This paper compared next-generation sequencing (NGS) metrics and identification of genetic variants in matched formalin-fixed paraffin-embedded (FFPE) and frozen specimens and investigated the effect of extraction method.
Bioinformatics and DNA-extraction strategies to reliably detect genetic variants from FFPE breast tissue samples.
Bhagwate AV, Liu Y, Winham SJ, McDonough SJ, Stallings-Mann ML, Heinzen EP, Davila JI, Vierkant RA, Hoskin TL, Frost M, Carter JM, Radisky DC, Cunningham JM, Degnim AC, Wang C
Purpose
Conclusion
Although raw coverage was comparable for FFPE and frozen specimens, the number of different molecular tags covered was lower and total number of called variants passing filter was higher in FFPE specimens than frozen specimens. This was largely due to an increase in variants that were low alternative allele frequency mutations (<5%) and the vast majority of these were unique to the FFPE specimen. The FFPE specimens also had a higher percentage of C>T variant than frozen specimens. The increase in the number of variants passing filter, the percentage of low frequency variants, percentage C>T variants, and the false discovery rate in FFPE specimens were all larger when extraction was with QIAamp than with GeneRead.
Studies (1)
Purpose
This study compared NGS metrics and identification of genetic variants in matched FFPE and frozen specimens and investigated the effect of extraction method. Four matched FFPE and frozen specimens from benign left breast specimens containing cysts, apocrine metaplasia, and proliferative changes were obtained from an archive (details of specimen processing not provided). Sections (10 µm) were taken from FFPE and frozen specimens and placed in tubes. DNA was extracted from FFPE specimens using the GeneRead DNA FFPE Kit or the QIAamp DNA FFPE Tissue Kit with an overnight lysis step. DNA was extracted from frozen sections using QIAGEN’s Gentra Puregene Kit. DNA was quantified using Qubit™ dsDNA BR Assay and the integrity verified using an Advanced Analytical Fragment Analyzer High Sensitivity Large Fragment Analysis Kit. QIAseq Human Breast Cancer Targeted Panel libraries were prepared and sequenced using an Illumina HiSeq 4000.
Summary of Findings
Raw coverage was comparable for FFPE and frozen specimens, but the number of different molecular tags covered was 2 to 4-fold higher in frozen than matched FFPE specimens. The total number of called variants passing filter was approximately five times higher in FFPE specimens extracted using the GeneRead DNA FFPE Kit than in frozen specimens (927-1,316 versus 248-291) and the number of variants passing filter in FFPE specimens was even higher when extracted using the QIAamp Kit (2,639-20,820). The FFPE specimens had a higher percentage of C>T variants than frozen specimens and this was particularly pronounced when extraction was with QIAamp (93-99% with QIAamp, 58-77% for GeneRead, and 37-43% for Frozen, P<1e-16). Importantly, the percentage variants that were low alternative allele frequency mutations (<5%) was high in FFPE specimens extracted using either method (76% for GeneRead and 94% for QIAamp). Further, most of these low frequency variants (allele frequency of <1%) were discordant between FFPE and frozen specimens with a false discovery rate of 94.8% and 69.8% for FFPE specimens extracted using QIAamp and GeneRead, respectively. Importantly, when the threshold was increased to an allele frequency greater than 5%, the FDR dropped to only 14.8% for FFPE specimens extracted using GeneRead and 17.4% for those extracted using QIAamp. The proportion of C>T transitions in FFPE specimens was much lower when only variants concordant with the frozen counterpart were considered.
Biospecimens
TissueBreastPreservation Types
FrozenFormalinDiagnoses
NeoplasticBenignPlatforms
DNANext generation sequencingPre-analytical Factors
Classification Pre-Analytical Factor Value(s) Biospecimen Preservation Type of fixation/preservation Formalin (buffered)FrozenAnalyte Extraction and Purification Analyte isolation method QIAGEN GeneRead DNA FFPE KitQIAGEN QIAamp DNA Mini KitNext generation sequencing Specific Data handling VAF threshold of <1%VAF threshold of <5%VAF threshold of >5%