Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

Detection of EGFR mutations in non-small cell lung cancer by droplet digital PCR.

Author(s): Williamson DFK, Marris SRN, Rojas-Rudilla V, Bruce JL, Paweletz CP, Oxnard GR, Sholl LM, Dong F

Publication: PLoS One, 2022, Vol. 17, Page e0264201

PubMed ID: 35202431 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   The purpose of this study was to evaluate the sensitivity of a plasma ddPCR assay for the detection of EGFR mutations from lung cancer patients by comparing to next generation sequencing results using FFPE tumor specimens from the same patients. The authors also investigated if the sensitivity differed among patients with different stage cancer or undergoing treatment. The authors also characterized the linearity of the assay and limit of detection using commercial DNA. This study utilized case-matched plasma specimens and FFPE tumor specimens from 183 patients with lung cancer.  No further details of specimen procurement and processing were provided. DNA was extracted from plasma using the QIAamp Circulating Nucleic Acid Kit and mutational status was determined by ddPCR. DNA was extracted from tumor enriched areas of ten 4 µm-thick sections of each FFPE specimen using the QIAamp DNA mini kit and mutational status was evaluated using the OncoPanel NGS assay. To investigate the concordance between the two assay types, DNA from twenty FFPE specimens was also assayed using the ddPCR assay. The linearity of the EGFR ddPCR assays was investigated using commercial DNA with a EGFR L858R, exon 19 deletion and wild type EGFR DNA at different total DNA concentrations and VAF.

   Summary of Findings:

   ddPCR measured DNA concentration was linear with DNA input for both the L858R and exon 19 deletion assays. As expected, as DNA concentration increased the number of mutations detected per droplet increased in a linear fashion when the VAF was low (≤10%); however, the proportion of mutations detected decreased when concentration exceeded the optimal loading concentration. While 5 of 10 reactions using wild-type specimens had a single false positive event for L858R, none had more than one false positive event, and the per droplet false positive rate was 0.003% for the L858R mutation. Based on this, the authors decided on a threshold of ≥3 events in duplicate reactions for L858R mutation positivity. The limit of detection for xxx was a VAF of 0.01% with 250 ng input, 0.1% with 125 ng or 62.5ng input, and 1% with 31.25 or 15.625 ng input. Importantly, VAF was very strongly correlated with DNA input (R2=0.986) when VAF was above the established limit of detection. No false positives were identified for the EGFR exon 19 deletion.  The limit of detection for the EGFR exon 19 deletion was a VAF of 0.1% with 62.5-250 ng input and 1% with 31.25-15.625 ng input. The VAF was very strongly correlated with DNA input when VAF was above the established limit of detection (R2=0.998).

    

   EGFR mutational status was 100% concordant between the two assays for the 20 tumor specimens analyzed  by both NGS and ddPCR. Of the 183 patients for whom a plasma and tumor specimen were available, 92 had an EGFR mutation identified by NGS in the tumor specimen, of these, only 49 had a mutation identified by ddPCR in the plasma specimen. However, four additional plasma specimens tested positive for the mutation but below the threshold of detection. No mutations were identified in the plasma specimen but not in the tissue specimen. The overall clinical sensitivity of the ddPCR assay in plasma for the EGFR mutational status of the tissue specimen by NGS was 53%. The sensitivity was 59% for patients with stage IV disease, but only 17% for patients with stage I or II cancer. The sensitivity was also higher in patients that were not undergoing treatment (71%), compared to those receiving tyrosine kinase inhibitors (44%) or any systemic therapy (including tyrosine kinase inhibitors) (37%). The authors also noted that the ddPCR assay had a faster turnaround time then the NGS assay.

   Biospecimens

   • Tissue - Lung
   • Bodily Fluid - Plasma

   Preservative Types

   • Formalin
   • None (Fresh)

   Diagnoses:

   • Neoplastic - Carcinoma

   Platform:

   Analyte	Technology Platform
   DNA	Next generation sequencing
   DNA	Digital PCR

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Preaquisition	Prognostic factor	Stage I-II Stage IV
   Preaquisition	Other drugs	Not undergoing treatment Tyrosine kinase inhibitors Systemic therapy including tyrosine kinase inhibitors
   Digital PCR Specific	Targeted nucleic acid	EGFR L858R EGFR Exon 19 del EGFR wildtype
   Digital PCR Specific	Template/input amount	250 ng 125 ng 62.5 ng 31.25 ng 15.625 ng
   Digital PCR Specific	Technology platform	NGS
   Biospecimen Acquisition	Biospecimen location	Lung tumor Plasma

   View more