key: cord-269045-i7vijtol
authors: Martínez‐Murcia, A.; Bru, G.; Navarro, A.; Ros‐Tárraga, P.; García‐Sirera, A.; Pérez, L.
title: Comparative in silico design and validation of GPS™ CoVID‐19 dtec‐RT‐qPCR test
date: 2020-07-29
journal: J Appl Microbiol
DOI: 10.1111/jam.14781
sha: 
doc_id: 269045
cord_uid: i7vijtol

AIMS: Providing a ready‐to‐use reverse transcriptase qPCR (RT‐qPCR) method fully validated to detect the SARS‐CoV‐2 with a higher exclusivity than this shown by early published RT‐qPCR designs. METHODS AND RESULTS: The specificity of the GPS™ CoVID‐19 dtec‐RT‐qPCR test by analysis of sequence alignments was approached and compared with other RT‐qPCR designs. The GPS™ CoVID‐19 dtec‐RT‐qPCR test was validated following criteria of UNE/EN ISO 17025:2005 and ISO/IEC 15189:2012. Diagnostic validation was achieved by two independent reference laboratories, the Instituto de Salud Carlos III, (Madrid, Spain), the Public Health England (Colindale, London, UK), and received the label CE‐IVD. The GPS design showed the highest exclusivity and passed all parameters of validation with strict acceptance criteria. Results from reference laboratories 100% correlated with these obtained by using reference methods and showed 100% of diagnostic sensitivity and specificity. CONCLUSIONS: The CE‐IVD GPS™ CoVID‐19 dtec‐RT‐qPCR test, available worldwide with full analytical and diagnostic validation, is the more exclusive for SARS‐CoV‐2 by far. SIGNIFICANCE AND IMPACT OF THE STUDY: Considering the CoVID‐19 pandemic status, the exclusivity of RT‐qPCR tests is crucial to avoid false positives due to related coronaviruses. This work provides of a highly specific and validated RT‐qPCR method for detection of SARS‐CoV‐2, which represents a case of efficient transfer of technology successfully used since the pandemic was declared.

Last 30th January, the Emergency Committee of the World Health Organization (WHO) under the International Health Regulations (IHR) declared an outbreak of pneumonia, lately named Corona Virus Disease 2019 , as a 'Public Health Emergency of International Concern' (PHEIC). The disease is caused by Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) and the first genome was rapidly provided (http://viro logical.org/t/novel-2019-coronavirus-genome/319). SARS-CoV-2 is a Betacoronavirus subgenus Sarbecovirus of group 2B, with similar characteristics than SARS-CoV, Bat-SARS-CoV and other Bat SARS-like-CoV (Ceraolo and Giorgi 2020; Jiang and Shi 2020; Lai et al. 2020; Lu et al. 2020; Zhou et al. 2020; Li et al. 2020) . A few weeks later, this novel coronavirus spread worldwide and forced WHO to declare a Pandemic on March 11, when more than 118 000 positives and 4291 deaths were already registered in 114 countries. Today, 06th July, the number of positive cases globally surpasses 11.5 million people with more than 530, 000 deaths. The aggressiveness of this global alarm has overwhelmed any forecast. A massive, reliable and rapid diagnosis, is undoubtedly vital and foremost priority for decision-making, which will facilitate public health interventions. Current molecular diagnostic tools for viral detection are typically based on the amplification of target-specific genetic sequences using the polymerase chain reaction (PCR). In acute respiratory infection, real-time PCR (socalled quantitative PCR; qPCR) is the gold-standard method, routinely used to detect causative viruses because it is the most sensitive and reliable method (Mackay et al. 2002; Drosten et al. 2003; Poon et al. 2004; Corman et al. 2012a; Corman et al. 2012b ). On the 17th January, WHO published the very first primers and probes for reverse transcriptase qPCR (RT-qPCR) developed by Corman et al. (2020) . They used known genomic data from SARS-CoV and SARS-CoV related (Bat viruses) to generate a non-redundant alignment. The candidate diagnostic RT-qPCR assay was designed upon the first SARS-CoV-2 sequence released, based on the alignment with known SARS-CoV sequences. Because only a single SARS-CoV-2 genome was available, the two monoplex PCR protocols (ORF1ab and N genes) designed to detect SARS-CoV-2 are also reactive to SARS-CoV and Bat SARS-like-CoV. A few days later, 23rd January, the same laboratory together with reference laboratories from the Netherlands, Hong Kong, France, United Kingdom and Belgium, added a third monoplex RT-qPCR (Corman et al. 2020) . Many laboratories worldwide are currently using this RT-qPCR protocol (Chan et al. 2020) , which was modified 15th March. The SARS-CoV-2 primer and probe sets were designed for the universal detection of SARS-like coronaviruses (N3 assay) and for specific detection of SARS-CoV-2 (N1 and N2 assays). Finally, authors from the Institut Pasteur, Paris, based on the first sequences of SARS-CoV-2 available on the GISAID database (Global Initiative on Sharing All Influenza Data), published a protocol for the detection of SARS-CoV-2 for two RdRp targets (IP2 and IP4) (Institut Pasteur 2020).

biotechnology-based companies recently developed kits for detection of SARS-CoV-2, based on RT-qPCR and provided easy transfer of technology to laboratories worldwide. A fully SARS-CoV-2-specific RT-qPCR thermostable kit was early launched on 27th January by Genetic PCR Solutions TM (GPS TM ), a brand of Genetic Analysis Strategies SL. (Alicante, Spain). The alignments used at that time included 13 SARS-CoV-2 genome sequences released by six different laboratories, deposited in GISAID and available since 19th January 2020. With the purpose to discriminate this new SARS-CoV-2 of present outbreak from previous related SARS, a second independent monoplex RT-qPCR test to detect any other non-SARS-CoV-2 was also produced and provided (not shown). On this study, we performed a deep analytical and diagnostic validation of the GPS TM COVID-19 dtec-RT-qPCR test, following the UNE/EN ISO 17025:2005 and ISO/IEC 15189:2012, respectively. A comparative analysis of the specificity (inclusivity and exclusivity) of the designed primers and probes with most previously published RT-qPCR methods is also here reported.

Partial alignments of 10 SARS-CoV-2 genomic sequences and these from strains of Bat-CoV, Bat SARS-like-CoV, SARS-CoV, Pangolin-CoV (ca. 18, 141 bp) were done. The corresponding phylogenetic tree ( Fig. 1 ) was obtained by Neighbour-Joining method (Saitou and Nei 1987) , with bootstrap values for 1000 replicates, using the MEGA 5.2.2 software (Tamura et al. 2011) .

In silico comparative analysis of primers/probes specificity The primers and probes of GPS COVID-19 dtec-RT-qPCR Test and the RT-qPCR designs recently published (Corman et al. 2020; Chan et al. 2020 (Table 1 ). An illustration of the mismatching of primers/probe sequences of the GPS CoVID-19 dtec-RT-qPCR Test, respect of the SARS-CoV-2, Bat SARS-like-CoV, SARS-CoV, Bat-CoV, and Pangolin-CoV groups is shown in Fig. 2 .

Assays using the GPS TM COVID-19 dtec-RT-qPCR kit (Alicante, Spain) were prepared and reaction mixtures were subjected to qPCR in a QuantStudio3 (ABI) as described in the manual provided. Internal, positive and ORF1ab

Hong Kong University Faculty of Medicine (Hong Kong)

ORF1b 

Journal of Applied Microbiology © 2020 The Society for Applied Microbiology negative PCR controls were included. Standard curve calibration of the qPCR was performed by preparing 10-fold dilution series containing 10 6 to 10 copies of standard template provided in the kit, but also using 5Á10 6 to 5Á10 copies of two complete synthetic RNA genomes from SARS-CoV-2 isolate Australia/VIC01/2020 (GenBank no.: MT007544.1) and isolate Wuhan-Hu-1 (GenBank no.: MN908947.3), provided by Twist Bioscience (South San Francisco, CA).

The SARS- Colindale, London, UK) with a sample panel of 195 specimens, including negative respiratory clinical specimens for SARS-CoV-2, as determined by the validated in-house PHE PCR assay (RdRP gene). Three dilutions of SARS-CoV-2 were used as positive material.

The phylogenetic relationships of selected SARS-CoV-2 genomes and other betacoronavirus as SARS-CoV, Bat SARS-like-CoV, Bat-CoV and Pangolin-CoV are shown in Fig. 1 . The analysis indicated that Bat-CoV RaTG13 and a sequence of Pangolin-CoV showed the highest sequence similarity to SARS-CoV-2 (96.70 and 90.74% respectively), while other Pangolin-CoV sequences available showed a lower homology (85.21%). For in silico specificity analysis, the sequences of primers and probes of all RT-qPCR designs available (Corman et al. 2020; Chan et al. 2020 CoVID-19 dtec-RT-qPCR Test, respect to the SARS-CoV-2 and selected sequences of Bat SARS-like-CoV, SARS-CoV, Bat-CoV and Pangolin-CoV is shown in Fig. 2 . The total number of mismatches were, 31 for Bat SARS-like-CoV (Fig. 2b) , 37 for SARS-CoV (Fig. 2c) , 40 for Bat-CoV (Fig. 2d) and 20 for Pangolin-CoV (Fig. 2e) , including considerable indels in probe and reverse primer sequences.

Analytical Table 2 . Empirical validation terms were evaluated for a minimum of 10 assays (15 in the case of LOD and LOQ) and results were subjected to stablished criteria for acceptance ( Table 2 ). The standard curve calibration of the qPCR was performed from 10-fold dilution series containing 10 6 to 10 copies of standard template (Fig. 3a,b) .

The in vitro inclusivity was assessed by testing two complete SARS-CoV-2 synthetic RNA genomes from isolate Australia/VIC01/2020 (GenBank no.: MT007544.1) and isolate Wuhan-Hu-1 (GenBank no.: MN908947.3). Results obtained from 5Á10 6 to 5Á10 copies are shown in Fig. 3c ,d. The slope and coefficient (R 2 ) values obtained were À3.534 and 0.9986, respectively, the F assay (0.014) was lower than the F fisher (5.318), and efficiency obtained was 93.1%, considering these values acceptable. The CV values, which assess the reliability of the method, ranged Journal of Applied Microbiology © 2020 The Society for Applied Microbiology from 0.53 to 1.31% for repeatability, and 0.59 to 1.83% for reproducibility. LOD for 10 copies was 100% reproducible and accuracy of LOQ for 10 copies was acceptable, as the t value (0.582) was lower than the theoretical value from a Student table (t Student = 2.145). Finally, the results obtained in the diagnostic validation of the GPS TM CoVID-19 dtec-RT-qPCR Test, carried out by the Instituto de Salud Carlos III (ISCIII), are shown in Table 3 . Full agreement was obtained by comparing data resulting from this commercial kit and these from the reference qPCR used by ISCIII. Consequently, 100% of diagnostic sensitivity and 100% of diagnostic specificity was assigned. The GPS TM kit was also evaluated by the Public Health England (PHE; Colindale, London, UK) and results completely correlated (100%) with these determined by the PHE in-house assay targeting the RNA-dependent RNA polymerase (RdRP) of SARS-CoV-2 (data not shown). Furthermore, the GPS TM kit obtained the CE-IVD label (in vitro diagnostics) under the corresponding sanitary licence from the Spanish Agency for Medicines and Health Products (Ministry of Health, Spain).

Only 3 months ago, an outbreak of severe pneumonia caused by the novel coronavirus SARS-CoV-2 started in Wuhan (China) and rapidly expanded to almost all areas worldwide. Due to the urgent need of getting tools for SARS-CoV-2 detection, several laboratories developed RT-qPCR methods. Primers and probes sequences were designed from the alignment of a single-first SARS-CoV-2 genome sequence with known SARS-CoV, and some of these protocols were published at WHO website (Corman et al. 2020; Chan et al. 2020 Figure 3 Quality control of the GPS TM CoVID-19 dtec-RT-qPCR test with data of six ranges of decimal dilution from 10 6 copies to 10 copies, and negative control. (a) Amplification plot and (b) a representative calibration curve with stats. Inclusivity of the GPS TM CoVID-19 dtec-RT-qPCR test using six ranges of decimal dilution from 5Á10 6 copies to 5Á10 copies, and negative control. Amplification plot of synthetic RNA of (c) Australian strain of SARS-CoV-2 (MT007544.1); and (d) Wuhan-Hu-1 strain of SARS-CoV-2 (MN908947.3). Department of Medical Sciences of Thailand 2020; Shirato et al. 2020). As the number of genomes available rapidly expanded during last January, the GPS TM CoVID-19 dtec-RT-qPCR test was based on a more specific target for SARS-CoV-2 detection, being this company one of the pioneers marketing a PCR-kit for the CoVID-19 worldwide.

The phylogenetic analysis indicated that SARS-CoV-2 sequences showed a high sequence homology (over 99.91-99.97%), and the closest relatives were strains of several betacoronaviruses, with a considerable sequence identity to Pangolin isolates, (Fig. 1) which confirms previous results (Ceraolo and Giorgi 2020; Zhou et al. 2020; Paraskevis et al. 2020; Xu et al. 2020; Zhang et al. 2020; Lam et al. 2020) . We have found that the single genome sequence of the Bat coronavirus RaTG13 isolated from Rhinolophus affinis in Wuhan, showed the highest homology level (96.70%) to SARS-CoV-2, as previously described (Ceraolo and Giorgi 2020; Jiang and Shi 2020; Zhou et al. 2020; Lam et al. 2020; Li et al. 2020a; Li et al. 2020) . However, because only a single Bat-CoV sequence with this high identity is available, and it was deposited after the outbreak started (27th January), the possibility of RNA contamination during genome sequencing should be ruled-out before take further conclusions. During the design of the GPS TM kit, one of the purposes of the present study was the in silico comparison (Table 1) with designed primers and probes so far published (Corman et al. 2020; Chan et al. 2020 ; Institut Pasteur 2020; Centers for Disease Control and Prevention from Atlanta 2020; National Institute for Viral Disease Control and Prevention from China 2020; Chu et al. 2020; Department of Medical Sciences of Thailand 2020; Shirato et al. 2020) . In overall, all qPCR designs were inclusive for SARS-CoV-2 as primers and probes showed a good matching. Only the probe for N gene designed by Chu et al. (2020) showed four mismatches, which may affect to its binding, particularly considering its short primary structure. In some cases, single nucleotide mismatching was observed in some primers, but none of them were located close to primer 3 0 -end. Considering all updated alignments, only the Australia/VIC01/2020 sequence showed a single mismatch with the GPS TM probe. Therefore, a full calibration was run using synthetic RNA genomes from Australia/VIC01/2020 isolate and the resulting Ct values correlated with this obtained from Wuhan-Hu-1 synthetic RNA genome (Fig. 3) , indicating that the mismatch in the probe has no effects on the amplification. In silico analysis for exclusivity was more complex, showing a wide range of discriminative power for the methods subjected to analysis (Table 1) . For instance, the two RT-qPCR designs IP2 and IP4 developed by Institut Pasteur seem to discriminate well between SARS-CoV-2 and other respiratory viruses as confirmed for a panel of specimens (Institut Pasteur 2020). The CDC from Atlanta (USA) designed three different primer/probes sets named N1, N2 and N3 (Centers for Disease Control and Prevention from Atlanta 2020). We found a low exclusivity in the N3 primer/probe, but a few weeks ago, this set was removed from the panel (https://www.who.int/emergencie s/diseases/novel-coronavirus-2019/technical-guidance/labo ratory-guidance). Both N1 and N2 showed a good level of mismatching with most coronaviruses except for some Pangolin-CoV sequences, which showed very few nucleotide differences. The RT-qPCR proposed by Corman et al. (2020) , designed to detect SARS-CoV-2, SARS-CoV and Bat SARS-like-CoV, is probably the most used worldwide. They suggest the use of an E gene assay as the first-line screening tool, followed by confirmatory testing with the two probes P1 and P2 in the RdRp gene assay. While P1 probe should react with both SARS-CoV-2 and SARS-CoV, P2 probe was considered specific for SARS-CoV-2. Although our in silico results confirmed that purpose for P1 (Table 1) , the RdRp_P2 assay may also react with some other coronaviruses. The CDC in China developed two RT-qPCR assays for ORF1ab and N genes (National Institute for Viral Disease Control and Prevention from China 2020). Both showed a good overall mismatching to consider them as exclusive, except for some Pangolin-CoV sequences. A similar conclusion may be taken for the N-gene RT-qPCR at the Ministry of Public Health of Thailand (Department of Medical Sciences from Thailand 2020). Data of Table 1 indicated that primer/probe of Chu et al. (2020) , may be reactive with SARS-CoV-2, SARS-CoV and Bat SARS-like-CoV. The exclusivity of the RT-qPCR design developed by Shirato et al. (2020) clearly resided in the reverse primer as showed seven mismatches with all SARS-related coronaviruses. Finally, Chan et al. (2020) developed three RT-qPCR assays targeting RdRp/Hel, S and N genes of SARS-CoV-2. They selected the RdRp/Hel assay as considered to give the best amplification performance and was tested in parallel with the RdRp-P2 from Charit e-Berlin (Corman et al. 2020) . All positive patients with the RdRp-P2 assay were positive with the RdRp/Hel design. However, 42 patients negative for the RdRp-P2 assay were positive with RdRp/Hel and they found that only RdRp-P2 assay, but not RdRp/Hel, cross-reacted with SARS-CoV culture lysates (Chan et al. 2020) . Above findings agreed with expected exclusivity derived from the present study. Additional comparative in vitro analysis was performed (Etievant et al. 2020; Jung et al. 2020; Nalla et al. 2020; Vogels et al. 2020) . Jung et al. (2020) indicated that primer/probes of ORF1ab from the CDC-China seems the most sensitive, and the N2 and N3 assays from the CDC-Atlanta were the most recommended. This partially disagrees with our findings as the N3 design may react with other coronaviruses different from SARS-CoV-2 (moreover, the N3 design was recently removed from the CDC panel). In the study by Nalla et al. (2020) , the specificity of methods from Charit e-Berlin and CDC-Atlanta was tested finding no false positive results but differences in the sensitivity. The most sensitive were N2 (CDC-Atlanta) and E (Charit e-Berlin). However, the present study indicates the RT-qPCR for E target may react with different SARS coronavirus. Finally, the kit GPS TM COVID-19 dtec-RT-qPCR Test showed the highest number of mismatches (i.e. 19-48) for all coronavirus sequences described so far, including these of Pangolin-CoV, which showed a range of 19-31 mismatches. In addition, considerable indels were discerned which enlarge even more the exclusivity of this design.

The GPS TM kit passed the analytical and diagnostic validation according to criteria of the UNE/EN ISO/IEC 17025:2005 and ISO/IEC 15189 ( Table 2 ). The analysis standard curve was repeated a minimum of 10 times and average values for all parameters were optimum according to standard limits. For reliability, the coefficient of variation obtained in all cases for both, repeatability and reproducibility, was always much lower than 10%. The LOD was tested with the usual protocol for 10 copies repeated 15 times with a positive result in all cases (100%). LOQ assays were performed in two sets of 15 tests for both, 100 and 10 copies of standard template. The LOQ measurement in both cases was validated with a t-Student test with a confidence interval of 95%. The kit received diagnostic validation by two different reference laboratories, ISCIII, Madrid, and PHE, London. The results shown in Table 3 indicated 100% of diagnostic sensitivity and 100% of diagnostic specificity was assigned. Currently, the kit is being used in several Spanish hospitals and diagnostic laboratories.

Obviously, at the time of designing the published RT-qPCR assays (Corman et al. 2020; Chan et al. 2020 ; Institut Pasteur 2020; Centers for Disease Control and Prevention from Atlanta 2020; National Institute for Viral Disease Control and Prevention from China 2020; Chu et al. 2020 ; Department of Medical Sciences of Thailand 2020; Shirato et al. 2020) , a lack of SARS-CoV-2 genomes available may explain the relatively scarce exclusivity found in some cases. Despite the greater or lesser in silico specificity of these primers and probes, the host specificity of Bat-CoV, Bat SARS-like CoV and Pangolin-CoV, together with the fact of that no human-SARS have been reported since 2004, all positive results obtained would be considered as SARS-CoV-2 infections ; National Institute of Allergy and Infectious Diseases 2020). However, RNA viruses may exhibit substantial genetic variability. Although efforts were made to design RT-qPCR assays in conserved regions of the viral genomes, variability resulting in mismatches between the primers and probes and the target sequences can result in diminished assay performance and possible false negative results. Primers and probes should be reviewed and updated according to new data, which will increase exponentially during the next few weeks/months.

and Prevention (CDC) (2020) 2019-Novel Coronavirus (2019-nCoV) Real-time rRT-PCR Panel Primers and Probes

Genomic variance of the 2019-nCoV coronavirus

Improved molecular diagnosis of COVID-19 by the novel, highly sensitive and specific COVID-19-RdRp/Hel real-time reverse transcription-polymerase chain reaction assay validated in vitro and with clinical specimens

Molecular diagnosis of a Novel coronavirus (2019-nCoV) causing an outbreak of pneumonia

Detection of a novel human coronavirus by real-time reverse-transcription polymerase chain reaction

Assays for laboratory confirmation of novel human coronavirus (hCoV-EMC) infections

Detection of 2019 novel coronavirus (2019-nCV) by realtime RT-PCR

RT-PCR Protocol for the Detection of 2019-nCoV. Thailand: Department of Medical Sciences, Ministry of Public Health

Identification of a novel coronavirus in patients with severe acute respiratory syndrome

Sensitivity assessment of SARS-CoV-2 PCR assays developed by WHO referral laboratories

Protocol: Real-Time RT-PCR Assays for the Detection of SARS-CoV-2. Par ıs: Institut Pasteur

The first disease X is caused by a highly transmissible acute respiratory syndrome coronavirus

Comparative analysis of primer-probe sets for the laboratory confirmation of SARS-CoV-2

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and coronavirus disease-2019 (COVID-19): the epidemic and the challenges

Identifying SARS-CoV-2 related coronaviruses in Malayan pangolins

Genetic evolution analysis of 2019 novel coronavirus and coronavirus from other species

Evolutionary history, potential intermediate animal host, and cross-species analyses of SARS-CoV-2

Genomic characterisation and epidemiology of 2019 novel coronavirus: implications for virus origins and receptor binding

Real-time PCR in virology

Internal validation of GPS TM MONODOSE CanAur dtec-qPCR kit following the UNE/ EN ISO/IEC 17025:2005 for detection of the emerging yeast Candida auris

National Institute for Viral Disease Control and Prevention (CDC) (2020) Specific Primers and Probes for Detection

NIAID (2020) United States of America

Full-genome evolutionary analysis of the novel corona virus (2019-nCoV) rejects the hypothesis of emergence as a result of a recent recombination event

A one step quantitative RT-PCR for detection of SARS coronavirus with an internal control for PCR inhibitors

The neighbor-joining method: a new method for reconstructing phylogenetic trees

Development of genetic diagnostic methods for novel coronavirus 2019 (n-CoV-2019) in Japan

MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods

Conformity assessment. General requirements for the competence of testing and calibration laboratories

Medical laboratoriesrequirements for quality and competence

Systematic comparison of two animal-to-human transmitted human coronaviruses: SARS-CoV-2 and SARS-CoV

Analytical sensitivity and efficiency comparisons of SARS-COV-2 qRT-PCR primer-probe sets

Probable Pangolin origin of SARS-CoV-2 ASSOCIATED with the COVID-19 outbreak

A pneumonia outbreak associated with a new coronavirus of probable bat origin

In memory of all the health professionals who gave their lives to care for their patients during this pandemic. We are most grateful to the scientists from the Instituto de Salud Carlos III, Madrid, Spain and the Public Health England, London, UK.

No conflict of interest declared.

All authors contributed equally to the work.