SARS-CoV-2 RNA Quantification Using Droplet Digital RT-PCR

Quantitative viral load assays have transformed our understanding of diseases. They hold similar potential to advance COVID-19 control and prevention, but SARS-CoV-2 tests are not yet widely available. molecular diagnostic tests, which typically employ real-time RT-PCR, yield semiquantitative results only. Droplet digital RT-PCR (RT-ddPCR) offers an attractive platform for RNA quantification. Eight primer/probe sets originally developed RT-PCR–based were evaluated use in RT-ddPCR; three identified as the most efficient, precise, sensitive RT-ddPCR–based For example, analytical efficiency E-Sarbeco set was approximately 83%, whereas assay precision, measured coefficient variation, 2% at 1000 input copies/reaction. Lower limits quantification detection this 18.6 4.4 copies/reaction, respectively. loads a convenience panel 48 COVID-19–positive specimens spanned 6.2log10 range, confirming substantial variation vivo. RT-ddPCR–derived E gene copy numbers further calibrated against cycle threshold values from commercial platform. This log-linear relationship can be used mathematically derive values, allowing wealth available test data harnessed address foundational questions biology. revolutionized ability manage diseases.1Mellors J.W. Rinaldo Jr., C.R. Gupta P. White R.M. Todd J.A. Kingsley L.A. Prognosis HIV-1 infection predicted by quantity virus plasma.Science. 1996; 272: 1167-1170Crossref PubMed Scopus (2309) Google Scholar, 2Mellors Viral-load provide valuable answers.Sci Am. 1998; 279: 90-93Crossref (18) 3Riddler S.A. Mellors dynamics measurement: implications therapy.AIDS Clin Rev. 1997; : 47-65PubMed 4Mellors Muñoz A. Giorgi J.V. Margolick J.B. Tassoni C.J. Saah A.J. Detels R. Phair J.P. Plasma CD4+ lymphocytes prognostic markers infection.Ann Intern Med. 126: 946-954Crossref (1720) 5Durante-Mangoni E. Zampino Portella G. Adinolfi L.E. Utili Ruggiero Correlates value first-phase hepatitis C kinetics during treatment.Clin Infect Dis. 2009; 49: 498-506Crossref (29) 6Chen Lin W. Shen F. Iloeje U.H. London W.T. Evans A.A. Past HBV predictor mortality morbidity HCC chronic liver disease prospective study.Am J Gastroenterol. 2006; 101: 1797-1803Crossref (232) Scholar Although SARS-CoV-2, quantitative could biology inform prevention measures.7Veyer D. Kernéis S. Poulet Wack M. Robillard N. Taly V. L'Honneur A.S. Rozenberg Laurent-Puig Bélec L. Hadjadj J. Terrier B. Péré H. Highly plasma sheds light on its clinical value.Clin 2020; ciaa1196Crossref (57) Scholar,8Bermejo-Martin J.F. González-Rivera Almansa Micheloud Tedim A.P. Domínguez-Gil et al.Viral is associated with critical illness dysregulated host response COVID-19.medRxiv. (doi:10.1101/2020.08.25.20154252)Google Most assays, however, detect one or more genomic targets using sequence-specific primers coupled fluorescent probe, only semiquantitative. These produce (Ct) readouts, represent PCR where sample began signal above background. each Ct decrement corresponds roughly twofold higher (due exponential nature amplification), cannot directly interpreted without calibration standard.9Han M.S. Byun J.H. Cho Y. Rim SARS-CoV-2: versus qualitative.Lancet 2021; 21: 165Abstract Full Text PDF (71) Scholar,10Lescure F.X. Bouadma Nguyen Parisey Wicky P.H. Behillil Gaymard Bouscambert-Duchamp Donati Le Hingrat Q. Enouf Houhou-Fidouh Valette Mailles Lucet J.C. Mentre Duval X. Descamps Malvy Timsit Lina van-der-Werf Yazdanpanah Clinical virological first cases Europe: case series.Lancet 20: 697-706Abstract (717) Rather, positive, indeterminate, negative based assay-specific cutoffs evolving guidelines. Due differences nucleic acid extraction method, target, other parameters, also comparable across technology platforms. quantification.11Liu Feng Zhang Guo Suo T. Hu Wang Huang Z. Xiong Chen Lan K. Analytical comparisons SARS-COV-2 qRT-PCR ddPCR multiple sets.Emerg Microbes Infect. 9: 1175-1179Crossref (75) Scholar,12Suo Liu Ullah Yang Sajid Deng Xu ddPCR: accurate tool low specimens.Emerg 1259-1268Crossref (218) Like employs target-specific making it relatively straightforward adapt assays. In ddPCR, reaction fractionated into 20,000 nL–sized droplets prior massively parallel amplification. At end point, droplet categorized positive (target present) absent), absolute target Poisson statistics. versatile has been mutation number determination human genome,13Salemi Falzone Madonna Polesel Cinà Mallardo Ascierto P.A. Libra Candido MMP-9 candidate marker BRAF inhibitors melanoma patients BRAF(V600E) detected circulating-free DNA.Front Pharmacol. 2018; 856Crossref verification following genome editing,14Findlay S.D. Vincent K.M. Berman J.R. Postovit L.M. A PCR-based method efficient highly specific screening edited cells.PLoS One. 2016; 11: e0153901Crossref (48) pathogens.15Lillsunde Larsson Helenius Digital (ddPCR) HPV 16, 18, 33 45 - short report.Cell Oncol (Dordr). 2017; 40: 521-527Crossref (32) 16Stevenson Wakeham Pan Kavanagh Millan Bell McLellan Graham S.V. Cuschieri suggests that correlates improved survival HPV-positive oropharyngeal tumours.J Virol. 129: 104505Crossref (9) 17Cao W.W. He D.S. Z.J. Zuo Y.Z. Chang Y.L. Z.G. Ye Shi Development porcine epidemic diarrhea virus.J Vet Diagn Invest. 32: 572-576Crossref (3) 18Persson Eriksson Lowther Ellström Simonsson Comparison between RT noroviruses oysters.Int Food Microbiol. 284: 73-83Crossref (31) 19Pinheiro-de-Oliveira T.F. Fonseca Camargos M.F. Laguardia-Nascimento de Oliveira A.M. Cottorello A.C.P. Goes-Neto Barbosa-Stancioli E.F. foot-and-mouth RNA.J Virol Methods. 259: 129-134Crossref (12) 20Bruner Simonetti F.R. Bender Kwon K.J. Sengupta Fray E.J. Beg Antar A.A.R. Jenike Bertagnolli L.N. Capoferri Kufera J.T. Timmons Nobles C. Gregg Wada Ho Y.C. Blankson J.N. Deeks S.G. Bushman F.D. Siliciano J.D. Laird G.M. R.F. approach measuring reservoir latent proviruses.Nature. 2019; 566: 120-125Crossref (280) 21Dong Zhou Niu Sheng Zhao Zhu Peng Xie Gao Dai Fang PCR.Talanta. 224: 121726Crossref (62) Several SARS-CoV-2—specific RT-ddPCR (11Liu Scholar,22Falzone Musso Gattuso Bongiorno Palermo C.I. Scalia Stefani Sensitivity assessment detection.Int Mol 46: 957-964Crossref (118) Scholar,23de Kock Baselmans Scharnhorst Deiman Sensitive multiplex RT-PCR.Eur Microbiol 807-813Crossref (27) manufacturer’s protocol) achieving high sensitivity some reports,12Suo Scholar,24Yu Yan Tang Ma Tan analysis infected patients.Clin 71: 793-798Crossref (437) 25Alteri Cento Antonello Colagrossi Merli Ughi Renica Matarazzo Di Ruscio Tartaglione Colombo Grimaldi Carta Nava Costabile Baiguera Campisi Fanti Vismara Fumagalli Scaglione Epis O.M. Puoti Perno C.F. Detection nasopharyngeal swabs suspected patients.PLoS 15: e0236311Crossref (73) 26Dang Yuan Fan Jin Lou qualitative analyses samples.Clin Chim Acta. 510: 613-616Crossref (24) few studies rigorously SARS-CoV-2–specific performance template. Furthermore, no knowledge values. Here, eight sets, RT-PCR,27World-Health-OrganizationSummary Available Protocols. World Health Organiation, Geneva, Switzerland2020Google RT-ddPCR. The authors linear equation relating RT-PCR–derived assay, LightMix Modular SARS-CoV (COVID19) E-gene conversion existing loads. study approved Providence Care/University British Columbia Simon Fraser University Research Ethics Boards under protocol H20-01055. assays27World-Health-OrganizationSummary assessed (Table 1). included Charité-Berlin (E-Sarbeco) set,28Corman V.M. Landt O. Kaiser Molenkamp Meijer Chu D.K. Bleicker Brünink Schneider Schmidt M.L. Mulders D.G. Haagmans B.L. van der Veer den Brink Wijsman Goderski Romette J.L. Ellis Zambon Peiris Goossens Reusken Koopmans M.P. Drosten 2019 novel coronavirus (2019-nCoV) RT-PCR.Euro Surveill. 25: 2000045Crossref (4173) Pasteur Institute RdRp IP2 IP4 (IP2 IP4, respectively),29Institut-Pasteur-ParisProtocol: Real-Time Assays SARS-CoV-2.2020Google Chinese Center Disease Control ORF N (China-ORF China-N, respectively),30Chinese-National-Institute-for-Viral-Disease-Control-and-PreventionSpecific Primers Probes Novel Coronavirus.2020Google Hong Kong (HKU-ORF HKU-N, respectively),31School-of-Public-Health-LKS-Faculty-of-Medicine-University-of-Hong-KongDetection Coronavirus Suspected Human Cases RT-PCR.2020Google US-CDC-N1 set.32Centers-for-Disease-Control-and-PreventionCDC 2019-Novel Diagnostic Panel.2020Google No changes sequences fluorophores made when transitioning Fluorescent quenchers employed however [ie, TAMRA, BlackBerry Quencher (BBQ)] changed dark IowaBlack internal ZEN quencher (3IABkFQ, Integrated DNA Technologies, Coralville, IA)], recommended (manufacturer’s protocol).Table 1SARS-CoV-2 Primer/Probe Sets Assessed Use RT-ddPCRSourceNameGene targetPrimer/probeSequenceCoordinatesCharité- BerlinE-SarbecoEForward5?-ACAGGTACGTTAATAGTTAATAGCGT-3?26,269–26,294Reverse5?-ATATTGCAGCAGTACGCACACA-3?26,381–26,360Probe5?-FAM-ACACTAGCC/ZEN/ATCCTTACTGCGCTTCG-3IABkFQ-3?26,332–26,357Pasteur InstituteIP2ORF1aForward5?-ATGAGCTTAGTCCTGTTG-3?12,690–12,707Reverse5?-CTCCCTTTGTTGTGTTGT-3?12,797–12,780Probe5?-HEX-AGATGTCTT/ZEN/GTGCTGCCGGTA-3IABkFQ-3?12,717–12,737IP4ORF1bForward5?-GGTAACTGGTATGATTTCG-3?14,080–14,098Reverse5?-CTGGTCAAGGTTAATATAGG-3?14,105–14,123Probe5?-FAM-TCATACAAA/ZEN/CCACGCCAGG-3IABkFQ-3?14,186–14,167China CDCChina-ORFORF1aForward5?-CCCTGTGGGTTTTACACTTAA-3?13,342–13,362Reverse5?-ACGATTGTGCATCAGCTGA-3?13,460–13,442Probe5?-FAM-CCGTCTGCG/ZEN/GTATGTGGAAAGGTTATGG-3IABkFQ-3?13,377–13,404China-NNForward5?-GGGGAACTTCTCCTGCTAGAAT-3?28,881–28,902Reverse5?-CAGACATTTTGCTCTCAAGCTG-3?28,979–28,958Probe5?-FAM-TTGCTGCTG/ZEN/CTTGACAGATT-3IABkFQ-3?28,934–28,953Hong UniversityHKU-ORFORF1bForward5?-TGGGGYTTTACRGGTAACCT-3?18,778–18,797Reverse5?-AACRCGCTTAACAAAGCACTC-3?18,849–18,872Probe5?-FAM-TAGTTGTGA/ZEN/TGCWATCATGACTAG-3IABkFQ-3?18,909–18,889HKU-NNForward5?-TAATCAGACAAGGAACTGATTA-3?29,145–29,166Reverse5?-CGAAGGTGTGACTTCCATG-3?29,179–29,198Probe5?-FAM-GCAAATTGT/ZEN/GCAATTTGCGG-3IABkFQ-3?29,254–29,236US CDCUS-CDC-N1NForward5?-GACCCCAAAATCAGCGAAAT-3?28,287–28,306Reverse5?-TCTGGTTACTGCCAGTTGAATCTG-3?28,358–28,335Probe5?-FAM-ACCCCGCAT/ZEN/TACGTTTGGTGGACC-3IABkFQ-3?28,309–28,332Coordinates Wuhan-Hu-1 (https://www.ncbi.nlm.nih.gov/genbank; GenBank accession MN908947.3).3IABkFQ, 3? Iowa Black Hole (Integrated Technologies); FAM, 6-carboxyfluorescein; HEX, hexachloro-fluorescein; ZEN, Technologies). Open table new tab Coordinates MN908947.3). 3IABkFQ, synthetic standards comprising six nonoverlapping 5000-base fragments equal quantities encoding (Control 2; https://www.ncbi.nlm.nih.gov/genbank; MN908947.3; Twist Biosciences, San Francisco, CA; supplied 1 million copies/fragment/?L). To avoid degradation, stored ?80°C thawed once, immediately before use, perform efficiency, sensitivity, dynamic range described herein. Moreover, mimic composition real biological specimen, all employing these supplemented consistent, physiologically relevant amount extracted pooled remnant SARS-CoV-2–negative (Supplemental Figure S1). Briefly, 1-mL aliquots transport medium NucliSens EasyMag (BioMerieux, Marcy-l’Etoile, France), eluted 60 ?L repooled. resulting material contained average 2200 cells/?L (as quantified RPP30 in33Kinloch N.N. Ritchie Brumme Dong Lawson Jones R.B. Montaner J.S.G. Leung Romney M.G. Stefanovic Matic Lowe Z.L. Suboptimal sampling probable cause false-negative results.J 222: 899-902Crossref (58) Scholar) 4400 RNAse P copies/?L extract in34Kinloch Shahid Evaluation swab collection techniques recovery participant experience: recommendations diagnostics.Open Forum 7: ofaa488Crossref (14) Scholar), concentrations line levels recovered swabs.33Kinloch Scholar,34Kinloch reactions performed combining template probe (900 nmol/L 250 nmol/L, respectively; IA) 1), One-Step Advanced Kit Supermix, Reverse Transcriptase, DTT (300 nmol/L) (all BioRad, Hercules, CA), XhoI restriction enzyme (New England Biolabs, Ipswich, MA), background (for only, see above), nuclease free water. Droplets generated Automated Generator (BioRad) cycled set–specific conditions (see below) (Figure Analysis QX200 Reader QuantaSoft software version 1.7.4 replicate wells merged analysis. set, acceptable thermal cycling temperature ranges reverse transcription annealing/extension determined modifying manufacturer-recommended default conditions, 42°C 50°C hour transcription); 95°C 10 minutes; 40 cycles (94°C 30 seconds followed 63°C minute); 98°C minutes, 4°C infinite hold. determine transcription, gradient 51.5°C while fixing step 52°C. Using optimized temperature, then identify ranges. Temperatures produced insufficient separation nonspecific amplification deemed unacceptable, those consecutive 95% confidence intervals (CIs) estimates outside maximal point estimate. quantify 100 copies. minimum (maximum four) technical replicates concentration, expressed estimate total CI derived replicates. As such, CIs around considered significantly different another. calculated dividing expected multiplying 100. Precision (CV), percentage, (LDR) interest serial 1:2 dilution series 114,286 1.2 chosen because crosses entire copies seeking protocol). Reactions duplicate. upper limit lower (LLOQ) defined boundaries concentration over linear. iteratively restricting regression ones maximized (R2) minimized residuals. Assay (LLOD), serially diluting 47.6 0.74 Between 6 18 dilution, analyzed probit regression. LLOD, through interpolation curve, probability 95%. Optimized applied SARS-CoV-2–positive submitted St. Paul's Hospital Virology Laboratory Vancouver, Canada testing Roche cobas assay. samples, acids re-extracted BioMerieux 50 ?L. Eluates aliquoted frozen single use. above. main goal characterize confounding platform, material, extracts re-tested uses set: 2019-nCoV (TIB Molbiol, Berlin, Germany), implemented LightCycler 480 (Roche Diagnostics, Basel, Switzerland).35Matic Li Champagne Practical challenges implementation saliva detection.Eur 447-450Crossref (20) Finally, responsive recent recommendation reported terms per cell equivalents,9Han cells/microliter previously described33Kinloch additionally copies/1000 cells. replicates, (Bio-Rad). precision CV, identifying residuals LLOD Correlation Spearman's rho (?). Where appropriate measure concordance, Lin's concordance correlation (?c) calculated. Statistical GraphPad Prism 8 (GraphPad Software, Diego, CA) Microsoft Excel 14.7.2 (Microsoft, Redmond, WA) < 0.05 statistically significant. Because vary sequence, amplicon length, annealing/extension. tolerant wide essentially zero