Practical Considerations for Using RNA Sequencing in Management of B-Lymphoblastic Leukemia: Malaysia-Singapore Acute Lymphoblastic Leukemia 2020 Implementation Strategy

Abstract

Despite the immense genetic heterogeneity of B-lymphoblastic leukemia [or precursor B-cell acute lymphoblastic leukemia (B-ALL)], RNA sequencing (RNA-Seq) could comprehensively interrogate its genetic drivers, assigning a specific molecular subtype in >90% of patients. However, study groups have only started to use RNA-Seq. For broader clinical use, technical, quality control, and appropriate performance validation are needed. We describe the development and validation of an RNA-Seq workflow for subtype classification, TPMT/NUDT15/TP53 variant discovery, and immunoglobulin heavy chain (IGH) disease clone identification for Malaysia-Singapore acute lymphoblastic leukemia (ALL) 2020. We validated this workflow in 377 patients in our preceding Malaysia-Singapore ALL 2003/Malaysia-Singapore ALL 2010 studies and proposed the quality control measures for RNA quality, library size, sequencing, and data analysis using the International Organization for Standardization 15189 quality and competence standard for medical laboratories. Compared with conventional methods, we achieved >95% accuracy in oncogene fusion identification, digital karyotyping, and TPMT and NUDT15 variant discovery. We found seven pathogenic TP53 mutations, confirmed with Sanger sequencing, which conferred a poorer outcome. Applying this workflow prospectively to the first 21 patients in Malaysia-Singapore ALL 2020, we identified the genetic drivers and IGH disease clones in >90% of patients with concordant TPMT, NUDT15, and TP53 variants using PCR-based methods. The median turnaround time was 12 days, which was clinically actionable. In conclusion, RNA-Seq workflow could be used clinically in management of B-cell ALL patients.