Abstract 7178: Advancing WGS and WTS diagnostic testing for pediatric leukemia: fast-tracking CNV, ploidy, and fusion oncogene detection in 5 days

Abstract Leukemia is the most common cancer in children, with ∼75% being acute lymphocytic leukemia (ALL) and most of the rest acute myeloid leukemia (AML). More than half of pediatric acute leukemias are characterized by chromosome translocations, 40% of B-ALLs by ploidy changes, and a subset of AMLs harbor adverse cytogenetic aberrations defined by large-scale copy number changes (LS-CNV). Whole genome (WGS) and whole transcriptome sequencing (WTS) enable comprehensive molecular diagnosis, but often take weeks. We introduce a rapid workflow integrating WGS and WTS for tumor profiling, using advanced bioinformatic tools to evaluate LS-CNVs, tumor ploidy, and gene fusions, yielding results within 5 working days. Samples with >30% tumor content were used for WGS and WTS. We developed tumor-only WGS-based copy number profiling (WGS-CNP) using the Illumina DRAGEN CNV pipeline and an internal post processing analysis to generate genome-wide CNV results within 2-3 hours post-sequencing. Further on, we developed and automated an algorithm which compares test samples to a normal diploid reference, factoring in DNA gains/losses identified by WGS-CNV analysis to calculate DNA index (DI) for tumor ploidy estimates within minutes. We also implemented Fuzzion2, a rapid pattern-matching algorithm, to identify fusion transcripts from WTS data within ∼1 hour. This integrated rapid approach was tested in a total of 117 AML and ALL cases previously diagnosed using standard WGS/WTS pipelines (CREST/CICERO/CONSERTING) and flow cytometry for DI, including 67 cases with characteristic fusion oncogenes, 40 ALLs defined by ploidy change, and 65 cases with LS-CNVs. For gene fusion detection, Fuzzion2 identified all expected fusion oncogenes with no false positives, according to the established criteria. For LS-CNV, findings between WGS-CNP and the standard pipeline were concordant in all tested cases. For tumor ploidy, WGS-DI showed values concordant with Flow-DI in 33/40 cases. The 7 remaining cases were identified as masked low-hypodiploid by WGS-based CNV and B-allele frequency profiling, of which Flow-DI values were above 1.1 representing the actual tumor DNA content due to genome doubling. When using post-genome doubling (4n) as the ploidy baseline to generate CNVs, the calculated WGS-DI was below 0.75, reflecting the original clone before genome doubling. Our pilot studies show that the WGS-CNP reliably reproduced previously observed LS-CNVs, accurately established tumor ploidy changes and distinguished masked low-hypodiploidy from hyperdiploidy. Together with the implementation of Fuzzion2 algorithm for gene fusion detection, this integrated rapid approach may detect 80-90% of disease-defining genetic alterations in pediatric acute leukemia with high sensitivity and specificity within 5 days, dramatically expediting the diagnostic process using WGS and WTS. Citation Format: Rebecca K. Voss, Maria Cardenas, Meiling Jin, Yen-Chun Liu, Alexa N. Siskar, Stephen V. Rice, Michael N. Edmonson, Jennifer L. Neary, Caleb Gallops, Jiali Gu, Jamie L. Maciaszek, Michael Rusch, Mark R. Wilkinson, David Rosenfeld, Priya Kumar, Jeffery M. Klco, Charles G. Mullighan, Jinghui Zhang, David A. Wheeler, Lu Wang. Advancing WGS and WTS diagnostic testing for pediatric leukemia: fast-tracking CNV, ploidy, and fusion oncogene detection in 5 days [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 7178.