Towards standardisation of cell-free DNA measurement in plasma: controls for extraction efficiency, fragment size bias and quantification

Abstract

Circulating cell-free DNA (cfDNA) is becoming an important clinical analyte for prenatal testing, cancer diagnosis and cancer monitoring. The extraction stage is critical in ensuring clinical sensitivity of analytical methods measuring minority nucleic acid fractions, such as foetal-derived sequences in predominantly maternal cfDNA. Consequently, quality controls are required for measurement of extraction efficiency, fragment size bias and yield for validation of cfDNA methods. We evaluated the utility of an external DNA spike for monitoring these parameters in a study comparing three specific cfDNA extraction methods [QIAamp circulating nucleic acid (CNA) kit, NucleoSpin Plasma XS (NS) kit and FitAmp plasma/serum DNA isolation (FA) kit] with the commonly used QIAamp DNA blood mini (DBM) kit. We found that the extraction efficiencies of the kits ranked in the order CNA kit > DBM kit > NS kit > FA kit, and the CNA and NS kits gave a better representation of smaller DNA fragments in the extract than the DBM kit. We investigated means of improved reporting of cfDNA yield by comparing quantitative PCR measurements of seven different reference gene assays in plasma samples and validating these with digital PCR. We noted that the cfDNA quantities based on measurement of some target genes (e.g. TERT) were, on average, more than twofold higher than those of other assays (e.g. ERV3). We conclude that analysis and averaging of multiple reference genes using a GeNorm approach gives a more reliable estimate of total cfDNA quantity.

Figure

Comparison of single and multiple reference gene normalisation for quantification of plasma cell free DNA

Fig. 1

Assessment of cell-free DNA (cfDNA) yield using four extraction methods. The mean yield ± one standard deviation from replicate extractions using plasma pool A (i) (with or without ADH) performed with the QIAamp circulating nucleic acid (CNA), NucleoSpin Plasma XS (NS) and FitAmp plasma/serum DNA isolation (FA) kits (n = 10) and the QIAamp DNA blood mini (DBM) kit (n = 9) is displayed relative to the mean yield of the DBM kit. The yield of cfDNA was quantified by quantitative PCR (qPCR) assays to TERT and ALUJ

Fig. 2

Assessment of extraction efficiency and fragment size bias of four extraction methods using an exogenous spike-in. Extraction efficiencies of the CNA, NS, FA and DBM kits are expressed as a percentage of input (10⁶ copies per millilitre of plasma) ± one standard deviation (n = 3 extractions for the DBM kit, n = 4 extractions for the other methods) for 115-, 461- and 1,448-bp fragments of the ADH plasmid spike-in. Significant differences between the yields of the ADH plasmid fragments are indicated: one asterisk, p < 0.05; four asterisks, p < 0.0001

Fig. 3

Assessment of qPCR inhibition by cfDNA extracts using two different extraction methods. Fragmented ADH plasmid (500 copies per reaction) was measured using the Adhβ assay (qPCR, n = 3) in both hydrolysis probe (a, b) and intercalating dye (c, d) assay formats in the presence of increasing concentrations of cfDNA extract produced using the CNA (a, c) and NS (b, d) kits and compared with control conditions (given a nominal concentration of 1 μL plasma per microlitre of extract owing to plotting on a log scale). Individual data points represent replicate qPCR assays. An increase in C _q indicates inhibition of qPCR versus the control condition. Three asterisks, p < 0.001

Fig. 4

Linearity and within-laboratory reproducibility of cfDNA extraction using a single method. Three independent extraction experiments performed on different days using the CNA kit were performed with 1, 2, 3 and 5 mL plasma pool B (n = 3 replicates per day). a The yield of cfDNA (copies per extraction) is compared with the volume of plasma (mL) processed per extraction using the CNA kit (n = 9 per volume). Data points correspond to individual extracts. b The within-laboratory reproducibility of extract yield per millilitre of plasma is compared for different input volumes and three independent experiments. Mean values ± one standard deviation are plotted for each day (n = 3)

Fig. 5

Comparison of cfDNA copy number of seven different reference genes in 17 donor samples measured by qPCR assays. The results for each reference gene are displayed with a different symbol for each target

Fig. 6

Comparison of fold differences between reference genes measured by qPCR and droplet digital PCR (dPCR). The ratios of TERT/ERV3 and RPPH1/ERV3 genomic copy numbers in extracts from 17 plasma samples are compared for qPCR-based and droplet-dPCR-based measurements. Box and whisker plots depict the median value (line), interquartile range (box) and upper and lower limits (whiskers) based on Tukey’s honest significant difference test. Outlier values are indicated as single points

Fig. 7

Comparison of a multiple reference gene approach for plasma cfDNA quantification with single gene measurements. The 95 % confidence interval (grey area) associated with a normalised geometric average cfDNA quantity calculated on the basis of three reference genes (TERT, RPPH1, ERV3) and three independent qPCR experiments is compared with mean estimates for each of the above-mentioned reference genes and ALUJ for each donor