Adapting clinical chemistry plasma as a source for liquid biopsies.
Author(s): Ding SC, Yu J, Liao T, Ahmann L, Yao Y, Ho C, Wang L, Pinsky BA, Gu W
Publication: Elife, 2026, Vol. 14, Page
PubMed ID: 42065375 PubMed Review Paper? No
Purpose of Paper
This paper investigated whether heparin separator tubes are suitable for cell-free DNA (cfDNA) analysis by comparing yield, fragment size, fragment end motif frequencies, viral loads, copy number alteration (CNA) profiles, estimated tumor fraction, and/or CpG methylation between case-matched plasma from blood collected in heparin separator, EDTA and Streck tubes and processed immediately; between case matched blood collected and stored (up to 24 h at 4°C, room temperature or 37°C) in heparin separator or EDTA tubes; and in case-matched leftover viral positive heparin and EDTA plasma specimens
Conclusion of Paper
Fragment size profiles and end motif frequencies were comparable among the heparin, EDTA and Streck plasma that were processed immediately, and methylation levels for CpG sites with ≥30x coverage and cell type proportion estimates were modestly to strongly correlated between the plasma types (ρ≥0.65, all pairs). The fragment size profiles were comparable between (a) the immediately processed specimen and (b) plasma from blood that was stored for 24 h prior to the first centrifugation in EDTA tubes at room temperature or 4°C and then stored for 7 days at 4°C before the second centrifugation and (c) plasma from blood stored in heparin tubes at 4°C, and then stored for 7 days at 4°C before the second centrifugation. However, when blood was stored in heparin tubes at room temperature prior to the first centrifugation, cfDNA fragment size decreased with pre-centrifugation storage duration, indicating degradation. Not surprisingly, DNA became extremely degraded when blood was stored in heparin tubes at 37°C for 24 h between centrifugation steps.
Analysis of case-matched leftover heparin and EDTA plasma, revealed that cfDNA yield (ng/mL plasma) was highly variable and only weakly correlated between case-matched pairs, but the ratio of read counts aligning to the human genome normalized to the lambda reads was very strongly correlated between EDTA and heparin plasma when lambda phage was spiked in during library preparation (R2=0.998). cfDNA from heparin plasma was more degraded than EDTA plasma, as evidenced by a shift in fragment size distribution and changes in the frequency of four of the six most frequent 4-mer motifs at the 5’ end of the cfDNA fragments. Nevertheless, viral loads, CNA profiles, and estimated tumor fraction were comparable and strongly to very strongly correlated between case-matched EDTA and heparin plasma (ρ=0.95, P<0.001; ρ=0.72-0.96, P<0.001; and ρ=0.77, P<0.001, respectively). When case-matched EDTA and heparin plasma were compared, CpG methylation levels were modestly correlated (ρ=0.63, P<0.001), and cell type proportion estimates were comparable.
Studies
-
Study Purpose
This study compared fragment size, fragment end motif frequencies, CpG methylation and estimated cell proportions between case-matched plasma from blood collected in heparin separator, EDTA and Streck tubes. Blood was collected from 5 healthy volunteers during a single venipuncture into heparin separator, EDTA and Streck tubes (further tube details not provided). Plasma was separated by centrifugation at 1,600×g for 10 min followed by 16,000×g for 10 min. Plasma aliquots were then stored at -80°C. cfDNA was extracted using the Maxwell RSC ccfDNA Plasma Kit on a Maxwell RSC 48 instrument. Sequencing libraries were constructed using the NEBNext Ultra II DNA Library Prep Kit, and sequenced on a NovaSeq X platform; fragment size distributions were calculated based on the frequency of fragments between 50 and 600 bp. Copy number alterations were identified using ichorCNA. DNA was converted using the NEBNext Enzymatic Methyl-seq Conversion Module, and FLEXseq DNA libraries were sequenced using the NovaSeq X platform. Tissue of origin was calculated based on the top 250 unmethylated CpG markers for each cell type.
Summary of Findings:
Fragment size profiles and end motif rankings of the 256 possible 4-mers were comparable among plasma isolated from K2EDTA, Streck, and heparin separator tubes . Methylation levels for CpG sites with ≥30x coverage and cell type proportion estimates were modestly to strongly correlated between heparin and EDTA (ρ=0.70 and ρ=0.65-0.70, respectively) and heparin and Streck specimens (ρ=0.69, ρ=0.65-0.69, respectively).
Biospecimens
Preservative Types
- Frozen
Diagnoses:
- Normal
Platform:
Analyte Technology Platform DNA Bisulfite conversion assay DNA Next generation sequencing Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Biospecimen Acquisition Type of collection container/solution K2EDTA tube
Streck tube
Heparin separator tube
-
Study Purpose
This study compared cfDNA yield, fragment size, and fragment end motif frequencies; viral loads; CNA profiles; estimated tumor fraction; and CpG methylation levels between case-matched leftover heparin and EDTA plasma specimens. A total of 38 leftover paired viral positive heparin and EDTA plasma specimens stored at -80°C were obtained for the study. cfDNA was extracted using the Maxwell RSC ccfDNA Plasma Kit on a Maxwell RSC 48 instrument and quantified by Qubit. Sequencing libraries were constructed using the NEBNext Ultra II DNA Library Prep Kit and sequenced on a NovaSeq X platform. The ratio of viral to human genome reads, fragment length distribution and end motif frequencies were calculated from the NEBNext sequencing data. Copy number alterations were identified using ichorCNA using a 500 kb bin size. DNA was converted using the NEBNext Enzymatic Methyl-seq Conversion Module and FLEXseq DNA libraries were sequenced using the NovaSeq X platform. Tissue of origin was calculated based on the top 250 unmethylated CpG markers for each cell type.
Summary of Findings:
cfDNA yield (ng/mL plasma) was highly variable among plasma specimens and was weakly correlated between case-matched EDTA and heparin plasma specimens (ρ=0.30, P=0.003). However, when lambda phage was spiked in during library preparation, the ratio of read counts aligning to the human genome normalized to the lambda reads was very strongly correlated between EDTA and heparin plasma (R2=0.998). While both EDTA and heparin plasma had fragment peaks at approximately 70 and 170 bp, there were more short fragments and a smaller 166 bp peak in heparin plasma than EDTA plasma specimens. Importantly, the frequency of four of the six most frequent 4-mer motifs at the 5’ end of cfDNA fragments differed between heparin and EDTA plasma. Viral loads were comparable and very strongly correlated between case-matched EDTA and heparin plasma (ρ=0.95, P<0.001). CNA profiles, including gains and losses, were similar for case-matched EDTA and heparin plasma, with strong correlations observed for each of the six specimens with identified copy number variation (ρ=0.72-0.96 P<0.001). The tumor fraction (estimated based on CNA) was also strongly correlated between case-matched EDTA and heparin plasma (ρ=0.77, P<0.001). CpG methylation levels were modestly correlated between case-matched EDTA and heparin plasma (ρ=0.63, P<0.001), and cell type proportion estimates were also comparable.
Biospecimens
Preservative Types
- Frozen
Diagnoses:
- Normal
Platform:
Analyte Technology Platform DNA Next generation sequencing Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Biospecimen Acquisition Type of collection container/solution K2EDTA tube
Heparin separator tube
Storage Storage temperature 4°C (up to 24 h) before the first centrifugation
Room temperature (up to 24 h) before the first centrifugation
4°C for 7 days between centrifugations
37°C for 24 h between centrifugations
Storage Storage duration 0, 1, 3, or 24 h at 4°C before first centrifugation and 7 days between centrifugations
1, 2, 3, or 24 h at room temperature before first centrifugation and 7 days between centrifugations
24 h at 37°C between centrifugations
0 h before and between centrifugations
Biospecimen Aliquots and Components Centrifugation Centrifugation delays investigated
-
Study Purpose
This study compared yield, fragment size, fragment end motif frequencies, viral loads, CNA profiles, estimated tumor fraction, and CpG methylation between case-matched leftover heparin and EDTA plasma specimens. A total of 38 leftover paired viral positive heparin and EDTA plasma specimens were obtained. cfDNA was extracted using the Maxwell RSC ccfDNA Plasma Kit on a Maxwell RSC 48 instrument and quantified by Qubit. Sequencing libraries were constructed using the NEBNext Ultra II DNA Library Prep Kit and sequenced on a NovaSeq X platform. The ratio of viral to human genome reads, fragment length distribution and end motif frequencies were calculated from the NEBNext sequencing data. Copy number alterations were identified using ichorCNA using a 500kb bin size. DNA was converted using the NEBNext Enzymatic Methyl-seq Conversion Module and FLEXseq DNA libraries were sequenced using NovaSeq X platform. Tissue of origin was calculated based on the top 250 unmethylated CpG markers for each cell type.
Summary of Findings:
cfDNA yield (ng/mL plasma) was highly variable among the specimens and was weakly correlated between case-matched EDTA and heparin specimens (ρ=0.30, P=0.003). However, when lambda phage was spiked in during library preparation, the ratio of read counts aligning to the human genome normalized to the lambda reads was very strongly correlated between EDTA and heparin plasma (R2=0.998). While both EDTA and heparin plasma had fragments peaks at approximately 70 and 170 bp, there were more short fragments and a smaller 166 bp peak in the heparin plasma than the EDTA plasma. Importantly, the frequency of four of the six most frequent 4-mer motifs at the 5’ end of the cfDNA fragments differed between heparin and EDTA plasma. Viral loads were comparable and very strongly correlated between case-matched EDTA and heparin plasma (ρ=0.95, P<0.001). The CNA profile including gains and losses was similar for case-matched EDTA and heparin plasma with strong correlations observed for each of the six specimens with identified copy number variation (ρ=0.72-0.96 P<0.001). The tumor fraction (estimated based on CNA) was also strongly correlated between case-matched EDTA and heparin plasma (ρ=0.77, P<0.001). CpG methylation levels were modestly correlated between the case-matched EDTA and heparin plasma (ρ=0.63, P<0.001) and the estimated cell type proportion estimates were also comparable.
Biospecimens
Preservative Types
- Frozen
Diagnoses:
- Other diagnoses
Platform:
Analyte Technology Platform DNA Next generation sequencing DNA Bisulfite conversion assay DNA Fluorometry Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Biospecimen Acquisition Anticoagulant EDTA
Heparin