CHARACTERIZING THE LANDSCAPE OF STRUCTURAL VARIANTS IN ADAMANTINOMATOUS CRANIOPHARYNGIOMA

Abstract INTRODUCTION: Adamantinomatous craniopharyngiomas (ACPs) are rare brain tumors that primarily occur in children and impact long-term morbidity and mortality. The canonical driver mutation for ACP growth occurs in CTNNB1 and leads to constitutive activation of the Wnt/β-catenin signaling pathway. In this study, we outline the genomic, transcriptomic, and structural variant (SV) landscape in a cohort of 41 ACP samples. METHODS: We performed whole-genome sequencing (WGS) and RNA-sequencing of 41 ACP samples. Matched normal samples were also characterized by WGS. Mutect2 was used to detect single nucleotide variants (SNVs) and indels, and copy number data was generated using the GATK pipeline. SvABA was used to perform SV analysis and to identify significantly recurrent breakpoints and juxtapositions. DESeq2 was used to perform differential gene expression analysis based on clinical and molecular annotation data. RESULTS: 29/41 (70%) of the ACP samples harbored missense mutations in exon 3 of CTNNB1, all of which have previously been reported in ACP tumors. SV analysis identified a median of 11.5 events per tumor. Overall, 9.7% of events were interchromosomal. Of the remainder, the majority (78.6%) were deletions. No SVs occurred within CTNNB1. A positive correlation (r = 0.533) was observed between the frequency of SVs and SNVs within samples. Analysis of significantly recurring breakpoints (SRBs) did not identify recurrent breakpoint events. Differential gene expression analysis comparing samples with and without CTNNB1 variants identified 2,143 differentially expressed genes with q-value < 0.05. CONCLUSION: This study identifies activating mutations in exon 3 of CTNNB1 in a large cohort of ACP samples. We also integrate SV and transcriptomic data to comprehensively investigate ACP tumor genomes and identify putative novel tumorigenic mechanisms that advance our understanding of ACP biology.