Virtual Library
Start Your Search
H.T. DeFedericis
Author of
-
+
P3.04 - Poster Session/ Biology, Pathology, and Molecular Testing (ID 235)
- Event: WCLC 2015
- Type: Poster
- Track: Biology, Pathology, and Molecular Testing
- Presentations: 2
- Moderators:
- Coordinates: 9/09/2015, 09:30 - 17:00, Exhibit Hall (Hall B+C)
-
+
P3.04-025 - DNA Extraction of Lung Cancer Samples for Advanced Diagnostic Testing (ID 3205)
09:30 - 09:30 | Author(s): H.T. DeFedericis
- Abstract
Background:
Tumor specimens are routinely formalin fixed and paraffin embedded (FFPE) prior to histologic evaluation. This process preserves the morphology and cellular features required for proper staining and microscopic review. However, this practice presents numerous challenges for the extraction of high quality DNA for advanced diagnostic testing that include nanoString and Next-Generation Sequencing (NGS) technologies. An extraction process that consistently produces sufficient DNA yield and fragment size from these difficult but most precious tissue samples is a requirement for any Molecular Pathology laboratory utilizing these platforms. The data presented here will compare the quantity and quality of DNA extracted using two methods, QIAGEN and Covaris, and success of downstream testing.
Methods:
FFPE tumor samples from a variety of tumor types, including lung, were macro-dissected using 14-guage needles, with 1 core extracted using the Covaris truXtract FFPE DNA isolation method and the other matched core using the QIAGEN DNeasy tissue kit. All samples were processed using manufacturer’s recommended instructions. DNA metrics were measured using Qubit (picogreen) and NanoDrop for yield and purity, followed by fragment size estimation on a 2100 BioAnalyzer (Agilent Technologies). A subset of matched DNA sample pairs were used as template for PGM AmpliSeq and MiSeq TSCA library preparation, followed by NGS. A subset of DNA sample pairs were also analyzed for copy number using the nanoString nCounter system.
Results:
DNA yields and fragment lengths were substantially higher for truXtract samples as compared to DNeasy when measured by picogreen quantitation and Bioanalyzer electrophoresis (Figure 1). A higher degree of successful advanced molecular diagnostic test results was also observed for the truXtract DNA samples, especially for the Illumina NGS system (improved clustering and coverage) and nCounter platform (improved counts) that prefer longer fragment lengths than Ion Torrent NGS. Figure 1Figure 1: 2100 Bioanalyzer traces of DNA prepared from three lung cancer FFPE samples; DNeasy (left) and TruXtract (right).
Conclusion:
FFPE tumor samples prepared using the truXtract FFPE DNA isolation kit provides an efficient system for generating high quality DNA samples from even the most difficult lung cancer specimens. The combination of improved yield and fragment size measured for nearly every sample tested suggests that even smaller biopsies can now be collected for advanced diagnostic testing.
-
+
P3.04-057 - Pyrosequencing VS NGS KRAS and EGFR Mutation Detection: A Head to Head Comparison in Lung Adenocarcinomas (ID 2712)
09:30 - 09:30 | Author(s): H.T. DeFedericis
- Abstract
Background:
Pyrosequencing is a popular method for detecting actionable somatic mutations. Most labs use pyrosequencing at an analytical sensitivity of 10%, potentially missing actionable mutations that have a low variant allele frequency (VAF) due to low neoplastic nuclear content or due to neoplastic heterogeneity. Furthermore, the cost-effectiveness of pyrosequencing rapidly decreases when numerous hotspots are interrogated simultaneously and scaleability is limited. Next-generation sequencing (NGS) is scaleable and has the capacity to detect mutations at VAFs less than 10%. The goals of this study were to perform NGS on a series of KRAS and EGFR cases that were “mutation negative” but had suspicious pyrosequencing peaks which were insufficient for a definite determination, and to review EGFR exon 19 deletion cases that were detected by NGS but missed by pyrosequencing.
Methods:
All the KRAS and EGFR pyrosequencing runs performed at Roswell Park Cancer Institute between July 2011 and November 2014 were manually reviewed. All actionable KRAS and EGFR variants that were found at a VAF of 4% or more and less than 10% and had remnant DNA were tested by a dual MiSeq/PGM platform NGS pipeline with a 3.6% VAF analytic sensitivity for FFPE tissues. We also included EGFR exon 19 cases that had discrepant findings between NGS and pyrosequencing.
Results:
Six lung adenocarcinomas with suspicious KRAS pyrograms were reviewed. By NGS, 4/6 were found to have activating codon 12 and 13 KRAS mutations (NGS VAF range 6-14%). Twelve lung adenocarcinomas with suspicious EGFR pyrograms or discrepant EGFR exon 19 pyrosequencing/NGS results were reviewed. By NGS, 4/12 were found to have actionable mutations, including 3 exon 19 deletions (NGS VAF range 5-24%) and 2 T790M resistance mutations (NGS VAF range 4-5%).
Conclusion:
Pyrosequencing lacks the analytic sensitivity to detect actionable KRAS and EGFR mutations with very low VAF and can entirely miss EGFR exon 19 deletions, even at a high VAF. NGS and can be optimized to detect single nucleotide alterations with a VAF less than 5% and can reliably detect EGFR exon 19 deletions. The capabilities of NGS can translate into improved clinical validity and clinical utility.
