Evaluation of a new molecular test for the detection of SARS-CoV-2 nucleic acid in salivary samples

Background Molecular testing is considered the gold standard for the detection of SARS-CoV-2. This study aimed to compare the performance of the P742H SARS-CoV-2 Nucleic Acid Multiplex Detection Kit in salivary samples, with respect to the 732HF Novel Coronavirus (2019-nCoV) Nucleic Acid Detection Kit and the TaqPath COVID-19 CEIVD RT-PCR Kit, used at University-Hospital of Padova, Italy. Methods One hundred twenty-four salivary samples selfcollected by healthcare workers (HCW) during the screening program at University-Hospital of Padova, Italy, from Oct to Nov 2022, were included in the study. RNA extraction was performed by Viral DNA and RNA Extraction Kit (Technogenetics, Lodi, Italy) and amplification by P742H and 732HF (Technogenetics, Lodi, Italy). RNA was extracted using MagNa Pure 96 DNA and Viral NA Small Volume Kit (Roche, Switzerland) for TaqPath analysis (Thermo Fisher Scientific, USA).


Introduction
For a long time, COVID-19 has been a worldwide health problem, for political, social, and clinical reasons, and therefore it has been treated with emergency dispositions.Thanks to the development of vaccines and the subsequent vaccine campaigns, the pandemic emergency had an arrest, reaching a higher population immunity globally.The World Health Organization (WHO) Emergency Committee agreed that the COVID-19 pandemic might be reaching an inflexion point, while still being a clinical issue that continues to have an impact on morbidity and mortality (1).Nevertheless, the diagnosis of infected subjects remains important especially in fragile patients, to prevent the clinical complications and pulmonary involvement, while screening strategies may be of relevance in specific contexts, such as in hospital settings and healthcare workers (HCW) monitoring, to reduce the spread of the infection (2)(3)(4).
The gold standard method for diagnosis of SARS-CoV-2 infection is based on the nucleic acid amplification tests (NAAT).Real-time PCR enables the detection and quantification of viral genes (5).The analysis of the presence of RNA viral sequences started with the extraction of nucleic acid from nasopharyngeal swabs (NPS) or saliva samples, followed by reverse transcription into cDNA and then by PCR amplification.The outcome is given in terms of positivity or negativity of infection based on the threshold cycle that results from the analysis of each gene analyzed in the panel, which could be automatically detected by advanced software (6,7).Other laboratory methods, such as immunoassays, are used to rapidly test SARS-CoV-2 presence in patients' samples (8).Recently, there has been proving evidence that NAAT integration with immunoassays detecting SARS-CoV-2 viral antigens could represent a promising cost-effective strategy for confining COVID-19 spread.However, in some circumstances requiring high sensitivity and specificity (e.g.patients receiving organ transplant, medical surgery, elderly fragile people), the utilization of NAAT can be favored with respect to antigen testing.Indeed, in a recent metaanalysis, it was demonstrated that rapid antigen tests using nasal or NPS showed a steady decline in sensitivity as the measures of sample viral load decline; the average sensitivity ranged from 34.3% to 91.3% in symptomatic participants, and from 28.6% to 77.8% for asymptomatic subjects (9).These results underlined that rapid antigen tests can be useful in detecting positivity of individuals with high viral load, while sensitivity remains too low for other settings (9).
In addition to different laboratory procedures, a variety of biological matrices can be used.Patients' sampling is done through NPS collection, especially for individuals with high tract respiratory infections.However, NPS have some drawbacks that must be considered.First of all, they cannot be performed individually by patients themselves, but they require expert HCW (10); secondly, they require the use of adequate personal protecting equipment to protect HCW, and they also may facilitate viral spread between individuals who undergo testing.Saliva is a valid alternative for several reasons, having the advantage of being self-collected by patients and enabling an easy sample handling (11).Moreover, saliva collection is non-invasive and can enhance patients' compliance and simplicity of collection, especially in screening settings (8,12), in addition to reducing the total costs by 25-30% with respect to NPS (13).Notably, some recent studies highlighted saliva results were concordant with NPS results both in qualitative and quantitative terms (8,12,13,14).Differently, for antigen detection rapid diagnostic test (Ag-RDT), saliva was described as less sensitive with respect to NPS (15).Furthermore, SARS-CoV-2 antigen levels in saliva decrease more rapidly than in NPS when analyzed with respect to the decline in viral load (8).
The aim of this study was to compare the performances of a new fast molecular method to detect the presence of SARS-CoV-2 in salivary samples of patients, the Technogenetics SARS-CoV-2 Nucleic Acid Multiplex Detection Kit (P742H), useful also as a confirmatory test for screening programs, with respect to two methods for NAAT, one from the same manufacturer and the other one in use at University-Hospital of Padova, Italy (AOPD).

Samples included in the analysis
For the aim of the study, 124 (43 males and 81 females) leftover salivary samples were randomly selected from the HCW ongoing screening program at AOPD, between October 24 th 2022 and Conclusion: The P742H method proved better performances than 732HF for salivary samples, both presenting the same amplification time.In addition, P742H results were comparable to those obtained through the highthroughput method TaqPath.

Extraction and amplification procedures
The 124 samples were analyzed in two analytic sessions, with half of the samples each, on the 17 th November and 24 th November 2022 respectively.Extraction and amplification were executed on the same day.The obtained results were subsequently compared with the respective results obtained from the ongoing screening program.for the samples to be included in the analysis.More specifically, the genes included in the P742H were RdRp gene, N gene and E gene, while the genes included in the 732HF were ORF1ab gene and N gene.For both kits, the amplification mix included a Reaction Solution, an Enzyme Mix and a Primer and Probe Mix.The amplification reaction was carried with 20 mL of amplification mix and 5 mL of RNA sample; each kit had a reaction profile of 45 amplification cycles.Moreover, each kit included one positive control and one negative control to confirm both the assays gave correct results.After the ampli-fication reaction, results were obtained from the analysis software.Each sample was considered valid if the internal control gene turned positive; moreover, each sample was considered either positive or negative for the analysis if the genes were amplified before or after cycle 43 respectively.For both P742H and 732HF, samples with negative IC and negative amplified genes were considered invalid.
Moreover, results were compared to those obtained through the method used at AOPD for screening program.The amplification reaction was carried out using TaqPath TM COVID-19 RT-PCR Kit (Applied Biosystems, Thermo Fisher Scientific, Massachusetts, USA) in QuantStudio5 thermocycler (Applied Biosystems, Thermo Fisher Scientific).The amplification reaction was a real-time PCR with TaqMan probes for the detection of the different genes included in the assay, which were ORF1ab gene, N gene and S gene.The amplification mix included the TaqPath 1-step Multiplex Master Mix (4X) and the COVID-19 Multiplex (20X), an Enzyme Mix and a Primer and Probe Mix.The amplification reaction was carried with 6 mL of amplification mix and 14 mL of RNA sample; the kit had a reaction profile of 40 amplification cycles.Moreover, two positive controls and one negative control were included to confirm the results of the assay.After the amplification reaction, results were obtained from the analysis software.Each sample was considered positive for the analysis if the genes were amplified before cycle 33.In order to test the appropriateness of salivary samples, the amplification of RNaseP gene was performed at the same time in the method routinely used at AOPD.Preliminary sample processing was the same, but the amplification reaction was specific.More in detail, 14 mL of extracted RNA were amplified with 8 mL of reaction mix, which contained home-designed Primers and Probes RNaseP mix (20X), AgPath Buffer (2X) and AgPath (AgPath-ID TM One-Step RT-PCR Kit) (Applied Biosystems, Thermo Fisher Scientific) in QuantStudio5 thermocycler.

Statistical analyses
All statistical analyses were performed by Stata v 16.2 (StataCorp, Lakeway drive, TX, USA).Median was used as descriptive statistics of Ct quantitative data.Cohen's kappa was used to measure interrater agreement.The module »diagt« was used to calculate sensitivity and specificity, and their 95% confidence intervals (95%CI).

Ethical statement
The study was conducted in accordance with the Declaration of Helsinki, and the Institutional Review Board of the University of Padova (protocol no.27444).
The agreement between the different amplification methods, highlighting the number of positive (P), negative (N) and invalid (I) samples obtained for each assay were reported in Supplementary Table I.
Figure 1 shows the threshold cycles (Ct) values of P742H, 732HF and TaqPath assays for all the evaluated genes.The Kruskall-Wallis test, adjusted by Dunn's method for multiple testing, underlined that differences of Ct values exist only between the two TaqPath genes and all the others genes of P742H and 732HF (p < 0.001 for all).There were not statistically significant differences across median Ct values of P742H and 732HF genes.

Discussion
Despite the national and international efforts for controlling and limiting the spread of SARS-CoV-2, COVID-19 still permanently remains an established infection in humans for a long time.The urgent need of rapid diagnosis of infected subjects pivotal in order to limit viral spread and prevent deaths of fragile subjects.NAATs are the gold standard methods for SARS-CoV-2 molecular detection, because of their elevated analytical sensitivity and specificity (16).Lippi et al. pointed out some challenges in providing routine molecular SARS-CoV-2 tests for screening contexts, especially using NPS.Difficulties can be attributed to recruiting staff for collecting NPS, and to obtaining the supplies needed for a large number of tests (17,18).Therefore, alternative sampling procedures have been tested.Self-collecting saliva was demonstrated to have comparable sensitivity and specificity to NPS (19,20), in addition to being a cost-effective and a simple way of specimen collection, thus enabling accurate large-scale SARS-CoV-2 surveillance testing (20).Furthermore, in subjects with higher viral load, in patients with critical conditions or waiting hospitalization at the emergency department, providing rapid results could be important.Molecular testing typically might require up to 4-6 hours to be completed (21).In addition, it possibly leaves laboratories with a huge backlog of samples to be processed when an elevated number of tests are required simultaneously (17).The implementation of novel rapid molecular tests may help to overcome the challenges described above, especially when patients' conditions require it.In addition, it has been demonstrated that strategies based on antigen (Ag) test, performed with laboratory equipment, coupled with real-time PCR testing in case of a first negative result offer better performances than Ag test alone, with a decreased cost with respect to real-time PCR alone (22).
In the present study, the SARS-CoV-2 Nucleic Acid Multiplex Detection Kit (P742H) has been evaluated and tested with respect to the Novel Corona virus (2019-nCoV) Nucleic Acid Detection Kit (732HF).After an RNA extraction (around 20 minutes), both assays, supplied by Technogenetics, require 60 minutes for giving qualitative (pos/neg) and quantitative (genes Ct) results for 64 samples.The two assays were further compared with a the TaqPath assay, routinely used at AOPD and, thus, considered as a reference method.Both assays were tested using 124 samples (95/124 positive, 76.6% for TaqPath assay); for P742H and 732HF one and four samples resulted as invalid, respectively, and the agreement was 97.5% (3 samples were discordant), with a Cohen's kappa of 0.930.Two of these samples were negative and one positive at TaqPath (which was exactly comparable to P742H).The efficiency in probe and primer design may affect the system per- formance, in addition to the amount of sample collected and the input volume of the specimen (23).Moreover, misinterpretation of samples may happen also because the amount of virus is below the detection limit of the method, leading to failure to detect a positive result (23).However, as pointed out by Fomenko et al. (24), for Ct above 35, the probability of obtaining a positive viral culture in infected subjects is very limited, virtually equal to zero for most of the molecular tests.
At last, fully automated systems ensure a more precise and standardized handling and testing of large numbers of samples, while manual work intrinsically carries HCW-specific limits (23).Indeed, from the quantitative Ct analysis of studied genes, no significant differences were found.
This study also included some limitations, such as the number of evaluated specimens, which was not elevated.However, the samples were evaluated with three different assays, confirming the solidity of the results.A further limitation is the lack of NPS comparison for individuals included in the study.However, salivary SARS-CoV-2 testing by TaqPath method was previously validated in a cohort of 6284 subjects, including 206 individuals with COVID-19, with positivity confirmed by NPS (13).

Conclusion
The development of novel, fast molecular kits for the diagnosis of SARS-CoV-2 infection is needed for the rapid and accurate identification of infected individuals (23,25).The SARS-CoV-2 Nucleic Acid Multiplex Detection Kit (P742H) resulted accurate and fast to be applied not only in high-risk individuals, but also in situations requiring a rapid and am accurate diagnosis of COVID-19.
After thawing samples at room temperature for 30 minutes, salivary samples underwent RNA extraction through the use of two extraction kits.The first extraction was performed using Viral DNA and RNA Extraction Kit (REF T014H version 1), provided by Technogenetics (Lodi, Italy), following manufacturer's recommendations with a dedicated Nucleic Acid Extractor machine (provided by Technogenetics).The second RNA extraction was performed with MagNA Pure 96 DNA and Viral NA Small Volume Kit (REF 06543588001 version 09) (Roche, Switzerland) in the nucleic acid extractor Magna Pure 96 (Roche) and then analyzed using the method used at AOPD for HCW screening program.RNAs extracted with the first method were amplified using the two different kits, the SARS-CoV-2 Nucleic Acid Multiplex Detection Kit (REF P742H version 3.0) (P742H) and the Novel Coronavirus (2019-nCoV) Nucleic Acid Detection Kit (REF P732HF version 1.0) (732HF), both provided by Technogenetics using a real-time PCR in Gentier 96 thermocycler (Technogenetics).Both amplification reactions are real-time PCRs that exploit TaqMan probes for the detection of the different genes included in the assay.The SARS-CoV-2 Nucleic Acid Multiplex Detection Kit (P742H) enables the recognition of three candidate genes and one internal control (IC) gene, while the Novel Coronavirus (2019-nCoV) Nucleic Acid Detection Kit (732HF) is based on the detection of two genes and one IC gene.The IC gene was used as a control of extraction and amplification reaction performances: it should give positive results

Figure 1
Figure 1 Threshold cycles (Ct) for all samples for the genes analyzed with P742H, 732HF and TaqPath assay, highlighting the median value for each gene.Maximum Ct values were 45 for P742H and 732HF and 40 for TaqPath.