Study: Comparative evaluation of 19 reverse transcription loop-mediated isothermal amplification assays for detection of SARS-CoV-2

Coronavirus disease 2019 (COVID-19) caused by SARS-CoV-2 has caused a global pandemics. To facilitate the detection of SARS-CoV-2 infection, various RT-LAMP assays using 19 sets of primers had been developed, but never been compared. 

We performed comparative evaluation of the 19 sets of primers using 4 RNA standards and 29 clinical samples from COVID-19 patients. Six of 15 sets of primers were firstly identified to have faster amplification when tested with four RNA standards, and were further subjected to parallel comparison with the remaining four primer sets using 29 clinical samples. Among these 10 primer sets, Set-4 had the highest positive detection rate of SARS-CoV-2 (82.8%), followed by Set-10, Set-11, and Set-13 and Set-17 (75.9%). Set-14 showed the fastest amplification speed (Tt value < 8.5 min), followed by Set-17 (Tt value < 12.5 min). Based on the overall detection performance, Set-4, Set-10, Set-11, Set-13, Set-14 and Set-17 that target Nsp3, S, S, E, N and N gene regions of SARS-CoV-2, respectively, were determined to be better than the other primer sets. Two RT-LAMP assays with the Set-4 primers in combination with any one of four other primer sets (Set-14, Set-10, Set-11, and Set-13) were recommended to be used in the COVID-19 surveillance.

Introduction

Coronavirus disease 2019 (COVID-19), caused by the newly discovered coronavirus SARS-CoV-2^1, 2, is rapidly spreading throughout the world, posing a huge challenge to global public health security. As of 20 September, 2020, it has infected over 30.6 million people, and resulted in at least 950,000 deaths globally. In the absence of effective antiviral drugs or efficacious vaccines, early diagnosis of SARS-CoV-2 infection is essential for the containment of COVID-19^3,4, without which it is impossible to timely implement intervention and quarantine measures, and difficult to track contacts in order to limit virus spread.

Nucleic acid testing of various approaches are widely used as the primary tool for diagnosing COVID-19^3,4. Among them, reverse transcription quantitative PCR (RT-qPCR) methods have been set as the gold standard for laboratory confirmation of SARS-CoV-2 infection because of their proven track record as being the most robust technology in molecular diagnostics^4,5,6. However, the RT-qPCR assay relies on sophisticated facilities with reliable supply of electricity and well-trained personnel in large general hospitals and health care facilities, or government labs (such as CDC), and it is relatively time-consuming (about 1.5–2 h). These limit its capacity in point-of-care settings. Moreover, visiting a clinical setting for testing increases the risk of spreading the virus. Therefore, an alternative, fast, simple, and sensitive point-of-care testing (POCT) is highly needed to facilitate the detection of SARS-CoV-2 infection in resource-limited settings^3,7.

Loop-mediated isothermal amplification (LAMP) is a promising POCT method with high sensitivity, specificity, and rapidity, and it is easy-to-use⁸. To overcome the limitation of RT-qPCR assay, a number of RT-LAMP assays using at least 19 sets of different primers had been developed in the last few months for the detection of SARS-CoV-2^{9,10,11,12,13,14,15,16,17,18,19}. Although these assays had proven sensitive and effective for the detection of SARS-CoV-2, how do they compare with each other have not been evaluated. In this study, we compared all 19 sets of SARS-CoV-2-specific RT-LAMP primers using the mismatch-tolerant LAMP system that is faster and more sensitive than the conventional ones^20,21, and screened the high-efficiency RT-LAMP assays for use in the detection of SARS-CoV-2.

Discussion

SARS-CoV-2 transmission mainly occurs in the early and progressive stages of COVID-19 disease during which the patients and virus carriers have higher viral load than that in recovery stage^22,23,24, and are generally more infectious. To contain the spread of the virus, early diagnosis is essential^3,4. It helps to trigger timely intervention (e.g. quarantine, lockdown, and contact tracing), and facilitates to optimize clinical management. It is clear that serological assays are not suitable for this purpose, because detectable antibodies always appear several days after infection. Therefore, viral RNA testing is the primary method for early diagnosis of COVID-19. Despite being the most robust diagnostic tests, RT-qPCR-based assays are more centralized in core facilities, and they are not amenable for large-scale monitoring for asymptomatic and pre-symptomatic virus carriers in point-of-care settings (e.g. community and home). Therefore, community- and/or home-based nucleic acid assays that allow individuals to test in the community, at home, or other point-of-care sites without having to visit hospitals are convenient tools for the detection of SARS-CoV-2 infection by the general public^3,7.

RT-LAMP assays are such needed tools^8,20,21. In fact, various LAMP assays have been developed that included at least 19 sets of primers targeting to different genomic regions of SARS-CoV-2, with reported high detection sensitivity ranging from 1 to 1200 copies per 25 µL reaction^{9,10,11,12,13,14,15,16,17,18,19}. However, these primers are never formally evaluated with clinical samples. The sensitivity and performance of a RT-LAMP assay are mainly determined by the primers, because other components of the reaction system are optimized and stable. Therefore, assessing the optimal RT-LAMP primer sets for the detection of SARS-CoV-2 infection is important for the selection of the best assay format to use for large field screening of COVID-19 patients.

Recently, the reaction system of RT-LAMP was further optimized to have higher sensitivity and faster amplification speed, even allowing the presence of few mismatched bases between primer and templates in a mismatch-tolerant version^20,21. The new optimized reaction system containing an additional 0.15 U of high-fidelity DNA polymerase is called as mismatch-tolerant LAMP. The inclusion of an additional amount of high-fidelity DNA polymerase makes it a higher applicability to highly variable viruses, and a 10–15 min faster reaction speed than the conventional LAMP method. Using this new version, we assessed 19 sets of SARS-CoV-2 RT-LAMP primers. Six sets of primers showing faster amplification speed were firstly selected from 15 sets of primers using 4 RNA standards, and then tested with other 4 primer sets using 41 clinical samples. Eight sets of primers showed either comparable or better performance than the other 2 sets of primers (Set-1 and Set-18) as determined by positive detection rate. Of the 8 sets of primers, six were further selected based on high positive detection rate and/or overall faster amplification speed (with mean Tt value less than 13 min). The six primer sets are Set-4, Set-10, Set-11, Set-13, Set-14 and Set-17 that correspond to Nsp3, S, S, E, N, and N genes of SARS-CoV-2, respectively.

Among selected assays, the N gene-based RT-LAMP assays (Set-14 and Set-17) had the fastest amplification speed, followed by Orf, S and E gene-based assay (Set-4, Set-10, Set-11 and Set-13). This result suggested that the N gene-based RT-LAMP assay was more sensitive in detecting SARS-CoV-2 than that based on other genes, consisting with results of RT-qPCR assays⁵. In this study, Set-4 had the highest positive detection rates than all other primer sets, and had a LOD of 3 copies per 25 µL reaction, obviously more sensitive than the previously reported sensitivity of more than 100 copies per 25 µL reaction (Table 1 and Fig. 4)^12,14. The sensitivity of Set-4 was comparable with highly sensitive primer sets Set-13 and Set-14 (less than 3 copies per 25 µL reaction)^16,17. In addition, the sensitivity of primer Set-11 was less than 50 copies per 25 µL reaction (data not shown), obviously low 200 copies/25 µL reaction reported in previous study¹². These indicated the mismatch-tolerant method significantly improved the detection sensitivity of RT-LAMP²⁰. In addition, two of our previously reported primers, Set-8 and Set-18, exhibited high sensitivities of 3–20 copies per 25 µL reaction and good performance in the detection of clinical samples under the mismatch-tolerant reaction condition^9,10, but they did not show better performance than other nine primer sets in this study. A reason might be that the use of the mismatch-tolerant reaction system generally improved the amplification efficiency of the primers reported by other groups²⁰.

The analyzed primer sets showed high specificity in that they did not amplify any SARS-CoV-2 negative clinical samples. Sequence alignment analyses further supported that the six sets of optimal primers had good specificity to SARS-CoV-2, albeit they might generate non-specific amplification for SARS-CoV due to a high degree of sequence identity. However, given the lethal nature of both SARS-CoV-2 and SARS-CoV²⁵, a non-specific positive result for SARS-CoV might also be of clinical importance.

Two nucleic acid assays targeting to different genes are suggested to be used in the detection of SARS-CoV-2 to avoid potential false-negative results⁵. Based on comparable performances, any two of the six optimal primer sets (Set-4, Set-10, Set-11, Set-13, Set-14 and Set-17) were recommend to be used in the detection of SARS-CoV-2. However, simultaneous use of Set-10 and Set-11, or Set-14 and Set-17 should be avoided because the former two sets target to the same S gene and the latter two sets target to the same N gene. In addition, because of highest positive detection rate and high sensitivity, Set-4 was strongly encouraged to be preferentially selected for the diagnosis of COVID-19 patients. Apart from the six recommend primer sets, other primers such as Set-2 and Set-5 also had good performance, and can also be used in the monitoring of COVID-19 infections.

Another advantage of our version of the RT-LAMP assay is that the results are easily visualized with a pH-sensitive indicator dye (e.g. cresol red and neutral red)²⁶. Moreover, a combination of a nucleic acid extraction-free protocol and a master RT-LAMP mix containing all reagents (enzymes, primers, magnesium, nucleotides, dye and additives), except the template, enables the development of a simple kit that can be used at home, or a community-based diagnosis center for the detection of COVID-19 infection^3,27.

In summary, we evaluated and selected six optimal primer sets from 19 sets of SARS-CoV-2 RT-LAMP primers through a comparative evaluation with clinical RNA samples from COVID-19 patients. Two RT-LAMP assays with the Set-4 primers and any one of the other four primer sets (Set-10, Set-11, Set-13 and Set-14) were recommended to be used in the COVID-19 surveillance to facilitate the early finding of asymptomatic and pre-symptomatic virus carriers in clinical and point-of-care settings, and the monitoring of environmental samples in the field.