新冠肺炎PCR檢測與抗原快篩探討

Studying PCR test and rapid antigen screening of COVID-19

陳珊民(Shan-Min Chen)

指導教授 ： 陳秀熙

共同指導教授 ： 許辰陽 ； 賴昭智(Chao-Chih Lai)

國立臺灣大學/公共衛生學院/公共衛生碩士學位學程/碩士(2021年)

https://doi.org/10.6342/NTU202104244

摘要

研究背景
新冠肺炎全球疫情自2020年於世界各地發生社區流行傳播以來，由於其高傳播力以及無症狀與症狀前傳播之特性，使罹病個案在感染後但臨床症狀尚未發生時即具有傳染力。此一特性也使新冠肺炎防疫往往需運用大規模非藥物防疫措施(non-pharmaceutical intervention，NPI)以降低社區傳播之風險，NPI雖在世界疫情初期顯現其對於疫情控制之功效，但對於社會經濟活動之衝擊使其無法作為長期防疫措施。因此運用快速且準確的檢測工具早期辨識出高病毒量且具傳播風險之感染個案即成為重要的防疫策略。傳統的Real-time PCR（real-time Polymerase chain reaction，即時定量聚合酶連鎖反應)為罹病檢測的黃金標準，在新冠肺炎疫情中也廣泛運用於診斷罹病個案，但進行Real-time PCR在時間、人力，與設備上皆須投入較多資源。抗原快速檢測由於檢測進行之便利與快速，可做為臨床上判斷是否罹病以及具有傳染力之重要工具，亦有利於在社區傳播情境中大規模於高風險地區對具有感染風險者進行篩檢，也因此成為新冠肺炎主要防疫措施之一。以抗原快速檢測作為社區主動監測之防疫措施於台灣尚未有實證評估結果，因此本研究旨在運用台北市立聯合醫院仁愛院區於台北爆發社區流行之檢測實證資料進行以下之研究目標：
1. 描述台北市以及鄰近區域COVID-19檢測陽性之區域分佈以及時間趨勢；
2. 評估抗原快篩檢測對於COVID-19包含敏感度與特異度之工具效度；
3. 評估不同Ct值對於抗原快篩檢測工具效度之影響；
4. 尋找抗原快篩檢測對於COVID-19個案臨床偵測的最適Ct值切點。
材料與方法
本研究運用隨時間改變之並行效度(time-varying concurrent validity)研究設計評估抗原快篩與核酸檢測包含敏感度以及特異度之工具效度。對於參與研究之個案同時對採樣檢體進行核酸檢測以及抗原快篩檢測，並且以核酸檢測做為黃金標準評估抗原快篩之工具效度。納入研究之個案為2021年5月及6月台北市爆發社區流行期間至本院設立之社區主動監測站民眾。本研究收集抗原快篩檢測結果、核酸檢結果、Ct值，與參與研究個案之個人特性如年齡、性別，接受採檢日期，以及居住地資料進行評估。核酸檢測之方法包含以半自動的核酸萃取儀加上手動式Real-time PCR檢測儀的半自動操作模式以及高通量自動化PCR儀器的全自動模式。本研究運用所收集的台北市社區主動監測實證資料評估社區傳播流行爆發期間隨時間改變之COVID-19盛行率、敏感度、特異度，以及各行政區別之COVID-19盛行狀況。並估算抗原快篩工具之貝氏因子。對於抗原快篩最適Ct值切點之評估，本研究以收集之實證資料建立羅吉斯回歸模式並估算其對應之接受者作業特徵曲線(receiver operation characteristics curve，ROC curve)，據以尋找抗原快篩臨床運用之最佳Ct切點值。
結果
於2021年5月到6月期間，本研究共計納入7643位個案，接受核酸PCR檢測以及抗原快篩，其中284位個案RT-PCR陽性，184位個案抗原快篩陽性。COVID-19盛行率為3.7% (95% CI: 3.3-4.1%)。抗原快篩陽性率為2.1% (95% CI: 2.1-2.8%)。抗原快篩之敏感度為58.1% (95% CI: 52.4-63.8%)，特異度為99.7% (95% CI: 99.6-99.9%)。抗原快篩工具之陽性概似比(positive likelihood ratio)為225，陰性概似比則為0.42。顯示抗原快篩陽性對於COVID-19個案偵測能力極佳。於5月期間，COVID-19盛行率為6.0% (95% CI: 4.9-7.0%)，抗原快篩敏感度為78.0% (95% CI: 70.5-85.4%)；6月期間之COVID-19盛行率則下降為2.9% (95% CI: 2.5-3.4%)，抗原快篩敏感度為44.0% (95% CI: 36.4-51.5%)。
社區主動監測之個案其Ct值由5月疫情初期之21漸次上升至6月初期之24以及6月底之30。抗原快篩敏感度之週別趨勢亦由5月初期之85% 下降至6月底之25%。抗原快篩特異度則均維持高於99%。COVID-19盛行率則以萬華區以及鄰近之大同區為最高，此盛行率隨時間下降。
運用羅吉斯回歸模式評估之結果顯示Ct值為主要影響抗原快篩陽性結果之因子，以AUC評估Ct值解釋程度達93.5% (95% CI: 90.9-96.2%)。抗原快篩運用之最佳Ct值切點為25，在此一切點值下，抗原快篩之敏感度與特異度可分別達82.1 (95% CI: 75.6-88.6)與99.4 (95% CI: 99.3-99.6)。
結論
本研究結果顯示於台北市運用抗原快篩作為社區主動監測之工具在5月疫情初期其敏感度與特異度可分別達78.0% (95% CI: 70.5-85.4%)與99.8% (95% CI: 99.7-100.0%)。抗原快篩之高特異使其具有良好之COVID-19個案偵測能力，抗原快篩之陽性概似比可達225，若對於高病毒量Ct值低於25個案，其敏感可達82.1% (95% CI: 75.6-88.6%)，陽性概似比則為137，顯示抗原快篩陽性可有效偵測高病毒量與高傳播風險個案，可運用於社區主動監測工具阻斷社區COVID-19傳播控制流行。

關鍵字

嚴重特殊傳染性肺炎 ； 嚴重急性呼吸道症候群冠狀病毒2型 ； 即時定量聚合酶連鎖反應 ； 抗原快速檢測 ； 循環數閾值

並列摘要

Background
The characteristics of high transmissibility and being infective prior to the occurrence of clinical symptoms render the containment of COVID-19 outbreak a challenging task. As the mainstay of containment measures, non-pharmaceutical interventions (NPIs) have been administered widely by countries worldwide since the early period of COVID-19 pandemic in 2020. Although NPIs have been proved to be effective in stamping out community-acquired outbreaks, the impacts on socioeconomical activities makes it a non-sustainable measure for containing the long-lasting COVID-19 pandemic.
In the face of community-based outbreaks of COVID-19 at large scale, diagnostic technologies are promising tools which can be used to identify infectives at earliest and to forestall the spread in community. As a gold standard, real-time polymerase chain reaction (Real-time PCR) has been used for detecting subjects infected by SARS-CoV-2. The rapid antigen test in a point-of-care setting is a feasible and handy tool that can be used to identify the infectives with high viral load and the likelihood of transmitting the pathogen on contact with susceptibles. Rapid antigen test can thus be used as a screening tool for active surveillance at community level during the period of community-acquired outbreak. Such an application in Taiwan is lacking of the support from empirical evidence. The aims of this study are thus to use the data collected from the RenAi Branch of Taipei City Hospital during the period of community-acquired outbreak
1. to depict the distribution of positives rates across the districts of Taipei city and the neighborhood areas;
2. to assess the tool validity of rapid antigen test for detecting COVID-19 cases by using the indicators of sensitivity and specificity;
3. to assess the impact of cutoff of Ct values on tool validity of rapid antigen test;
4. to find the optimum cutoff of Ct value for the clinical use of rapid antigen test to detect COVID-19 cases.
Material and Methods
This study applied a time-varying concurrent validity design to evaluated the tool validity of rapid antigen test for the identification of SARS-CoV-2 infection. The population enrolled in this study consists of the subjects attending the active surveillance station of the RenAi Branch, Taipei City Hospital. Only subjects without clinical symptoms and signs of COVID-19 who attended the active surveillance station between May and June, 2021, when the COVID-19 community-acquired outbreak struck Taipei, were included in this study. Attendants received both rapid antigen test and RT-PCR in parallel to identify subjects infected by SARS-CoV-2. In addition to the binary results of rapid antigen test and RT-PCR, Ct values, date of the tests performed, and the personal attributes including residential area, age, and sex were also collected.
On the basis of the data collected during the period with community-acquired outbreak in Taipei, the prevalence of COVID-19 in Taipei and in each district were assessed. By using the time stamped data collected during outbreak, the time trends of the prevalence, sensitivity, and specificity for rapid antigen test were also explored. The optimal cutoff of Ct value for the clinical use of rapid antigen test in detecting the infectives with high viral load were determined by using logistic regression model in conjunction with the receiver operation characteristics (ROC) curve.
Results
A total of 7643 subjects attending the active surveillance station of RenAi branch, Taipei City Hospital between May and June, 2021 were enrolled in this study. Among the enrolled population, 284 RT-PCR positive cases and 184 rapid antigen positive tests were identified, which give the prevalence for COVID-19 and positive rate for rapid antigen test of 3.7% (95% CI: 3.3-4.1%) and 2.1% (95% CI: 2.1-2.8%), respectively. The prevalence rate of COVID-19 was highest for the epicenter of Wanhua district, which decreased gradually for the neighbor areas of the epicenter. The sensitivity and specificity of rapid antigen test were estimated as 58.1% (95% CI: 52.4-63.8%) and 99.7% (95% CI: 99.6-99.9%), respectively, with the positive and negative likelihood ratio of 225 and 0.42, respectively, suggesting the rapid antigen test as a useful tool for rule-in the diagnosis of COVID-19.
During the initial period of outbreak in May, the prevalence was 6.0% (95% CI: 4.9-7.0%), which reduced to 2.9% (95% CI: 2.5-3.4%) in June. The sensitivity and specificity of rapid antigen test in May were estimated as 78.0% (95% CI: 70.5-85.4%) and 99.8 (95% CI: 99.7-100%), respectively. The corresponding estimates in June were 44.0% (95% CI: 36.4-51.5%) and 99.7% (95% CI: 99.6-99.9%), respectively.
Regarding the viral load captured by the Ct value of RT-PCR, the average Ct values were 21, 24, and 30 for confirmed cases detected in May, early June, and the end of June, respectively, which show a decreasing trend of the viral load following the containment of the community-acquired outbreak in Taipei.
The optimum cutoff of Ct value for the clinical use of rapid antigen test in detecting infectives was estimated as 25 by using a logistic regression model. The sensitivity and specificity were estimated as 82.1% (95% CI: 75.6-88.6%) and 99.4 (95% CI: 99.3-99.6), respectively, which gives the area under ROC curve of 93.5% (95% CI: 90.5-96.2%).
Conclusion
Supported by the empirical evidence of community-acquired outbreak in Taipei, the sensitivity and specificity of rapid antigen test were estimated as 78.0% (95% CI: 70.5-85.4%) and 99.8% (95% CI: 99.7-100.0%), respectively, during the period of initial COVID-19 outbreak in May. For the infectives with high viral load and a Ct lower than 25, the sensitivity was 82.1% (95% CI: 75.6-88.6%). Our results suggest that rapid antigen test is a useful tool for the identification of infectives with high viral load and transmissibility and can be used as a screening tool during community-acquired outbreak for the containment of outbreaks at earliest.

並列關鍵字

Coronavirus disease 2019 ； COVID-19 ； SARS-CoV-2 ； Real-time PCR ； Rapid antigen test ； cycle threshold value

參考文獻

1.Zhu H, Zhang H, Xu Y, Lassakova S, Korabecna M, Neuzil P. PCR past, present and future. Biotechniques. 2020;69(4):317-25.

2.Udugama B, Kadhiresan P, Kozlowski HN, Malekjahani A, Osborne M, Li VYC, et al. Diagnosing COVID-19: The Disease and Tools for Detection. ACS Nano. 2020;14(4):3822-35.

3.Becker-Andre M, Hahlbrock K. Absolute mRNA quantification using the polymerase chain reaction (PCR). A novel approach by a PCR aided transcript titration assay (PATTY). Nucleic Acids Res. 1989;17(22):9437-46.

4.Chelly J, Kaplan JC, Maire P, Gautron S, Kahn A. Transcription of the dystrophin gene in human muscle and non-muscle tissue. Nature. 1988;333(6176):858-60.

5.Carter LJ, Garner LV, Smoot JW, Li Y, Zhou Q, Saveson CJ, et al. Assay Techniques and Test Development for COVID-19 Diagnosis. ACS Cent Sci. 2020;6(5):591-605.