Cytomegalovirus (CMV) and Epstein-Barr virus (EBV) are ubiquitous, large, double-stranded DNA herpes viruses that infect most humans by older age. The seroprevalence of CMV ranges from 30% to 97% (Staras et al., 2006), and EBV seropositivity prevalence is approximately 95% (Kuri et al., 2020). Both viruses are preventable causes of morbidity and mortality in populations who have undergone solid organ transplant (Razonable and Hayden, 2013, Abbate et al., 2016, Ljungman et al., 2017), are HIV positive (Deayton et al., 2004), are COVID positive (Simonnet et al., 2021), or critically ill (Libert et al., 2015, Griffiths and Reeves, 2021). In addition, EBV is an established oncogenic virus for nasopharyngeal cancer and Burkitt lymphoma, and recently Bjornevik et al. provided convincing evidence that EBV is causal in the development of multiple sclerosis (Bjornevik et al., 2022). CMV is primarily a concern among pregnant women (Sapuan et al., 2022) and the severely immunocompromised (Griffiths and Reeves, 2021). Precise quantitation of low viral activity is crucial for monitoring EBV and CMV reactivation in severely immune compromised patients to determine the initiation of antiviral therapy (Abbate et al., 2016, Deayton et al., 2004, Engelmann et al., 2008). Viral monitoring in the clinical setting can be achieved through the detection of cell-free viral DNA in plasma or serum (DNAemia). EBV-DNAemia in plasma has been used to estimate the severity or prognosis of EBV-associated lymphoid malignancies and response to therapy of intractable lymphomas (Kimura and Kwong, 2019). Close monitoring of low viral loads benefits these patient populations.

Beyond clinical applications, there is a need to deploy low level viral testing in epidemiologic studies. CMV IgG seropositivity has been associated with increased all-cause mortality (Chen et al., 2020), cardiovascular disease (Wang et al., 2017, Nikitskaya et al., 2016), and cancer mortality (Wesley et al., 2021, Gkrania-Klotsas et al., 2013). However, IgG is a crude marker of any past infection and cannot distinguish latency from active viral infection, including reactivation from latency. There is a need to determine whether DNAemia can be used in epidemiologic research, where low levels of viral activity are anticipated.

There are two available techniques to quantify DNAemia: digital PCR (dPCR), which is typically chip or droplet-based, and quantitative PCR (qPCR). dPCR is a preferred method, particularly for large scale studies, as it is possible to achieve absolute quantitation without the use of a standard curve. dPCR is based on the division of the PCR master mix (all components including DNA or RNA targets) into thousands of partitions. PCR amplification of target genes then occurs in each individual partition (Salipante and Jerome, 2020). These partitions can be created using several different mechanisms, such as emulsified microdroplets suspended in oil (droplet), manufactured microwells, or microfluidic valving (Salipante and Jerome, 2020). Quantification of target genes is estimated based on a Poisson distribution, by calculating the ratio of positive partitions over the total number of partitions (Majumdar et al., 2017). However, reagent and platform differences in dPCR may lead to imprecise measurements in samples with low-level viremia. Factors that may impact the performance of dPCR for low level quantitation include cell free DNA matrix (plasma or serum), dPCR platform (chip or droplet based), and low DNA mass. The aim of the current study is to compare the performance of CMV and EBV DNAemia testing across dPCR platforms and blood matrices in a population experiencing low level viral activity.