARS-CoV-2 culture positivity provides a significant advancement on the transmissibility-based approach beyond the detection of SARS-CoV-2 genomic RNA, and warrants further studies on the development of better diagnostic strategy.

Diagnosis of COVID-19 has relied on detection of SARS-CoV-2 viral RNA by real-time reverse transcription PCR (RT-PCR). This method is rapid and highly sensitive, but the viral RNA may be detectable for many weeks after clinical recovery [1, 2] , which could unnecessarily prolong patients' isolation. In addition, we do not know, on a real-time basis, the viability of SARS-CoV-2 in patients confirmed by conventional RT-PCR of SARS-CoV-2, so some unnecessary contact tracing with quarantine is inevitable. Therefore, the development of a rapid diagnostic test for viable SARS-CoV-2 is important for conserving resources, and compliance with quarantine and isolation polices.

Culture-based isolation of SARS-CoV-2 virus is the best indicator of the presence of replicating virus, but it is difficult, labor-intensive, and time-consuming. In addition, it is further complicated by the need for Biosafety Level-3 facilities. Meanwhile, coronaviruses have a unique mechanism of discontinuous transcription involving the synthesis of subgenomic RNA [3] . Coronaviruses subgenomic RNAs contain a nested set of negativesense RNAs from the 3' end of the virus genome joined to a common leader sequence of about 70 nucleotides derived from the 5' end of the genomic RNA [4, 5] . Therefore, the detection of subgenomic RNAs may better reflect replication-competent virus than detection of conventional genomic RNA [6] . However, there have been limited studies with regard to the correlation between replicative SARS-CoV-2 and subgenomic RNAs in respiratory samples [6] [7] [8] . We had a good opportunity to study COVID-19 patients of various severities including immunocompromised patients who were expected to shed viable virus for a prolonged period. We therefore evaluated the utility of subgenomic RNAs for diagnosing viable SARS-CoV-2 in patients with COVID-19.

J o u r n a l P r e -p r o o f

Respiratory specimens were collected from patients with confirmed COVID-19 who agreed to multiple serial sampling and were admitted to Asan Medical Center, a 2,700-bed tertiary hospital in Seoul, Republic of Korea, between February 2020 and December 2020. Diagnosis of COVID-19 was confirmed with nasopharyngeal (NP) swab samples by real-time reverse transcription polymerase chain reaction (RT-PCR) for the RdRp, N, and E genes of SARS-CoV-2 using the Allplex TM 2019-nCoV assay (Seegene, Seoul, Republic of Korea) or a

PowerChek 2019-nCoV real-time PCR kit (Kogene Biotech, Seoul, Republic of Korea). NP swab samples, sputum and saliva samples were collected from the patients between the initial day of admission and the day of discharge, up to 50 days after symptom onset. NP swab samples in viral transport medium, and sputum and saliva samples collected in an airtight container, were aliquoted and stored in a -80°C deep freeze until use. This study was reviewed and approved by the ethics committee of the institutional review board of Asan Medical Center (IRB No. 2020-0297), and all participants gave written informed consent.

The collected respiratory samples were inactivated at 65℃ for 30 min in a special negative pressure laboratory. Viral RNA was extracted from the respiratory specimens using a Table 2 ). RT-PCR reactions were performed as described in Supplementary Materials.

Culture-based isolation of SARS-CoV-2 from respiratory specimens was performed by a plaque assay in a Biosafety Level 3 laboratory at Korea University College of Medicine, Seoul, Republic of Korea. Detailed procedures are described in Supplementary Materials.

Categorical variables were compared using Fisher's exact test, and continuous variables with the Mann Whitney U-test, as appropriate. The diagnostic performance for each test was calculated and compared by using logistic regression with generalized estimating equation to account for the clustering effects. All tests of significance were 2-tailed; p values of ≤ 0.05 were considered significant. IBM SPSS Statistics for Windows, version 23.0 (IBM Corp., Armonk, NY, USA) was used for these statistical analyses.

Twenty confirmed COVID-19 patients were included in this study, and the clinical characteristics of these patients are presented in Table 1 

A total 189 respiratory specimens from the 10 COVID-19-confirmed patients were analyzed.

There were 74 nasopharyngeal swab specimens, 62 sputum specimens, and 53 saliva specimens, and the collection times ranged from 1 to 49 days after symptom onset. The N and S genes were detected by genomic RNA RT-PCR assays (limits of detection: 2.6 Log copies/ml of specimen) in 160 (N gene) and 161 (S gene) of the 189 specimens. Subgenomic RNA was detected in 108 (N gene) and 107 (S gene) specimens, and viral loads in the positive subgenomic RNA-specimens were 6.21±1.32 log copies/ml and the viral loads in the negative subgenomic RNA specimens 3.35±0.67 log copies/ml (p < 0.001, Mann Whitney Utest). Sixty-two specimens were successfully cultured and viral loads were 6.76±1.16 log copies/ml; 93 specimens were not cultured and the viral loads were 4.26±1.24 log copies/ml The prediction performances for viable SARS-CoV-2 were analyzed with the culture-valid 155 samples ( Table 2 ). All of the 62 culture-positive specimens were positive for genomic and subgenomic RNA, so the sensitivity of detection of genomic and subgenomic RNA in culture-positive cases was 100% (95% CI, 93-100%) ( Figure 1 and Table 2 ). For the N gene, 21/93 and 60/93 culture negative specimens were negative for genomic and subgenomic RNA; thus the specificities of genomic and subgenomic RNA for culture negative specimens were 23% (95% CI, 14-34%) and 65% (95% CI, 53-75%), respectively (p value < 0.001). For the S gene, the specificities for genomic and subgenomic RNAs among the culture negatives were 17% (16/93, 95% CI, 11-26%) and 68% (95% CI, 56-78%), respectively (p value < 0.001).

Depending on the sample site, NP swab samples showed the highest specificity of subgenomic RNA, 69% (N gene, 22/32, 95% CI, 54-80%) and 72% (S gene, 23/32, 95% CI, 57-83%), and saliva samples showed the lowest specificity of subgenomic RNA, 54% (N gene, 13/27, 95% CI, 41-67%) and 58% (S gene, 14/24, 95% CI, 44-71%)( Table 2 ). This study included 3 patients with immunocompromised conditions. The sensitivity and specificity of genomic and subgenomic RNA in patients with or without immunocompromised condition were presented in Supplementary The distributions of positive subgenomic RNA and culture test results as a function of number of days after symptom onset are shown in Figure 2 and Table 3 Table 5 ). The corresponding data for subgenomic RNAs were detected from 23 of 25 specimens with viral loads ≥ 7.0 log copies/ml, 28 of 32 specimens with viral loads 6.0-6.99 log copies/ml, 30 of 38 specimens with viral loads 5.0-5.99 log copies/ml, 9 of 26 specimens with viral loads 4.0-4.99 log, and 2 of 29 specimens with viral loads 3.0-3.99 log copies/ml, none of 39 specimens with viral loads < 3.0 log copies/ml based on copy number of the S gene are also presented in Supplementary Table 5 .

The idea of subgenomic RNA detection is not novel. Two studies have reported that J o u r n a l P r e -p r o o f replicative virus determined by viral culture and subgenomic RNA detection was not found beyond 8 or 9 days after symptom onset in patients with mild illness [6, 7] . However, viral shedding has been observed for longer in patients with severe disease [9] [10] [11] . Van Kampen et al. detected culturable virus up to 20 days after symptom onset in patients with moderate-tosevere disease [12] . We extensively investigated whether the subgenomic RNA detect can Taken together, our study results indicate that the assay for subgenomic RNA predicts the duration of infective viral shedding accurately, as its duration was similar to that of culturable virus and its highest positive rate at the earliest period from the symptom onset.

Most of specimens with viral loads ≥ 6.0 log copies/ml were positive for both culture and Table 5 ). Although, in the cell culture system we used, viable virus could still be detected in dilutions containing only 1 PFU/mL [13] , some invalid culture result could not rule out the possibility that some viable virus might only be detected by subgenomic RNA PCR. However, subgenomic SARS-CoV-2 RNA tightly associated with membrane structures might be protected from cellular RNases, so the detection of subgenomic RNAs from infected patients may not always be indicative of viable virus [8] . In addition, the previous study on the comparison of levels of viral subgenomic RNAs to viral genomic RNAs in the same sample indicated the decreasing ratio from specimens in patients with the late course of COVID-19 [7] . So, it is interesting whether a certain cut-off value of the ratio of viral subgenomic RNAs to viral genomic RNAs may more exactly reflect the viable viral shedding. Further studies are needed in this area.

The detection of genomic RNA has greatly contributed for the early diagnosis of patients with SARS-CoV-2 infection. However, the genomic RNA detection that cannot differentiate viable virus from non-viable virus inevitably results in the prolonged isolation of patients with SARS-CoV-2 infection and the unnecessary quarantine of contacts with those with SARS-CoV-2 infection who have less chance to transmit the virus. The recent studies that the rapid antigen test (RAT) but not genomic RNA detection was correlated with SARS-CoV-2 viral culture [14, 15] . However, many other studies raised the issues on the suboptimal sensitivity of RAT [16, 17] . In this context, our findings showing that the subgenomic RNA detection had comparable sensitivity to the genomic RNA detection and higher specificity for the viable virus than the genomic RNA detection further facilitate our venture toward the transmissibility-based approach instead of the diagnosis-based approach. For example, the surveillance genomic PCR test may be positive in asymptomatic individuals. If these persons do not have any known exposure to the confirmed SARS-CoV-2 infected patients, the J o u r n a l P r e -p r o o f subgemoic RNA detection may be useful for the epidemiologic investigation and isolation policy. In addition, the subgenomic RNA detection may objectively guide the de-isolation policy for immunocompromised patients with COVID-19 or those with severe COVID-19.

Our study has some limitations. First, the subgenomic RNA PCR assay we developed was not quantitative. A quantitative assay such as real-time PCR might provide more reliable information about the presence of viable virus. We are currently developing real-time PCR assays against subgenomic RNA. Second, some may argue that the cell culture procedure that is considered the gold standard for detecting viable SARS-CoV-2 is not so sensitive that underestimation of viable virus is possible. Therefore, a clinical parameter such as symptom duration might be more valuable; hence recent guidelines have symptom cessation as the indicator for discontinuing transmission-based precautions. We therefore analyzed positivity for subgenomic RNA, genomic RNA and cell culture as a function of time of symptom onset (Table 3 and Supplementary Table 4 ). However, it is difficult to estimate the time of symptom onset in some patients due to pauci-or null symptoms. Thus, our evidence that detection of subgenomic RNA is closely correlated with symptom duration may facilitate objective testbased discontinuation of precautions. Finally, we did not evaluate the diagnostic performance of various genetic variants including alpha-, beta,-gamma-, delta-variants because there were limited reported cases in South Korea during the study period.

In conclusion, this study shows that subgenomic RNA was detected for a few days after viral culture turns negative, but the mean duration of viral shedding assessed by subgenomic RNA detection was much more similar to that of virus culture in moderate-to-critical patients than that assessed by detecting genomic RNA. Therefore, subgenomic RNA could be a useful surrogate for predicting the duration of infectious viral shedding. 

Temporal profiles of viral load in posterior oropharyngeal saliva samples and serum antibody responses during infection by SARS-CoV-2: an observational cohort study

Dynamic profile of RT-PCR findings from 301 COVID-19 patients in Wuhan, China: a descriptive study

A contemporary view of coronavirus transcription

Continuous and discontinuous RNA synthesis in coronaviruses

A unifying structural and functional model of the coronavirus replication organelle: tracking down RNA synthesis

SARS-CoV-2 Virus Culture and Subgenomic RNA for Respiratory Specimens from Patients with Mild Coronavirus Disease

Virological assessment of hospitalized patients with COVID-2019

SARS-CoV-2 genomic and subgenomic RNAs in diagnostic samples are not an indicator of active replication

Chronological changes of viral shedding in adult inpatients with COVID-19 in Wuhan, China

Analysis of factors affecting the prognosis of COVID-19 patients and viral shedding duration

Prolonged persistence of SARS-CoV-2 RNA in body fluids

Duration and key determinants of infectious virus shedding in hospitalized patients with coronavirus disease-2019 (COVID-19)

Duration of Culturable SARS-CoV-2 in Hospitalized Patients with Covid-19

Antigen-Based Testing but Not Real-Time Polymerase Chain Reaction Correlates With Severe Acute Respiratory Syndrome Coronavirus 2

Viral Culture

Evaluation of Abbott BinaxNOW Rapid Antigen Test for SARS-CoV-2 Infection at Two Community-Based Testing Sites -Pima County

Ag Test for the Detection of SARS-CoV-2 Infection

Surveillance in the United States: Insights From the National Football League Occupational Health Program