NIH, National Cancer Institute, Division of Cancer Treatment and Diagnosis (DCTD) NIH - National Institutes of Health National Cancer Institute DCTD - Division of Cancer Treatment and Diagnosis

Multicenter Evaluation of Circulating Cell-Free DNA Extraction and Downstream Analyses for the Development of Standardized (Pre)analytical Work Flows.

Author(s): Lampignano R, Neumann MHD, Weber S, Kloten V, Herdean A, Voss T, Groelz D, Babayan A, Tibbesma M, Schlumpberger M, Chemi F, Rothwell DG, Wikman H, Galizzi JP, Bergheim IR, Russnes H, Mussolin B, Bonin S, Voigt C, Musa H, Pinzani P, Lianidou E, Brady G, Speicher MR, Pantel K, Betsou F, Schuuring E, Kubista M, Ammerlaan W, Sprenger-Haussels M, Schlange T, Heitzer E

Publication: Clin Chem, 2019, Vol. , Page

PubMed ID: 31628139 PubMed Review Paper? No

Purpose of Paper

The purpose of this paper was to investigate the effects of collection tube type and extraction using various methods at different sites on the cell-free DNA yield, fragment size profile, integrity, and variant allele frequency (VAF) from plasma pools spiked with mononucleosomal DNA. The effects of quantification method and methods for determination of VAF were also investigated.

Conclusion of Paper

Although the initial cfDNA concentration of the plasma pool was slightly higher in PAXgene tubes than Streck tubes, the fragment size profile, yields, integrity, and VAFs were generally comparable among specimens collected in the different tube types. Although extraction with the MAXwell AX1115 Kit resulted in the least variation, use of any of the three Qiagen kits resulted in higher yields and integrity indices (based on ratio of Alu 187 bp to Alu 60 bp PCR products) and less variation in the detected variant allele frequency, especially when quantified by qPCR. Overall, the Chemagic and Maxwell AS1480 kits were also found to result in lower yields and integrity indices than the Qiagen kits. Use of Alu PCR or either of the ddPCR assays rather than Qubit for quantification resulted in a high level of overall concordance and low variability. VAFs were generally very strongly correlated between next-generation sequencing (NGS) and ddPCR.

Studies

  1. Study Purpose

    The purpose of this study was to investigate the effects of collection tube type and extraction method on the cfDNA yield, fragment size profile, integrity, and VAF from plasma pools spiked with mononucleosomal DNA. The effects of cfDNA quantification method and methods for determination of VAF were also investigated. Plasma pools were obtained from blood drawn from 15 healthy donors into Streck Cell-Free DNA BCT tubes and from 28 healthy donors into PAXgene Blood ccfDNA tubes. Blood was spiked with mononucleosomal DNA from one of two different non-small cell lung cancer cell lines, each harboring a homozygous TP5320 mutation. Each of 15 laboratories was sent three replicate plasma pools from each tube type (PAXgene and Streck) with one spike-in and two replicate plasma pools from each tube type with the other spike-in. DNA was extracted using one of six methods: QIAamp Circulating Nucleic Acid Kit (4 sites), QIAsymphony circulating DNA Kit (4 sites), QIAamp MinElute ccfDNA Kit (4 sites), Maxwell RSC ccfDNA Plasma Kit AX1115 (3 sites), Maxwell RSC ccfDNA Plasma Kit AS1480 (1 site), and Chemagic CNA 4k Kit special (1 site) following the manufacturer protocols. Eluates were shipped on dry ice to four locations: Bayer AG, QIAGEN, Medical University of Graz, and TATAA Biocenter Ab in Sweden. At Bayer, cfDNA was quantified using Qubit High Sensitivity Kit, integrity was assessed by bioanalyzer, and VAF as well as quantification were determined by ddPCR amplification of mutant and wildtype TP53. At QIAGEN, cfDNA was quantified using the real-time PCR-based Quantiplex Pro assay. At TATAA Biocenter, DNA integrity was assessed using Alu PCR of 60 and 187 bp products. At Medical University of Graz, the VAF was determined by NGS.

    Summary of Findings:

    The initial cfDNA concentration of the plasma pool was 17.06 ng/mL for the pool from Streck tubes and 21.3 ng/mL for the pool from PAXgene tubes. The size distribution analysis showed a fragment peak at 140 bp (spiked-in mnDNA) and 166 bp (donor-derived cfDNA). Inter-laboratory variability in quantification was greater than intra-laboratory variability and variability was higher using Qubit than the Alu PCR or either of the ddPCR assays. The highest variability in yield was observed when extraction was with the QIAamp CNA Kit and the lowest variation when extracted with the Maxwell AX1115 Kit, but the yields as determined by PCR-based methods were higher using the Qiagen kits than the MAXwell AX1115 Kit. The lowest yields occurred using the Chemagic Kit. The yield was found to be higher when specimens extracted with the Maxwell 1115 Kit were quantified by Qubit rather than a PCR-based method and those extracted with QIAamp CNA were quantified using Qubit rather than QuantiplexPro but the differences were significant only in Streck plasma (P<0.05, all) and not PAXgene plasma. Bioanalyzer analysis showed both expected peaks in all specimens and the absence of high molecular weight DNA, but the presence of two peaks was most prominent in specimens extracted using QIAamp or QIAsymphony. The authors attribute this to the relatively higher concentration of the cfDNA resulting in use of more cfDNA. The integrity index of the DNA as determined using the Alu PCR assay (ratio of Alu 187 bp to Alu 60 bp PCR products) was significantly higher using the Qiagen kits (QIAamp CNA, QIAsymphony Kits, and QIAamp MiniElute) (1.30-1.46, all) than for the Maxwell AX1115 Kit (1.34-1.35), regardless of tube type (P<0.05, all). The integrity indices for the Maxwell AS1480 and Chemagic Kits were slightly lower than for the Qiagen kits but still higher than with the Maxwell AX1115 Kit. Like yields, VAF showed more inter-laboratory variability than intra-laboratory variability but were strongly to very strongly correlated between NGS and ddPCR (r=0.7979–0.9854). The VAFs were highly concordant among the Qiagen extraction methods, quantification methods, and tube types for both spike-in sets. However, the VAF was much higher and much more variable in specimens from both tube types extracted using the Maxwell AX1115 Kit and quantified by ddPCR and a similar trend toward higher VAF was also observed for spike-in set 1 when extracted using the Maxwell AS1480 Kit and quantification by ddPCR. The VAFs were generally lower in specimens extracted using the Chemagic Kit, regardless of tube type.

    Biospecimens
    Preservative Types
    • Streck/BCT
    • PAXgene
    Diagnoses:
    • Normal
    Platform:
    AnalyteTechnology Platform
    DNA Automated electrophoresis/Bioanalyzer
    DNA Fluorometry
    DNA Real-time qPCR
    DNA Next generation sequencing
    DNA Digital PCR
    Pre-analytical Factors:
    ClassificationPre-analytical FactorValue(s)
    Biospecimen Acquisition Type of collection container/solution Streck Cell-Free DNA BCT
    PAXgene Blood ccfDNA tube
    Biospecimen Preservation Type of fixation/preservation PAXgene
    Blood collection tube additive
    Digital PCR Specific Technology platform NGS
    Fluorometry Specific Technology platform Qubit
    ddPCR
    Real-time PCR
    Analyte Extraction and Purification Analyte isolation method QIAamp Circulating Nucleic Acid kit
    QIAsymphony circulating DNA kit
    QIAamp MinElute ccfDNA kit
    Maxwell RSC ccfDNA Plasma Kit AX1115
    Maxwell RSC ccfDNA Plasma Kit AS1480
    Chemagic CNA 4k kit special kit

You Recently Viewed  

News and Announcements

  • Most Downloaded SOPs in 2024

  • New Articles on the GTEx Project are Now FREELY Available!

  • Just Published!

  • More...