Independent thesis Advanced level (degree of Master (One Year)), 20 credits / 30 HE credits
ABSTRACT Background: Human Papilloma Virus positive (HPV+) Oropharyngeal Squamous Cell Carcinoma (OSCC), dominated by tonsillar cancer (TSCC) and base of tongue cancer (BOTSCC) has low mutation-frequency and better survival for younger than older patients.Aim: To examine if HPV+ TSCC and BOTSCC have distinct gene-mutation profiles, for 50 often-mutated genes in cancer, in younger compared to older patients and to test and compare different variant callers to get a deeper understanding of the data.Materials and methods: DNA had previously been extracted from 299 formalin-fixed-paraffin-embedded (FFPE) tumor biopsies and 13 normal samples, and sequenced on the Ion Proton sequencer, a NGS (Next-Generation Sequencing) platform. Alignment and variant calling had been performed via the Ion Torrent Suite software v5 (ITS), and Torrent Variant Caller (TVC).UPPMAX, a High-Performance-Computational cluster (HPC) at Uppsala University was used for storing and computing of the sequenced data. Parallel-processing was used to optimize repetitive steps, saving days of computation time. The descriptive analysis, graphical data representations, and more in-depth analysis, were done in R.Initially, variant calling was performed for 13 tumor/normal paired samples using the novel MuTect2 software from the Genome Analysis Toolkit (GATK) toolset. Variant annotation and statistical analysis was performed on all the 13 paired sequenced samples, using SnpEff. Due to a poor overlap between the above MuTect2 and TVC, after adequate filtering TVC, MuTect, Strelka and VarScan2 were also utilized for comparison.Result and Conclusion: Having only 13 normal samples, normal-tumor paired variant calling to distinguish germ line and somatic variants could not be performed. To obtain an approximation of the amount of germline variation in our cohort, additional variant callers were used for the tumor normal pairs. The data obtained with the TVC caller formed the basis for further analysis of the tumor samples.Notably, comparisons of TVC with MuTect2 output revealed major discrepancies and limited overlap. However, when comparing MuTect2 with MuTect, Strelka, or VarScan 2 regarding overlaps with TVC – the overlap were still limited, but higher degrees of overlaps were disclosed between Strelka, MuTect and MuTec2, indicating that MuTect2 was a successful further development and successor of MuTect.Evidence for presence of distinct gene-mutation profiles correlating to age could not be obtained in the analyzed tumor cohort with the software tool kits applied. Statistical analysis using Wilcoxon-Mann Whitney test did not support a hypothesis of distinct age related mutations for any of the 50 genes analyzed in this tumor cohorts.The highest p-value for the Wilcoxon-Mann Whitney test was for the gene APC, at p ~ 0.054 hints at a possible connection. However, more extensive research with more samples sequenced is necessary to confirm or reject this correlation.
2016. , 61 p.