Microfluidic nano-scale qPCR enables ultra-sensitive detection of SARS-CoV-2

Authors:
Preprint
Publisher website
Google Scholar
Abstract
Background: A major challenge in controlling the COVID-19 pandemic is the high false-negative rate of the commonly used standard RT-PCR methods for SARS-CoV-2 detection in clinical samples. Accurate detection is particularly challenging in samples with low viral loads that are below the limit of detection (LoD) of standard one- or two-step RT-PCR methods. Methods: We implement a three-step approach for SARS-CoV-2 detection and quantification that employs reverse transcription, targeted cDNA preamplification and nano-scale qPCR based on the Fluidigm 192.24 microfluidic chip. We validate the method using both positive controls and nasopharyngeal swab samples. Results: Using SARS-CoV-2 synthetic RNA and plasmid controls, we demonstrate that the addition of a preamplification step enhances the LoD of the Fluidigm method by 1,000-fold, enabling detection below 1 copy/μl. We applied this method to analyze 182 clinical NP swab samples previously diagnosed using a standard RT-qPCR protocol (91 positive, 91 negative) and demonstrate reproducible detection of SARS-CoV-2 over five orders of magnitude (< 1 to 10 6 viral copies/μl). Crucially, we detect SARS-CoV-2 with relatively low viral load estimates (<1 to 40 viral copies/μl) in 17 samples with negative clinical diagnosis, indicating a potential false negative rate of 18.7% by clinical diagnostic procedures. Conclusion: The three-step nano-scale RT-qPCR method can robustly detect SARS-CoV-2 in samples with relatively low viral loads (< 1 viral copy/μl) and has the potential to reduce the false negative rate of standard RT-PCR-based diagnostic tests for SARS-CoV-2 and other viral infections. Summary: We test, implement and report the results of a microfluidic RT-qPCR assay system involving sequential RT, preamplification and nano-scale qPCR that can robustly detect SARS-CoV-2 in clinical samples with viral loads less than 1 copy/ul.
All Related Versions

This publication has 21 references indexed in Scilit: