Performance verification of five commercial RT-qPCR diagnostic kits for SARS-CoV-2

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which has caused a global pandemic beginning in 2020, can be detected by reverse-transcription polymerase chain reaction (RT-PCR). However, owing to the urgent need for a large number of detection kits, the time spent researching and developing these kits has been shortened during the pandemic, and the kits that are being used commercially have not undergone full and independent evaluation. To ensure the accuracy of SARS-CoV-2 test results, performance verification of commercial Real-Time quantitative PCR (RT-qPCR) kits is required.

The performance of five commercial RT-qPCR diagnostic kits for SARS-CoV-2 used in China was evaluated using a coronavirus disease 2019 (COVID-19) RNA liquid performance verification reference product-manufactured by Guangzhou Bondson (BDS) Biotechnology Co., Ltd.,Guangzhou, China-that uses droplet digital RT-PCR technology combined with fluorescence quantitative PCR. The five kits of Novel Coronavirus 2019-nCoV nucleic acid detection kit (RT-qPCR method) evaluated were Da An (Da An Gene Co., Ltd. of Sun Yat-sen University), Liferiver (Shanghai ZJ Bio-Tech Co., Ltd.), Kinghawk (Beijing Kinghawk Pharmaceutical Co., Ltd.), eDiagnosis (Wuhan Easy Diagnosis Biomedicine Co., Ltd.), and Maccura (Maccura Biotechnology Co., Ltd.). Performance verification criteria included the coincidence rate, limit of detection (LoD), cross-reactivity, precision, and anti-interference. Finally, through the BDS performance verification reference product kit, clinical samples are used to verify its clinical diagnostic efficacy.

The coincidence rate was 100% for all kits except for Kinghawk, which was 95%. The LoD for Da An, eDiagnosis and Maccura was 250copies/mL, and it was 1000 copies/ml for Liferiver. Kinghawk was not able to detect its advertised LoD of 500 copies/ml. The cross-reactivity test results were all negative. Moreover, all kits had a coefficient of variation less than 5%; however, Liferiver showed the best precision. Da An, Liferiver, and eDiagnosis showed higher sensitivity to the nucleocapsid (N) gene than they did to the open reading frame (ORF) 1ab genes. Anti-interference results for all five kits were positive. The results of clinical diagnostic efficacy were that the specificity of the four kits was 1.000 (0.877-1.000), the sensitivity of Da An was 1.000 (0.850-1.000), Liferiver was 0.964 (0.798-0.998), Maccura was 0.893 (0.706-0.972), and eDiagnosis was 0.857 (0.664-0.953).

All commercial RT-qPCR diagnostic kits for SARS-CoV-2 passed the BDS performance verification, except for Kinghawk (batch No:20200608113) which failed to detect the LoD of 500 copies/mL. Da An and Liferiver have excellent clinical diagnostic specificity and sensitivity. This study can provide guidance for the selection or optimization of RT-qPCR diagnostic test kits for SARS-CoV-2.