Saliva‐based Proteinase K method: A rapid and reliable diagnostic tool for the detection of SARS‐COV‐2 in children

Early and accurate detection of viruses in children might help prevent transmission and severe diseases. In this study, the severe acute respiratory syndrome coronavirus 2 (SARS‐COV‐2) detection in children was evaluated using saliva specimens with a Proteinase K (PTK)‐based RNA preparation, as saliva collection is a simple and noninvasive procedure, even in young children, with fewer concerns about sample contamination. The saliva‐based PTK and the conventional paired nasopharyngeal aspiration (NPA)‐based detection methods were compared between COVID‐19‐positive and ‐negative children. In addition, the detection rate for SARS‐COV‐2 and the difference between admission and discharge by the saliva‐based PTK method was tested in COVID‐19 patients. The diagnostic accuracy of the saliva‐based PTK method was 98.8% compared to NP swab‐based reverse transcriptase polymerase chain reaction. Saliva samples showed high sensitivity (94.1%) and specificity (100%) when using the PTK method. Furthermore, the saliva‐based PTK method significantly reduced the test processing time by 2 h. Notably, Ct values at discharge increased in saliva samples compared with those at admission, which might indicate patients' clinical conditions or virus activity. In conclusion, the saliva‐based PTK implemented in this study streamlines RNA extraction, making the process faster, safer, and more cost‐effective, demonstrating that this method is a rapid and reliable diagnostic tool for SARS‐CoV‐2 detection in children.

With the loosening of social distancing interventions and the reopening of schools, children are exposed to the danger of the coronavirus disease 2019 (COVID-19).They could also be reservoirs that spread the severe acute respiratory syndrome coronavirus 2 (SARS-COV-2).The presentation of COVID-19 in children was less severe or asymptomatic in general.Infants might present serious illness, and older children might subsequently develop severe illnesses such as multisystem inflammatory syndrome in children. 1 Yet, there are still limitations in vaccination and antiviral treatment for young children.Therefore, early detection and diagnosis of SARS-COV-2 in children, even those presenting as asymptomatic, is necessary to prevent SARS-COV-2 transmission and a severe COVID-19 prognosis.
Although other rapid platforms and alternative specimen sources have been evaluated, detecting SARS-COV-2 using nasopharyngeal (NP) specimens by reverse transcriptase polymerase chain reaction (RT-PCR) is the gold standard for diagnosis. 2,3The diagnostic steps for collecting specimens with NP swabs, extracting RNA from the sample, analyzing through RT-PCR, and reporting the results require adequate skills and are time-consuming.In particular, the collection of NP specimens from children is an arduous task. 2,4th the demand for reducing the processing time and simplifying the process for the diagnosis of COVID-19, this study attempted to implement a faster and simpler detection process by assessing comparatively easily procurable saliva specimens from children using Proteinase K (PTK)-based RNA preparation method.
Comparing SARS-CoV-2 detection from patient-matched NP and saliva samples revealed significant sensitivity and specificity in the detection of the virus in saliva.Moreover, the PTK method also showed good overall performance throughout the test.The findings from this study demonstrate that saliva is reliable for detecting SARS-COV-2 in children, even with the PTK method for RNA extraction.

| Study design
The sample size was calculated using the formula for comparing two independent proportions, which is used to estimate sample size for studies comparing the sensitivity and/or specificity of two tests of unpaired design. 5 Initially, NP swabs from children were delivered in a transport medium from each hospital.According to the manufacturer's instructions, RNA was prepared using prefilled kits and an automated nucleic acid purification instrument, MagNA Pure 96 (Roche, Switzerland).Viral RNA was detected by using Allplex™2019-nCoV Assay (Seegene Inc., Seoul, Korea) for amplification of the RNAdependent RNA polymerase (RdRp), N genes specific for SARS-CoV-2, and E genes for all Sarbecovirus, including SARS-CoV-2.The cycle threshold (Ct) values from the RT-qPCR were measured; C t values < 35 were reported as positive, and C t values between 35 and less than 40 were considered indeterminate. 6liva from all participants was collected using a sterile collection tube without any transport medium.Self-collected saliva samples up to 500 μL were obtained at least 30 min after the absence of food or water as previously described. 4If self-collection was infeasible, researchers collected saliva samples from children under 2 years of age through cotton swabs or oral suction.

| Sample preparation
Collected NPA and saliva samples were processed by RNA extraction with the QIAamp Viral RNA Mini kit (QIAGEN) or the PTK-based RNA preparation method.The PTK method was slightly modified based on a previous protocol. 7To process samples with PTK, 50 μg of PTK solution (Thermo Fisher Scientific) was added to 200 μL of the saliva sample and vortexed vigorously for 1 min.Then, the PTK activity was inactivated by heating for 15 min at 95°C.

| Multiplex RT-qPCR assay
For quantification of the SARS-CoV-2 genome, a multiplex RT-qPCR assay was conducted using CFX-96 real-time PCR detection system (Bio-Rad Laboratories).To prepare the RT-qPCR master mix, a total of 5 μL of primer-probe sets were added to 10 μL of buffer and 1 μL

| Comparison of SARS-COV-2 detection between samples and methods
On admission, 51 patients with a confirmed COVID-19 prognosis were provided with a single set of paired NPA and saliva samples.
And 32 patients provided 2 sets of paired samples on both admission and discharge.In the negative control group, 198 children were evaluated (Figure 1A).When the different samples collected and detection methods used for COVID-19 patients were compared, mean C t values for E genes and RdRp genes of SARS-COV-2 were lower in the NPA samples than in the saliva samples when processed by the conventional method.This was also the case when using the PTK method without RNA extraction (Figure 1B−E).Notably, C t values for E genes and RdRp genes of SARS-COV-2 in saliva did not cross over 40, the upper limit of diagnostic value.
Moreover, compared to the conventional method using NPA, the overall sensitivity of saliva using the PTK method was 94.1%.In addition, the specificity of the PTK method using saliva was also high, with a 98.8% diagnostic accuracy in this study (Table 2).

T A B L E 1
We also compared mean C t values for the E and RdRp genes of SARS-COV-2 in paired saliva samples in COVID-19 patients.
As shown in Figure 2A,B, the mean C t values for both E and RdRp genes were higher at discharge compared to those at admission.

| DISCUSSION
The COVID-19 pandemic caused by SARS-COV-2 has exerted considerable influence on the world.This unprecedented worldwide respiratory infection has changed the healthcare system, including diagnostics, new drug development, and people's lifestyles.Recently, this pandemic has transitioned into an endemic. 8However, the unpleasant impacts of SARS-COV-2 still remain, as evidenced by the variants of mutation and its persistent infectivity. 9 this study, we tested saliva as a reliable specimen in children to confirm SARS-COV-2 infection using the PTK method.Although this study was conducted in a relatively small population, multiple centers in Korea participated in this experiment and included very young children.
The saliva collection did not require any specific medium or materials.The time for saliva collection from the subjects took under 30 min, even in very young children.Moreover, the time for RNA extraction using the PTK method took only approximately 15 min.0][11][12] and different types of samples. 12,13The PTK method streamlines RNA extraction, making the process faster and more cost-effective.It could also quantify the viral loads with high output and diagnostic power.Using a PTK-containing tube during saliva collection could make the following process safer and more accurate without re-opening and re-capping the virus-containing tube.In addition, this study could help saliva garner attention as a potential alternative specimen for diagnosing respiratory infections, despite the previously reported inconsistent diagnostic accuracy of saliva-based SARS-CoV-2 being dependent on the studied population or clinical settings. 3,4,14,15reover, saliva collection is a noninvasive, easy, specimen acquirement method that requires no specific preparation, transportation, or storage.
Even in the case of very young children (under 2 years old) who find it challenging to spit or self-collect, an older caregiver or medical staff can collect a sample by holding cotton or cotton swabs in the young children's mouths.Furthermore, the saliva-based method might raise fewer concerns about sample contamination (e.g., nosebleeds during the NPA process), which can reduce the confounding factor.
Collectively, the saliva-based PTK method could be a potential diagnostic tool for new emerging pathogens, such as SARS-COV-2.
Although further research is required to test this method's effectiveness in the detection of other respiratory viruses, it offers potential value in point-of-care settings, and can even be applied to children.Resources.Hye Young Lee: Supervision.Dong Hyun Kim and Jae Myun Lee: Writing-review and editing, project administration.

AUTHOR CONTRIBUTIONS
A minimum of 246 samples per site were required to achieve 95% confidence.A total of 251 children were enrolled from May to October 2021.The PCR-negative children were outpatients who visited the COVID-19 screening clinic at Severance Children's Hospital or Yongin Severance Hospital.They provided saliva samples when they tested for SARS-COV-2 by NP swabs at the same time.The PCR-positive children were in-patients admitted to an isolation ward of Inha University Hospital due to COVID-19.Paired NP aspiration (NPA) and saliva samples were collected at admission and discharge from all the children in the study.They were stored at 4°C for up to 5 days until the analysis.
of enzyme mix provided by Luna ® Universal One-Step RT-qPCR kit (NEW ENGLAND BioLabs ® Inc.).The primer and probe set for RNAdependent RNA polymerase was designed based on the sequence in Genebank (Accession number: NC_045512) (RdRp, F: ATGAGCT TAGTCCTGTTGCACTAC, R: ACC TCCCTTTGTTGTGTTGTAGTAAG, P1: FAM-ATGTCTTGTGCTGCCGGTACTAC-BHQ1, P2: FAM-TGC TTGCACTGATGACAATGCGTTA-BHQ1).For envelope gene, we modified the E gene targeting set recommended by WHO for optimization (E, F: GAAGAGACAGGTACGTTAATAGTTAATAGCGT, R: ATATTGCAGCAGTACGCACACA, P: HEX-ACACTAGCCATCC TTACTGCGCTTCG-BHQ1).The sensitivity and specificity of primer-probe sets were confirmed using the Nucleotide Basic Local Alignment Search Tool (Blastn) provided by the National Center for Biotechnology Information.Meanwhile, a primer-probe set for RNase P (RNase P, F: AGATTTGGACCTGCGAGCG, R: GAGCGGCTGTCTC CACAAGT, P: Cy5-TTCTGACCTGAAGGCTCTGCGCG-BHQ-3) used as an internal control was obtained from the previously published article. 7After mixing 5 μL of the processed samples to the master mix, amplification was performed according to the manufacturer's instruction for 40 cycles.C t values < 40 were reported as positive.

2. 4 |
Statistical analysisBaseline characteristics of patients were compared using the Mann-Whitney U test or Fisher's exact test, as appropriate.Normal distribution was determined by Kolmogorov-Smirnov test.Numerical variables were expressed as the mean and standard deviation (SD).Numerical parameters with non-normal distributions were presented as the median and interquartile range (IQR).A p < 0.05 was considered statistically significant.Statistical software (SPSS, version 23.0, SPSS Inc.) was used for all analyses.Graphs were generated by GraphPad Prism (version 8.0, GraphPad Software).

3 | RESULTS 3 . 1 |
Characteristics of the subjects in the study Of the 251 children who participated in the study, 51 were confirmed COVID-19 patients.The median length of admission was 10 days (IQR 8-12 days) in the positive group.Most children (n = 171, 68.1%) were tested for SARS-COV-2 due to COVID-19-related symptoms, such as fever or respiratory symptoms.There were no children with gastrointestinal symptoms reported.The youngest patient providing a saliva sample was 4 months of age, and there were seven young children tested under 2 years of age.There were no severe adverse reactions reported during saliva sample collection in any of the children.Clinical characteristics are listed in Table1.

1
Study Flowchart (A) and comparison between the mean cycle threshold (C t ) values for the E (B, D) and RdRp (C, E) genes based on the RNA extraction method in the nasopharyngeal aspirate (NPA) and the saliva of COVID-19 confirmed patients.

2
Comparison between the mean cycle threshold (C t ) values for the E (A) and RdRp (B) genes in paired saliva specimens detected using the Proteinase K (PTK)-based method, between the time of admission and discharge.
Clinical characteristics of the patients.
The mean C t values for both E and RdRp genes in the saliva-based PTK method were higher than those of the NPA-based RT-PCR method, but the diagnostic accuracy of the saliva-based SARS-COV-2 PTK method was 98.8% compared to NP swab-based RT-PCR test.Moreover, the sensitivity and specificity of the saliva-based PTK method were also high, 94.1% and 100%, respectively.Among COVID-19 patients, the mean C t values for both E and RdRp genes detected by the saliva-based PTK method were higher at discharge.The difference in the C t values between diagnosis and discharge by the saliva-based PTK method might help estimate the patients' clinical conditions or virus activity.Ever since Vogels et al. suggested the saliva-based PTK method (SalivaDirect) as a simplified and flexible platform to detect SARS-COV-2, 7 various studies have discussed comparisons among different kinds of methods Jung Kim: Conceptualization, data curation, formal analysis, writing-original draft preparation.Pil-gu Park: Methodology, investigation.Su Jin Hwang and Kyeo Re Han: Investigation, data curation, formal analysis.Seung Jun Bang, Jae Hwa Jung, Eun Bin Kown, Eun Kyung Sul, Kyung Chul Song, and Joon-sik Choii:T A B L E 2Comparison of the detection rate of SARS-COV-2 between the conventional method using nasopharyngeal aspirates samples and the Proteinase K (PTK)-based method using saliva samples.
a C t values < 40 were reported as positive.