Rapid Implementation of Severe Acute Respiratory Syndrome Coronavirus 2 Emergency Use Authorization RT-PCR Testing and Experience at an Academic Medical Institution

An epidemic caused by an outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in China in December 2019 has since rapidly spread internationally, requiring urgent response from the clinical diagnostics community. We present a detailed overview of the clinical validation and implementation of the first laboratory-developed real-time RT-PCR test offered in the NewYork-Presbyterian Hospital system following the Emergency Use Authorization issued by the US Food and Drug Administration. Nasopharyngeal and sputum specimens (n = 174) were validated using newly designed dual-target real-time RT-PCR (altona RealStar SARS-CoV-2 Reagent) for detecting SARS-CoV-2 in upper respiratory tract and lower respiratory tract specimens. Accuracy testing demonstrated excellent assay agreement between expected and observed values and comparable diagnostic performance to reference tests. The limit of detection was 2.7 and 23.0 gene copies per reaction for nasopharyngeal and sputum specimens, respectively. Retrospective analysis of 1694 upper respiratory tract specimens from 1571 patients revealed increased positivity in older patients and males compared with females, and an increasing positivity rate from approximately 20% at the start of testing to 50% at the end of testing 3 weeks later. Herein, we demonstrate that the assay accurately and sensitively identifies SARS-CoV-2 in multiple specimen types in the clinical setting and summarize clinical data from early in the epidemic in New York City. An epidemic caused by an outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in China in December 2019 has since rapidly spread internationally, requiring urgent response from the clinical diagnostics community. We present a detailed overview of the clinical validation and implementation of the first laboratory-developed real-time RT-PCR test offered in the NewYork-Presbyterian Hospital system following the Emergency Use Authorization issued by the US Food and Drug Administration. Nasopharyngeal and sputum specimens (n = 174) were validated using newly designed dual-target real-time RT-PCR (altona RealStar SARS-CoV-2 Reagent) for detecting SARS-CoV-2 in upper respiratory tract and lower respiratory tract specimens. Accuracy testing demonstrated excellent assay agreement between expected and observed values and comparable diagnostic performance to reference tests. The limit of detection was 2.7 and 23.0 gene copies per reaction for nasopharyngeal and sputum specimens, respectively. Retrospective analysis of 1694 upper respiratory tract specimens from 1571 patients revealed increased positivity in older patients and males compared with females, and an increasing positivity rate from approximately 20% at the start of testing to 50% at the end of testing 3 weeks later. Herein, we demonstrate that the assay accurately and sensitively identifies SARS-CoV-2 in multiple specimen types in the clinical setting and summarize clinical data from early in the epidemic in New York City. The novel coronavirus severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a member of the Betacoronavirus genera in the subfamily Coronavirinae, which are known to cause respiratory illness and gastroenteritis in humans.1Lorusso A. Calistri P. Petrini A. Savini G. Decaro N. Novel coronavirus (SARS-CoV-2) epidemic: a veterinary perspective.Vet Ital. 2020; 56: 5-10PubMed Google Scholar,2Wei X. Li X. Cui J. Evolutionary perspectives on novel coronaviruses identified in pneumonia cases in China.Natl Sci Rev. 2020; 7: 239-242Crossref PubMed Scopus (34) Google Scholar An outbreak of respiratory disease caused by SARS-CoV-2, first detected in Wuhan, China, in December 2019, rapidly spread to other countries, including the United States.3Chen L. Li Q. Zheng D. Jiang H. Wei Y. Zou L. Feng L. Xiong G. Sun G. Wang H. Zhao Y. Qiao J. Clinical characteristics of pregnant women with Covid-19 in Wuhan, China.N Engl J Med. 2020; 382: e100Crossref PubMed Scopus (401) Google Scholar,4Chen N. Zhou M. Dong X. Qu J. Gong F. Han Y. Qiu Y. Wang J. Liu Y. Wei Y. Xia J. Yu T. Zhang X. Zhang L. Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study.Lancet. 2020; 395: 507-513Abstract Full Text Full Text PDF PubMed Scopus (14509) Google Scholar New York City in particular became an epicenter of the pandemic.5Goyal P. Choi J.J. Pinheiro L.C. Schenck E.J. Chen R. Jabri A. Satlin M.J. Campion Jr., T.R. Nahid M. Ringel J.B. Hoffman K.L. Alshak M.N. Li H.A. Wehmeyer G.T. Rajan M. Reshetnyak E. Hupert N. Horn E.M. Martinez F.J. Gulick R.M. Safford M.M. Clinical characteristics of COVID-19 in New York City.N Engl J Med. 2020; 382: 2372-2374Crossref PubMed Scopus (1602) Google Scholar Given the devastating impact on the health care system and the need for accurate and quick diagnosis of SARS-CoV-2 infection, the US Food and Drug Administration (FDA) has established a rapid pathway for using laboratory-developed tests that was outlined in a guidance document published on February 29, 2020 (https://www.fda.gov/media/135659/download, last accessed October 22, 2020). According to this guidance, SARS-CoV-2 testing could be performed by Clinical Laboratory Improvement Amendments–certified high-complexity laboratories under Emergency Use Authorization (EUA), according to a set of recommendations regarding the minimum validation required for ensuring the analytical and clinical validity of the test. Details of the test and validation had to be submitted by the laboratory to the FDA through an EUA application within 15 days of initiating testing, after which testing could continue provisionally until a decision by the FDA was rendered. The CDC and the New York State Department of Health had designed and manufactured new test kits for SARS-CoV-2. However, few laboratories were able to get access to these reagents or had the required instruments, which were also not available in our institution. Limited access to SARS-CoV-2 RNA reference control material presented another significant hurdle to the validation process. The FDA EUA announcement required laboratories to procure SARS-CoV-2 RNA from the World Reference Center for Emerging Viruses and Arboviruses (WRCEVA) or the NIH Biodefense and Emerging Infections Research Resources Repository. The scale of demand for diagnostic testing and the shortage of supplies led to the need for high-throughput testing that could be readily implemented in a variety of laboratories. Herein, we describe the validation and implementation of an EUA test in respiratory tract, including nasopharyngeal (NP) and sputum specimens, using research use only RealStar SARS-CoV-2 Reagent (altona Diagnostics GmbH, Hamburg, Germany). We also detail workflow considerations and results using the test from the early days of the coronavirus disease 2019 (COVID-19) outbreak in New York City (March 11, 2020, through March 31, 2020) for upper tract respiratory (URT) specimens and after the peak number of cases (April 17, 2020, to May 15, 2020) for lower respiratory tract (LRT) specimens. SARS-CoV-2 RNA reference material from WRCEVA was obtained from the University of Texas Medical Branch (strain USA_WA1/2020, lot TVP 23156) for use in clinical evaluation and limit of detection (LOD) studies. Clinical samples used for the initial validation were residual NP and sputum samples from routine clinical testing of patients suspected of having respiratory tract infections and pooled archived frozen (NP and sputum) samples used as matrix for generating contrived specimens. Reactive clinical NP samples also included four SARS-CoV-2–positive samples tested by the New York City Department of Health and Mental Hygiene. NP samples were tested for the presence of 21 common respiratory viruses using the BioFire FilmArray Respiratory Pathogen 2 panel (BioFire Diagnostics, LLC, Salt Lake City, UT). Moreover, to further evaluate the performance of the test, a total of 30 NP and 20 sputum patient specimens were also analyzed in parallel using the Roche cobas 6800 (Roche Diagnostics, Indianapolis, IN) or the Cepheid GeneXpert (Cepheid, Sunnyvale, CA) and the Hologic Panther Fusion system (Hologic, Inc., Marlborough, MA), respectively. Additional retrospective analysis of patient characteristics was performed on consecutive URT (n = 1694) and LRT (n = 141) specimens obtained from patients with high suspicion for COVID-19 who were treated at NewYork-Presbyterian Hospital campuses between March 11, 2020, and March 31, 2020, and between April 17, 2020, and May 15, 2020, respectively. The Institutional Review Board Committee at Weill Cornell Medicine approved this study. All specimens were initially processed using Biosafety level 2 biosafety measures (CDC, https://www.cdc.gov/coronavirus/2019-ncov/lab/lab-biosafety-guidelines.html, last accessed October 22, 2020). An off-board lysis viral inactivation step was performed on 200 μL of NP swab viral transport media and was followed by automated extraction of total nucleic acid using the QIAsymphony DSP Virus/Pathogen Mini Kit coupled on the QIAsymphony SP (Qiagen, Germantown, MD) to produce a resulting eluate volume of 60 μL. For lysis inactivation, specimens were mixed with ACL buffer, followed by vortexing with a lysis mix consisting of internal control (IC), carrier RNA, proteinase K, and AVE and ATL buffers, and then incubated at 68°C for 15 minutes. The IC consisted of proprietary artificial RNA/DNA molecules with no homologies to any other known sequences, and was included with the RT-PCR reagents. For sputum, 100 μL of specimen was first treated with 0.3% dithiothreitol solution (1:1 ratio) and incubated at 37°C for 30 minutes to reduce viscosity. One-step to and real-time RT-PCR of the viral and were and IC was performed using μL total nucleic acid eluate and the RealStar SARS-CoV-2 real-time RT-PCR Kit (altona Diagnostics on the in a total reaction volume of 30 μL. and detection were performed using for the and of RNA, and IC in a in which the and the or only the gene or the gene SARS-CoV-2 RNA were detected within the first of were The presence of gene was as to any viral and IC was as The FDA EUA guidance four performance consisting of and clinical For the to were performed with at by WRCEVA RNA WRCEVA RNA reference approximately NP and sputum RNA obtained from patient NP or sputum P. and use diagnostics for the detection of SARS-CoV-2 2020; Scopus Google Scholar was with 20 for of and For the a total of NP and 60 sputum) and NP and 30 sputum) specimens were specimens were patient specimens tested by or contrived specimens that were by WRCEVA RNA pooled NP or sputum specimen RNA as of the contrived clinical specimens were at a of to with the of samples the assay testing FDA guidance the for test performance as agreement at to and agreement at other and P. and use diagnostics for the detection of SARS-CoV-2 2020; Scopus Google Scholar and were performed by Diagnostics Additional were performed using NP samples that tested by the BioFire FilmArray Respiratory Pathogen 2 panel for other coronaviruses as from the by the including and were performed using and for for Scholar was through a using the following and Laboratory MD) with to F. with to Scholar on pooled NP specimens with WRCEVA RNA reference with a viral of gene per reaction to gene copies per reaction to demonstrated an accurate and response of detection for NP specimens and four of detection for sputum specimens and analysis was to the NP data after an of testing were performed at and gene copies per and the to 2.7 gene copies per reaction at detection rate and analysis were performed on sputum at and 20 gene copies per and a lower with an of 23.0 gene copies per reaction at detection All 20 of 20 NP and of sputum tested at as of for NP and with at copies per 20 (NP) and specimens were tested at the limit of detection of and 100 gene copies per internal not nasopharyngeal viral transport in a new of detection (LOD) of nasopharyngeal (NP) and sputum by Additional were performed using World Reference Center for Emerging Viruses and Arboviruses RNA reference material in pooled NP specimen eluate and sputum specimen eluate to the and of were at gene copies per reaction and at gene copies per for NP and of were at 100 gene copies per reaction and at gene per for sputum specimens. An were performed at and gene copies per reaction for NP specimens and and 20 gene copies per reaction for sputum specimens. analysis to be 2.7 gene copies per reaction for NP specimens and 23.0 gene copies per reaction for sputum specimens at detection rate An 20 (NP) and specimens were tested at the limit of detection of and 100 gene copies per respectively. internal not nasopharyngeal viral transport The in analysis for (altona Diagnostics demonstrated of the and gene and and with SARS-CoV-2 published in on All last accessed October 22, 2020) and Center for as of March analysis for demonstrated of the and with the of was between the and a of in the and of and and in the in of P. and in cases of was not a as only the or was samples (n = (n = (n = and (n = detected by BioFire FilmArray Respiratory Pathogen 2 panel tested for SARS-CoV-2 on the RealStar real-time RT-PCR Clinical evaluation in the detection of SARS-CoV-2 in specimens contrived by WRCEVA RNA reference material (n = pooled NP viral transport media or sputum and four patient samples tested by New York City Department of Health and Mental 2 and The patient at (n = the of to to and clinical evaluation were performed for sputum specimens the by the also with 2 and All archived sputum and NP specimens that tested on the BioFire FilmArray Respiratory Pathogen 2 panel also tested on the SARS-CoV-2 real-time RT-PCR assay of Accuracy by and sputum SARS-CoV-2 contrived specimens with RNA and sputum SARS-CoV-2 contrived specimens with RNA swab clinical samples tested by New York State Department of Health as a of the initial NP clinical samples by reference tests after SARS-CoV-2–positive specimens with lower respiratory tract samples pooled sputum patient evaluation of the RealStar SARS-CoV-2 real-time RT-PCR assay using automated total nucleic acid followed by real-time RT-PCR the and coronavirus and of values are for The number of contrived through RNA pooled RNA eluate or of clinical or patient and specimens is also with the of specimens that were as or NP and sputum SARS-CoV-2 contrived specimens with RNA NP swab clinical samples tested by New York State Department of Health as a of the initial NP clinical samples by reference tests after SARS-CoV-2–positive specimens with lower respiratory tract samples pooled sputum patient in a new Clinical evaluation of the RealStar SARS-CoV-2 real-time RT-PCR assay using automated total nucleic acid followed by real-time RT-PCR the and coronavirus and of values are for The number of contrived through RNA pooled RNA eluate or of clinical or patient and specimens is also with the of specimens that were as or Additional performed with patient specimens after testing demonstrated that of the NP and 20 sputum SARS-CoV-2 specimens that tested with the test also with the RealStar real-time RT-PCR test. The values were and for the NP and sputum the 20 SARS-CoV-2–positive reference specimens, tested were samples with values The specimen tested on the Roche test only and on the test, that the had viral values are not available for the NP and sputum specimens to New York State Department of Health and and University Salt Lake City, respectively. The was of of on and patient NP swab specimens were tested using US Food and Drug Administration Emergency Use Authorization SARS-CoV-2 RT-PCR on the Roche cobas 6800 or Cepheid GeneXpert The values are from on the gene (altona Diagnostics SARS-CoV-2 2 and gene four NP specimens were tested by EUA sputum specimens were tested at reference laboratory SARS-CoV-2 assay on Panther 20 specimens, were contrived samples obtained by a of to from a patient lower respiratory tract to SARS-CoV-2 by a SARS-CoV-2 EUA assay pooled sputum of the samples and were patient sputum not New York State Department of in a new patient NP swab specimens were tested using US Food and Drug Administration Emergency Use Authorization SARS-CoV-2 RT-PCR on the Roche cobas 6800 or Cepheid GeneXpert The values are from on the gene (altona Diagnostics SARS-CoV-2 2 and gene four NP specimens were tested by EUA sputum specimens were tested at reference laboratory SARS-CoV-2 assay on Panther 20 specimens, were contrived samples obtained by a of to from a patient lower respiratory tract to SARS-CoV-2 by a SARS-CoV-2 EUA assay pooled sputum of the samples and were patient sputum not New York State Department of the early days of the in New York City, we performed 1694 tests on NP specimens specimens and NP specimens from 1571 values were not for the and IC between or samples The number of tests with or results were and in NP swab The values for and IC in samples were 23.0 to to and to of the gene as a for viral the upper respiratory tract specimens could be high = = 20 to = and = P. Satlin M.J. A. M. Chen A. T. A. M. H. M. of high-throughput reaction for the detection of severe acute respiratory syndrome coronavirus 2020; PubMed Scopus Google Scholar these initial 3 weeks of testing, of samples could be as having to high viral patients with testing, only had results on the test, including patients who first tested tested and patients who had detected in specimen (NP or not the the patients who from to were initially tested at the testing performed within 3 had viral had viral and had high viral The patient tested as an and then with high viral as an days later. of the patients in the data set presented with and were tested for suspected with SARS-CoV-2, and patients in the and were for SARS-CoV-2 as a to be required with health care consisted of patients with a positivity of from test to and in laboratory to were to and to respectively. The of tests with detected SARS-CoV-2 increased as the weeks and at approximately 50% from March 2020, to March 2020 of the samples were from the (n = followed by (n = and (n = and the positivity rate was in the with 50% of patients with detected SARS-CoV-2 = was a significant in the = and = with lower of detected in patients and of patients had detected older patients = tested with patients = = this was not the with patients in the = was after patients from the and who were of = was no significant in the of tests in = of = = = and other were not used the test for An tests were performed at at other these were not included in the values are in in a new For and other were not used the test for An tests were performed at at other these were not included in the values are in respiratory tract specimens, including sputum, and were and tested using the RealStar SARS-CoV-2 test April 17, of May 15, 2020, sputum, 30 and specimens had from with and SARS-CoV-2. results were for The values for and IC in LRT samples (n = were to to and to values were not for the and IC between and samples The number of LRT samples per was to which was lower compared with the number of URT specimens tested to Given the on clinical characteristics were not for LRT specimens. Given the increased rate of SARS-CoV-2 and the of any available the FDA a pathway on February 29, 2020, that laboratories to laboratory-developed tests to this diagnostic and rapid implementation of a clinical test was an in the early of the the spread of the was Diagnostics the RealStar SARS-CoV-2 research use only assay as as the disease spread to on February R. A. A. N. A. of laboratory diagnostics for the novel coronavirus SARS-CoV-2 in February 2020; Scopus Google Scholar The reagents were designed as a dual-target rapid detection of and RNA in a the test was and to with a comparable to other RT-PCR R. A. A. N. A. of laboratory diagnostics for the novel coronavirus SARS-CoV-2 in February 2020; Scopus Google M. R. A. T. J. L. G. J. M. M. H. of 2019 novel coronavirus by real-time 2020; PubMed Scopus Google J. M. E. M. L. A. H. P. the analytical performance of SARS-CoV-2 diagnostic 2020; PubMed Scopus Google Scholar were also able to use the reagents with a of extraction and real-time instruments, for in R. A. A. N. A. of laboratory diagnostics for the novel coronavirus SARS-CoV-2 in February 2020; Scopus Google M. R. A. T. J. L. G. J. M. M. H. of 2019 novel coronavirus by real-time 2020; PubMed Scopus Google J. M. E. M. L. A. H. P. the analytical performance of SARS-CoV-2 diagnostic 2020; PubMed Scopus Google Scholar the access for compared with other EUA tests in that rapid validation of types other NP could be Accuracy of NP and sputum samples in our laboratory excellent agreement between the expected and observed results for contrived clinical samples and patient specimens tested by this we were able to of the samples within the The reference (Roche and a compared with the test, which be at in by the of initial volume used for testing compared with the required for the Roche 6800 and Cepheid P. Satlin M.J. A. M. Chen A. T. A. M. H. M. of high-throughput reaction for the detection of severe acute respiratory syndrome coronavirus 2020; PubMed Scopus Google Scholar test is for the detection and of SARS-CoV-2 in and of a respiratory to early of The for the NP and sputum samples were 2.7 and 23.0 gene copies per a analytical for the NP specimens compared with the analytical of this assay by specimen is compared with the in the by other R. A. A. N. A. of laboratory diagnostics for the novel coronavirus SARS-CoV-2 in February 2020; Scopus Google M. R. A. T. J. L. G. J. M. M. H. of 2019 novel coronavirus by real-time 2020; PubMed Scopus Google J. M. E. M. L. A. H. P. the analytical performance of SARS-CoV-2 diagnostic 2020; PubMed Scopus Google Scholar be by extraction types and the analysis performed in our study. The in the between the types is to the high of sputum, of nucleic acid L. A. E. F. F. M. G. F. A. G. F. and detection of SARS-CoV-2 RNA using the COVID-19 2020; PubMed Scopus Google Scholar the analytical was lower in sputum LRT samples were to test for the compared with URT and other types in COVID-19 Y. R. R. Han G. of SARS-CoV-2 in types of clinical 2020; Scholar the of these validation routine SARS-CoV-2 testing with the RealStar SARS-CoV-2 real-time RT-PCR assay was on March 11, Given the and of WRCEVA SARS-CoV-2 control we were able to the validation within a followed by a testing evaluation the to total of 1694 URT and LRT specimens from 1571 and were The lower number of LRT compared with URT specimens that LRT testing to patients who of NP LRT or to patients with high suspicion for COVID-19 with testing by RT-PCR NP the of patients tested 3 weeks by our results were in older males compared with and which has by N. J. Chen X. Chen H. Wang D. Liu N. Liu D. Chen G. Zhang Y. Li D. Li J. H. Zhang L. Zhang J. of COVID-19 in 2020; Full Text Full Text PDF PubMed Scopus Google Liu Wang Wang carrier acute respiratory and pneumonia to severe acute respiratory syndrome coronavirus 2 and 2020; PubMed Scopus Google Scholar women to a of compared with an that has to the of T. Zhang J. Wang T. Cui P. Chen Jiang J. Zhou J. Wang L. Y. J. X. X. Wang in Wuhan, China: of and on SARS-CoV-2 2020; Google Scholar this older women were to test for SARS-CoV-2 compared with However, the in detection rate between older and women was after patients of the of of positivity to of patients to and the of designed studies. the only tested for SARS-CoV-2, which is to the obtained for women at at other NewYork-Presbyterian Hospital D. M. D. for SARS-CoV-2 in women for Engl J Med. 2020; 382: PubMed Scopus Google Scholar in the number of tests were by this was early in the epidemic in New York City. The had tests patients to be compared with The of tests increased and at 50% at 3 weeks the with testing on other positivity as high as to as the epidemic peak in this patients tested after initial results in the had to be of increased viral on days after or specimen N. G. J.B. Dong Dong We T. A. N. as a cause of COVID-19 diagnostic test 2020; PubMed Scopus Google Scholar we the clinical and implementation of an FDA EUA real-time RT-PCR test for SARS-CoV-2 in our a to in tests. We also detailed the clinical and testing characteristics of the first of COVID-19 patients to our the early days of the viral outbreak in New York City. We of the and health care who performed and in testing at the clinical laboratories of NewYork-Presbyterian Cornell Reference Center for Emerging Viruses and for viral RNA control and Diagnostics for of reagents and values for and internal control are for the Nasopharyngeal (NP) or were tested from March 11, 2020, to March 2020; NP or swab specimens were tested from March 11, 2020, to March 2020; and only NP swab specimens were tested from March 2020, to March was no significant in values for the gene = gene = or IC = between these upper respiratory tract respiratory tract specimens were tested April 17, 2020, and results specimens until May 15, was no significant in values for the gene = gene = or IC = between or types sputum samples were with with