A Comparative Analysis of Clinical Utility of Cytogenetics and Copy-Number-Integrated Risk Stratification Scores in a Prospective Cohort of Pediatric B-Cell Acute Lymphoblastic Leukemia.

Abstract

The present study compares and evaluates the clinical utility of three published copy-number alteration (CNA)- and cytogenetics-integrated risk group classifiers in a prospective cohort of pediatric B-cell acute lymphoblastic leukemia (B-ALL) patients. All cases underwent CNA testing via digital (n = 185) or conventional multiplex ligation probe amplification (MLPA; n = 22), and the data were integrated with genetics according to the three classifier systems. Out of 207 pediatric B-ALL patients enrolled, a primary genetic abnormality was noted in 87% (181/207) of the patients, with high hyperdiploidy being the most common (31%). The overall CNA frequency was 54% (112/207), with CDKN2A/2B deletion being most common (28.5%). Poor-risk CNAs in all three integrated risk classifiers correlated significantly with high TLC (>50 × 10⁹/L), NCI-HR, MRD>0.01%, ICiCLe-HR, and event. In addition, the two-year EFS and OS were worse for the UKALL-integrated poor-risk group (56.7%; p = <0.0001 and 68.3%; p = 0.012), whereas the digital MLPA-combined poor-risk group and the PersonALL IKZF1 high-risk group had poor two-year EFS (55.3%; p = 0.038) and (39.5%; p = 0.015), respectively. Multivariate analysis revealed the UKALL poor-risk integrated group (GEN-PR) to predict relapse (HR: 48.58; p = 0.014), thereby highlighting that the UKALL-CNA-integrated classifier is more likely to precisely identify a subset of intermediate-risk cytogenetic cases with intermediate- or poor-risk CNAs that require escalated treatment.