Nucleic acid testing (NAT) is widely adopted for diagnosing infectious diseases and genetic disorders. Through the prolonged global experience with the COVID-19 pandemic (Barakat et al., 2024, Carter et al., 2020), the drive has been strong for securing cost effectiveness and shorter turnaround times without sacrificing test accuracy. To achieve this, the detection and quantification of multiple genes with a single polymerase chain reaction (PCR) using duplex or even multiplex assays (primer–probe sets) rather than a singleplex assay has been widely accepted (Cassinari et al., 2021, Yin et al., 2021, de Kock et al., 2021, Zhou et al., 2021, Nyaruaba et al., 2021, Zhang et al., 2020, Park et al., 2020). However, as the assay design and PCR conditions may have a significant impact on test results, multiplex experiments present a challenge, and therefore a thorough optimization process is required for accurate quantitative measurement and for ideal NAT performance. For example, measurement results are affected by PCR assay details including the primer and probe sequences as well as their melting temperature (Kinloch et al., 2020, Vogels et al., 2020, Park et al., 2021, Ricchi et al., 2017, Long and Berkemeier, 2021).

Several molecular assays have been developed to detect SARS-CoV-2 RNA, but especially in the early stages of the pandemic, it was unclear to many clinical, research, and public health laboratories which assays should be employed and whether the data were comparable. Many studies were subsequently conducted on the analysis efficiency and sensitivity of various assays commonly used for diagnosing SARS-CoV-2, including those developed by the German Charité Institute of Virology (Charité), the United States Center for Disease Control (US CDC), and the China CDC (Zhen et al., 2020, Etievant et al., 2020, Jung et al., 2020, Nayar et al., 2021). From these studies, for example, the sensitivity of the Charité RdRp assay, which has a large difference in the melting temperature (Tm) values between forward and reverse primers, was found to be low compared to other assays (Etievant et al., 2020, Jung et al., 2020). This highlights that, in order to design an appropriate assay, the Tm values of the forward and reverse primers and probes generally need to be optimized. In addition, ΔG (Gibbs free energy change) values can be checked to test the possibility of forming primer heterodimers and hairpin structures (Mann et al., 2009). In duplex PCR reactions which a doubled number of primers, the formation of heterodimers between primers needs to be avoided (Park et al., 2020, Xie et al., 2022).

Among various NAT methods, quantitative PCR (qPCR) is used as the gold standard for most molecular diagnostic kits (Jin et al., 2020, FDA, 2024, MacKay et al., 2020). But many recent studies have focused on the measurement of SARS-CoV-2 using digital PCR (dPCR) as enables more sensitive and precise testing (Falzone et al., 2020, Van Poelvoorde et al., 2021, Tan et al., 2021, Suo et al., 2020, Liu et al., 2020, Duong et al., 2021, Mio et al., 2021). As such, dPCR methods have reached technical maturity in commercial instruments and are therefore gaining popularity in both research and clinical laboratories. Unlike conventional qPCR, dPCR does not require a calibration curve for the absolute quantification of the target sequence and the provision of copy number concentration (Vogelstein and Kinzler, 1999). Commercial dPCR platforms can be classified by their partitioning method. In droplet-based dPCR, the PCR reactions are divided into oil-based partitions using microfluidics, whereas chip- and nanoplate-based dPCR achieve partitioning using a physical plate equipped with a number of microwells.

In this study, we demonstrate an optimization process for duplex SARS-CoV-2 RNA assays in reverse transcription (RT)-qPCR and RT-dPCR settings. We started with combinations of assays, including WHO-recognized and in-house developed assays, using KRISS 111–10–507 (batch 2) SARS-CoV-2 reference material (Lee et al., 2022). The performance of the Charité RdRp assay as first improved dramatically by correcting a few bases in the primer sequences. It was also found that heterodimer formation in RdRp and S duplex reactions negatively impact qPCR results, but much less so in dPCR. We then present a thorough optimization process for RT-dPCR that leads to the achievement of appropriate measurement results even when suboptimal duplex assay conditions are employed. In addition to demonstrating accurate quantitative measurement of SARS-CoV-2, the results of this study can be applied to NAT for various other RNA and DNA targets.