The effect of formalin is one factor very difficult to control within a laboratory. The samples received from the biobank may have been treated very differently by different individuals, formalin treatment time leading to very different levels of DNA quality. Similarly, methylation and deamination can result in sequencing errors, the fragmentation that is induced may affect the quantification and the yield extracted and this effect on quality can subsequently affect the amplifiability of the DNA which may provide difficulties during library preparation and a PCR step. FFPE tissues are fragmented and chemically modified making them challenging to use in molecular studies.
Compared with frozen samples, NGS data from FFPE samples had smaller library insert sizes, greater coverage variability, and an increase in C to T transitions that was most pronounced at CpG dinucleotides, suggesting interplay between DNA methylation and formalin-induced changes.
Figure 1. Comparison of clinical targeted next-generation sequence data from formalin-fixed and fresh-frozen tissue specimens Spencer 2013
Similarly, Hedegaard et al. (2014) below showed that 3 months storage resulted in less efficient DNA extraction as well as:
The results show a 70-80% concordance between FFPE and FF samples of less than three years old, however when the samples are over three years the concordance dramatically decreases.
Figure 2. Single nucleotide variants detected in DNA-Exome-Seq data from the paired FF/FFPE samples. The percentage of common (grey), exclusively FF (white) and exclusively FFPE (red) single nucleotide variants are shown as bars referring to the left axis. The average coverage is shown as stars for FF (white) and FFPE (red) on the right axis. Patient ID and number of years of storage are shown below.