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

Cell-free DNA analysis in healthy individuals by next-generation sequencing: a proof of concept and technical validation study.

Author(s): Alborelli I, Generali D, Jermann P, Cappelletti MR, Ferrero G, Scaggiante B, Bortul M, Zanconati F, Nicolet S, Haegele J, Bubendorf L, Aceto N, Scaltriti M, Mucci G, Quagliata L, Novelli G

Publication: Cell Death Dis, 2019, Vol. 10, Page 534

PubMed ID: 31296838 PubMed Review Paper? No

Purpose of Paper

This study explored the potential effects of cancer diagnosis and plasma volume on cell-free DNA (cfDNA) yield, and of cfDNA volume on library output, coverage, and limit of detection. The study also examined the concordance between mutations identified in cfDNA and those identified in tissue biopsies, as well as investigated if cfDNA sequencing of healthy individuals could identify future cancer occurrence.

Conclusion of Paper

As expected, cfDNA yield was significantly higher in plasma from cancer patients than healthy controls and significantly correlated with the volume of plasma used, regardless of cancer status.  The amount of cfDNA used was correlated to NGS library output and inversely correlated to the limit of detection, consequently the median molecular coverage was higher and limit of detection lower in specimens from cancer patients than healthy controls. Despite a duration of 0-70 months between biopsy and tumor collection, 70% of mutations were concordant between the specimen types. While not significant, violin plots revealed a trend to more concordant mutations in specimens stored for shorter durations.  Sequencing of cfDNA from healthy individuals identified clinically relevant germline and/or hotspot mutations in 7 of 55 patients.  Upon follow-up 1-10 years later (mean of 8.5 years), 3 of these patients were diagnosed with fibroadenoma or hyperplasia of the breast, two with breast cancer, and one with a non-breast solid tumor.

Studies

  1. Study Purpose

    This study explored the potential effects of cancer diagnosis and plasma volume on cfDNA yield, and of cfDNA volume on library output, coverage, and limit of detection. The study also examined the concordance between mutations identified in cfDNA and those identified in prior tissue biopsies, as well as investigated if sequencing of cfDNA in healthy individuals could identify future cancer occurrence. Blood was collected from 114 healthy donors undergoing a screening mammography, nine breast cancer patients, and 54 lung cancer patients into either K2EDTA or BCT tubes. Plasma was obtained after two centrifugations (twice at 1600 x g for 30 min at room temperature) and stored at -80°C.  cfDNA was extracted from 0.4-5.5 mL plasma using MagMax Cell-Free Total Nucleic Acid Isolation Kit. Yield was determined using the Qubit dsDNA HS assay kit and integrity assessed using the Agilent High Sensitivity D1000 ScreenTape System. NGS libraries were prepared from 2.5-10.5 ng of cfDNA from 55 healthy donors and 40 cancer patients using HeliXmoker, HeliXgyn, and HeliXafe workflows and sequenced on an Ion S5 instrument. Fresh 10 µm FFPE sections of biopsies obtained at diagnosis were obtained for eight of the breast cancer patients and thirty of the non-small cell lung cancer patients. DNA was extracted from xylol deparaffinized sections using the RecoverAll Extraction Kit and quantified using the Qubit 2.0 dsDNA HS assay.  Oncomine Solid Tumor Assay were sequenced on an Ion S5 instrument and results were compared to those obtained HeliXgyn protocol (breast cancer patients) and HeliXmoker protocol (NSCLC patients).

    Summary of Findings:

    As expected, the cfDNA concentration was significantly higher in plasma collected from cancer patients than healthy controls (P=0.006). The total yield of cfDNA was significantly correlated with the volume plasma used for both healthy controls (ρ=0.244, P=0.0089) and cancer patients (ρ=0.587, P<0.0001).  The NGS library output was correlated with the amount of cfDNA used as input for both healthy controls (ρ=0.349, P=0.0088) and cancer patients (ρ=0.699, P<0.0001). The limit of detection was also correlated with cfDNA input for both healthy controls (ρ=-0.551, P<0.0001) and cancer patients (ρ=-0.790, P<0.0001). While the percentage of mapped reads did not differ among specimens from healthy controls and cancer patients, specimens from cancer patients had a higher median molecular coverage  and a lower limit of detection compared to healthy individuals (P<0.001), but this may be attributable to the higher cfDNA concentration observed among cancer patients. When concordance was examined among the subset of specimens for which sequencing was performed for both cfDNA and tissue biopsies, , concordance was 71%, with 26% of these specimens carrying additional clinically relevant mutations detected in cfDNA. The most common concordant mutations in breast cancer patients were in PIK3CA, AKT1 and ERBB3, with TP53, ESR1 and BRAF mutations more likely to be found solely in plasma, but it is important to note that blood was collected 0-70 months after biopsy. Though the trend was nonsignificant, violin plots revealed concordance between specimen types increased when specimens were stored for shorter durations. In NSCLC, EGFR mutations were most likely to be concordant, but detected discordant EGFR deletion mutations were more likely to occur in tissue specimens while EGFR substitution mutations were more likely to occur in plasma. Sequencing of cfDNA from healthy individuals identified clinically relevant mutations in 7 specimens, of which 3 had only germline mutations at >40%, 2 only had hotspot variants and 2 had both germline and hotspot variants. Upon follow-up 1-10 years later (mean of 8.5 years), 3 of these patients were diagnosed with fibroadenoma or hyperplasia of the breast (two with hotspot mutations and one with germline), two with breast cancer (both with germline and hotspot mutations), and one with a non-breast solid tumor (only germline mutation).

    Biospecimens
    Preservative Types
    • Frozen
    • Streck/BCT
    • Formalin
    Diagnoses:
    • Neoplastic - Carcinoma
    • Normal
    Platform:
    AnalyteTechnology Platform
    DNA Automated electrophoresis/Bioanalyzer
    DNA Next generation sequencing
    DNA Fluorometry
    Pre-analytical Factors:
    ClassificationPre-analytical FactorValue(s)
    Preaquisition Diagnosis/ patient condition Healthy
    Cancer (Lung or Breast)
    Next generation sequencing Specific Template/input amount 2.5-105.5 ng
    Biospecimen Aliquots and Components Aliquot size/volume 0.4-2.0 mL
    1.5-5.5 mL
    Biospecimen Acquisition Biospecimen location Plasma
    Lung tumor
    Breast tumor

You Recently Viewed  

News and Announcements

  • Most Downloaded SOPs in 2024

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

  • Just Published!

  • More...