Development of a novel duplex real-time PCR assay for plasma DNA quantification and its comparison to other methods

Abstract
BACKGROUND: It has long been known that plasma DNA is present in healthy and diseased individuals and many investigators have been attracted to this emerging field. However, a direct comparative analysis of the abundant plasma DNA data available is often prevented by variability in sample storage, sample processing, and protocols for DNA extraction. Therefore, a more accurate and stable assay with lower uncertainty is needed in conjunction with a multi-center study. METHODS: In this study, a duplex real-time PCR assay with an internal control was developed as a novel method for quantification of plasma DNA. The detect limit, precision, accuracy and stabilitity of this novel assay was then evaluated and compared to other two popular plasma DNA quantification methods by real-time PCR with external standards and by direct fluorescent PicoGreen staining. Furthermore, the plasma DNA concentrations of 1,187 healthy Chinese adults have been determined by the novel assay. RESULTS: The detect limits of novel duplex real-time PCR, real-time PCR with external standards and PicoGreen assay were 0.1 ng/ml (R2 = 0.96, P < 0.0001), 0.1 ng/ml (R2 = 0.96, P < 0.0001) and 1.0 ng/ml (R2 = 0.91, P < 0.0001), respectively. For plasma with concentration of 1.0 ng/ml, the CV values of duplex real-time PCR assay, real-time PCR with external standards and Picogreen assay were 53.0%, 60.3% and 96.7%, respectively. In the recovery test, there was a higher correlation of the plasma DNA concentrations determined by the duplex real-time PCR assay with the input DNA levels (R2 = 0.97, P < 0.0001) than those of the real-time PCR with external standards (R2 = 0.85, P < 0.0001) and PicoGreen assay (R2 = 0.92, P < 0.0001). Although remarkable differences in plasma DNA concentrations determined by the real-time PCR with external standards and PicoGreen assay existed in samples purified by the three different protocols (P = 0.0041 and 0.0000), there was no statistically significant difference in plasma DNA concentrations among the methods by using our duplex real-time PCR assay (P = 0.5821). The median plasma DNA concentration of females (16.9 ng/ml) was significantly lower than males (22.6 ng/ml; Mann-Whitney two-sample rank sum test, P < 0.0001). Within the 95% confidence interval, the normal reference interval of the plasma DNA concentration was 0~50 ng/ml for males and 0~40 ng/ml for females. CONCLUSIONS: This newly developed plasma DNA quantification method with the power of eliminating variables introduced during plasma sample processing allows more sensitive, repetable and accurate quantitative measurement, and has a very promising future in clinical application for diagnosis and disease monitoring.