Transcriptomic molecular risk classification for children with B cell acute lymphoblastic leukemia

The National Cancer Institute awarded The Children's Hospital of Philadelphia $350,860 on September 1, 2026, through a Cooperative Agreement (UH2CA312695) under the Cancer Detection and Diagnosis Research program (CFDA 93.394) to implement exome capture-based RNA sequencing for real-time clinical risk stratification in pediatric B-cell acute lymphoblastic leukemia (B-ALL) treatment.

The recipient will establish centralized diagnostic RNA sequencing-based molecular classification across three CLIA/CAP-certified reference laboratories serving the Children's Oncology Group (COG). The work addresses limitations in current risk stratification approaches, which rely on resource-intensive multimodal testing strategies that cannot detect prognostically relevant genomic lesions (PAX5, IKZF1) or transcriptomic signatures (ETV6::RUNX1-like, KMT2A-like, ZNF384-like). RNA sequencing as a stand-alone test will detect prognostic molecular B-ALL subtypes and identify oncogenic drivers including gene fusions, single nucleotide variants, and copy number abnormalities. The project will define the analytical performance of the RNA sequencing assay and harmonize it across reference laboratories to enhance risk stratification for the next generation of NCTN-supported COG clinical trials for B-ALL patients.

Performance takes place in Philadelphia, Pennsylvania. The period of performance runs September 1, 2026, through August 31, 2027.

Abstract

A primary goal in the treatment of B-cell acute lymphoblastic leukemia (B-all) is precise and accurate assessment of relapse risk to allocate therapy appropriately to prevent relapse in patients at highest risk while minimizing treatment-related morbidity in patients at lowest risk of relapse. In the children's oncology group (COG), current risk stratification identifies limited prognostic genetic biomarkers through a multimodal testing strategy that is resource-intensive, requires multiple samples from each patient, and uses diagnostic tools that cannot detect many genomic lesions (PAX5, IKZF1) that carry prognostic relevance nor prognostic transcriptomic signatures (ETV6::RUNX1-like, KMT2A-like, and ZNF384-like). Thus, our current risk stratification approach is insufficient to appropriately assign risk-stratified therapy. Recent retrospective studies have shown that RNA sequencing (RNASEQ) as a stand-alone test allows detection of prognostic molecular B-all subtypes and identification of oncogenic drivers including gene fusions, single nucleotide variants (SNVS), and copy number abnormalities (CNAS). The overarching goal of this proposal is to implement exome capture-based RNASEQ for real-time clinical use in risk stratification in the next generation of nctn-supported COG clinical trials for patients with B-all. We will establish centralized diagnostic RNASEQ-based molecular classification across three clia/cap-certified laboratories which will serve as COG reference testing sites. We hypothesize that RNASEQ will accurately and efficiently define prognostically relevant molecular B-all subsets and, when integrated with current clinical genomic testing, will enhance risk stratification. We will test this hypothesis with specific aims focused on defining the analytical performance of the RNASEQ assay and harmonizing IT across reference laboratories (aim 1, UH2 phase); and performing clinical validation through centralized molecular classification for patients enrolled on a clinical trial (aim 2, UH3 phase). In aim 1, the three reference laboratories will perform RNASEQ on a panel of 104 carefully selected cases where each sample is aliquoted and processed independently in the three laboratories to assess analytical performance for transcriptomic classification and fusion detection. In aim 2, banked specimens from 300 patients enrolled and treated on the currently accruing nctn/COG AALL1732 trial will be selected based on demographic, clinical, and molecular features as well as response and outcomes and will be sent to the three reference laboratories for end-to-end testing and transcriptomic classification to validate our centralized RNASEQ platform. We will combine these data with clinically available genomic testing to provide integrated risk stratification as intended for real-world use. The nctn and COG have consistently established the clinical standard for the diagnosis and treatment of childhood all. The results of this work will not only improve risk stratification for patients on clinical trials but will establish a new gold standard for the molecular characterization of all children with B-all in the united states.