Nucleic acid extraction from formalin-fixed paraffin-embedded cancer cell line samples: a trade off between quantity and quality?

BACKGROUND: Advanced genomic techniques such as Next-Generation-Sequencing (NGS) and gene expression profiling, including NanoString, are vital for the development of personalised medicines, as they enable molecular disease classification. This has become increasingly important in the treatment of cancer, aiding patient selection. However, it requires efficient nucleic acid extraction often from formalin-fixed paraffin-embedded tissue (FFPE). METHODS: Here we provide a comparison of several commercially available manual and automated methods for DNA and/or RNA extraction from FFPE cancer cell line samples from Qiagen, life Technologies and Promega. Differing extraction geometric mean yields were evaluated across each of the kits tested, assessing dual DNA/RNA extraction vs. specialised single extraction, manual silica column based extraction techniques vs. automated magnetic bead based methods along with a comparison of subsequent nucleic acid purity methods, providing a full evaluation of nucleic acids isolated. RESULTS: Out of the four RNA extraction kits evaluated the RNeasy FFPE kit, from Qiagen, gave superior geometric mean yields, whilst the Maxwell 16 automated method, from Promega, yielded the highest quality RNA by quantitative real time RT-PCR. Of the DNA extraction kits evaluated the PicoPure DNA kit, from Life Technologies, isolated 2–14× more DNA. A miniaturised qPCR assay was developed for DNA quantification and quality assessment. CONCLUSIONS: Careful consideration of an extraction kit is necessary dependent on quality or quantity of material required. Here we provide a flow diagram on the factors to consider when choosing an extraction kit as well as how to accurately quantify and QC the extracted material. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12907-016-0039-3) contains supplementary material, which is available to authorized users.