Abstract
Analysis of tumoral RNA from bone marrow (BM) biopsy is essential for diagnosing childhood B-cell acute lymphoblastic leukemia (B-ALL), risk stratification, and monitoring, by detecting fusions and gene expression patterns. However, frequent BM biopsies are invasive and traumatic for patients. Small extracellular vesicles (sEVs) circulating in blood contain a variety of biomolecules, including RNA, that may contribute to cancer progression, offering a promising source of non-invasive biomarkers from liquid biopsies. While most EV studies have focused on small RNAs like microRNAs (miRNAs), the role of longer RNA species, including messenger RNAs (mRNAs), long non-coding RNAs (lncRNAs), and circular RNAs (circRNAs), remains underexplored despite their demonstrated potential for risk-based patient stratification when starting from BM biopsies. We used immuno-purification to isolate sEVs from peripheral blood at diagnosis in B-ALL patients and cell model-based conditioned culture medium (CCM) with ETV6::RUNX1 and TCF3::PBX1 fusions. Using whole-transcriptome sequencing targeting transcripts over 200 nt and a novel data analysis pipeline, we identified 102 RNA transcripts (67 mRNAs, 16 lncRNAs, 10 circRNAs, 4 pseudogenes, and 5 others) in patient-derived sEVs. These transcripts could serve as biomarkers for two distinct molecular subgroups of B-ALL, each with different risk profiles at diagnosis. This is the first study characterizing the long transcriptome in blood-derived sEVs for childhood B-ALL, highlighting the potential use of circulating RNAs for improved risk-based stratification.