First, we set up a framework using the dPCR technology to quantify cfDNA and cfRNA from human blood plasma in order to compare cfDNA/cfRNA co-purification kit performance. To that end, we optimized two dPCR duplex assays that are designed to quantify both cfDNA and cfRNA, by ensuring that primers and probes are located within a highly abundant exon. Next, we applied our optimized workflow to evaluate the co-purification performance of two manual and two semi-automated kits using different plasma input volumes (0.06–4 mL).

Sample collection was approved by the ethics committee of Ghent University Hospital (registration number B670201733701) and written informed consent was obtained from the healthy donors. Venous blood was collected from healthy donors in 2 different blood collection tubes: BD Vacutainer Plastic K2EDTA tube (EDTA; Becton Dickinson and Company, 367525) and Vacuette Tube 9 mL 9NC Coagulation sodium citrate 3.2% (citrate; Greiner Bio-One, 455322). Immediately after blood draw, blood collection tubes were gently inverted five times and tubes were transported to the laboratory for immediate plasma preparation. Platelet-depleted plasma was prepared within two hours after blood draw, by means of two sequential centrifugation steps (two times 2500g for 15 min) on a Centrifuge 5804 (Eppendorf, 5804000013) with Rotor A-4-44 (Eppendorf, 5804709004) and appropriate adapters (Eppendorf, 5804753003). Plasma was snap frozen in liquid nitrogen and stored at − 80 °C immediately after preparation. Hemolysis was assessed by determining the absorbance at 414 nm, i.e., the levels of free heamoglobin, by spectral analysis using a NanoDrop 1000 Spectrophotometer (Thermo Fisher Scientific). Absorbances are ranging from 0.103 to 0.153.

Nucleic acids were extracted with six different commercially available (co-)purification kits by following the manufacturers’ manual: (1) miRNeasy Serum/Plasma Advanced Kit (MIRA; Qiagen, 217204), (2) QIAamp ccfDNA/RNA Kit (abbreviated to CCF; Qiagen, 55184), (3) the iCatcher Circulating cfDNA/cfRNA 4000 kit (CatchGene, AC30400) in combination with iCatcher 12 Automated Nucleic Acid Purification System (CAT; CatchGene, IC1200) (4) MagNA Pure 24 Total NA Isolation Kit with the cfNA ss 2000 and cfNA ds 2000 purification protocols (Roche, 07658036001) in combination with the MagNA Pure 24 instrument (MAP; Roche, 07290519001), (5) Maxwell ccfDNA LV Plasma Kit (Promega, AS1480) in combination with the Maxwell RSC Instrument (MAX; Promega, AS4500), and (6) miRNeasy Serum/Plasma Kit (MIR; Qiagen, 217184) (Fig. 1). The MIRA, CCF, MAP and CAT kits were included in the study to examine their capacity to co-purify cfDNA and cfRNA, while MAX and MIR kits served as a reference for cfDNA only and cfRNA only extractions, respectively, as these kits are routinely used in our department for these applications. For each kit, different plasma input volumes, ranging between 0.06 and 4 mL, from two donors (three in case of CAT) were tested with the maximum recommended elution volume (Table 1). Eluates were stored at − 80 °C until further processing.

Eluates of each of the four cfDNA/cfRNA co-purification kits (MIRA, CCF, MAP and CAT) were split in two equal parts, using one part for cfDNA quantification and the other part for cfRNA quantification (Fig. 1). The cfDNA part was immediately used for quantification with digital PCR without any further processing. For the cfRNA part, DNA was removed using HL-dsDNase (ArcticZymes, 70800-202) and Heat & Run 10× Reaction Buffer (ArcticZymes, 66001). Briefly, 1 µl HL-dsDNase and one tenth of the RNA input volume as reaction buffer were added, and incubated for 10 min at 37 °C, followed by 5 min inactivation at 55 °C. Subsequently, reverse transcription was performed using the iScript Advanced cDNA Synthesis Kit for RT-qPCR (Bio-Rad, 1725038) to enable digital PCR-based quantification of the cDNA (~ cfRNA). The full eluate of the MIR kit underwent DNA removal and cDNA synthesis using the same protocol, serving as a cfRNA only control. The volume of the MAX eluate was reduced to 20 µl by means of vacuum centrifugation (Eppendorf, Concentrator Plus, program V-AQ at 30 °C). As previous findings indicated incompatibilities of the MAP elution buffer for downstream DNase treatment [15], and to reduce elution volume, the MAP eluate was further purified and concentrated using Vivacon 500 2000 MWCO Hydrosart ultrafiltration columns (Sartorius, VN01H91) prior to DNA removal [15]. cDNA and cfDNA was stored at − 20 °C until further processing.

For unbiased quantification of both cfDNA and cfRNA, two digital PCR duplex assays were designed each targeting a single, well covered exon of two highly abundant genes in healthy donor plasma (based on RNA sequencing data from the Extracellular RNA Quality Control study [15]): CAVIN2 (HEX)/NRGN (FAM) and AIF1 (FAM)/B2M (HEX). Designing a primer pair within a single exon allows the usage of the same assay for both cfDNA and cfRNA quantification (Table 2).

PCR primers were picked using the Primer3Plus tool (with default settings, except amplicon size range of 60–100 nucleotides [17]). The performance of the primers was thoroughly evaluated in silico. To determine the primer specificity, BiSearch e-PCR (with default settings, except for mismatch string: 1233333333333333 [18]) and the UCSC tool [19] were used. Subsequently, the OligoEvaluator tool [20] was used to check for secondary structure formation and GC content. Lastly, the SBT tool [21] was used to predict the melting temperature (Tm). The hydrolysis probe sequence was picked manually in between forward and reverse primer, aiming for a Tm of at least 3 °C higher than the primers’ Tm. Primers and probes were ordered with Integrated DNA Technologies (IDT, Leuven, Belgium). Probes were ordered as double-quenched hydrolysis probes with optional LNA nucleotides to enhance the Tm and purified by HPLC. Primers were purified by standard desalting. All oligonucleotides were resuspended in TE buffer (10 mM Tris–HCl (pH 8.0), 0.1 mM EDTA) to 100 µM (primers) and 10 µM (probes) and stored at − 20 °C. The primer efficiency and specificity was also validated on a dilution series of Quantitative PCR Human Reference Total RNA (Agilent technologies, 750500), reverse transcribed to cDNA.

Digital PCR was performed using the QX100 Droplet Digital PCR system (Bio-Rad, California, USA), according to the manufacturer’s protocol with minor modifications (250 nM primer and 100 nM probe reaction mix concentration). Each 20 µl dPCR reaction contains 10 µl 2 × ddPCR Supermix for Probes (Bio-Rad, 1863010), 2 µl of primer and probe mix (with a total of 4 primers and 2 probes per duplex assay) and 2—8 µl template. After pipetting 20 µl sample mix and 70 µl droplet generation oil (Bio-Rad, 1863005) in the cartridge (Bio-Rad, 1864008), droplets were generated by means of the Bio-Rad QX100 Droplet Generator. Droplets were then transferred from the cartridge to a 96-well plate (Bio-Rad, 12001925) and a thermocycling program was performed on a C1000 Touch Thermal cycler (Bio-Rad): 95 °C for 10 min, followed by 40 cycles of 15 s on 95 °C and 1 min at 56.9 °C (optimized annealing temperature for both duplexes by means of a gradient dPCR). Finally, reactions were heated to 98 °C for 10 min and then cooled down to 12 °C before transferring the plate to the QX100 Droplet Reader (Bio-Rad). Each plate also included a positive control (PC) and negative no template control (NTC). QuantaSoft Analysis Pro Software Version 1.3.2.0 was used to calculate the number of copies per µl in the dPCR reaction by manual thresholding, followed by copies per µl eluate concentration determination (Additional file 6: Fig. S3) to enable comparison among kits.

To determine the added value of the combined analysis of cfDNA and cfRNA, separate quantification of cfDNA and cfRNA was compared to combined quantification of both. As this is a proof-of-principle experiment, only one kit (MIRA0.06) was used to co-purify nucleic acids from plasma of three healthy donors (platelet-free plasma). Eluates were split into three equal parts to quantify cfDNA, cfRNA and cfDNA/cfRNA using the two dPCR duplex assays (Additional file 7: Fig. S4). For cfDNA/cfRNA combined quantification, only cDNA synthesis was performed on the eluate, as such containing both cfDNA and cDNA, while for cfRNA only quantification, a DNase treatment was performed prior to cDNA synthesis. For cfDNA quantification, the eluate remained untouched.