key: cord-0871540-m7sdgkzh
authors: Schellenberg, John J.; Ormond, Margaret; Keynan, Yoav
title: Extraction-free RT-LAMP to detect SARS-CoV-2 is less sensitive but highly specific compared to standard RT-PCR in 101 samples
date: 2021-02-16
journal: J Clin Virol
DOI: 10.1016/j.jcv.2021.104764
sha: 8661a473a4acd85ca22ad51744e394a3706635cf
doc_id: 871540
cord_uid: m7sdgkzh

The current scale of public and private testing cannot be expected to meet the emerging need for higher levels of community-level and repeated screening of asymptomatic Canadians for SARS-CoV-2. Rapid point-of-care techniques are increasingly being offered to fill the gap in screening levels required to identify undiagnosed individuals with high viral loads. However, rapid, point-of-care tests often have lower sensitivity in practice. Reverse transcription loop-mediated isothermal amplification (RT-LAMP) for SARS-CoV-2 has proven sensitive and specific and provides visual results in minutes. Using a commercially available kit for RT-LAMP and primer set targetting nucleocapsid (N), we tested a blinded set of 101 archived nasopharyngeal (NP) swab samples with known RT-PCR results. RT-LAMP reactions were incubated at 65 °C for 30 minutes, using heat-inactivated nasopharyngeal swab sample in viral transport medium, diluted tenfold in water, as input. RT-LAMP agreed with all RT-PCR defined negatives (N = 51), and all positives with cycle threshold (Ct) less than 20 (N = 24), 65% of positives with Ct between 20-30 (N = 17), and no positives with Ct greater than 30 (N = 9). RT-LAMP requires fewer and different core components, so may not compete directly with the mainline testing workflow, preserving precious central laboratory resources for those with the greatest need. Careful messaging must be provided when using less-sensitive tests, so that people are not falsely reassured by negative results, but this caveat must be weighed against the clear benefits of reliably identifying those with high levels of virus in prioritized samples at the point of care.

 Relative to RT-PCR in 101 samples, extract-free RT-LAMP had a specificity of 100% and sensitivity of 77%, with all strong positives (Ct < 24) detected, indicating that the most infectious individuals would be correctly identified with this assay.

 Using a positive control at a known concentration, the limit of detection for RT-LAMP was determined to be 50,000 copies/ml, comparable to other studies and consistent with concentrations at which SARS-CoV-2 can be cultured, an indicator of transmissibility. In an unprecedented scientific feat, nucleic acid amplification tests for SARS-CoV-2 were   published 21 days after the Chinese communicable disease control team arrived in Wuhan on 31 December 2019 [1] , based on complete genome sequences published 10 days earlier [2] . Hundreds of molecular diagnostic tests for SARS-CoV-2 have been introduced since then [3] [4] [5] . The main diagnostic targets rely on specific host antibodies, viral proteins and viral RNA, each with its own specific benefits and limitations in accurately detecting SARS-CoV-2 during its infectious period, in order to more efficiently prevent its spread [6] . The best, most sensitive, tests require high levels of laboratory expertise and specialized facilities that understandably increase expense and turnaround time. This is especially true when challenged with recent, unprecedented testing volumes.

Advice on whether to test people "without symptoms" (which represents a range of presymptomatic, peri-symptomatic, sub-symptomatic and truly asymptomatic phenotypes) wavers as much over time as actual demand for testing by people "without symptoms" [7, 8] .

"Asymptomatic" testing has been characterized as wasteful, since only a small percentage come back positive, especially when investing the most precise, most expensive and most timeconsuming test available. Testing programs that incorporate parallel strategies for detecting infection in those without symptoms have been demonstrated at large scale in Iceland [9] and more recently country-wide in Slovakia [10].

Several jurisdictions have approved and applied point-of-care assays that can rapidly indicate infection outside the laboratory. These include rapid antibody-detecting cassettes that use serum or a few drops of peripheral blood, rapid antigen tests on nasopharyngeal samples, and all-in-one RT-PCR platforms such as the Xpert® Xpress SARS-CoV-2 (Cepheid®), which provides highly sensitive and specific results relative to other RT-PCR-based workflows in a fraction of the time [11, 12] . However, a key limitation is lack of scalability, with each (very expensive) machine only able to run one or a few samples at a time. Fluctuating demand for cartridges across all jurisdictions makes overall access unpredictable.

More recently, a number of nucleic acid isothermal amplification techniques for viral detection have been described, including loop-mediated isothermal amplification (LAMP) [13-15], using a 4-or 6-primer set and incubated at 65˚C for 30-60 minutes, recombinase-polymerase amplification (RPA) [16] , using multiple enzymes and incubated at 42˚C for 15-20 minutes, and many others [17] [18] [19] . Isothermal assays have been shown to detect SARS-CoV-2 with very high sensitivity and specificity, and underlie several tests recently authorized for emergency use, including the ID NOW from Abbott, and CRISPR-based detection by SHERLOCK [20] (currently not approved for sale outside the United States).

The ID NOW compares favourably to GeneXpert [21] , but also shares its disadvantages in terms of single sample per cartridge and single-source cartridges. Some kits, such as Twist Bioscience's RPA-based assays [22] and the SHERLOCK LAMP/CRISPR assays, do not use special machines or cartridges, requiring only strip tubes or plates and p10 filter tips, hence can easily be scaled up to conduct hundreds or thousands of tests. New England Biolabs' Colorimetric WarmStart LAMP kit can also be scaled and is available in Canada. In this study, we evaluated how a commercial colorimetric RT-LAMP kit combined with published SARS-CoV-2 primer sets and heat-treated, diluted sample compares with standard RT-PCR analysis.

J o u r n a l P r e -p r o o f

Archived sample set -A total of 101 leftover nasopharyngeal swab samples in viral transport medium, from both symptomatic and asymptomatic individuals attending for testing at walk-in or drive-through facilities in Winnipeg, Canada, tested for SARS-CoV-2 by RT-PCR using E gene primers and purified RNA template on the Cobas or Panther Fusion platforms, were used to validate a recently described RT-LAMP assay [23] . Comparisons were based on 50 RT-PCR positive samples, with fluorescence threshold values between 10 and 37 cycles, and 51 RT-PCR negative samples.

Sample processing and heat treatment -After assigning blinded sample identifiers to reduce potential bias in interpretation, samples were thawed and briefly spun down in a minicentrifuge to collect cells and debris. An aliquot of 60 l, drawn from the bottom of the tube near the pelleted material, was transferred to a 1.7 ml Eppendorf tube, labelled with the blinded code, then incubated in a heating block at 95˚C for 5 minutes. Serial tenfold dilutions of sample were prepared in nuclease-free water (New England Biolabs, Whitby, ON).

WarmStart LAMP Kit (New England Biolabs) and a published LAMP primer set targeting the N gene of SARS-CoV-2 (Table 1) , previously shown to have high sensitivity [23, 24] 

Sensitivity, specificity and predictive values -Sensitivity of the RT-LAMP assay using raw, heat-inactivated sample was 77% compared to RT-PCR of purified RNA extracts ( Table 2) .

Samples with the lowest Ct values (<22) were all bright yellow by RT-LAMP, even when diluted up to 1000-fold, indicating a similar range of detectable concentration to the positive control. All RT-PCR negatives were also RT-LAMP negative ( Table 2 ). This indicates that RT-LAMP may be most useful to more quickly detect those with high viral loads.

Interpretation of colour change -Only samples that were bright yellow after 30 minutes incubation were considered positive ( Figure 1 ). In some cases, a partial colour change resulted, from magenta to pink to orange, but did not become yellow within the period of observation (20- 30 minutes, regardless of dilution factor down to 1,000 copies). This partial colour change may indicate a weak positive (all samples do eventually turn yellow), a failed reaction (not indicating anything about the sample itself), or a true negative, depending on the reason for the weak change in pH. In this study, 9 samples initially resulted in an orange colour and all were negative when re-tested (8 were from PCR-negative samples, and the remaining sample was PCR-positive with a Ct value of 29).

Controls and limit of detection -Negative controls were reliably magenta throughout the experiments ( Figure 2 ) but did eventually get contaminated. Routine cleanup with DNAse and careful separation of pre-amplification and amplification/post-amplification areas are required (unless you use NEB's more expensive dUTP kit) to prevent the massive DNA contamination that LAMP is prone to. The MERS-CoV DNA negative control at 50,000 and 500,000 copies/ml did not produce any yellow colour, however 5,000,000 copies/ml resulted in a positive reaction (not shown). SARS-CoV-2-N DNA positive control produced a bright yellow reaction within 30 minutes down to 50,000 copies/ml, but remained magenta at lower dilutions, indicating that this is the lower limit of detection of this test under current conditions (Figure 2 ).

In this study, we confirm that one of the fastest, simplest, cheapest and most scalable protocols available for detecting SARS-CoV-2 nucleic acid is so far reliable for strong positives only, but highly specific compared to RT-PCR detection. Other studies have shown similar J o u r n a l P r e -p r o o f reduced sensitivity but high specificity [23, [25] [26] [27] [28] . One early study found that RT-LAMP was more sensitive than RT-PCR at detecting virus in an asymptomatic carrier monitored for several days [29] . RT-LAMP has been tested extensively at large scale in the UK and has recently been shown to have less-than-expected sensitivity (<50%) when rolled out to large populations as part of Operation Moonshot [30] . However, similarly to the current study, sensitivity was greatest in strong positives, indicating that only those with high viral loads may reliably be detected. Recent work has shown that SARS-CoV-2 could not be cultured from samples with Ct values greater than 24 and/or longer than 8 days past symptom onset [31] . This observation indicates that the RT-LAMP assay described here would detect all of those most likely to have culturable (and therefore infectious) virus, as has been shown when the two have been compared directly [25] . Several jurisdictions use the Panbio Ag Rapid Test Device to identify positives early, assuming that negative results should be confirmed [34] [35] [36] . One recent study showed that none of those who were test discordant (positive by RT-PCR but negative using Panbio) had culturable virus in their sample [34] , indicating potentially reduced transmissibility by individuals defined as "false negatives" in current protocols. Other studies have indicated that sensitivity of the Panbio devices drops in people with no symptoms [36] , or when using self-collected specimens such as nasal swabs and saliva [37] .

The impact of mutations on our ability to accurately detect SARS-CoV-2 with primers based on older genome sequences is an area of growing concern [38], requiring constant vigilance and updating of primers being used. Mutations across the genome (including in the N gene targetted in this report) are regularly reported around the world [39] , and can conveniently be monitored online (www.covidcg.org) [40] . Other strategies to mitigate the risk of false negative RT-LAMP tests due to mutations, for example by targetting more than one gene (standard in RT-PCR assays), have also recently been reported [41] .

Molecular diagnostic techniques that use raw or minimally processed sample are fastest, but less sensitive, since the background chemistry of a raw sample may introduce uncontrolled variability [25, 42] . Potential inhibitors or other contaminants may reduce the reliability of both positive and negative results. This concern is somewhat lessened in the context of the current report, which we have confirmed is highly specific and can amplify target molecules even in highly diluted samples. However, the current assay's limit of detection (50,000 copies/ml, similar to other studies [25] ) does decrease the likelihood of detecting a person with low viral load. In contrast, Sherlock Biosciences claims on its website (www.sherlock.bio) that its LAMP-based assay can detect 7,000 copies per ml VTM or more than 7 times fewer than this report.

Is the risk of missing someone with a low viral load greater than making a person with a high viral load wait days to find out they are positive? In all cases, expectations about what a negative test result actually means must be properly managed. In this study, RT-LAMP with raw, heat-treated sample was ~75% sensitive compared to nasopharyngeal RT-PCR, meaning that a false negative result can be expected 1 out of 4 times the test is performed. Therefore, careful explanation should be provided as part of pre-test and post-test counselling that a single negative test does not mean a person is free of SARS-CoV-2.

Another way to have more confidence in less sensitive tests is by repeated testing, increasing confidence that a negative test result is not just due to random error, transience of viral shedding or low test sensitivity [43] . Regular testing is facilitated when methods are fast, cheap, based on saliva, extract-free and easy to conduct in diverse settings, identifying the earliest timepoint at which a person is infectious, and increasing the immediacy and impact of contact tracing. Building an expanded repertoire of testing platforms that do not all depend on the same instruments, reagents and laboratory infrastructure is an essential strategy to improve surge capacity [33] . Ideally, new testing modalities should not compete with central laboratory tests for equipment, consumables or personnel, all of which are presently stretched very thin. Core costs and processing time of extraction-free RT-LAMP are similar to RT-PCR workflows (Table 4 ), but can be done without expensive machines or specialized labs, by trained non-specialists with oversight and support. This frees nurses and other front-line staff for less technical duties and preserves public health testing resources for those most in need.

In conclusion, scalable rapid tests, such as the one evaluated in this study, may efficiently detect individuals with high viral loads at the point of care. Our findings suggest RT-LAMP could be useful as a screening mechanism for prioritized samples within the existing test chain, reliably identifying those with highest virus concentrations ahead of the standard RT-PCR workflow, and able to be scaled to any required number of tests per day. Further work must focus on improving sensitivity, incorporating saliva or other self-collected samples to facilitate repeated testing, and triangulating evidence from different testing modalities [44] to better ascertain an individual's infectious period.

J o u r n a l P r e -p r o o f J o u r n a l P r e -p r o o f J o u r n a l P r e -p r o o f J o u r n a l P r e -p r o o f J o u r n a l P r e -p r o o f https://nationalpost.com/opinion/opinion-put-an-end-to-asymptomatic-testing-before-thesystem-gets-overwhelmed, Accessed Oct. 1, 2020. 

Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCR

Genomic characterization of the 2019 novel humanpathogenic coronavirus isolated from a patient with atypical pneumonia after visiting Wuhan

Center for Health Security, Bloomberg School of Public Health

COVID-19 In Vitro Diagnostic Devices and Test Methods Database. 2020

Coronavirus Test Tracker: Commercially Available COVID-19 Diagnostic Tests. 360Dx.com; 2020

COVID-19 diagnostic approaches: different roads to the same destination

Put an end to asymptomatic testing, before the system gets overwhelmed. National Post

Evaluation of clinical applicability of reverse transcription-loop-mediated isothermal amplification assay for detection and subtyping of Influenza A viruses

Recombinase polymerase amplification: Basics, applications and recent advances

RNA detection with high specificity and sensitivity using nested fluorogenic Mango NASBA

Performance characteristics of a high throughput automated transcription mediated amplification test for SARS-CoV-2 detection

Room Temperature Isothermal Colorimetric Padlock Probe Rolling Circle Amplification for Viral RNA Detection

Point-of-care testing for COVID-19 using SHERLOCK diagnostics

Comparison of Cepheid Xpert Xpress and Abbott ID Now to Roche cobas for the Rapid Detection of SARS-CoV-2

An enhanced isothermal amplification assay for viral detection

A colorimetric RT-LAMP assay and LAMPsequencing for detecting SARS-CoV-2 RNA in clinical samples

Rapid Molecular Detection of SARS-CoV-2 (COVID-19) Virus RNA Using Colorimetric LAMP. medRxiv

Optimizing direct RT-LAMP to detect transmissible SARS-CoV-2 from primary patient samples

Simple, Inexpensive, and Mobile Colorimetric Assay COVID-19-LAMP for Mass On-Site Screening of COVID-19

Rapid detection of COVID-19 coronavirus using a reverse transcriptional loop-mediated isothermal amplification (RT-LAMP) diagnostic platform

Detecting SARS-CoV-2 at point of care: preliminary data comparing loop-mediated isothermal amplification (LAMP) to polymerase chain reaction (PCR)

Development and Clinical Application of a Rapid and Sensitive Loop-Mediated Isothermal Amplification Test for SARS-CoV-2

Rapid test missed over 50% of positive cases in Manchester pilot

Predicting infectious SARS-CoV-2 from diagnostic samples

COVID-19 Diagnostics: How Do Saliva Tests Compare to Swabs? The Scientist

Evaluation of a rapid antigen test (Panbio TM COVID-19 Ag rapid test device) for SARS-CoV-2 detection in asymptomatic close contacts of COVID-19 patients

Evaluation of the PanbioTM rapid antigen test for SARS-CoV-2 in primary health care centers and test sites

Panbio antigen rapid test is reliable to diagnose SARS-CoV-2 infection in the first 7 days after the onset of symptoms

Evaluation of the rapid antigen test Panbio COVID-19 in saliva and nasal swabs in a population-based point-ofcare study

FDA Issues Alert Regarding SARS-CoV-2 Viral Mutation to Health Care Providers and Clinical Laboratory Staff (Press release

Genome-wide analysis of SARS-CoV-2 virus strains circulating worldwide implicates heterogeneity

COVID-19 CG: Tracking SARS-CoV-2 mutations by locations and dates of interest

Enhancing colorimetric loopmediated isothermal amplification speed and sensitivity with guanidine chloride

Rapid SARS-CoV-2 testing in primary material based on a novel multiplex LAMP assay

Test, re-test, re-test': using inaccurate tests to greatly increase the accuracy of COVID-19 testing

Combined Point-of-Care Nucleic Acid and Antibody Testing for SARS-CoV-2 following Emergence of D614G Spike Variant

Evaluation of ID NOW and RT-PCR for detection of SARS-CoV-2 in an ambulatory population

An extended laboratory validation study and comparative performance evaluation of the Abbott ID NOW TM COVID-19 assay in a coastal California tertiary care medical center

Evaluation of the Panbio Covid-19 rapid antigen detection test device for the screening of patients with Covid-19

Real-life validation of the Panbio COVID-19 antigen rapid test (Abbott) in community-dwelling subjects with symptoms of potential SARS-CoV-2 infection

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.J o u r n a l P r e -p r o o f