Comparison of five cell-free DNA isolation methods to detect the EGFR T790M mutation in plasma samples of patients with lung cancer.
Author(s): Solassol J, Quantin X, Larrieux M, Senal R, Audran P, Grand D, Mangé A, Diamandis EP, Vendrell JA
Publication: Clin Chem Lab Med, 2018, Vol. , Page
PubMed ID: 29626410 PubMed Review Paper? No
Purpose of Paper
This paper compared the effects of cell-free DNA (cfDNA) extraction method on the yield of cell-free DNA from plasma pools measured by four different quantification methods, on the detection of mono- and di-nucleosome bound cfDNA peaks, and on mutation detection.
Conclusion of Paper
cfDNA yields were dependent on the extraction kit used and quantification method, but regardless of extraction method no genomic DNA contamination was observed. By fluorometry, the highest concentrations were found using Norgen or Qiagen kits for extraction, but by real-time PCR the highest concentrations were with the Norgen or Zymo Research kits and the lowest with the Qiagen kit. Quantified yields by TapeStation were much higher than by other methods for two pools, which the authors state makes it an inaccurate assay for cfDNA quantification. The authors hypothesize the differences between quantification methods for some kits but not others is reflective of the amount of PCR inhibition. TapeStation peaks corresponding to mono-nucleosome bound cfDNA were observed in all plasma pools regardless of extraction method, but the di-nucleosome –bound cfDNA peak was found in all 6 pools only when extraction was with the Zymo Research kit. While extraction method had no effect on detection of mutations in the pool with a high mutational burden, in the two pools with low mutational burdens mutations, were only properly detected when the Norgen or Zymo kits were used. Using the Bioo Scientific kit, mutations were found in two pools where the mutations were not expected based on matched tissue specimens indicating potential false positives.
Studies
-
Study Purpose
This study compared the effects of cell-free DNA extraction method on the yield of cf DNA from plasma pools measured by four different quantification methods, on the detection of mono- and di-nucleosome bound cfDNA peaks, and on mutation detection. Six pooled plasma specimens were produced by pooling the thawed plasma of 24 adenocarcinoma patients such that each pool represented the plasma of 4 patients chosen based on mutation status. No details of plasma processing or storage were provided. cfDNA was obtained from 1 mL of plasma using the Maxwell RCS ccfDNA Plasma Kit, Norgen Plasma/Serum Cell-Free Circulating DNA purification kit, Zymo Research Quick-cfDNA Serum and Plasma kit, Qiagen QIAamp circulating nucleic acid kit, or the Bioo Scientific NextPrep-Mag cfDNA Isolation kit following the manufacturers’ protocols. Cell-free DNA concentrations were determined using the Qubit dsDNA HS assay kit, by real-time PCR amplification of a 70 bp amplicon, by TapeStation, or by droplet digital PCR (ddPCR).
Summary of Findings:
While no genomic DNA contamination was observed for any of the extraction methods, cfDNA yields varied among extraction kits and quantification methods. When concentrations were assessed by Qubit fluorometry, higher concentrations were found using Norgen (median of 0.49 ng/µL) or Qiagen (median of 0.39 ng/µL) kits for extraction than the other kits (P<0.001, all). As determined by amplification of a 70-bp amplicon, concentrations of cfDNA were highest when extraction was with the Norgen (median of 0.32 ng/µL) or Zymo Research kits (median of 0.31 ng/µL) and lowest when extracted using the Qiagen kit (median of 0.23 ng/µL). Quantified yields measured by TapeStation were much higher than those quantified by other methods for two of the plasma pools, which the authors indicate makes it an inaccurate assay. Importantly, there were no significant differences between methods in quantified cfDNA concentration in specimens extracted using the Maxwell and Zymo Research kits. Consequently, the authors hypothesize that these kits lead to reduced PCR inhibition. Using TapeStation, the peak corresponding to mono-nucleosome bound cfDNA was observed in all plasma pools regardless of extraction method. In contrast, di-nucleosome–bound cfDNA was found in all six pools when extraction was with the Zymo Research kit, in five pools extracted using the Norgen kit, three pools extracted using the Maxwell kit, two extracted using the Bioo scientific kit and only one of the six pools extracted using the Qiagen kit. The expected epidermal growth factor receptor (EGFR) T790M and Del19 mutations were detected in the plasma pool with the highest mutational load, regardless of the cfDNA extraction method, but detection was only possible in the two pools with low levels of the T790M mutation when cfDNA was extracted using the Norgen or Zymo kits. Using the Bioo Scientific kit, mutations were found in two pools where the mutations were not expected based on matched tissue specimens, indicating potential false positives.
Biospecimens
Preservative Types
- Frozen
Diagnoses:
- Neoplastic - Carcinoma
Platform:
Analyte Technology Platform DNA Real-time qPCR DNA Automated electrophoresis/Bioanalyzer DNA Fluorometry DNA Digital PCR Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Analyte Extraction and Purification Analyte isolation method Maxwell RCS ccfDNA Plasma Kit
Norgen Plasma/Serum Cell-Free Circulating DNA purification kit
Zymo Research Quick-cfDNA Serum and Plasma kit
Bioo Scientific NextPrep-Mag cfDNA Isolation kit
Qiagen QIAamp circulating nucleic acid kit
Digital PCR Specific Targeted nucleic acid EGFR T790M
EGFR Del19
Fluorometry Specific Technology platform Real-time PCR
ddPCR
TapeStation
Qubit