ID #1068 Investigating Genome-Wide Germline Variants in Paediatric Patients with CNS Tumour

Abstract

Germline alterations in cancer predisposition genes (CPGs) are associated with increased risk of developing brain tumours. The reported prevalence of pathogenic/likely pathogenic (P/LP) germline variants in central nervous system (CNS) tumours differs greatly from 8.6-35.5%, primarily attributed to the inclusion of patients with recurrent tumours or specific tumour diagnoses (∼40% of SHH-MB carrying germline pathogenic variants), and many cancer predisposition syndromes/CPGs are known to be associated with brain tumours (e.g. NF1, TP53, APC). Our PREDICT study recruited 230 patients (age < 21 years) newly diagnosed with cancer across NSW from 2021-2023, which included 29 patients with CNS tumours. Whole genome sequencing (WGS) data from those patients (and their biological parents if available) was analysed against ∼1200 cancer-related genes using our comprehensive bioinformatics pipelines to investigate genome regions beyond exonic small alterations, including single nucleotide variants (SNVs) and small insertions/deletions (indels), canonical and non-canonical splicing variants, regulatory element variants, and structural variants (SVs)/copy number variations (CNVs). At least 1 reportable clinically relevant variant was detected in 8 patients with CNS tumour (n = 8/29, 27.6%), and novel candidate variants were identified in 8/29 patients (27.6%), with 3 patients carrying both clinical and research variants. Only 2 loss-of-function variants (6.9%) were concordant with the cancer type (ELP1 in medulloblastoma SHH subtype and NF1 in bilateral optic glioma) and the other 7 clinically relevant variants were classified as risk factors (e.g. heterozygous SNV in BUB1B and MLH1 detected in a patient with GBM). Multiple germline alterations were detected in pilocytic astrocytoma including one patient with maternally inherited stop-gain in FAN1 and missense variant in DYRK1A, and paternally inherited CNV deletion in FANCA. Predicted high-impact variants in cytoskeleton gene TPM4 were detected in one patient with pilocytic astrocytoma and one patient with DMG. High detection rate of clinical variants in the small cohort of brain tumour highlighted the importance of comprehensive WGS analysis; however further characterisation of variants classified as risk factors or novel variants with unknown significance is urgently needed to clarify their roles in brain cancer development.