key: cord-1048799-nzgsvxc7
authors: Miller, Eric W.; Lamberson, Celeste M.; Akabari, Ratilal R.; Riddell, Scott W.; Middleton, Frank A.; Nasr, Michel R.; Smith, Scott C.; Sperber, Steven M.
title: Development and Validation of Two RT-qPCR Diagnostic Assays for Detecting SARS-CoV-2 Genomic Targets Across Two Specimen Types.
date: 2022-02-04
journal: J Mol Diagn
DOI: 10.1016/j.jmoldx.2021.12.010
sha: cdf0b927653241316093369d6fe168fcee21b72b
doc_id: 1048799
cord_uid: nzgsvxc7

Following the outbreak and subsequent pandemic of COVID-19, clinical diagnostic laboratories worldwide sought accurate and reliable testing methodologies. However, many laboratories were, and still are, hindered by a number of factors, including an unprecedented demand for testing, reagent and laboratory supply shortages, and availability of qualified staff. In order to respond to these concerns, two separate laboratory developed tests (LDTs) were validated for SARS-CoV-2 detection using two different specimen types. Additionally, these assays target different genomic regions of SARS-CoV-2, allowing for viral detection and mitigating genetic variation. Lower limit of detection (LoD) and clinical evaluation studies demonstrated detection of SARS-CoV-2 at 500cp/mL with nasopharyngeal and saliva samples. These multiplexed RT-qPCR assays, while based on modified CDC, NYSDOH and WHO EUA tests, allow for higher throughput and rapid turnaround time, benefiting patients, clinicians and communities as a whole. These cost-effective tests also use readily obtainable reagents, circumventing commercial assay supply chain issues. The LDTs described here have improved patient care and are highly adaptable should the need arise at other clinical diagnostic laboratories. Furthermore, the foundation and design of these assays could potentially be modified in the future for COVID-19 variant detection or other RNA-based viral detection tests.

Primers and probes targeting two regions of the SARS-CoV-2 RNA-dependent RNA polymerase 171 (RdRp) gene (Accession: NC_045512.2) (IP2 and IP4) were designed by the National 172

Reference Center (CNR) for Respiratory Viruses at the Institut Pasteur in Paris, France 173 (https://www.who.int/docs/default-source/coronaviruse/real-time-rt-pcr-assays-for-the-detection-174 of-sars-cov-2-institut-pasteur-paris.pdf?sfvrsn=3662fcb6_2, last accessed: November 3, 175 2021). 13 Both probes targeting these regions employ a 5'-HEX probe and 3'-IABkFQ quencher. 176

Peptidylprolyl Isomerase A (PPIA) serves as an internal control, with primers and probes pre-177 designed by Integrated DNA Technologies to detect all splice variants across exons 4-5 178 (Accession: NM_021130). The PPIA probe contains a 5'-Cy5 probe and 3'-IAbRqSp quencher. 179

All primers and probes were synthesized by IDT (Coralville, IA) ( Table 2) . 180 181 Specimen Collection 182 183 Nasopharyngeal specimens were collected using flocked swabs (BD, Franklin Lakes, NJ and 184 Hardy Diagnostics, Santa Maria, CA) by standard methods and transferred to tubes containing 185 1-3mL UTM (universal transport media) or VTM (viral transport media). All nasopharyngeal 186 specimens were transported at room temperature and stored at 4°C for 72 hr or -80°C for >72 187 hr. Saliva specimens were collected using the ORAcollect RNA collection device (DNA 188 Genotek, Ottawa, ON, Canada). For the saliva samples, different regions of the mouth were 189 self-swabbed under the supervision of a healthcare professional. Swabs were then transferred 190 to the collection tube and mixed with stabilizing reagent. Upon specimen receipt, saliva swabs 191 were incubated at 60°C for two hours to heat-inactivate SARS-CoV-2 and enhance protease 192 activity, allowing for specimens to be handled outside of a biological safety cabinet. Saliva 193 specimens were transported and stored at room temperature where they are stable for up to 194 one week as indicated by the manufacturer. 

Reverse transcription of RNA eluates, cDNA synthesis and target amplification were performed 219 using an Applied Biosystems 7500 Fast Dx Real-Time PCR Instrument (Thermo Fisher 220 Scientific). For each reaction, 5µL of RNA eluate was combined with 5µL of 4x TaqPath 1-Step 221 RT-qPCR Master Mix (Thermo Fisher Scientific), 1µL 20x primer/probe mix and 9µL nuclease-222 free H2O. The 20x primer/probe mix was generated to allow for 0.9µM working concentrations 223 of N1 and N2 forward and reverse primers, 0.25µM working concentrations of N1-FAM and N2-224 Using a Bio-Rad CFX96 thermal cycler (Bio-Rad, Hercules CA), RNA eluates were reverse 240 transcribed to cDNA and RdRp gene targets were amplified. For each reaction, 5µL of RNA 241 eluate was combined with 5µL 4x Bio-Rad Reliance Supermix, 1µL 20x primer/probe mix and and IP4-HEX probes, 0.3µM PPIA forward and reverse primers, and 0.083µM of PPIA-Cy5 245 probe. The cycling protocol included a reverse transcription cycle at 50°C for 20 minutes, an 246 enzyme inactivation and polymerase activation cycle at 95°C for 10 minutes, and 45 cycles of 247 amplification at 95°C for 10 seconds followed by 59°C for 20 seconds. The RNA extraction 248 control, positive control and no template control were prepared in the same manner as for the 249 SARS-CoV-2 Nucleocapsid assay. Table S1 ). An automatic threshold of 500 RFUs for the 257 HEX signal was set for the SARS-CoV-2 RdRp assay. Ct values were also exported as a flat 258 file (CSV) and imported into the worksheet (Supplementary Table S2 The analytical sensitivity of the SARS-CoV-2 Nucleocapsid LDT was assessed by performing a 283

LoD study using a Ct value of 40 as a cutoff for positive specimens. This resulted in a LoD of 284 500cp/mL for both the N1 and N2 gene targets using the MagMAX Viral/Pathogen II Kit and the 285 KingFisher Flex in 100% (25/25) of nasopharyngeal specimens tested (Table 3 and Figure 1 ). 286

Additionally, in the same assay but using the NucleoMag Viral Kit, a LoD of 500cp/mL was 287 achieved in 96% (24/25) of nasopharyngeal specimens tested ( was achieved with positive nasopharyngeal specimens ( Table 7) . The three discordant samples 310

were subsequently re-tested using the TaqPath EUA (Thermo Fisher Scientific, Waltham, MA) 311 and found to be negative. 312

In order to determine analytical sensitivity when comparing the SARS-CoV-2 Nucleocapsid LDT 314 to the Wadsworth EUA, two Wadsworth EUA runs and three SARS-CoV-2 Nucleocapsid LDT 315 runs were performed (Table 8) . Each run utilized a series of SARS-CoV-2 dilutions in negative 316 patient RNA eluates, ranging from 0.5cp/µL -50,000cp/µL. Upon comparison, acceptable 317 correlation levels were achieved for both the N1 and N2 reactions, demonstrating successful 318 assay efficiency. Moreover, this supports our multiplexing design as any competitive inhibition 319 between primer pairs and/or probes was not observed. This also led us to confidently 320 hypothesize that a LoD of 500cp/mL would most feasible as Ct values >30 were observed in 321 reactions with lower SARS-CoV-2 concentrations.

A LoD study for the SARS-CoV-2 RdRp LDT yielded similar results. Regardless of which 324 extraction kit was used with the KingFisher Flex (MagMAX Viral/Pathogen II Kit or NuceloMag 325 Viral Kit), virus was detected at 500cp/mL in 100% (24/24) of saliva specimens (Table 9 and 326 Figure 2 ). Moreover, the same study was performed using nasopharyngeal specimens and 327 detected SARS-CoV-2 at a level of 500cp/mL in 96% of all specimens for both extraction kits 328 (Table 10) . 329 330 A clinical evaluation for both nasopharyngeal and saliva specimens using the SARS-CoV-2 331

RdRp LDT was also performed. RNA from each specimen type was extracted using the 332

KingFisher Flex and both the MagMAX Viral/Pathogen II Kit and the NucleoMag Viral Kit. Saliva 333 specimen concordance was evaluated based on results from the Clarifi COVID-19 Test Kit EUA 334 (Table 11 ). The Clarifi COVID-19 Test Kit EUA uses RNA from samples extracted with the 335 Zymo Research RNA Isolation Kit and primers and probes targeting the same genomic regions 336 of RdRp, albeit at different concentrations. Nasopharyngeal specimen concordance was 337 compared to results from the Wadsworth EUA using RNA from specimens previously extracted 338 using the EZ1 Advanced XL and EZ1 Viral Mini Kit 2.0. All negative (27/27) and positive (44/44) 339 saliva specimens were 100% concordant, while 92.5% (37/40) concordance was observed for 340 positive nasopharyngeal specimens and 100% (30/30) concordance in negative 341 nasopharyngeal specimens (Table 12) . Furthermore, the three discordant positive samples 342 were found to be negative using the TaqPath EUA. 343 344

RdRp LDT allow for detection of the virus at an acceptable LoD regardless of specimen origin 349 (nasopharyngeal or saliva). We initially developed and validated the SARS-CoV-2 350

Nucleocapsid LDT as our laboratory was specifically testing nasopharyngeal specimens and we 351 needed to implement a multiplexed diagnostic assay. The singleplex assay (Wadsworth EUA) 352 initially required more time, supplies and technical work. The SARS-CoV-2 RdRp LDT was 353 developed and validated following the implementation of the SARS-CoV-2 Nucleocapsid LDT. 354

Saliva specimens are more readily obtainable and more often preferred than nasopharyngeal 355 specimens. Additionally, this also allowed for circumvention of supply chain issues should As new variants of SARS-CoV-2 have emerged, we confirmed that both the SARS-CoV-2 379

Nucleocapsid LDT and SARS-CoV-2 RdRp LDT targets were unaffected by these mutations. 16 380

We have performed sequence analyses on all major SARS-CoV-2 variants and found only that 381 C.37, which holds a P13L mutation in the N gene, may affect the SARS-CoV-2 Nucleocapsid 382 LDT while the SARS-CoV-2 RdRp LDT remains unaffected. Additionally, this effect may be 383 negligible as it only involves one nucleotide change in the probe-binding region for the N1 384 reaction while the N2 reaction is not affected. Several other assays, including many rapid tests, 385 target regions which include mutations found in these variants, possibly impeding accurate 386 clinical diagnosis of SARS-CoV-2 infection. 17 Nonetheless, variants of SARS-CoV-2 could 387 result in potential limitations for these assays. It should be noted that as new variants are 388 identified over time, the targets used in these assays may require modification. 389

The SARS-CoV-2 Nucleocapsid LDT and SARS-CoV-2 RdRp LDT have also demonstrated 391 higher analytical sensitivity when compared to EUA PCR-based rapid tests. In several separate 392 instances, we were able to detect the presence of SARS-CoV-2 in both nasopharyngeal and 393 saliva specimens while other FDA-EUA rapid tests did not detect SARS-CoV-2 in these same 394

samples. However, it should be noted that low-level positives are indeed subject to random 395 sampling errors and are not always indicative of a difference in analytical or clinical sensitivity. 396 397 Limitations for both assays include test turnaround time and ample training. Upon specimen 398 receipt and preparation, the extraction process requires approximately 40 minutes on the EZ1 about 1.5 hours in length, followed by the time required for proper analysis, interpretation and 401 review. These assays also require training on multiple instruments using different software for 402 analysis. Nonetheless, there is robust flexibility found in the SARS-CoV-2 RdRp LDT as it 403 allows for two different specimen types to be tested simultaneously. Additionally, the use of 404 automated liquid-handlers enhances the process in a more rapid and efficient manner with less 405 room for human error. However, it should be noted that these systems rely on several 406 consumable materials and supply chain issues could impact their utilization. Furthermore, the flexibility in regards to specimen type. Diagnostic laboratories could introduce these assays into 416 their workflows in an efficient manner with instruments and reagents that are often found in 417 many molecular laboratories. This also aids in supply acquirement as many EUA rapid tests 418 require the use of reagents specific to the manufacturer and supply chain issues can hamper 419 turnaround time. Moreover, these assays could be modified in the future to aid in the diagnosis 420 of other viral targets that require nucleic acid-based testing. Overall, the SARS-CoV-2 421

Nucleocapsid LDT and the SARS-CoV-2 RdRp LDT are cost-effective, analytically-specific and 422

analytically-sensitive assays that serve as accurate molecular diagnostic tests for COVID-19. (ZeptoMetrix) from SARS-CoV-2 isolate: USA-WA1/2020 was supplied at a concentration of 545 50,000cp/mL and transferred to nasopharyngeal specimen aliquots resulting in a concentration 546 of 500cp/mL. Nasopharyngeal aliquots were extracted using the KingFisher Flex Extraction 547 instrument and NucleoMag Viral kit and subsequently tested for SARS-CoV-2 using the SARS-548

CoV-2 Nucleocapsid LDT. A LoD of 500cp/mL was detected in 96% of extracted specimens. 549

All 24 Coronoavirus 2 (SARS-CoV-2) External Run Control (ZeptoMetrix) from SARS-CoV-2 isolate: 555 USA-WA1/2020 was supplied at a concentration of 50,000cp/mL and transferred to 556 nasopharyngeal specimen aliquots resulting in a concentration of 1,000cp/mL. Nasopharyngeal 557 specimen aliquots were extracted and subsequently tested for SARS-CoV-2 using the SARS-558 599 Table 10 . SARS-CoV-2 was detected at a LoD of 500cp/mL in 96% of nasopharyngeal 600 specimens tested. RNA was extracted from nasopharyngeal specimens using the KingFisher 601

Flex Extraction instrument and either the MagMAX Viral/Pathogen II kit or the NucleoMag Viral 602

kit. Specimens were subsequently tested for SARS-CoV-2 using the SARS-CoV-2 RdRp LDT. 603

All Ct Values and interpretations are listed for each specimen along with the mean for each 604 sample set. One sample resulted as invalid due to no detection of PPIA. The presumptive 605 positive sample was the result of no PPIA detection in conjunction with positive RdRp detection. 

An overall concordance level of 92.5% was achieved when comparing the SARS-614

CoV-2 RdRp LDT to the Wadsworth EUA. A clinical evaluation was performed comparing the 615

SARS-CoV-2 RdRp LDT with the KingFisher Flex Extraction Instrument and MagMax

Viral/Pathogen II Kit and the Wadsworth EUA using the EZ-1 Advanced XL instrument with Viral

Ct values and concordance results are listed for each patient sample