Evaluation of automatic cell free DNA extraction metrics using different blood collection tubes.
Author(s): Andersson D, Kristiansson H, Luna Santamaría M, Zafar H, Mijakovic I, Torinsson Naluai Å, Ståhlberg A
Publication: Sci Rep, 2025, Vol. 15, Page 19364
PubMed ID: 40461680 PubMed Review Paper? No
Purpose of Paper
The paper compared cell-free DNA (cfDNA) levels and fragment size in plasma from blood collected in EDTA, Streck, Norgen and PAXgene tubes and stored for 0, 48 and 168 h before centrifugation. The effect of isolating plasma using single versus two-step centrifugation was also investigated. Blood was collected from healthy individuals, and cfDNA was quantified using fluorometry, single-locus real-time PCR and multi-locus real-time PCR quantification; fragment size was analyzed by real-time PCR and capillary electrophoresis.
Conclusion of Paper
cfDNA levels quantified by single locus real-time PCR were strongly correlated with those determined by fluorometry (r=0.79, P<2.2e-16) and very strongly correlated with those from multi-locus real-time PCR (r=0.96, P<2.2e-16). Plasma cfDNA levels were dependent on collection tube type. Immediately processed plasma from EDTA and Streck tubes had comparable cfDNA levels, which were higher than plasma collected in Norgen or PAXgene tubes that were immediately processed. Levels of cfDNA in plasma were significantly higher relative to immediately processed plasma when blood was stored for ≥48 h before centrifugation in EDTA or PAXgene tubes and when stored for 168 h in EDTA, PAXgene tubes or Streck tubes, but were unaffected by storage of blood for 168 h in Norgen tubes. Correlations in cfDNA levels among time points were strongest for Streck plasma (r=0.42-0.62), followed by Norgen and PAXgene plasma (r=0.17-0.54 and 0.21-0.46, respectively) and were lowest for EDTA plasma (r=0.08-0.39). cfDNA levels of immediately processed plasma were strongly correlated when PAXgene and EDTA plasma were compared (r=0.71) and modestly correlated between EDTA and Norgen or Streck plasma (r=0.52 and 0.53, respectively) and between Norgen and PAXgene plasma (r=0.54). When blood was stored for 48 h, correlations between cfDNA levels between any two tube types were weak (r=0.14-0.27). Interestingly, among specimens stored for 168 h before centrifugation, modest correlations in cfDNA levels were observed between Streck plasma and Norgen or PAXgene plasma (r=0.42 and r=0.50, respectively). As expected, the ratio of long to short DNA amplicons in plasma (an indication of contamination due to cell lysis) increased with the duration of pre-centrifugation storage of blood in EDTA, PAXgene or Norgen tubes, but no significant differences were observed in the ratio of long to short DNA amplicons in plasma among specimens stored for ≤168 h in Streck tubes. Capillary electrophoresis confirmed this corresponded to an increase in long DNA, particularly in EDTA plasma. The ratio of long to short DNA amplicons and DNA levels were generally lower when plasma was obtained by dual centrifugation rather than single centrifugation, but the significance of the difference was limited to blood that was stored in Norgen tubes (P≤0.01 for the 48 h time point; P≤0.0001 for all other timepoints)). The authors report that there was no correlation between cfDNA levels and patient age (unspecified range) or gender.
Studies
-
Study Purpose
The study compared cell-free DNA (cfDNA) levels and fragment size in plasma from blood collected in EDTA, Streck, Norgan and PAXgene tubes and stored for 0, 48 and 168 h before centrifugation. Effects associated with isolating plasma with a single versus two centrifugation steps were also investigated. Blood was collected from 23 healthy volunteers into BD Vacutainer PPT Plasma Preparation Tubes, Norgen cf-DNA/cf-RNA Preservative Tubes, PAXgene Blood ccfDNA Tubes, and Streck Cell-Free DNA BCTs. Specimens were collected from 20 of the volunteers twice (a mean of 47.6 days apart). Tubes were stored at room temperature for 0, 48 and 168 h before plasma separation. Plasma was separated by centrifugation of K2EDTA, PAXgene, and Streck tubes at 2000 g for 10 min and Norgen tubes at 2000 g for 20 min. To investigate the effect of a second centrifugation, some plasma aliquots were then centrifuged at 16,000 g for 15 min at 4°C. Isolated plasma was stored at -80°C for 83-345 days before DNA extraction from K2EDTA, Norgen and Streck plasma with the QIAsymphony DSP Circulating DNA Kit and DNA extraction from PAXgene plasma using the QIAsymphony PAXgene Blood ccfDNA Kit. cfDNA was quantified by Qubit dsDNA high sensitivity (HS) Assays and by real-time PCR amplification of PDGFRA (74 bp), FLI1 (445 bp), and 60 and 187 bp fragments of Alu. cfDNA was concentrated and fragment size was analyzed using the Agilent HS Large Fragment 50 kb Kit on a fragment analyzer.
Summary of Findings:
cfDNA levels quantified by single locus real-time PCR was strongly correlated with levels determined by fluorometry (r=0.79, P<2.2e-16), and very strongly correlated with those determined by multi-locus real-time PCR (r=0.96, P<2.2e-16). Plasma cfDNA levels were dependent on collection tube type. cfDNA levels were lower in immediately processed plasma collected in Norgen (P≤ 0.0001, both) or PAXgene tubes (P≤0.0001, both) relative to immediately processed plasma from EDTA and Streck tubes. When plasma was immediately processed, cfDNA levels were lower in plasma from Norgen than Paxgene tubes (P≤ 0.0001), but comparable in plasma from Streck and EDTA tubes. Plasma cfDNA levels were significantly higher when blood was stored for 48 h compared to those stored for 0 h before centrifugation in EDTA or PAXgene tubes (P≤ 0.0001 and P≤ 0.01, respectively); when stored for 168 h rather than 0 h before centrifugation in EDTA, PAXgene tubes or Streck tubes (p ≤ 0.0001, P≤ 0.001 and P≤ 0.01, respectively); and when stored for 168 rather than 48 h in EDTA or PAXgene tubes (P≤ 0.0001 and P≤ 0.05, respectively). Not surprisingly, correlations among storage time points were strongest for Streck plasma (r=0.42-0.62), followed by Norgen and PAXgene plasma (r=0.17-0.54 and 0.21-0.46, respectively) and were weakest for EDTA plasma (r=0.08-0.39). When blood was immediately processed, cfDNA levels were strongly correlated between PAXgene and EDTA plasma (r=0.71), and modestly correlated between EDTA and Norgen or Streck plasma (r=0.52 and 0.53, respectively) and between Norgen and PAXgene plasma (r=0.54). When blood was stored for 48 h, cfDNA levels were weakly correlated for comparisons between any two tube types(r=0.14-0.27). Interestingly, modest correlations in cfDNA levels were observed between Streck plasma and Norgen or PAXgene plasma that were stored for 168 h before centrifugation (r=0.42 and r=0.50, respectively). As expected, the ratio of long to short DNA amplicons in plasma (an indication of contamination due to cell lysis) increased with the duration of pre-centrifugation storage of EDTA blood from 0.02 at 0 h to 0.11 at 48 h (P≤0.0001) and to 0.13 at 168 h (P≤0.0001). Significant increases in the ratio of long to short DNA amplicons in plasma were also observed following storage in PAXgene tubes (0.011 at 0 h to 0.110 after 48 h, P≤0.0001; 0.110 after 48 h to 0.258 after 168 h, P≤0.0001) and after storage in Norgen tubes for 168 h (0.003 at 0 h and 0.050 after 168, P≤0.001), but no significant differences were observed in the ratio of long to short DNA amplicons in plasma among specimens stored for ≤168 h in Streck tubes. The ratio of long (contaminating) to short (cfDNA) DNA amplicons in plasma was significantly correlated with the level of DNA in EDTA and PAXgene plasma (r=0.500, P=4.4e-9 and r=0.230, P=0.0098, respectively), but no significant correlations were observed in Norgen or Streck plasma. Capillary electrophoresis confirmed an increase in long DNA, particularly in EDTA plasma. The ratio of long to short DNA amplicons and DNA levels were generally lower when plasma was obtained by dual centrifugation rather than single centrifugation, but significance was limited to blood stored in Norgen tubes (P≤0.01 for the 48 h time point; P≤0.0001 for all other timepoints). The authors report that cfDNA levels were not correlated to patient age (unspecified range) or gender.
Biospecimens
Preservative Types
- Streck/BCT
- PAXgene
- None (Fresh)
- Other Preservative
Diagnoses:
- Normal
Platform:
Analyte Technology Platform DNA Real-time qPCR DNA Fluorometry DNA Automated electrophoresis/Bioanalyzer Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Storage Storage duration 0 h
48 h
168 h
Biospecimen Aliquots and Components Centrifugation Centrifugation delays investigated
Different number of centrifugation steps compared
Real-time qPCR Specific Technology platform Multi-locus PCR
Single loci PCR
Fluorometry
Capillary electrophoresis
Single-locus
Biospecimen Acquisition Type of collection container/solution EDTA tube
Streck tube
Norgen tube
PAXgene tube