DNA Library Prep from FFPE Samples

Even though the advantages of using formalin-fixed, paraffin-embedded (FFPE) tissue samples outweigh the disadvantages, sequencing DNA from FFPE samples can be tricky. Invaluable resources such as tissue repositories that store FFPE samples safely for decades can be instrumental in retrospective studies. Unfortunately, the processes of formalin fixation and paraffin embedding inflict multiple types of damage on nucleic acids, which don’t take kindly to being cross-linked, dehydrated, fragmented, and chemically modified. After extracting the sample DNA, preparing a DNA library for next-generation sequencing (NGS) requires some know-how for optimal results. This article covers challenges and advice for preparing high-quality sequencing libraries from FFPE samples.

Repairing damaged samples

The DNA damage and low yield typical of FFPE samples presents multiple challenges. DNA sustains different types of injuries during the fixation and embedding processes (e.g., chemical modifications, crosslinking, heating, dehydration) and subsequent extraction can inflict further damage, all of which can hinder library preparation. Damaged DNA can result in mutations, i.e., oxidative damage (G to T mutation) and cytosine deamination (C to T mutation), and abasic sites can lead to ligase or polymerase blockage. “Low library conversion and yield typically limit the utility of the data from these samples as coverage can often be low and not uniform, challenging various mutation detection applications,” says Maggie Heider, NGS Development Scientist at New England Biolabs. “Additionally, certain damaged bases can be incorporated into sequencing libraries through end repair or PCR, resulting in artifactual mutations and confounding accurate patient mutation detection.”

Library prep tools designed specifically for FFPE samples can increase success by repairing DNA damage and making the conversion from DNA to library more efficient. For example, the NEBNext UltraShear^® FFPE DNA Library Prep Kit contains a mixture of enzymes (the NEBNext FFPE DNA Repair v2) that repair damage to DNA, including cytosine deamination, oxidative damage, abasic sites, blocked ends, nicks, and gaps. “Addressing these damage types can help improve conversion, since some damage types can block polymerase or ligase activity,” says Heider. “Additionally, repairing damage to bases—such as cytosine deamination and oxidative damage—reduces the level of false-positive mutations detected in these libraries.”

NEB’s kit also contains NEBNext UltraShear for optimized fragmentation, which increases mapping rates and reduces chimeric reads. Repairing damage prior to library fragmentation enables “a sample-quality-agnostic fragmentation condition,” says Heider. “Library yield and size is maintained, resulting in high library complexity for whole genome or target capture workflows.”

Another example of reagents that are optimized to prepare libraries from damaged DNA such as FFPE samples is Agilent Technologies’ SureSelect Max DNA Reagent Kits, which can prepare libraries from as little as 10 ng DNA, with an efficient and streamlined workflow. “Agilent’s library prep kits [also] incorporate molecular barcodes to the libraries at the ligation step to enable low allele frequency detection, as well as provide recommendations on optimal fragmentation time and gDNA input, based on sample quality,” says Ronda Allen, Vice President and General Manager of the Life Sciences and Diagnostics Markets Group at Agilent. “For example, the SureSelect Max Kits use one tube enzymatic fragmentation, dA-tailing, and end-repair to streamline the library preparation process and maintain library complexity.”

Navigating low input amounts and chimeric reads

Amplifying sequences without bias is especially important for low input samples. The PCR master mix reagents included in NEB’s NEBNext UltraShear FFPE DNA Library Prep Kit are resistant to overamplification. This allows “a single PCR cycle condition across different sample qualities or inputs to obtain high library yield from all samples without introducing coverage bias,” says Heider.

Integrated DNA Technologies’ (IDT) xGen™ cfDNA & FFPE library prep kit is designed specifically for challenging or low input sample types, such as DNA from FFPE samples, or circulating cell-free DNA. “By using sequential, single-strand splint-ligation, the kit optimizes sample conversion efficiency, enabling more of the DNA in the sample to be converted into library and more complex library to be sequenced,” says Brittany Niccum, Senior Commercial Product Manager, NGS, at Integrated DNA Technologies.

Chimeric reads are another pitfall to beware when using DNA from FFPE samples. DNA fragments can anneal with single-stranded overhangs, which then show up as chimeric reads in sequencing. IDT’s kit features a new ligase that prevents the formation of DNA chimeras. “Both the ligase and the sequential, single-strand splint-ligation maximize the number of DNA fragments that are converted into library, resulting in more unique DNA molecules converting into usable library,” says Niccum. “This, coupled with thoughtful adapter modifications and in-line unique molecular identifiers, provides reduced dimer formation and error correction, enabling researchers to obtain diverse NGS libraries and perform ultra-low variant calling.”

Help from QC and automation

Optimizing library prep protocols is easier using quality control tools to monitor DNA integrity. Automated electrophoresis systems, such as Agilent’s TapeStation and Fragment Analyzer, can tell you valuable information about DNA integrity at different points in the library prep process. “Following DNA extraction, quality metrics such as DIN (DNA integrity number) and GQN (genomic quality number) provide objective assessments of the sample's size distribution,” says Allen. “These metrics not only evaluate DNA quality but also establish empirical cutoffs to distinguish between high-quality and suboptimal samples.” Such detailed feedback can help scientists make important decisions about sample quality, library prep protocols, and sequencing.

Another tool adapted to solve the thorny problems presented by FFPE samples is automation. The consistency provided by automated systems for library prep, such as Agilent’s Magnis NGS Prep System and Bravo NGS Workstation, is only one of many benefits they offer. “These systems are designed to handle FFPE samples efficiently, incorporating automated barcode checking and self-checking routines to ensure high-quality results,” says Allen. “By leveraging these advanced tools and technologies, Agilent helps researchers overcome the inherent challenges of working with FFPE samples, enabling them to generate high-quality DNA libraries for next-generation sequencing.”