Comparison of two digital PCR methods for EGFR DNA and SARS-CoV-2 RNA quantification

Abstract

Background: The COVID-19 pandemic caused by the severe acute SARS-CoV-2 virus has undeniably highlighted the importance of reliable nucleic acid quantification. Digital PCR (dPCR) is capable of the absolute quantification of nucleic acids.

Method: By using the droplet dPCR (QX200) and the digital real-time PCR (LOAA), the copy numbers were compared via multiple assays for three distinct targerts; EGFR DNA, SARS-CoV-2 and HIV-1 RNA.

Results: The droplet dPCR and digital real-time PCR showed similar copy numbers for both DNA and RNA quantification. When the limit of detection (LOD) and limit of quantitation (LOQ) of each method were estimated for DNA and RNA targets, the digital real-time PCR showed a higher sensitivity and precision especially with low copy number targets.

Conclusion: The breath of nucleic acid testing in diagnostic applications continues to expand. In this study we applied common diagnostic targets to a novel digital real-time PCR methodology. It performed comparably to the established dPCR method with distinctive advantages and disadvantages for implementing in laboratories. These rapidly developing dPCR systems can be applied to benefit the accurate and sensitive nucleic acid testing for various clinical areas.

Keywords: Digital PCR; Digital real-time PCR; EGFR; HIV-1; SARS-CoV-2.

Fig. 1

Optimization of the digital real-time PCR platform LOAA. A: Images of the LOAA cartridge and wells within it (left), and representative results including real-time amplification curves and a two-dimensional scatter chart of FAM (channel 1) and FRET (channel 2) signals (right). B–G: Digital real-time PCR amplification curves using a set of assays distinguishing the T790M mutation (FAM) from the wild-type (FRET). FAM channel (B and D) and FRET channel (C and E) with indicated conditions. Digital real-time PCR amplification curves using the following optimal concentrations of primers and probes for two assays (different quenchers): forward primer: reverse primer: FAM probe: FRET probe as 20:20:4:20 pmol per reaction (F and G). Blue lines indicate a fixed fluorescence intensity of 300.

Fig. 2

Results of EGFR copy number concentration measured by droplet digital PCR (QX200) and digital real-time PCR (LOAA) using 1 ng of genomic DNA isolated from cancer cell lines A549 and NCI-H1975. A and B: Graphs comparing copy numbers of T790M and L858R EGFR mutations and the corresponding wild-type copies in A549 (A) and NCI-H1975 (B). Circles indicate the copy number from the QX200 and crosses indicate the copy number from the LOAA. C: Copy number ratio of each EGFR mutation over wild-type in NCI-H1975 genomic DNA by instrument. Error bars indicate the standard deviation (SD) at each data point with the mean of replicated measurements (n ≥ 5). Significant difference is tested by t-test (p > 0.05, not significant).

Fig. 3

Linearity assessment of T790M mutation quantification using the LOAA and QX200. A: LOAA results for the T790M mutation showing amplification curves for each partition over cycles (upper) and end-point two-dimensional scatter plots (lower). The red bar indicates the position of the threshold. B: Representative QX200 results showing end-point one-dimensional amplitude graphs. The ordinate scales indicate the fluorescent amplitude. The pink line indicates the threshold, above which are positive droplets (blue) containing at least one copy of the target DNA and below which are negative droplets (gray) without any target DNA. C: Ratios of copy number concentrations of the T790M mutation over wild-type in serially diluted templates. Significant difference is tested by t-test (p > 0.05, not significant). Error bars indicate the SD at each data point. D and E: Log-scale graphs showing the linear correlation between the input DNA amount and copy number concentrations of T790M mutation (D) and wild-type allele (E) from the QX200 (red) and LOAA (blue). Error bars indicate the SD with the mean of replicated measurements (n ≥ 5). F: LOD and LOQ values for T790M quantification using the QX200 and LOAA (n > 10 per sample).

Fig. 4

Quantification of SARS-CoV-2 and HIV RNA. A–D: Representative SARS-CoV-2 amplification curves and scatter plots from LOAA (left four panels: positive and NTC) and one-dimensional plots from QX200 (rightmost) using four different assays: RdRP-1 (A), RdRP-2 (B), E (C), and N (D). E and F: Representative HIV-1 amplification curves and scatter plots from LOAA (left four panels: positive and NTC) and one-dimensional plots from QX200 (rightmost) using two different assays: A assay (E) and B assay (F). The red bars and pink lines indicate the set thresholds. G and H: Graphs comparing copy number results for SARS-CoV-2 (G) and HIV-1 (H) using the QX200 (circles) and LOAA (crosses). Error bars indicate the SD with the mean of replicated measurements (n ≥ 3). Significant difference analyzed by t-test is labeled (**p < 0.01) and otherwise not significant p > 0.05.

Fig. 5

Linearity assessment of RNA quantification. A: LOAA results showing amplification curves for each partition over cycles (upper) and scatter plots (lower). Red bars indicate the set threshold. B: Representative QX200 results showing an end-point one-dimensional amplitude plot. The pink line indicates the set threshold. C: Log-scale graph showing the linear correlation between the input SARS-CoV-2 RNA amount and the copy number concentrations from QX200 (red) and LOAA (blue). D: LOD and LOQ values for SARS-COV-2 quantification using the QX200 and LOAA (RdRP-1 assay). E: Log-scale graph showing the linear correlation between input HIV-1 RNA amount and copy number concentrations from QX200 (red) and LOAA (blue) using A assay. F: LOD and LOQ values for HIV-1 quantification using the QX200 and LOAA (A assay). Error bars indicate the SD with the mean of replicated measurements (n > 10 per sample).