Abstract

Abstract DNA copy number variation (CNV) has been identified in many cancers. Recent studies have shown that CNV data, in combination with other high-throughput techniques, can provide novel biological insight to tumor biology. Although a number of CNV whole-genome studies have been performed on melanoma using cell lines and frozen tumors, very few have used formalin fixed paraffin embedded (FFPE)-derived material. Melanoma tumors are small and rarely frozen, which means that FFPE tumor blocks would be a very valuable source of genomic material to be used for CNV analysis if the methodologies were adequately developed. It has been shown that reliable next generation sequencing data can be produced from FFPE-derived DNA in lung cancers (Wood et al., 2010). Here, we report a study showing that DNA from FFPE melanoma tumor blocks can be used for CNV detection on a next generation sequencing platform. DNA from 20 FFPE melanoma samples was used to prepare genomic libraries, which were sequenced on the Illumina GAIIx instrument. Successful library preparation occurred at DNA levels of at least 30ng, while current protocols generally suggest 50ng with high quality material and 1μg generally. The use of SPRIworks (Beckman Coulter) and Bravo (Agilent) robots was compared with the manual method for library preparation with input DNA mass of 30-150ng. Libraries prepared manually were always of better quality and yield. We also investigated means of improving library preparation. Melanin co-purifies with DNA and inhibits downstream reactions, and indeed lower coverage was seen in our samples if they were heavily pigmented. To overcome the adverse effects of melanin, 10 DNA samples (5 pairs) were used for library preparation with and without bovine serum albumin (BSA). BSA has been shown to bind to melanin, and as a result the melanin does not inhibit the DNA polymerase reaction. All 5 libraries with BSA were of higher quality and yield when compared to those prepared without BSA. All comparisons were made after running the libraries on an Agilent DNA chip. Currently available data from the GAIIx were used to develop the analysis pipeline. The first 10 samples were multiplexed together and run twice, while the remaining 10 were multiplexed at 5 samples per lane and run once. The average coverage obtained was 0.10x with a range of 0.03x-0.11x. We observed a strong effect on mapping quality based upon the end-cycle quality scores, leading us to adopt an aggressive data trimming procedure. More samples from the Leeds Melanoma Cohort will soon be prepared and sequenced on the Illumina HiSeq2000 instrument. We aim to identify if tumors with driver BRAF mutations have different genomic signatures when compared to those with NRAS mutations. Finally, genomic data will be combined with clinical data and we will check for the prognostic significance of CNV changes identified in our samples. Citation Format: Anastasia Filia, Alastair Droop, Mark Harland, Tim Bishop, Julia Newton-Bishop. Using formalin-fixed paraffin-embedded melanoma tumors for the detection of copy number variation by next generation sequencing. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 3140. doi:10.1158/1538-7445.AM2013-3140

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call