The purpose of this paper was to determine the effects of formalin fixation, extraction method, storage duration, tissue type, and tumor status on DNA and RNA next generation sequencing (NGS) results using formalin-fixed, paraffin-embedded (FFPE) and frozen specimens.
Next-generation sequencing of RNA and DNA isolated from paired fresh-frozen and formalin-fixed paraffin-embedded samples of human cancer and normal tissue.
Hedegaard J, Thorsen K, Lund MK, Hein AM, Hamilton-Dutoit SJ, Vang S, Nordentoft I, Birkenkamp-Demtröder K, Kruhøffer M, Hager H, Knudsen B, Andersen CL, Sørensen KD, Pedersen JS, Ørntoft TF, Dyrskjøt L
Purpose
Conclusion
RNA and DNA yields from FFPE specimens were dependent on the extraction method, and RNA and DNA integrity were adversely affected by FFPE block storage. NGS library construction from FFPE specimens resulted in more duplication and a high failure rate for DNA but not RNA. FFPE DNA generated a lower percentage of mapped reads, more unaligned ends, more non-perfectly aligned reads, and more transition errors (T<->C and A<->G) than matched frozen specimens, and these differences increased with block storage. For FFPE RNA, a higher percentage of non-perfect matches and a lower percentage of total exon mapping were observed than with the frozen specimens, but the percentage of non-mapped reads was comparable. While the expression data between matched FFPE and frozen specimens was strongly correlated, 1494 genes were found to be differentially expressed between frozen and FFPE specimens in all three tissue types. Some of these genes overlapped with genes that were differentially expressed between normal adjacent and tumor specimens, but the specimens still clustered based on diagnosis.
Studies (4)
Purpose
The purpose of this study was to determine the effect of storage duration and extraction kit on the yield and integrity of DNA and RNA from FFPE specimens. The effect of extraction kit was tested using 16 specimens including normal liver and skin, reactive tonsil and liver, lung, breast and bladder carcinomas.
Summary of Findings
RNA yield and integrity were highest when extraction was performed using the miRNeasy FFPE or ExpressArt kits rather than the Nucleospin kit, and DNA yields were highest using the QIAamp FFPE DNA kit rather than the Nucleospin kit. Co-purification of RNA and DNA using either QIAsymphony or RecoverAll resulted in low DNA and RNA yields. RNA and DNA integrity were adversely affected by FFPE block storage duration, but RNA fragments of up to 4000 bp could still be isolated from blocks stored for up to 13 months.
Biospecimens
TissueBladderTissueTonsilTissueLiverTissueSkinTissueBreastTissueLungPreservation Types
FrozenFormalinDiagnoses
NeoplasticCarcinomaOther diagnosesNormalPlatforms
DNAFluorometryDNASpectrophotometryRNAAutomated electrophoresis/BioanalyzerRNASpectrophotometryPre-analytical Factors
Classification Pre-Analytical Factor Value(s) Storage Storage duration 13 months2 months241 months63 monthsAnalyte Extraction and Purification Analyte isolation method ExpressArt FFPE RNA readyNucleospin FFPE DNA kitNucleospin FFPE RNAQIAamp DNA FFPE TissueQIAsymphony DSP DNA mini kitQiasymphony DSP RNA kitRecoverAll Total Nucleic Acid Isolation Kit for FFPEmiRNeasy FFPE
Purpose
The purpose of this study was to determine the effects of storage duration, preservation method, and library construction optimization on DNA library creation and DNA NGS results. Construction of the library was tested using 19 paired colorectal carcinoma specimens stored 2-13 years and 43 FFPE specimens including prostate, colon, bladder and liver carcinomas, healthy skin, normal liver and reactive tonsil specimens.
Summary of Findings
NGS libraries were only successfully made from 18 of 61 FFPE specimens, with insufficient yields most often occurring when the DNA was degraded. The efficiency of NGS library construction was lower for DNA from freshly made FFPE blocks (stored <3 months) than from frozen specimens and declined further with FFPE block storage. The authors state the decreased efficiency occurs during the amplification step and may be attributed to formalin modifications of the DNA and can be partially compensated for by increasing the number of amplification cycles. Importantly, the PCR duplication rate was higher when FFPE specimens were used than when frozen specimens were used (60-85% versus 30%) resulting in fewer unique reads, unless half as many specimens were used for each lane (3 specimens/2 lanes instead of 3 specimens/1 lane). Increasing the input DNA at any stage or changing the polymerase to a uracil friendly polymerase had no effect on the duplication rate. FFPE specimens generated a lower percentage of mapped reads, more unaligned ends, more non-perfectly aligned reads, and more transition errors (T<->C and A<->G) than matched frozen specimens, but these differences were small in specimens stored for <3 years and increased with block storage. Importantly, for specimens stored <3 years, 70-80% of variants were found both in the frozen and FFPE specimen, but when stored longer, the number of variants found just in FFPE specimens increased while the percentage found just in frozen specimens remained constant. 8 of the 11 FFPE specimens with known KRAS/BRAF mutations were sequenced, and all the expected mutations were found. Two additional KRAS hotspot mutations, not found by Sanger sequencing, were identified at a frequency of <2%, indicating higher sensitivity.
Biospecimens
TissueBladderTissueProstateTissueColorectalTissueLiverTissueTonsilTissueSkinPreservation Types
FormalinFrozenDiagnoses
NeoplasticCarcinomaNormalOther diagnosesPlatforms
DNANext generation sequencingPre-analytical Factors
Classification Pre-Analytical Factor Value(s) Biospecimen Preservation Type of fixation/preservation Formalin (buffered)FrozenStorage Storage duration 13-15 months2-3 months241-244 months60-62 monthsNext generation sequencing Specific Nucleic acid amplification KAPA HiFi Uracil+ DNA PolymeraseTruSeq DNA library Taq PolymeraseNext generation sequencing Specific Template/input amount 1 library1 µg2 libraries2 µg500 ng
Purpose
The purpose of this study was to determine the effects of storage duration, preservation method, and library construction optimization on RNA library creation and RNA NGS results. Construction of the library was tested using 19 paired colorectal carcinoma specimens stored 2-13 years and 43 FFPE specimens including prostate, colon, bladder and liver carcinomas, healthy skin, normal liver and reactive tonsil specimens. rRNA was removed using Ribo-Zero, which, the authors state, did not work consistently between libraries and tissues.
Summary of Findings
RNA NGS libraries were successfully created from all FFPE specimens, regardless of block storage duration, but the insert sizes were smaller for FFPE specimens and further decreased with increased FFPE block storage. The authors attempted to chemically degrade RNA to get consistent insert sizes, but they were not able to achieve this goal and suggest omitting the fragmentation step for RNA degraded below 200 nt. Treatment of the RNA with DNAse and Exonuclease I did not affect the distribution or strand specificity of NGS data. For both FFPE and frozen specimens, 61% were mapped to the transcriptome in the forward direction, 8% to the transcriptome in the reverse direction, 10% to rRNA (forward or backward) and 21% were unmapped. For FFPE specimens, higher levels of duplication and percentages of non-perfect matches and a lower percentage of total exon mapping were observed than with the frozen specimens, but the percentages of non-mapped reads were comparable between preservation methods.
Biospecimens
TissueBladderTissueProstateTissueSkinTissueTonsilTissueLiverTissueColorectalPreservation Types
FormalinFrozenDiagnoses
NeoplasticCarcinomaNormalOther diagnosesPlatforms
RNANext generation sequencingPre-analytical Factors
Classification Pre-Analytical Factor Value(s) Biospecimen Preservation Type of fixation/preservation Formalin (buffered)FrozenStorage Storage duration 13-15 months2-3 months241-244 months60-62 monthsAnalyte Extraction and Purification Nucleic acid digestion DNAse and Exonuclease I treatedNo DNase treatmentNext generation sequencing Specific Nucleic acid amplification KAPA HiFi Uracil+ DNA PolymeraseTruSeq DNA library Taq PolymeraseNext generation sequencing Specific Template modification FragmentedNot fragmented
Purpose
The purpose of this study was to determine the effects of preservation method, tissue type and tumor status on NGS RNA expression analysis. This study included 12 matched FFPE and frozen colon tumor specimens with 6 matched normal adjacent specimens stored 2-13 years, 8 matched FFPE and frozen bladder tumor specimens stored 5-9 years, 10 frozen bladder carcinoma tumors, and 7 matched FFPE and frozen prostate tumor specimens stored 2-11 years.
Summary of Findings
The expression data in matched FFPE and frozen specimens was strongly to very strongly correlated (0.90±0.05), with no obvious effect of storage. A total of 7078, 2984 and 6333 genes were found to be differentially expressed (false discovery rate <0.05) in colon, bladder and prostate specimens, respectively, with slightly more genes showing reduced expression in FFPE when compared to frozen tissue. A total of 1494 genes were found to be differentially expressed between frozen and FFPE specimens in all three tissue types. Of these, 1054 (71%) were reduced in FFPE versus frozen specimens in all three tissue types. 1207 and 471 genes were found to be differentially expressed between colorectal tumor and normal adjacent specimens in FFPE and frozen specimens, respectively. Of these, 66/1207 (FFPE) and 28/471 (frozen) were also found to be affected by formalin fixation. Expression of 12 gene signature for bladder carcinoma was highly correlated for some matched FFPE and frozen specimens, but for others, it was not correlated which the authors attribute to tumor heterogeneity.
Biospecimens
TissueBladderTissueProstateTissueColorectalPreservation Types
FrozenFormalinDiagnoses
NeoplasticCarcinomaNeoplasticNormal AdjacentPlatforms
RNANext generation sequencingPre-analytical Factors
Classification Pre-Analytical Factor Value(s) Preaquisition Diagnosis/ patient condition CancerousNormal adjacentBiospecimen Acquisition Biospecimen location BladderColonProstateBiospecimen Aliquots and Components Biospecimen heterogeneity Intratumoral sampling (exact positions not specified)Biospecimen Preservation Type of fixation/preservation Formalin (buffered)FrozenStorage Storage duration 2-13 years