Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

Comparison of clinical targeted next-generation sequence data from formalin-fixed and fresh-frozen tissue specimens.

Author(s): Spencer DH, Sehn JK, Abel HJ, Watson MA, Pfeifer JD, Duncavage EJ

Publication: J Mol Diagn, 2013, Vol. 15, Page 623-33

PubMed ID: 23810758 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   The purpose of this study was to compare NGS success and data for 27 cancer-related genes between FFPE and frozen specimens and to compare NGS and microarray results. DNA was extracted from 16 matched snap-frozen OCT embedded and deparaffinized FFPE specimens (fixed overnight) using the QIAamp Micro DNA kit. All specimens had been stored for 7-12 years. Libraries were constructed using equal amounts of DNA and pooling for sequencing across 3 lanes.

   Summary of Findings:

   The authors report that FFPE specimens yielded at least 1 µg of DNA (quantified by Qubit) and PCR products of at least 200 bp. Significantly more NGS reads were generated with FFPE than frozen specimens (26.8 million versus 17.9 million, p=0.04), but compared with frozen specimens, FFPE specimens generated a lower percentage of mapped reads (98.9% versus 99.3%, p=0.001) and properly mapped reads (97.7% versus 98.6%, p=0.03), more unaligned bases per read (4.6 versus 2.9, p=0.02), a lower average insert size (177 versus 222, p=0.000001) and a higher coefficient of variance in coverage (53% versus 46%, p=0.002). While FFPE specimens had slightly more discrepancies per read than frozen specimens (0.29 versus 0.27), the difference was not significant. The percentages of mapped on-target reads and unique reads, overall coverage, and mean coverage across exon sequences for each target were not affected by preservation method. When a coverage depth of >50 was applied, of the 4407 variants identified, only 56 were unique to the frozen specimen and 70 to the FFPE specimen. 64 of these discordant variants were found in >10 reads in both specimen types, indicating the discrepancy is due to heterogeneity. Importantly, transition mutations were responsible for a higher percentage of discrepancies in FFPE specimens than frozen specimens (39% versus 31%, p=0.00000064). An average of 54 insertions/deletions were identified per subject and all were concordant between matched specimens. Importantly, 98.0% and 98.9% of variant calls made by NGS were concordant with those identified by microarray using FFPE and frozen specimens, respectively, with discordance arising from a homozygous call on microarray and a low NGS frequency (10-20%).

   Biospecimens

   • Tissue - Lung

   Preservative Types

   • OCT
   • Formalin

   Diagnoses:

   • Neoplastic - Carcinoma

   Platform:

   Analyte	Technology Platform
   DNA	Next generation sequencing
   DNA	Fluorometry
   DNA	SNP assay
   DNA	PCR
   DNA	DNA microarray

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Biospecimen Preservation	Type of fixation/preservation	Formalin (buffered) OCT
   Next generation sequencing Specific	Technology platform	Micrarray
   PCR Specific	Length of gene fragment	105 bp 239 bp 299 bp 411 bp
   PCR Specific	Targeted nucleic acid	GAPDH

   View more

2. Study Purpose

   The purpose of this study was to determine the effects of cold ischemia and fixation duration on mutation detection by NGS and to compare results with those generated using a matched snap frozen, OCT embedded specimen. A single uterine adenomyosis was divided into pieces which were frozen immediately in OCT, subjected to up to 48 h cold ischemia in room temperature saline before formalin-fixation, or immediately fixed in formalin for up to 72 h. DNA was extracted using the QIAamp Micro DNA kit. 12 million reads were selected at random for comparisons.

   Summary of Findings:

   The authors state that room temperature ischemia of ≥24 h or fixation for ≥48 h resulted in loss of PCR products ≥239 bp. Fixation for ≥48 h or 48 h of room temperature cold ischemia did not affect the number of mapped reads and slightly increased on-target reads, but decreased the number of unique reads and the number of specimens meeting coverage levels for targeted positions. Importantly, variant detection decreased by 5-6% with fixation for 72 h or 48 h of room temperature cold ischemia. This loss was particularly evident when coverage was <50X. Using the frozen specimen as standard, the false discovery rate after fixation for ≥48 h or 48 h of room temperature cold ischemia was around 1%. C to T transitions increased with increasing fixation duration.

   Biospecimens

   • Tissue - Uterus

   Preservative Types

   • OCT
   • Formalin

   Diagnoses:

   • Other diagnoses

   Platform:

   Analyte	Technology Platform
   DNA	PCR
   DNA	Next generation sequencing

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Biospecimen Preservation	Type of fixation/preservation	Formalin (buffered) OCT
   Biospecimen Acquisition	Cold ischemia time	0 h 24 h 48 h
   PCR Specific	Length of gene fragment	105 bp 239 bp 299 bp 411 bp
   PCR Specific	Targeted nucleic acid	GAPDH
   Biospecimen Preservation	Time in fixative	0 h 24 h 48 h 72 h

   View more