Abstract 3322: Diagnosis and monitoring of pediatric cancer patients using low-pass whole genome and targeted sequencing of cell-free DNA

Abstract Cell-free DNA (cfDNA) assays are transforming cancer care by enabling minimally invasive diagnosis, minimal residual disease (MRD) monitoring and by guiding risk stratification and therapy. However, progress in pediatric cancers is limited due to a) limited amounts of cfDNA b) differences in the cancer driver landscape - higher prevalence of copy number alterations (CNA) and fusions with few point mutations and c) small sample sizes. Here, we addressed some of these challenges by performing low-pass whole genome (LPWGS) and/or targeted sequencing (TS) of cfDNA on a cohort of 125 pediatric solid (n=78) and brain (n=47) tumor patients enrolled either at diagnosis, after therapy or relapse. Plasma was collected from patients with sarcoma, renal, hepatic and germ-cell tumors; cerebrospinal fluid (CSF) during surgery from children with low and high grade gliomas, ependymomas, craniopharyngiomas and embryonal tumors; and aqueous humor for retinoblastoma. WGS libraries were constructed with the xGen Prism DNA Library Kit using 5ng of input cfDNA and sequenced (Illumina NextSeq 500) to a mean coverage of 4.1x. CNAs were evaluated using ichorCNA and compared with clinical microarray results from primary tumors. Selected cases with previously clinically reported BRAF, EWSR1 and FOXO1 canonical gene fusions and driver mutations from our OncoKids assay were analyzed using TS (Twist Bioscience) sequenced to a mean coverage of 582x. In treatment naive cases that had an aberration detected in the primary tumor, we observed an overall positivity rate in 67% (72/107) of cases using cfDNA. For a subset of 45 cases with at least 10% circulating-tumor (ct) DNA, we compared LPWGS results with microarray data from the primary tumor and identified 88/92 (96%) of the clinically reported events suggesting its utility as a diagnostic tool. In adult cancers, proportion of short (<150bp) to long fragments (S/L ratio) using LPWGS was shown to correlate with ctDNA concentration and consequently detectability. Therefore, we categorized tumor types as low-burden if the median S/L ratio was below the median value across all tumor types and as high-burden otherwise. Interestingly, ctDNA detectability rates were significantly higher in high-burden tumors (27/52 vs 28/37, Fisher’s exact P-value = 0.03) suggesting that S/L ratios could define ctDNA detectability. Though anecdotal in evidence, LPWGS demonstrated CNAs distinctly present in primary and metastatic lesions and in at least one case, revealed ctDNA in plasma, 9 months prior to relapse. These findings suggest cfDNA may be more representative of tumor clonal composition and utilized for MRD detection. These findings affirm the clinical utility of liquid biopsy assays to diagnose and monitor a variety of pediatric solid and brain tumors. Clinical trials and larger cohorts are necessary for ctDNA guided risk and therapeutic stratification of pediatric cancers. Citation Format: Venkata Yellapantula, Eirini Christodoulou, Katrina O’Hollaran, Jianling Ji, Liya Xu, Jesse L. Berry, Anya Zdanowicz, Leo Mascarenhas, James Amatruda, Dejerianne Ostrow, Nicholas Chapman, Jason Chu, Mark Krieger, Peter Chiarelli, Pan Yachen, Moiz Bootwalla, Xiaowu Gai, Fariba Navid, Jaclyn A. Biegel. Diagnosis and monitoring of pediatric cancer patients using low-pass whole genome and targeted sequencing of cell-free DNA [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3322.