Abstract 3388: Copy number alteration in primary melanoma

Abstract

Abstract The systematic analysis of the genomic characteristics of melanoma primaries has been infeasible given the limited size of these formalin-fixed paraffin embedded (FFPE) lesions. Further studying a clearly ascertained, deeply phenotyped patient population allows meaningful extrapolation of the prevalence of distinct genomic features and investigation of the association of these features with lifestyle exposures and germline profiles. For these reasons, we focused on a recently recruited patient cohort, the Leeds Melanoma Cohort (LMC) consisting of 2184 recruits from a distinct geographical region of Northern England; the only selection applied to recruitment was to only include those whose melanoma was less than 0.75 mm Breslow thickness in the early years of recruitment. Tumor samples from 703 cases have been analyzed for genome-wide gene expression (Pozniak et al, this meeting). For this study, we wanted to explore the prevalence of copy number alterations (CNA) and the mutation profile of tumors (reflecting the association of C>T mutations with UV exposure, the primary environmental risk factor) and so we have adopted a next-generation sequencing (NGS) approach. Tumor blocks were recovered from local pathology laboratories for those cases already deceased at the time that this CNA analysis was initiated plus those surviving more than 5 years post diagnosis at this same time. Blocks at risk of being compromised for clinical purposes by sampling for this research study were excluded from sampling; this amounted to almost 50% of identified samples. A total of 333 NGS libraries were sequenced on an Illumina GAII or HiSeq sequencer to produce >100bp paired-end reads (either 5 or 1 per lane). DNA reads were aligned and mapped achieving approximately 1.8x coverage (9.4x for those at 1 per lane); the number of reads falling within each 10kb window was adjusted for GC content and mappability and compared to a composite normal FFPE sample to determine local copy number. We adopted various approaches to assess the informativeness of the methodology. (i) Replicate samples showed high reproducibility (all p < 0.0001). (ii) Focused analysis of the CDKN2A region with segmentation analyses to determine CNA breakpoints yielded high quality, biologically plausible data revealing multiple copy number events. 67% of samples showed no CNA across the 4 Mb region covering CDKN2A while, as expected, only a small proportion of CNAs did not involve CDKN2A. Four common distinct patterns were observed. Copy number loss ranged from 20kb to > 4 Mb. (iii) A germline CNA could reliably be identified from the tumor data. (iv) CDKN2A expression levels correlated with estimated copy number. We have shown that small FFPE melanomas can be characterized for genomic alterations using NGS opening up the potential for studies associating epidemiological and germline profile with the resulting tumor. 1. Newton-Bishop, J.A., et al. (2009) J Clin Oncol, 27, 5439-5444. Funded by Worldwide Cancer Research and Cancer Research UK. Citation Format: D. Timothy Bishop, Anastasia Filia, Alastair Droop, Joey Diaz, Mark Harland, Jon Laye, Juliette Randerson-Moor, Julia A. Newton-Bishop. Copy number alteration in primary melanoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3388. doi:10.1158/1538-7445.AM2017-3388