Application of targeted capture high depth sequencing and transcriptome sequencing in childhood B-cell acute lymphoblastic leukemia / 白血病·淋巴瘤

ABSTRACT

Objective:To investigate the differences among targeted capture high depth sequencing (Panel-seq), transcriptome sequencing (RNA-seq) and traditional detection methods in cytogenetic and molecular genetic typing of childhood B-cell acute lymphoblastic leukemia (B-ALL) and their significances.Methods:The clinical data of 152 newly diagnosed childhood B-ALL cases in Guangzhou Women and Children's Medical Center from September 2020 to December 2021 were retrospectively analyzed. Along with traditional cytogenetic and molecular detection methods including karyotyping, fluorescence in situ hybridization (FISH) and 43 kinds of fusion gene quantitative screening for traditional cells and molecular genetic detection, both Panel-seq and RNA-seq were also performed. Panel-seq covered more than 600 genes with common mutations in hematological tumors, from which fusion genes and gene mutations were both analyzed. RNA-seq was used to analyze fusion genes, gene mutations, gene expression, and copy number variation at the chromosome level. High hyperdiploid karyotype was estimated by using gene expression profile clustering and copy number variations. The cytogenetic typing results of all detection methods were also analyzed.Results:Among 152 patients, 93 cases were males and 59 cases were females, with the median age of 4.0 years (0.8-13.0 years). The median blast cell ratio was 0.855 (0.215-0.965). The traditional detection methods could identify 4 cases (2.6%) with BCR-ABL1, 2 cases (1.3%) with CRLF2 gene-related fusion, 27 cases (17.8%) with ETV6-RUNX1, 1 case (0.7%) with iAMP21, 5 cases (3.3%) with MLL rearrangement, 8 cases (5.3%) with TCF3-PBX1 and 22 cases (14.5%) with high hyperdiploid karyotype. Panel-seq could identify 4 cases (2.6%) with BCR-ABL1, 2 cases (1.3%) with CRLF2 gene-related fusions, 27 cases (17.8%) with ETV6-RUNX1, 3 cases (2.0%) with MEF2D gene-related fusions, 1 case (0.7%) with MEIS1-FOXO1, 5 cases (3.3%) with MLL rearrangement, 5 cases (3.3%) with PAX5 gene-related fusions, 8 cases (5.3%) with TCF3-PBX1 fusions, 4 cases (2.6%) with ZNF384 gene-related fusions, and 2 cases (1.3%) with IKZF1 N159Y mutations. Among 152 patients, 1 case with MLL rearrangement didn't receive RNA-seq detection because of sample quality; in other 151 B-ALL cases, 1 case (0.7%) with ACIN1-NUTM1, 4 cases (2.6%) with BCR-ABL1, 3 cases (2.0%) with CRLF2 gene-related fusions, 8 cases (5.3%) with DUX4 gene-related fusions, 27 cases (17.9%) with ETV6-RUNX1, 3 cases (2.0%) with MEF2D gene-related fusions, 1 case (0.7%) with MEIS1-FOXO1, 4 cases (2.6%) with MLL rearrangement, 5 cases (3.3%) with PAX5 gene-related fusions, 1 case (0.7%) with ZMIZ1-ABL1, 8 cases (5.3%) with TCF3-PBX1,4 cases (2.6%) with ZNF384 gene-related fusions, 61 cases (40.4%) with hyperdiploid karyotypes, and 2 cases (1.3%) with IKZF1 N159Y mutations were detected; RNA-seq had obvious advantage in detecting fusion gene and hyperdiploid karyotype. The cytogenetic and molecular genetic typing rates of traditional method, Panel-seq and RNA-seq were 45.4% (69/152), 40.1% (61/152) and 87.4% (132/151), respectively. The combination of the three could identify 89.5% (136/152) of childhood B-ALL patients.Conclusions:The combination of Panel-seq and RNA-seq can increase the detection rate of genetic abnormality in childhood B-ALL, which provides a more accurate molecular genetic classification for B-ALL and the basis for treatment guideline and prognosis judgement.