key: cord-1046539-38p19ge7 authors: Huang, Pei-Ying; Yin, Xin; Huang, Yue-Ting; Ye, Qi-Qing; Chen, Si-Qing; Cao, Xun-Jie; Xie, Tian-Ao; Guo, Xu-Guang title: Evaluation of CRISPR-Based Assays for Rapid Detection of SARS-CoV-2: A Systematic Review and Meta-Analysis date: 2022-04-20 journal: Yonsei Med J DOI: 10.3349/ymj.2022.63.5.480 sha: 0a9c20527b78c475702752ef29d908dcdd8636ac doc_id: 1046539 cord_uid: 38p19ge7 PURPOSE: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the pathogen of coronavirus disease 2019. Diagnostic methods based on the clustered regularly interspaced short palindromic repeats (CRISPR) have been developed to detect SARS-CoV-2 rapidly. Therefore, a systematic review and meta-analysis were performed to assess the diagnostic accuracy of CRISPR for detecting SARS-CoV-2 infection. MATERIALS AND METHODS: Studies published before August 2021 were retrieved from four databases, using the keywords “SARS-CoV-2” and “CRISPR.” Data were collected from these publications, and the sensitivity, specificity, negative likelihood ratio (NLR), positive likelihood ratio (PLR), and diagnostic odds ratio (DOR) were calculated. The summary receiver operating characteristic curve was plotted for analysis with MetaDiSc 1.4. The Stata 15.0 software was used to draw Deeks’ funnel plots to evaluate publication bias. RESULTS: We performed a pooled analysis of 38 independent studies shown in 30 publications. The reference standard was reverse transcription-quantitative PCR. The results indicated that the sensitivity of CRISPR-based methods for diagnosis was 0.94 (95% CI 0.93–0.95), the specificity was 0.98 (95% CI 0.97–0.99), the PLR was 34.03 (95% CI 20.81–55.66), the NLR was 0.08 (95% CI 0.06–0.10), and the DOR was 575.74 (95% CI 382.36–866.95). The area under the curve was 0.9894. CONCLUSION: Studies indicate that a diagnostic method based on CRISPR has high sensitivity and specificity. Therefore, this would be a potential diagnostic tool to improve the accuracy of SARS-CoV-2 detection. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was identified as the pathogen of the coronavirus disease 2019 (COVID- 19) , and it has caused more than 1.45 million deaths worldwide by November 30, 2020. 1 Patients infected with SARS-CoV-2 may exhibit symptoms such as shortness of dyspnea, high fever, and pneumonia, which are fatal for vulnerable individuals. 2 Coronavirus-infected inpatients are more likely to develop acute respiratory failure, pulmonary embolism, or septic shock, resulting in death. 3 Moreover, with the sharply increasing number of infected people and limited assays currently, the development of efficient, rapid, accurate, and sensitive SARS-CoV-2 sensing tools is urgent for public health in the world. 4 Molecular tests and serological tests have been implemented for COVID-19 diagnosis to detect viral RNA and anti-SARS- CoV-2 immunoglobulins, respectively. 5 For molecular diagnostic tests, the collection of upper nasopharyngeal swabs is recommended by the US Centers for Disease Control and Prevention. So far, reverse transcription-quantitative PCR (RT-qP-CR) has widely been used as the reference standard for the detection of viral RNA in SARS-CoV-2. [6] [7] [8] [9] However, it requires well-trained personnel and advanced equipment, which limits the application of RT-qPCR, especially in resource-constrained developing countries. [10] [11] [12] Metagenomic next-generation sequencing is another molecular test to identify SARS-CoV-2, but the sensitivity of this method is restricted by the influence of the human host background. 13 On the other hand, the serology tests, including immunochromatographic analysis and enzyme-linked immunosorbent assay (ELISA), are not sufficiently accurate in detecting SARS-CoV-2. 4 In addition, asymptomatic patients are considered to play a major role in the spread of the virus. 14 These factors increase the need for effective, cheap, and rapid alternative methods. 4 The clustered regularly interspaced short palindromic repeats (CRISPR)-CRISPR-associated proteins (Cas) system shows strong collateral activity against single-stranded RNA and DNA targets through molecular immune mechanisms, providing highly accurate methods of nucleic acid detection. 15 The mechanism of the detection system is the specific binding and cleavage activity of CRISPR-Cas. Once the primers for reverse transcription loop-mediated isothermal amplification or reverse transcription recombinase polymerase amplification recognize the specific regions of the SARS-CoV-2 genome, the targeted nucleic acid is amplified at a constant temperature. The guide RNAs then target SARS-CoV-2 E, N, or Orf1ab amplicons with the base-pairing pattern at attomolar sensitivity, ensuring the amplified nucleotide cleaved by the Cas nuclease accurately. The target nucleotide is finally identified on the detection platform with fluorescence tracking. 16, 17 Therefore, CRISPR is a more efficient and suitable point-of-care diagnostic method than RT-qPCR, considering its sequence-specific detection method and isothermal amplification approaches. [18] [19] [20] In this study, we conducted a systematic review and metaanalysis to assess the diagnostic accuracy of CRISPR in detecting SARS-CoV-2 infection, evaluate the quality of available evidence, and perform an in-depth analysis regarding the related research. This study was conducted according to the PRISMA guidelines. 21 We selected four databases, including PubMed, Embase, Cochrane Library, and Web of Science, and searched for data using "SARS-CoV-2" and "CRISPR" as keywords. All of the scientific papers were published before August 2021, without language restriction. All synonyms of the above-mentioned keywords were also included in the search formula for more comprehensive literature. The retrieved publications were independently selected by four researchers. Based on the predetermined inclusion and exclusion criteria, data were extracted by analyzing the titles, abstracts, and full texts of the studies. All disagreements were resolved through discussion and consultation. The publications that met all of the following criteria were included based on preset conditions: 1) the investigators' experimental objectives included the role of CRISPR in the diagnosis of COVID-19 infection; 2) the study type was a diagnostic accuracy test, and the diagnostic accuracy was evaluated by comparing the index to be tested with the standard reference method; and 3) the data provided by the study could identify true positive (TP), false positive (FP), true negative (TN), false negative (FN), or sensitivity and specificity. Exclusion criteria were as follows: 1) studies that were animal experiments; 2) studies where the reference method was not mentioned; 3) letters, conference abstracts, reviews, editorials, or erratum; and 4) duplicated publications or those with no description of the available data. The EndNoteX9 software was used for file management and data extraction from articles. Excel standardized electronic data entry form was used to pool the required information, including the author's name, publishing year, study type, sample size, reference standard, and indicators. In addition, the diagnostic features of SARS-CoV-2 were extracted along with TP, FP, TN, and FN. We reviewed the extracted information, resolved all disagreements through negotiated discussion, and excluded duplicate data. Four investigators evaluated the quality of the included studies independently in accordance with the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) guidelines, 22 regarding four main steps: case selection, trial to be assessed, reference standard, and case process and progress. The assessment of all four components was applied to analyze the risk of bias, while the assessment of the first three components was applied for the evaluation of clinical applicability. Issues with other iconic study designs were included in the risk of biased judgments, which were related to the potential for judicial bias. Responses of "Yes, " "No, " or "Indeterminate" corresponded to a risk of bias rating of "Low, " "High, " or "Indeterminate, " based on the questions included in each section. We used the MetaDiSc 1.4 software (Ramony Cajal Hospital, Madrid, Spain) for statistical analysis following standard methods, and used the Stata 15.0 software (StataCorp LLC, College Station, TX, USA) to draw Deeks' funnel plot and test funnel plot symmetry as well as publication bias. Spearman correlation coefficient and Cochran-Q were performed to analyze the heterogeneity of the included data, and a fixed-effects model or random-effects model was selected based on the result value. The sensitivity, specificity, positive likelihood ratio (PLR), and negative likelihood ratio (NLR) were calculated and analyzed by drawing a forest plot using MetaDiSc 1.4. The effect value and its 95% confidence interval (CI) were shown in the forest plot. In addition, the area under the curve (AUC) was calculated using the summary receiver operating characteristic (SROC) curve to obtain the specificity and sensitivity. Then, the total efficiency of CRISPR was assessed using diagnostic odds ratio (DOR) and AUC. The Review Manager 5.3 software (The Nordic Cochrane Centre, Copenhagen, Denmark) was used to evaluate the quality of the included studies. After searching through all four literature databases, we obtained 547 related documents, from which 374 were selected after the removal of duplicated publications. A total of 156 studies were removed for their uncorrelated titles or "CRISPR detection" not mentioned in their abstracts. We read through the text afterwards, and 188 studies were excluded for various reasons. Finally, 30 articles were selected with a total of 38 groups of data (Fig. 1) . 6, 10, 16, 17, The effect-indicator proposed in each literature was involved in the composition of the data extracted. Table 1 shows the characteristics of these studies in detail. The quality of the included studies was evaluated by analyzing the data in terms of case selection, index detection, reference standard, and case process and progress using Review Manager 5.3. Fig. 2A summarizes the results of the QUADAS-2 assessment, and Fig. 2B shows the independent quality assessment of each study. The results indicated that for case selection, seven studies had a risk of bias due to the unclear case-control study design and the unknown inclusion of consecutive or randomized case conditions. In the index test field, three studies were at higher risk since the interpretation of the index test results was made when the reference standard results were known. Both the reference standard field and the flow rate and time were considered to have a low risk of bias. Overall, the sensitivity of CRISPR in the diagnosis of COVID-19 Identification of studies via databases and registers PRISMA 2020 flow diagram for new systematic reviews which included searches of databases and registers only (Fig. 3A) , and the specificity was 0.98 (95% CI 0.97-0.99, I 2 =65.0%) (Fig. 3B ). As shown in the chart in Fig. 3C , the AUC was 0.9894. The PLR was 34.03 (95% CI 20.81-55.66, I 2 =66.0%) (Fig. 4A) , and the NLR was 0.08 (95% CI 0.06-0.10, I 2 =14.0%) (Fig. 4B) . The value of the pooled DOR was 575.74 (95% CI 382.36-866.95) (Fig. 4C) . In the threshold effect analysis, the Spearman correlation coefficient was 0.024, and the p-value was 0.888 (p>0.05). More-over, the SROC curve (Fig. 3C ) did not have a "shoulder arm" distribution. Therefore, we concluded that there was no threshold effect in the included studies. A forest map was used to plot the ratio following a random pattern. The heterogeneity in non-threshold effects was low (Fig. 4C ): Cochran-Q=38.80, p=0.3884 (p>0.05), inconsistency=4.6% (inconsistency<50%). The Deeks' funnel plot (Fig. 5) was made using the Stata 15.0 software to identify publication bias in the included publications (p=0.457>0.1), and it showed no potential publication bias for the included studies. Currently, RT-qPCR assays are the recommended molecular diagnostic tools to detect COVID-19 infection. 7, 8 However, they come with a high demand for equipment and skillful lab technicians. 10 In contrast, the CRISPR-based nucleic acid detection platforms have the combined advantages of conventional RNAtargeting technologies, and a fluorescence readout or a lateralflow readout can be used to analyze the results in an hour, with a setup time of less than 15 min. 49 The simplicity of operation, remarkable specificity, and high efficiency make CRISPR-based diagnostics the new avenues for sensitive, robust, and rapid detection of viral pathogens. In this study, we performed a pooled analysis, and the results of the systematic review and meta-analysis indicated that CRISPR had an overall sensitivity of 0.94 (95% CI 0.93-0.95) and an overall specificity of 0.98 (95% CI 0.97-0.99) in detecting SARS-CoV-2. The value of the pooled DOR was 575.74. The AUC was 0.9894 and was close to 1. Based on these results, we can infer that the diagnosis of COVID-19 by using CRISPR was highly accurate. Several studies have evaluated the accuracy of immunochromatographic analysis and ELISA for SARS-CoV-2 detection. 50 A test strip for the detection of SARS-CoV-2 IgG/IgM-combined antibody based on immunochromatography has been developed by Liao, et al., 51 with a sensitivity of 92.9% and a specificity of 98.7%. However, the antibody-positive rate in the first week of infection was only 77.3% and reached 100% on day 9. Another study reported by Beavis, et al. 52 evaluated an ELISA assay to detect SARS-CoV-2 IgA and IgG antibodies. The sensitivity of IgA ELISA was 82.9% and the specificity was 88.4%, while the sensitivity of IgG ELISA was 67.1% and the specificity was 97.7%. Although these assays are fast and easy to operate 50, 53 compared to CRISPR, infection-generated antibodies are detectable at later stages in the disease, which is not conducive to early disease screening. 51, 54 In addition, if the sample is heatinactivated, the effective concentration of the antibody would decrease and probably give false-negative results. 55 Meanwhile, according to Beavis, et al., 52 ELISA assay tended to have a lower sensitivity and specificity compared to CRISPR. Therefore, CRISPR is a valid and appropriate instrument for detecting SARS-CoV-2. Furthermore, to minimize the sources of heterogeneity, this study implemented strict criteria for the inclusion and exclusion of the studies. In the threshold effect analysis, the Spearman correlation coefficient was found to be 0.024 (<0.6) and the p-value was 0.888 (p>0.05), which indicated the lack of threshold effect in the included studies. However, I 2 values of pooled sensitivity (52.8%), specificity (65.0%), PLR (66.0%), which exceeded 50%, suggested the presence of heterogeneity from non-threshold effects. Subgroup analysis was performed to investigate the heterogeneity caused by different types of Cas enzyme used, Cas12 and Cas13, but no statistically significant results were obtained. Instead, we found that gene targets and readout modes might be the possible sources of underlying heterogeneity. 6, 10, 23 Moreover, the Deeks' funnel plot (p=0.457 >0.1) indicated that no publication bias was possibly subsistent. The present systematic review and meta-analysis also had several limitations. First, we only extracted data from the literature published in the four select English databases, and ignored some negative results without statistical significance or unpublished data. This may lead to defects in the comprehensiveness of the current study and more publication bias. Second, the detection capability of the reference methods may not necessarily be more reliable than that of CRISPR. The reference methods could also provide false-positive results, thereby leading to underestimation of the specificity of the CRISPR method. Finally, there were no remarkable changes in subgroup analyses. This study can be improved with the accumulation of more clinical data in the future. With more COVID-19 cases being reported every day worldwide, there may be more studies supporting our theory and, at the same time, having important implications for the diagnosis of COVID-19. In summary, CRISPR has proven to be a rapid, sensitive, and specific method to detect SARS-CoV-2. It can provide reliable information for clinical laboratory tests and contribute to pointof-care diagnostics where simplicity and cost-effectiveness are needed. This technology is expected to become the major auxiliary diagnostic method for COVID-19 in the near future. All data generated or analyzed during this study are included in this published article and its supplementary information files. Insights to SARS-CoV-2 life cycle, pathophysiology, and rationalized treatments that target COVID-19 clinical complications A guide to COVID-19: a global pandemic caused by the novel coronavirus SARS-CoV-2 Comparison of the characteristics, morbidity, and mortality of COVID-19 and seasonal influenza: a nationwide, populationbased retrospective cohort study COVID-19 diagnosis-A review of current methods Pathophysiology, transmission, diagnosis, and treatment of coronavirus disease 2019 (COVID-19): a review Clinical validation of a Cas13-based assay for the detection of SARS-CoV-2 RNA Analytical sensitivity and efficiency comparisons of SARS-CoV-2 RT-qPCR primer-probe sets Detection and analysis of nucleic acid in various biological samples of COVID-19 patients A rapid advice guideline for the diagnosis and treatment of 2019 novel coronavirus (2019-nCoV) infected pneumonia Ultra-sensitive and high-throughput CRISPR-p owered COVID-19 diagnosis Reverse-transcription PCR (RT-PCR) Real-time PCR for mRNA quantitation Metagenomic next-generation sequencing of nasopharyngeal specimens collected from confirmed and suspect COVID-19 patients Clinical and immunological assessment of asymptomatic SARS-CoV-2 infections Negative nasopharyngeal and oropharyngeal swabs do not rule out COVID-19 CRISPR-Cas12-based detection of SARS-CoV-2 Contamination-free visual detection of SARS-CoV-2 with CRISPR/Cas12a: a promising method in the point-of-care detection iS-CAN: an RT-LAMP-coupled CRISPR-Cas12 module for rapid, sensitive detection of SARS-CoV-2 Advances in CRISPR-Cas systems for RNA targeting, tracking and editing Nucleic acid detection with CRISPR-Cas13a/C2c2 Preferred reporting items for a systematic review and meta-analysis of diagnostic test accuracy studies: the PRIS-MA-DTA statement QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies Rapid and sensitive detection of COVID-19 using CRISPR/Cas12a-based detection with naked eye readout, CRISPR/Cas12a-NER Detection of SARS-CoV-2 with SHERLOCK one-pot testing Rapid detection of 2019 novel coronavirus SARS-CoV-2 using a CRISPR-based DETECTR lateral flow assay. medRxiv sensitive and visual detection of SARS-CoV-2 using all-in-one dual CRISPR-Cas12a assay MeCas12a, a highly sensitive and specific system for COVID-19 detection Streamlined inactivation, amplification, and Cas13-based detection of SARS-CoV-2 Detection of severe acute respiratory syndrome coronavirus 2 and influenza viruses based on CRISPR-Cas12a Comparative performance of CRISPR-Cas12a assays for SARS-CoV-2 detection tested with RNA extracted from clinical specimens A smartphoneread ultrasensitive and quantitative saliva test for COVID-19 An engineered CRISPR-Cas12a variant and DNA-RNA hybrid guides enable robust and rapid COVID-19 testing A scalable, easy-to-deploy protocol for Cas13-based detection of SARS-CoV-2 genetic material Machine learning-driven and smartphone-based fluorescence detection for CRISPR diagnostic of SARS-CoV-2 Cornelissen M, van den Akker E. Rapid, sensitive, and specific severe acute respiratory syndrome coronavirus 2 detection: a multicenter comparison between standard quantitative reverse-transcriptase polymerase chain reaction and CRISPR-based DETECTR Pointof-care CRISPR-Cas-assisted SARS-CoV-2 detection in an automated and portable droplet magnetofluidic device Evaluation of a lyophilized CRISPR-Cas12 assay for a sensitive, specific, and rapid detection of SARS-CoV-2. Viruses 2021 Sensitive quantitative detection of SARS-CoV-2 in clinical samples using digital warm-start CRISPR assay Detection of SARS-CoV-2 by CRISPR/Cas12a-enhanced colorimetry Fluorescence polarization system for rapid COVID-19 diagnosis One-tube SARS-CoV-2 detection platform based on RT-RPA and CRISPR/Cas12a Isothermal amplification and ambient visualization in a single tube for the detection of SARS-CoV-2 using loop-mediated amplification and CRISPR technology CRISPR/Cas12a technology combined with RT-ERA for rapid and portable SARS-CoV-2 detection Sensitive and easy-read CRISPR strip for COVID-19 rapid point-of-care testing An ultralocalized Cas13a assay enables universal and nucleic acid amplificationfree single-molecule RNA diagnostics opvCRISPR: onepot visual RT-LAMP-CRISPR platform for SARS-cov-2 detection Simultaneous dual-gene diagnosis of SARS-CoV-2 based on CRISPR/Cas9-mediated lateral flow assay Rapid, ultrasensitive, and highly specific diagnosis of COVID-19 by CRISPRbased detection SHERLOCK: nucleic acid detection with CRISPR nucleases SARS-CoV-2 and the COVID-19 disease: a mini review on diagnostic methods Development and clinical application of a rapid SARS-CoV-2 antibody test strip: a multi-center assessment across China Evaluation of the EUROIMMUN anti-SARS-CoV-2 ELISA assay for detection of IgA and IgG antibodies Development and clinical application of a rapid IgM-IgG combined antibody test for SARS-CoV-2 infection diagnosis Strategies and perspectives to develop SARS-CoV-2 detection methods and diagnostics Heat inactivation of serum interferes with the immunoanalysis of antibodies to SARS-CoV-2 Conceptualization: Xu-Guang Guo. Data curation: Pei-Ying Huang.