Abstract

Introduction

Despite advances in diagnosis and treatment, B-precursor acute lymphoblastic leukemia (B-ALL) remains the most common childhood cancer and one of the leading causes of cancer-related death in children and adolescents. Although B-ALL is highly curable, approximately 10 - 20% of children diagnosed with B-ALL still do not respond to the current treatment protocols. Minimal residual disease (MRD) at the end of induction of remission is strongly associated with prognosis. Therefore there is an urgent need to understand the molecular mechanisms underpinning MRD and to identify biomarkers for the development of novel and more effective therapeutic strategies. This project was undertaken to determine whether molecular perturbation in patients with positive MRD at day 46 differs from those with negative MRD in different subtypes of B-ALL and to identify biological pathways dysregulated. We hypothesized that gene expression profiles differ significantly between patients with positive MRD at day 46 and patients with negative MRD.

Methods

We analyzed publicly available gene expression data derived from samples obtained from 189 patients with B-ALL (47 with positive MRD at day 46 and 142 with negative MRD). The data was downloaded from the NCBI’s Gene Expression Omnibus (GEO) database under accession number GSE33315. Patients were classified into seven subtypes of B-ALL which are hyperdiploid, ETV6-RUNX1, MLL rearrangement, hypodiploid, BCR-ABL1, TCF3-PBX1 and others (no detectable recurring genetic abnormalities). Samples from patients with BCR-ABL1 were excluded due to a different prognosis and treatment approach. Patients with TCF3-PBX1 were excluded due to the small sample size; leaving 165 patients in the analysis (35 with positive MRD at day 46 and 130 with negative MRD). We analyzed gene expression data using both supervised and unsupervised analysis. Supervised analysis was performed between patients with positive MRD and negative MRD for each subtype of B-ALL. Unsupervised analysis using hierarchical clustering was performed on significantly differently expressed genes (P < 0.005) to identify functionally related genes with similar patterns of expression profiles. Pathway analysis was performed using the Ingenuity Pathways Analysis (IPA) system to identify biological pathways that are dysregulated in response to positive MRD in different subtypes of B-ALL.

Result

Comparison of gene expression profiles between positive MRD and negative MRD revealed significantly differentially expressed genes between the two groups. The numbers of significantly (P < 0.005) differentially expressed genes for hyperdiploid, ETV6-RUNX1, MLL rearrangement, hypodiploid and others were 93, 82, 87, 140 and 289 genes; respectively. The identified genes included BCL2, BECN1, CBFB, IKZF1, PAX5, SH2B3 and TOX which are known to be associated with B-ALL. Unsupervised analysis using hierarchical clustering and GO analysis revealed similarity in patterns of gene expression within subtypes of B-ALL and functional relationships among the identified genes. Among the identified genes included genes involved in cell death and survival, cellular development and DNA replication, recombination, and repair. Network and Pathway analysis revealed multi-gene regulatory networks and key biological pathways including granzyme B signaling, TCA cycle II and B cell receptor signaling. Pathway analysis also revealed upstream regulators including RB1, CDKN2A and TP53 which have been reported to be involved in the hypodiploid subtype, a subtype characterized with poorer prognosis.

Conclusion

Although the sample size is small, our analysis demonstrates that molecular perturbation significantly differs between pediatric B-ALL patients with positive MRD and those with negative MRD, and that these differences are subtype-specific. The results further demonstrate that biological pathways are dysregulated in response to MRD status and that use of gene expression analysis has the promise to stratify patients on the basis of MRD status and to identify potential biomarkers.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.