Clinical Lab Directors

Nucleic acids from FFPE specimens are highly fragmented, difficult to recover and may be chemically altered

High quality nucleic acids are vital to obtaining accurate, reproducible test results. But the standard formalin fixation and paraffin embedding (FFPE) specimen is well documented to be a source of lower quality, fragmented DNA with chemically-induced artifactual changes. (1)

Today, nucleic acids submitted for molecular testing (especially for the increasingly common Next Generation Sequencing, or NGS) are sourced from solid tissue specimens that have undergone FFPE. That preparation results in DNA fragments which are well documented to be harder to recover, highly degraded and altered at the molecular level.

Recover Cells Not Exposed to Formalin

Why would you want to run your molecular testing or NGS on suboptimal nucleic acids? Instead, you can now perform NGS on DNA and RNA from fresh cells untouched by formalin, if your hospital uses the Crow’s Nest™ Biopsy Catchment System (hereafter referred to as the Crow’s Nest). The Crow’s Nest does not objectively expose the secondary specimen recovered, the dislodged tumor cells, to formalin, resulting in a far superior specimen for molecular testing than that recovered by current approaches that try to remediate nucleic acid quality from FFPE specimens.

If you are a clinical lab director interested in finding out if the hospital or hospital systems you serve could implement a Crow’s Nest program and thereby provide you with fresh, high-quality starting material for molecular testing, read on.

Measuring quality and quantities of available DNA with this new method

Preliminary data using a prequel version of the Crow’s Nest yielded promising findings regarding the quality and quantities of available DNA captured from Dislodged Tumor Cells (DDTCs).

DDTCs were collected in an interventional radiology (IR) suite during scheduled core needle biopsy procedures. The specimens were transported to a pathology laboratory wherein the DDTCs and tissue core were separated. The parent tissue core underwent routine FFPE.

Once the DDTC specimens were collected, they were transferred to a laboratory core facility where the DNA was extracted using a commercially available kit. Quantitation was evaluated using a Quibit fluorometer and qualitative metrics assessed by molecular weight (MW) on an Agilent Tape Station.

If you are a clinical lab director interested in finding out if the hospital or hospital systems you serve could implement a Crow’s Nest program and thereby provide you with fresh, high-quality starting material for molecular testing, read on.

Genomic sequencing is of course more informative with longer DNA chains as a substrate

The DDTC specimens were then run through an illumina next-generation sequencer (NGS), and additional qualitative features (Phred score and Per Base N content [PBNC]) obtained. High MW DNA (between 15 to 18 kilobase-pairs) was recovered in each case.

In contrast, traditionally-processed FFPE specimens have recovered DNA fragments typically <500 base-pairs (bps) in size. Crow’s Nest prequel results show nucleic acid chains 30-36x larger than is expected in FFPE specimens with size limited by use of an extraction protocol not designed for ultra-high MW DNA recovery.

The Genomic Quality Number (GQN), which is the calculated percentage of the total DNA in a sample that is above a set threshold (e.g., typically 500 bps for FFPE specimens) and reported in a range from 0 (worst) to 10 (best), was either 9.9 or 10 in the samples tested. This compares favorably to the wide variation in values from 2.5 to 7 that have been reported in FFPE specimens.

Sequencing metrics have provided confirmatory data 

The Phred score, the numeric value reflective of the quality and confidence of the nucleic acids generated during NGS, were consistently ≥35, indicating a base calling accuracy of 99.97%. Additionally, the Phred score was ≥35 throughout the entire sequence read, in contrast to the dip towards the tail end of the sequence that is normally observed. 

The PBNC metric that measures the ability of a sequencer to designate a base as either adenine, guanine, cytosine, or thymidine was consistently <<1% (Figure 4C – FastQC per Base N Content).

All together, these findings support the Crow’s Nest premise that recovering DNA before and without exposure to formalin yields data with high-quality starting material for molecular studies.

Confirmation that the recovered DDTCs are diagnostic and neoplastic in nature comes from the finding of Tier 1 (TSC2 gene mutations) and Tier 2 (p53 gene mutations) alterations in the DNA of the preliminary data samples

Microtrauma’s built-in bias toward friable tumor cells

This finding supports the Crow’s Nest principal underlying hypothesis: Microtrauma from tumor tissue biopsies has a built-in bias that more tumor cells than normal cells will become dislodged because of tumor cells’ “friable” or invasive nature within the host tissue.

Use the Crow’s Nest in your hospital. The device takes advantage of one of the underlying gross observations in pathology laboratories about tumor specimens -- that because of their expansile and destructive nature as they invade the surrounding tissue, they are less adherent to the surrounding tissue or friable. The absence of the complex fibroconnective tissue framework that keeps normal or non-malignant cells in place and in highly regimented arrangements within tissues or organs, is not present in tumors. Hence, the microtrauma associated with biopsy procedures will tend to provide a higher percentage of already loosened or friable tumor cells in a sample if equal parts tumor and normal are sampled. This means that your lab can benefit from improved small biopsy specimen processing and increase utilization in high throughput molecular testing platforms (such as Thermo Fisher’s Genexus platform or an Illumina next-generation sequencer (NGS)). Contact us to discuss how.


1) Do H, Dobrovic A. Sequence artifacts in DNA from formalin-fixed tissues: causes and strategies for minimization. Clin Chem 2015;61(1):64-71.