Fresh-frozen tissue is the “gold standard” biospecimen type for next-generation sequencing (NGS). However, collecting frozen tissue is usually not feasible because clinical workflows deliver formalin-fixed, paraffin-embedded (FFPE) tissue blocks. Some clinicians and researchers are reticent to embrace the use of FFPE tissue for NGS because FFPE tissue can yield low quantities of degraded DNA, containing formalin-induced mutations. We describe the process by which formalin-induced deamination can lead to artifactual cytosine (C) to thymine (T) and guanine (G) to adenine (A) (C:G > T:A) mutation calls and perform a literature review of 17 publications that compare NGS data from patient-matched fresh-frozen and FFPE tissue blocks. We conclude that although it is indeed true that sequencing data from FFPE tissue can be poorer than those from frozen tissue, any differences occur at an inconsequential magnitude, and FFPE biospecimens can be used in genomic medicine with confidence:

DNA extracted from formalin-fixed, paraffin-embedded tissue sections is often inadequate for sequencing, due to poor yield or degradation. We optimized the proteinase K digest by testing increased volume of enzyme and increased digest length from the manufacturer’s protocol using 54 biospecimens, performing the digest in centrifuge tubes. Doubling the quantity of proteinase K resulted in a median increase in yield of 96%. Applying the optimized proteinase K protocol to sections deparaffinized on microscope slides generated a further increase in yield of 41%, but only at >50,000 epithelial tumor cells/section. DNA yield now correlated with (χ² = 0.84) and could be predicted from the epithelial tumor cell number. DNA integrity was assayed using end point multiplex PCR (amplicons of 100–400 bp visualized on a gel), quantitative PCR (qPCR; Illumina FFPE QC Assay), and nanoelectrophoresis (DNA Integrity Numbers [DINs]). Generally, increases in yield were accompanied by increases in integrity, but sometimes qPCR and DIN results were conflicting. Amplicons of 400 bp were almost universally obtained. The process of optimization enabled us to reduce the percentage of samples that failed published quality control thresholds for determining amenability to whole genome sequencing from 33% to 7%.

The optional RNase digest that is part of many DNA extraction protocols is often omitted, either because RNase is not provided in the kit or because users do not want to risk contaminating their laboratory. Consequently, co-eluting RNA can become a “contaminant” of unknown magnitude in a DNA extraction. We extracted DNA from liver, lung, kidney, and heart tissues and established that 28–52% of the “DNA” as assessed by spectrophotometry is actually RNA (depending on tissue type). Including an RNase digest in the extraction protocol reduced 260:280 purity ratios. Co-eluting RNA drives an overestimation of DNA yield when quantification is carried out using OD 260 nm spectrophotometry, or becomes an unquantified contaminant when spectrofluorometry is used for DNA quantification. This situation is potentially incompatible with the best practice guidelines for biobanks issued by organizations such as the International Society for Biological and Environmental Repositories, which state that biospecimens should be accurately characterized in terms of their identity, purity, concentration, and integrity. Consequently, we conclude that an RNase digest must be included in DNA extractions if pure DNA is required. We also discuss the implications of unquantified RNA contamination in DNA samples in the context of laboratory accreditation schemes.

To evaluate the stability of RNA and microRNA (miRNA) in PAXgene-fixed paraffin-embedded tissue blocks after 7 years’ storage. RNA and miRNA were extracted from PAXgene-fixed paraffin-embedded (PFPE) blocks in 2009 then stored at −80°C. Seven years later, RNA and miRNA were again extracted from the same blocks. RNA and miRNA integrity in the 2009 and 2016 extractions were compared using RNA integrity number (RIN), paraffin-embedded RNA metric (PERM), reverse transcription polymerase chain reaction (RT-PCR) for different amplicon lengths, and quantitative RT-PCR (qRT-PCR) for three mRNA and three miRNA targets. In PFPE blocks, mRNA was poorer in 2016 extractions compared to the 2009 extractions in all blocks and all assays applied, with transcripts degrading at different rates in the same blocks. For miRNA, qRT-PCR showed no statistically significant differences between 2009 and 2016 extractions. mRNA in PFPE tissue blocks degrades at room temperature storage over 7 years.

Although there are thousands of formalin-fixed paraffin-embedded (FFPE) tissue blocks potentially available for scientific research, many are of questionable quality, partly due to unknown preanalytical variables. We analyzed FFPE tissue biospecimens as part of the National Cancer Institute (NCI) Biospecimen Preanalytical Variables program to identify mRNA markers denoting cold ischemic time. The mRNA was extracted from colon, kidney, and ovary cancer FFPE blocks (40 patients, 10–12 hr fixation time) with 1, 2, 3, and 12 hr cold ischemic times, then analyzed using qRT-PCR for 23 genes selected following a literature search. No genes tested could determine short ischemic times (1–3 hr). However, a combination of three unstable genes normalized to a more stable gene could generate a “Cold Ischemia Score” that could distinguish 1 to 3 hr cold ischemia from 12 hr cold ischemia with 62% sensitivity and 84% specificity.