The impact of comparative genomic hybridization/single‐nucleotide polymorphism microarray in risk stratification of pediatric acute lymphoblastic leukemia

Abstract

AbstractBackgroundThe objective of this study is to assess the concordance and added value of combined comparative genomic hybridization plus single‐nucleotide polymorphism microarray (CGH/SNP) analyses in pediatric acute lymphoblastic leukemia (ALL) risk stratification compared to conventional cytogenetic methods.ProcedureThis is a retrospective study that included patients aged 1–18 years diagnosed with de novo ALL at Sainte‐Justine Hospital between 2016 and 2021. Results from conventional cytogenetic and molecular analyses were collected and compared to those of CGH/SNP.ResultsA total of 135 ALL patients were included. Sample failures or non‐diagnostic analyses occurred in 17.8% cases with G‐banding karyotypes versus 1.5% cases with CGH/SNP. The mean turnaround time for results was significantly faster for CGH/SNP than karyotype with 5.8 versus 10.7 days, respectively. The comparison of ploidy assessment by CGH/SNP and G‐banding karyotype showed strong concordance (r = .82, p < .001, r2 = .68). Furthermore, G‐banding karyotype did not detect additional clinically relevant aberrations that were missed by the combined analysis of CGH/SNP and fluorescence in situ hybridization. The most common gene alterations detected by CGH/SNP were deletions involving CDKN2A (35.8%), ETV6 (31.3%), CDKN2B (28.4%), PAX5 (20.1%), IKZF1 (12.7%), and copy‐neutral loss of heterozygosity (CN‐LOH) of 9p (9.0%). Among these, only ETV6 deletion was found to have a significant prognostic impact with superior event‐free survival in both univariate and multivariate analyses (adjusted hazard ratio 0.08, 95% confidence interval: 0.01–0.50, p = .02).ConclusionCGH/SNP provided faster, reliable, and highly concordant results than those obtained by conventional cytogenetics. CGH/SNP identified recurrent gene deletions in pediatric ALL, of which ETV6 deletion conferred a favorable prognosis.