Using NGS panels to sequence FFPE cancer tumor tissue
Next generation DNA sequencing (AKA next gen sequencing, or NGS) is an enormously powerful tool for cancer research, with numerous companies such as Illumina, ArcherDx, Pillar BioSciences, and Accuragen developing specific kits to identify DNA mutations in specific regions, known as cancer hotspots, in the human genome. These NGS cancer gene panels can be used for diagnostic purposes, or to profile cancer tumors to inform treatment options, or simply to aid in cancer research.
One common source of material for NGS with cancer panels is formalin fixed, paraffin embedded (FFPE) tissue from biopsies or resected tumors. While this tissue is readily available, one may ask if there are any other reasons why FFPE tissue might be useful. Are there specific benefits for using FFPE tissue for NGS (FFPE Sequencing)? How does it compare to other sample types? Will FFPE tissue provide suitable DNA for sequencing? Let's take a look at the potential value of performing NGS on this sample type.
Are FFPE samples a better option than frozen samples?
Frozen samples are a phenomenal source of high quality DNA, but they’re not easily available. Its costly to maintain the infrastructure to keep resected cancer tissue frozen for long periods of time. Freezers can also fail - destroying entire sample collections. FFPE tissue samples on the other hand, can be kept in a cabinet or a drawer for years. Hospitals and research labs have been keeping collections of FFPE tissues for decades actually, creating a tremendous resource for researchers to draw on. In addition, hospital pathologists are accustomed to creating blocks of FFPE tissue from cancers from patients currently under treatment. This makes FFPE tissue samples a great alternative because they're there. For some tumors, they can be the only available source of DNA.
FFPE sequencing
may be the only available tool for providing genomic information about many cancers.
Does FFPE sequencing result in lower quality data?
Immersing a block of tissue in formalin, and then embedding it in paraffin, does affect the quality of the DNA inside. It will generally be broken into fragments and chemically modified. Fragmenting the DNA generally won't affect the sequencing process, unless the fragments are extremely small. The chemical modifications to the actual DNA nucleotides can be a big problem. A particularly vexing issue is that formalin removes an amine group from Cytosine and makes it appear to be a Thymine - causing a large number of C to T transitions appear in sequenced FFPE DNA. This problem is sufficiently prevalent that virtually all commercial NGS cancer panels have protocols for identifying or eliminating these spurious sequence changes. The FFPE process can also cause single nucleotide variations (SNVs) or insertions/deletions of nucleotides (indels) in the DNA sequences. However, these mutations seem to be random and evenly distributed across different DNA fragments – meaning they can be identified. The easiest way of correcting these mutations is by making sure the sequencing depth of coverage is high. Depth of coverage refers to the fact that many copies of each fragment is sequenced. The current consensus is that using coverage levels of 80 times or more is preferred in the analysis of FFPE material. This means that on average, 80 copies of each nucleotide in the DNA strand will have been sequenced. Formalin also chemically crosslinks the DNA strands together - making them impossible to sequence. This crosslinking can be corrected by heating the DNA for an extended time, however this will also degrade the DNA. Choosing the right balance of time and temperature can address this problem.
So the short answer is yes, FFPE tissue yields poor quality DNA - but researchers have figured out how to compensate.
Is the quality of the FFPE sample affected by its age?
The age of the FFPE sample will affect its quality to a degree, but the way the samples were collected and the way they’ve been stored seems to be more important. Samples over 14-21 years old do tend to yield poorer DNA sequences, but it is unclear if this is because of the tissue age, or if this is due to the fixation and embedding process, or now long a piece of tissue sat before being immersed in formalin. At any rate FFPE sequencing of older samples has a been performed many times quite successfully.
How much FFPE tissue needed to make NGS successful?
The amount of FFPE tissue needed for NGS varies based on the quality of the sample, and whether you are trying to sequence the entire genome at a high depth of coverage, or if you are looking at a targeted region or regions within the genome. Generally, however, a few milligrams or tens of milligrams of tissue will be enough to generate useful data.
How can FFPE samples be used to help treat cancer?
An early, accurate diagnosis of cancer is critical to achieving a successful treatment. Sequencing biopsy material can aid in diagnosis if visual inspection of an FFPE tissue slide by a pathologist does not yield a definitive answer. Targeted NGS gene panels can often pinpoint cancer-associated DNA mutations. Also. a specific type of NGS that looks at methylated DNA can assess whether cancer-associated genes are being activated or shut down. Lastly, researchers are identifying mutations that potentially can inform doctors regarding prognosis, or treatment options.
In summary, FFPE sequencing is emerging as an extremely powerful tool to help doctors confirm a positive cancer diagnosis, inform them as to the best way to treat a cancer, or at the very least help them give a more accurate prognosis.
