Current concepts in pediatric Philadelphia chromosome-positive acute lymphoblastic leukemia

Abstract

The t(9;22)(q34;q11) or Philadelphia chromosome creates a BCR-ABL1 fusion gene encoding for a chimeric BCR-ABL1 protein. It is present in 3-4% of pediatric acute lymphoblastic leukemia (Ph(+) ALL), and about 25% of adult ALL cases. Prior to the advent of tyrosine kinase inhibitors (TKI), Ph(+) ALL was associated with a very poor prognosis despite the use of intensive chemotherapy and frequently hematopoietic stem-cell transplantation (HSCT) in first remission. The development of TKIs revolutionized the therapy of Ph(+) ALL. Addition of the first generation ABL1 class TKI imatinib to intensive chemotherapy dramatically increased the survival for children with Ph(+) ALL and established that many patients can be cured without HSCT. In parallel, the mechanistic understanding of Ph(+) ALL expanded exponentially through careful mapping of pathways downstream of BCR-ABL1, the discovery of mutations in master regulators of B-cell development such as IKZF1 (Ikaros), PAX5, and early B-cell factor (EBF), the recognition of the complex clonal architecture of Ph(+) ALL, and the delineation of genomic, epigenetic, and signaling abnormalities contributing to relapse and resistance. Still, many important basic and clinical questions remain unanswered. Current clinical trials are testing second generation TKIs in patients with newly diagnosed Ph(+) ALL. Neither the optimal duration of therapy nor the optimal chemotherapy backbone are currently defined. The role of HSCT in first remission and post-transplant TKI therapy also require further study. In addition, it will be crucial to continue to dig deeper into understanding Ph(+) ALL at a mechanistic level, and translate findings into complementary targeted approaches. Expanding targeted therapies hold great promise to decrease toxicity and improve survival in this high-risk disease, which provides a paradigm for how targeted therapies can be incorporated into treatment of other high-risk leukemias.

Keywords: BCR–ABL1; acute lymphoblastic leukemia; chemotherapy; hematopoietic stem-cell transplantation; tyrosine kinase inhibition.

Figure 1

Frequency of BCR/ABL1 rearrangement in pediatric and adult ALL. Cytogenetic abnormalities in pediatric (>1 year) and adult patients with ALL are shown (5). The majority of children <1 year of age carry a rearrangement of the MLL-gene and are not included in this graph. Favorable cytogenetic abnormalities are represented in green, neutral in blue, and unfavorable cytogenetics are represented in yellow/red. Favorable cytogenetics (high hyperdiploidy, ETV6–RUNX1) decrease, while the frequency of BCR/ABL1 rearrangement increases with age. The higher percentage of unfavorable cytogenetics substantially contributes to inferior outcomes in adult versus pediatric ALL.

Figure 2

Structure of the most common BCR–ABL1 fusion genes. Domain structure of wild type BCR and wild type ABL1 protein, as well as retained domains in the three most common BCR–ABL1 variants, p230, p210, and p190. OD: oligomerization domain (coiled-coil domain) mediating oligomerization, Tyr177: tyrosine 177, which, when phosphorylated, serves as a docking site for the adaptor protein GRB-2; SH2-domain: SRC homology 2 (binding to phosphorylated tyrosine residues, including BCR exon 1), SH3-domain: SRC homology 2 (binding to proline rich peptides). SH1-domain: SRC homology 1 (ABL1 catalytic domain); GEF-domain: guanine nucleotide exchange factors (G-protein signaling); E1: exon 1 of ABL1, contains the inhibitory N-terminal “cap” that binds the catalytic domain (SH1) of ABL1 and prevents autophosphorylation; NLS: nuclear localization signal.

Figure 3

BCR–ABL1 signaling pathways. Downstream signaling pathways activated by BCR–ABL1. Numerical references 1–6 denote classes of inhibitors in Table 1. I: imatinib; D: dasatinib, N: nilotinib.

Figure 4

B-cell development and transcription factors mutated in Ph⁺ ALL. Differentiation stage-dependent expression (blue bars) and function of the three major B-cell developmental regulators mutated in Ph⁺ ALL, Ikaros, Pax5, and EBF1 (106, 107). Ikaros expression is detected early in hematopoietic development and appears to have a role in shutting down stem-cell programs and nudging cells toward lymphoid development. Ikaros expression is maintained through B-cell development. Complete loss of Ikaros in murine models leads to a differentiation block at the LMPP stage and complete absence of the entire B-cell lineage (red block). A severe reduction allows the development of B-cell progenitors, but maturation is blocked at the Pro-B stage (orange block). EBF1 is turned on in common lymphoid progenitors (CLPs) and controls lineage specification to the B-cell lineage. Loss of EBF1 in murine models leads to a differentiation block at the Pro-B stage (red block). Pax5 is turned on the latest and maintains lineage commitment. Loss of Pax5 causes a differentiation block at the Pro-B-cell stage. Neither Pax5 nor EBF1 appear to have a role in silencing hematopoietic stem-cell programs, which may explain why IKZF1 mutations are associated with a poor prognosis, while PAX5 and EBF1 mutations do not predict adverse outcomes. HSC: hematopoietic stem cells; LMPP: lymphoid-primed multipotent progenitors; CLP: common lymphoid progenitors.

Figure 5

BCR–ABL1 TKD mutations. Location of the BCR–ABL1 tyrosine kinase domain mutations listed in Table 2. The T315I mutation, which causes resistance against imatinib, dasatinib, and nilotinib is depicted in red. P-loop: phosphate-binding loop; A-loop: activation loop.

Figure 6

Survival of children with Ph⁺ ALL treated with imatinib + chemotherapy on COG AALL0031. (A) Event-free survival (early follow-up) of cohort 5 of AALL0031 treated with the MTD of imatinib in combination with chemotherapy compared to historic controls. Thirteen of the 44 patients went on to receive a matched related HSCT, while the remaining 31 patients received chemotherapy only. Introduction of imatinib onto a backbone of standard chemotherapy dramatically improved the outcome of Ph⁺ ALL. (B) Disease-free survival (5.2-year median follow-up) of patients treated as per AALL0031 cohort 5 based on transplant status. Chemotherapy only – 24 patients, matched related HSCT on study – 13 patients, unrelated HSCT off study – 6 patients. HSCT did not offer additional benefit compared to chemotherapy + imatinib.