Abstract 1509: Longitudinal profiling of high-risk pediatric malignancies using a multiomics approach

Henry J. Martell,Avanthi T. Shah,Alex G. Lee,Bogdan Tanasa,Stanley G. Leung,Aviv Spillinger,Heng-Yi Liu,Inge Behroozfard,Phuong Dinh,María V. Pons Ventura,Florette K. Hazard,Arun Rangaswami,Sheri L. Spunt,Norman J. Lacayo,Tabitha Cooney,Jennifer G. Michlitsch,Anurag K. Agrawal,Marcus R. Breese,Alejandro Sweet-Cordero

DOI: https://doi.org/10.1158/1538-7445.am2023-1509

IF: 11.2

2023-04-04

Cancer Research

Abstract:Abstract For many pediatric cancer patients, commonly used gene-panel sequencing tests yield few actionable results, partly due to the complex genomic alterations present. We hypothesized that an unbiased approach, combining whole-genome (WGS) and RNA sequencing (RNAseq), could overcome this and lead to a more comprehensive understanding of these diseases. While prior studies have evaluated WGS and RNAseq in pediatric cancers, few focused primarily on metastatic or relapsed disease. We also placed special focus on longitudinal profiling of patients, including with additional deep sequencing, to capture tumor evolution at the primary and metastatic sites, and to quantify the utility of resampling. We assembled a cohort of 191 high-risk pediatric oncology patients, including solid tumors, CNS tumors, and leukemias/lymphomas. We have representation of patients with relapsed/refractory disease (68), metastatic disease at diagnosis (10), rare diagnoses (19), prior cancer history, and estimated overall survival <50%. We characterized 280 samples with WGS (tumor ~60X; germline ~30X) and/or RNAseq (tumor, polyA selected, ≥20 million reads), including multiple samples taken from 85 patients at different time points (diagnosis, resection, relapse, etc.). Variants (SNVs), structural rearrangements (SVs), mutational signatures, and copy-number alterations (CNAs) were identified using WGS. RNAseq was used to profile gene expression outliers, gene fusions, and expression of variants identified by WGS. The integrated results were used to prioritize potentially actionable variants for each patient. For 20 patients (44 samples), we performed targeted deep sequencing of the DNA (~500X) to profile tumor evolution that cannot be captured by WGS. Multiple sampling from the same patient identified drastic spatial and temporal differences in the genomes and transcriptomes of these tumors. Using the Jaccard index as a measure of concordance between samples shows dynamic changes between samples collected at different time points across multiple modalities (range 0-1, 1 is identical); SNVs ranged from 0.01-0.79, SVs 0.01-0.73, major CNAs 0.07-0.99, minor CNAs 0.38-0.99, up expression outliers 0.12-0.56, down expression outliers 0.04-0.54, and fusions 0-1. Potentially biologically significant differences in therapy-induced mutations by platinum agents were also observed, highlighting the impact of therapy on tumor evolution. Clonal architectures were extracted from deep resequencing and show extensive spatial, temporal, and metastatic heterogeneity in these rare and highly aggressive malignancies that is not captured by WGS alone. Identifying clinically relevant evolution remains a challenge in most patients, but our results suggest that resampling of pediatric tumors at relapse or metastasis will be important for the effectiveness of targeted therapies in the future. Citation Format: Henry J. Martell, Avanthi T. Shah, Alex G. Lee, Bogdan Tanasa, Stanley G. Leung, Aviv Spillinger, Heng-Yi Liu, Inge Behroozfard, Phuong Dinh, María V. Pons Ventura, Florette K. Hazard, Arun Rangaswami, Sheri L. Spunt, Norman J. Lacayo, Tabitha Cooney, Jennifer G. Michlitsch, Anurag K. Agrawal, Marcus R. Breese, Alejandro Sweet-Cordero. Longitudinal profiling of high-risk pediatric malignancies using a multiomics approach [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 1509.

oncology