Skip to main content

Performance characteristics of a polymerase chain reaction-based assay for the detection of EGFR mutations in plasma cell-free DNA from patients with non-small cell lung cancer using cell-free DNA collection tubes.

May T, Clement MS, Halait H, Kohlmann A, Kohlmann M, Lai J, Lee N, Li-Sucholeiki X, Meldgaard P, Joshi S, Scudder S, Shrestha N, Sorensen B, Kiral M, O'Donnell P

Publication: PLoS One (2024) • Vol 19 • Page e0295987PMID: 38593164View on PubMed

Purpose

This paper compared the successful detection of EGFR mutationsbetween specimens collected in Roche Cell-Free DNA (cfDNA) Collection Tubes and K2EDTA tubes, among plasma from blood specimens collected in cfDNA tubes that were inverted 2-58 times, and among plasma obtained from cfDNA tubes using different centrifugation protocols.  The stability of specimens collected in cfDNA tubes was also investigated by comparing mutation detection after whole blood was stored for 25 hours at 32˚C followed by storage at 27˚C for ≤8 days or after plasma was stored at 32˚C for ≤25 hours, 2-8˚C for ≤8 days, or at -25˚C to -15˚C or -70˚C for ≤13 months with three freeze-thaw cycles.  The reproducibility of the assay and the effects of contamination with high levels of hemoglobin, triglycerides, bilirubin, albumin, or the preservative solution were also investigated.

Conclusion

EGFR mutation status was 100% concordant between tube types for specimens from patients with mutation-positive NSCLC, patients with mutation-negative NSCLC, and healthy patients (spiked with EGFR mutations).  EGFR mutation status was unaffected by any of the tested storage conditions. While at least 4 inversions of the Roche cfDNA tubes were necessary to prevent hemolysis, the number of tube inversions did not affect the detection of mutations in plasma from blood spiked with sheared DNA containing EGFR mutations. EGFR mutations were detected in the plasma from the spiked blood regardless of the centrifugation protocol used to separate the plasma. The previously identified limit of detection for K2EDTA plasma was verified using plasma from Roche cfDNA tubes, and in paired spiked specimens diluted to the limit of detection and analyzed separately; the agreement in the EGFR status between these specimens was ≥95.8%. When spiked blood from healthy patients was placed in three different lots of Roche cfDNA tubes and analyzed at three different sites with two different operators on three non-consecutive days, the agreement in EGFR status between specimens was ≥98.6%, with a coefficient of variation for cycle threshold value of 4.4-13.7% across all mutations. There was no effect of high levels of hemoglobin, triglycerides, bilirubin, albumin, or the preservative solution on the detection of EGFR mutations using the Cobas EGFR test.

Studies (1)

Purpose

This study compared EGFR mutation detection between specimens collected in Roche Cell-Free DNA (cfDNA) Collection Tubes and K2EDTA tubes, among plasma from blood specimens collected in cfDNA tubes that were inverted 2-58 times, and among plasma obtained from cfDNA tubes using different centrifugation protocols. The stability of specimens collected in cfDNA tubes was also investigated by comparing EGFR mutation detection after whole blood was stored for 25 hours at 32˚C followed by storage at 27˚C for ≤8 days or after plasma was stored at 32˚C for ≤25 hours, 2-8˚C for ≤8 days, or at -25˚C to -15˚C or -70˚C for ≤13 months with three freeze-thaw cycles. The reproducibility of the assay and the effects of contamination with high levels of hemoglobin, triglycerides, bilirubin, albumin, or the preservative solution were also investigated. To compare tube types, blood was collected from 17 patients with mutation-positive NSCLC, 34 patients with mutation-negative NSCLC, and 20 healthy patients (spiked with EGFR mutations) into Roche Cell-Free DNA (cfDNA) Collection Tubes and K2EDTA tubes. To investigate specimen stability, blood from patients (unspecified number) with NSCLC was stored for 25 h at 32˚C followed by storage at 27˚C for 4, 7, or 8 days and plasma was stored at 32˚C for 5 or 25 hours, at 2-8˚C for 4 or 8 days, or at -25˚C to -15˚C or –70˚C for 31 days or 13 months with 3 freeze-thaw cycles (details not provided).  Potential effects of tube inversions were investigated by inverting tubes of whole blood collected from healthy volunteers for 0, 2, 4, 6, 8, or 28 times, then adding sheared DNA from a cell line that contains EGFR mutations, and inverting the tubes 2, 4, 6, 8, 10, or 30 times, respectively; plasma was subsequently separated. The effect of centrifugation speed was investigated by mixing whole blood from healthy volunteers DNA from a cell line that contains EGFR mutations; separating  plasma by centrifugation at 2600 g for 9 or 16 min, at 2600 g for 12.5 min, or at 1100 g for 9 or 16 min; and analysis with a Cobas EGFR system. The limit of detection was verified using plasma from patients with NSCLC that was spiked at the limit of detection (75 or 100 copies/mL) with EGFR mutations and was then validated by creating two spiked specimens and analyzing them separately. Assay reproducibility was tested by spiking blood from healthy donors into three different lots of Roche cfDNA tubes and analyzing samples at three different sites with two different operators on three non-consecutive days. To test for effects of interfering substances, hemoglobin (2 g/L), triglycerides (33 g/L), albumin (60 g/L), bilirubin (0.2 g/L), or Roche cfDNA preservative solution (4×) were added to plasma from healthy volunteers (spiked with EGFR mutations).  For analysis, EGFR mutations were profiled by the real-time PCR-based Cobas EGFR test and were validated in K2EDTA plasma specimens by NGS using an Illumina MiSeq System.

Summary of Findings

EGFR mutation status was 100% concordant between tube types for specimens from patients with mutation-positive NSCLC, patients with mutation-negative NSCLC, and healthy patients (spiked with EGFR mutations).  EGFR mutation status was unaffected by any of the tested conditions (storage of whole blood for 25 hours at 32˚C followed by storage at 27˚C for ≤8 days; storage of plasma at 32˚C for ≤25 hours, 2-8˚C for ≤8 days, or at -25˚C to -15˚C or -70˚C for ≤13 months with 3 freeze-thaw cycles) (details not provided). While at least 4 inversions of the Roche cfDNA tubes were necessary to prevent hemolysis, the number of tube inversions did not affect the detection of mutations in plasma from blood that was spiked with DNA from a cell line that contains EGFR mutations. EGFR mutations were detected in plasma from the spiked blood regardless of the centrifugation protocol used to separate the plasma. The previously identified limit of detection for K2EDTA plasma was verified using plasma isolated from blood collected in Roche cfDNA tubes, and in paired spiked specimens that were diluted to the limit of detection and analyzed separately; the agreement in EGFR status between specimens was ≥95.8%. When spiked blood from healthy patients was placed in three different lots of Roche cfDNA tubes and analyzed at three different sites with two different operators on three non-consecutive days the agreement between specimens was ≥98.6%, with a coefficient of variation for cycle threshold value of 4.4-13.7% across all mutations. There was no effect of high levels of hemoglobin, triglycerides, bilirubin, albumin, or the preservative solution on the detection of EGFR mutations using the Cobas EGFR test.

  • Biospecimens

    FluidBlood
    FluidPlasma
  • Preservation Types

    FrozenNone (Fresh)Streck/BCT
  • Diagnoses

    NeoplasticCarcinoma
    Normal
  • Platforms

    DNAReal-time qPCR
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    PreaquisitionDiagnosis/ patient condition
    Healthy
    Mutation negative NSCLC
    Mutation positive NSCLC
    Biospecimen AcquisitionType of collection container/solution
    K2EDTA tube
    Roche Cell-Free DNA (cfDNA) Collection Tube
    Biospecimen Aliquots and ComponentsBiospecimen components
    4 times Roche cfDNA preservative solution
    Added albumin (60 g/L)
    Added bilirubin (0.2 g/L)
    Added hemoglobin (2 g/L)
    Added triglycerides (33 g/L)
    Biospecimen Aliquots and ComponentsBiospecimen mixing
    Inverted 0 times then twice
    Inverted 2 times then 4 times
    Inverted 28 times then 30 times
    Inverted 4 times then 6 times
    Inverted 6 times then 8 times
    Inverted 8 times then 10 times
    Biospecimen Aliquots and ComponentsCentrifugation
    Centrifugation delays investigated
    Multiple durations compared
    Multiple speeds compared
    StorageFreeze/thaw cycling
    0 cycles
    3 cycles
    StorageStorage duration
    Blood for 25 h at 32˚C followed by storage at 27˚C for 4 day
    Blood for 25 h at 32˚C followed by storage at 27˚C for 7 day
    Blood for 25 h at 32˚C followed by storage at 27˚C for 8 days
    Plasma for 13 months at –25˚C to –15˚C with 3 freeze -thaw cycles
    Plasma for 13 months at –70˚C with 3 freeze -thaw cycles
    Plasma for 25 h at 32˚C
    Plasma for 31 days at –25˚C to –15˚C with 3 freeze -thaw cycles
    Plasma for 31 days at –70˚C with 3 freeze -thaw cycles
    Plasma for 4 days at 2-8˚C
    Plasma for 5 h at 32˚C
    Plasma for 8 days at 2-8˚C
    StorageStorage temperature
    Blood at 32˚C for 25 h followed by storage at 27˚C for 4 days
    Blood at 32˚C for 25 h followed by storage at 27˚C for 7 days
    Blood at 32˚C for 25 h followed by storage at 27˚C for 8 days
    Plasma at 2-8˚C for 4 days
    Plasma at 2-8˚C for 8 days
    Plasma at 32˚C for 25 h
    Plasma at 32˚C for 5 h
    Plasma at –25˚C to –15˚C for 13 months with 3 freeze -thaw cycles
    Plasma at –25˚C to –15˚C for 31 days with 3 freeze -thaw cycles
    Plasma at –70˚C for 13 months with 3 freeze -thaw cycles
    Plasma at –70˚C for 31 days with 3 freeze -thaw cycles