Clinical Manifestations and Treatment of Childhood Acute Lymphoblastic Leukemia

Serial risk-directed clinical trials have optimized the combination of chemotherapeutic agents and, along with advances in supportive care, have led to current cure rates of childhood acute lymphoblastic leukemia (ALL) exceeding 85% compared with those of less than 10% in the 1960s. However, because ALL is the most common cancer in children, relapsed ALL remains a leading cause of death from disease in this age group. Over the last two decades, minimal residual disease (MRD) has become the most important determinant in risk stratification. Prophylactic cranial irradiation has been successfully omitted from virtually all patients in several frontline protocols, and the Abelson (ABL) tyrosine kinase inhibitors (TKIs) imatinib and dasatinib have revolutionized the treatment of patients with Philadelphia chromosome (Ph)-positive ALL. Significant advances in biotechnology, such as next-generation sequencing (NGS), have led to a deeper understanding of the molecular pathways involved in ALL, which in turn helped identify new ALL subtypes and their driver mutations. Precision medicine ALL strategies based on inherited and leukemia-specific genomic features, host pharmacogenomics, and targeted molecular and immunologic treatment approaches are increasingly implemented to improve cure rates and reduce toxicity. The remarkable success of bispecific T-cell engagers and chimeric antigen receptor modified T cells (CAR-T cells) targeting CD19, once considered only as treatment of relapsed or refractory B-ALL, has led to their testing in upfront clinical trials, while changing the way practitioners assess allogeneic hematopoietic stem cell transplant (allo-HSCT).

Epidemiology

ALL accounts for approximately 75% of all cases of childhood leukemia, and is the most common pediatric cancer, representing 23% of cancer diagnoses among children younger than 15 years of age. About 3600 children and adolescents are diagnosed with ALL each year in the United States, with an annual rate of 30 to 40 per million. The peak incidence of ALL occurs around 3 to 5 years of age. This young age peak historically has been associated with major periods of industrialization in many countries, suggesting a causative role for environmental changes. ALL, especially T-cell subtype, occurs more frequently in boys than in girls.

Certain inherited genetic and acquired factors are associated with the development of ALL, but most patients have no recognized risk factors. Children with constitutional trisomy 21 (Down syndrome) are up to 15 times more likely to develop leukemia than children without Down syndrome. Genetic instability and DNA repair disorders (e.g., Bloom syndrome, ataxia-telangiectasia, Fanconi anemia) are also associated with an increased risk of developing ALL. Among identical twins, if one is diagnosed with ALL during the first year of life, the risk that the other twin will develop ALL is over 70%, approaching 100% in twins with a single monochorionic placenta. The extraordinarily high concordance rate in monozygotic infant twins compared with dizygotic infant twins or non-twinned siblings results from the metastasis of leukemic cells from one twin to the other through shared placental circulation. Highest in infancy, this risk diminishes with increasing age. If the first twin develops ALL by 5 to 7 years of age, the risk to the second twin is approximately twice that in the general population, regardless of zygosity. After the age of 7 years, the risk to the unaffected twin is similar to that for persons in the general population. The majority of infants with ALL have a chromosome translocation that results in the fusion of the KMT2A (formerly MLL) gene at 11q23 with a variety of partner genes, the most common of which in B-ALL is AF4 . Identical twin infant pairs with concordant ALL share the same acquired KMT2A gene rearrangements.

Several genetic, dietary, and environmental factors have been proposed to modify the risk of leukemia initiation. Children with various congenital immunodeficiency diseases, including Wiskott-Aldrich syndrome, congenital hypogammaglobulinemia, and ataxia-telangiectasia, have an increased risk of developing lymphoid malignancies, as do patients undergoing chronic treatment with immunosuppressive drugs. The loss of cellular immune surveillance capability for tumor antigens and the inability to self-regulate lymphoproliferative processes may contribute to malignant transformation in these patients. Absence of exposure to common infections in the first year of life is associated with a higher risk of developing ETV6-RUNX1 –positive or hyperdiploid ALL (>50 chromosomes) in older children. A possible explanation is that in the absence of infection-driven modulation of the naive immune network in infants, subsequent infectious exposures could result in a highly dysregulated response to infections in older children contributing to leukemogenesis. Exposure to ionizing radiation and certain toxic chemicals could also facilitate the development of acute leukemia. Other evidence linking most environmental exposures to risk of childhood ALL has largely been inconsistent. Causation pathways are likely to be multifactorial and it is probable that the risk of ALL from environmental exposure is influenced by germline genetic variations through the co-inheritance of multiple low-risk variants.

Pathobiology

Acute leukemias comprise a group of clonal disorders of maturation at an early phase of hematopoietic differentiation. ALL subtypes are a heterogeneous group of malignancies with distinctive immunophenotype and molecular pathogenesis that result in varying clinical characteristics and response to therapy. Accurate pathobiologic diagnosis is not only important for prognostic stratification, but can also help define patient-specific therapeutic approaches.

Lymphoblastic leukemias can arise from either B- or T-cell mutant hematopoietic progenitor cells capable of indefinite self-renewal. T-cell ALL can be classified into several distinct genetic subgroups that correspond to specific T-cell development stages and is frequently associated with translocations of T-cell receptor genes on chromosome 14q11 or 7q34 with other gene partners. Early T-cell precursor (ETP) ALL comprises 12% to 15% of T-ALL and is characterized by immature genetic and immunophenotypic features (CD1a-negative, CD8-negative, and CD5-weak and the co-expression of stem-cell or myeloid markers), and gene expression reminiscent of double-negative 1 thymocyte that retains the ability to differentiate into both T-cell and myeloid, but not B-cell lineages. The discovery of its mutational spectrum recapitulated that of acute myeloid leukemia (AML) by whole genome sequencing (see discussion in Genomics of ALL) and global transcription profile similar to that of normal hematopoietic stem cells, and granulocyte macrophage precursors suggest that this subtype of leukemia is a stem-cell leukemia. Most B-lineage leukemias are early precursor B cells, expressing CD19 and CD10 (or cALLa, the common acute leukemia antigen) but lacking surface or cytoplasmic immunoglobulin. The mature B-cell ALL, or Burkitt cell leukemia, is stratified separately and not included in this chapter.

Genomics of Acute Lymphoblastic Leukemia

Most cases of ALL harbor driver genetic aberrations. In B-ALL in particular, a number of recurrent structural chromosomal anomalies have been identified that are of pathogenic and prognostic importance. Recent advances in global genome analysis have enabled the identification of recurring alterations in genes and pathways with key roles in cell growth and tumorigenesis. The identification of such lesions may help further refine risk stratification and could represent targetable genetic vulnerabilities.

B-Acute Lymphoblastic Leukemia Genomics

Chromosomal aneuploidy is common in B-ALL. Hyperdiploidy (modal chromosome number 51 to 65 or DNA index of >1.16) is present in 20% to 25% of pediatric patients with B-ALL, most commonly in younger children with decreasing frequency with age. Hyperdiploidy is associated with a more favorable outcome and is incorporated into the risk stratification algorithm of several childhood cancer consortia worldwide. Studies from the Pediatric Oncology Group (POG) and subsequently the Children’s Oncology Group (COG) have found that patients with trisomies of chromosomes 4 and 10 (double trisomies, DT) in particular (with or without meeting the definition of hyperdiploidy) have excellent outcomes with standard therapy, thus the COG incorporates DT as a favorable cytogenetic lesion in its risk stratification.

Hypodiploidy (modal chromosome number <44 or DNA index of <0.81) is a relatively rare (1% to 2%) cytogenetic lesion associated with a particularly poor outcome in B-ALL. While data based on small numbers suggests patients with hypodiploidy who clear bone marrow (BM) MRD by the end of induction therapy have reasonably good outcomes with chemotherapy alone, analysis of a larger COG cohort revealed the outcomes were overall dismal and not improved by HSCT in first complete response (CR). Thus, there is a dire need for novel therapeutic options for patients with hypodiploidy B-ALL.

Chromosomal translocations are also prevalent in B-ALL, a number of which have prognostic importance. Fusion of the ETV6 and RUNX1 genes by translocation t(12;21)(p12;q22) occurs in 20% to 25% of childhood B-ALL, and is associated with favorable outcome. Conversely, rearrangements of KMT2A on chromosome 11q23 are associated with poor prognosis. KMT2A rearrangements drive approximately 80% of infant B-ALL, and KMT2A r infant ALL is a particularly poor prognosis disease. KMT2A rearrangements are also found in 3% to 5% of older children with B-ALL, and while associated with increased MRD and inferior outcomes, the risk of relapse can largely be mitigated with intensification of chemotherapy.

Another recurrent structural chromosomal lesion conferring an increased relapse risk is intrachromosomal amplification of chromosome 21 (iAMP21) which is present in 1% to 3% of pediatric B-ALL cases. Similar to KMT2A rearrangements, outcomes are greatly improved with appropriate intensification of chemotherapy. The TCF3-HLF fusion gene results from another, rare (<1%) translocation in B-ALL, t(17;19). B-ALL patients with TCF3-HLF fusions characteristically present with coagulopathy and hypercalcemia and have a dismal prognosis, thus warranting novel therapies and consideration of HSCT in first CR.

The t(9;22) Philadelphia chromosome (Ph+) is another recurrent and well-characterized genomic driver of B-ALL. This translocation results in the BCR-ABL1 fusion gene. A number of potent and specific TKIs targeting BCR-ABL1 such as imatinib and dasatinib have been developed, and their continuous administration in combination with standard ALL chemotherapy has dramatically improved the outcome of Ph+ B-ALL.

In more recent years, gene expression studies of large B-ALL cohorts identified a subset of Ph-negative patients characterized by a gene expression profile highly similar to Ph+ B-ALL, subsequently designated Ph-like B-ALL (or BCR-ABL1 -like) B-ALL. Whereas Ph+ B-ALL steadily increases with age, the incidence of Ph-like disease peaks in young adulthood. In B-ALL ∼10% of children under 10 years, ∼20% of adolescents, and greater than 30% of young adults have Ph-like disease with rates then decreasing with increasing age. Ph-like B-ALL is associated with poor prognosis, particularly in the AYA (adolescent and young adult) population. B-ALL patients of Hispanic ethnicity and Native American genetic ancestry have significantly higher rates of Ph-like disease and an association has been identified between the risk of Ph-like B-ALL and specific germline polymorphisms of GATA3 . While Ph-like B-ALL lacks the BCR-ABL1 kinase fusions, a number of lesions driving Ph-like B ALL have been identified via seminal genomic studies. The identified causative lesions are characterized by the activation of a number of different kinase signaling pathways. Many of the activated kinase pathways can be selectively inhibited by targeted small molecule TKIs, thus providing a possible approach to improve the historically inferior outcomes of affected patients.

The most common subsets of Ph-like ALL are those with aberrantly increased expression of the gene, cytokine receptor-like factor 2 ( CRLF2 ) secondary to fusion with either IGH or P2RY8 . Thymic stromal lymphopoietin receptor (TSLPR) is the cell surface receptor protein encoded by CRLF2. As a cell surface protein, TSLPR is readily detected by flow cytometry . Thus, cell surface expression of TSLPR on diagnostic flow cytometry is suggestive of Ph-like disease with a driving CRLF2 fusion, though confirmatory testing by fluorescence in situ hybridization (FISH) is required to confirm the diagnosis. The signaling pathway responsible for the Ph-like gene expression signature in CRLF2 -rearranged B-ALL is the pro-proliferative, pro-survival, anti-apoptotic Janus kinase/signal transducers and activators of transcription (JAK)/STAT) pathway. TSLPR forms a heterodimeric cell surface receptor with IL7 receptor α (IL7Rα) which activates the JAK/STAT signaling pathway when bound by ligand. Overexpression of CRLF2 in Ph-like B-ALL results in constitutive JAK/STAT signaling hyperactivation. CRLF2 rearranged B-ALL is also characterized by frequent activating JAK1/2 mutations, highlighting the critical role of JAK signaling in disease pathogenesis. CRLF2 alterations are enriched in B-ALL patients of Hispanic ethnicity and occur in approximately 50% of the B-ALL cases in children with Down syndrome.

Another, less frequent subset of Ph-like B-ALL with aberrant JAK/STAT activation are those with rearrangements of the erythropoietin receptor gene ( EPOR ) that produce a truncated protein lacking the negative regulatory domain. Other, rare, JAK/STAT activating lesions are recurrently identified in Ph-like B-ALL, including IL7R insertion or deletions, fusions of JAK2 , and SH2B3 mutations/deletions. As JAK/STAT activation drives a large subset of Ph-like B-ALL, ongoing trials are testing if the JAK1/2 inhibitor, ruxolitinib, in combination with standard chemotherapy will be tolerable and efficacious for the treatment of JAK-activated Ph-like B-ALL (NCT02723994, NCT03117751, NCT02420717).

Another 3% to 5% of pediatric B-ALL cases have kinase fusions resulting in constitutive activation of ABL-class kinases including, ABL1/2 , PDGFRA/B, CSF1R , and LYN . ABL-class Ph-like disease may be effectively targeted by ABL kinase TKIs such as imatinib and dasatinib. The combination of ABL TKIs and chemotherapy is being explored for the treatment of ABL-class Ph-like disease (NCT02883049, NCT03117751, and NCT02420717). Rarely, lesions of other targetable activating signaling kinases such as FLT3, FGFR1 , and NTRK3 drive Ph-like A-ALL, suggesting the possibility to improve outcome with kinase inhibition.

The era of genome sequencing has essentially defined the landscape of genetic lesions driving B-ALL. Recent work has uncovered recurrent genomic lesions in pediatric B-ALL characterized by distinct gene expression programs and clinical characteristics, of potential prognostic and therapeutic relevance. These include cases with combined deregulation of DUX4 , a homeobox transcription factor, and ERG , a member of the erythroblast transforming-specific (ETS) transcription factor family (∼7%); rearrangement of the zinc-finger protein 384 gene ( ZNF384 ) (∼4% to 5%); and monocyte enhancer factor D2 (MEF2D ) rearrangements (∼3% to 4%). MEF2D -rearrangements are more common in older children, whereas ZNF384 rearrangements are enriched in younger B-ALL cases and tend to have weak CD10 expression and variable expression of myeloid and stem markers on immunophenotyping and gene expression analyses. DUX4-ERG dysregulated B-ALL has been associated with a favorable prognosis, and small studies suggest MEF2D -rearranged B-ALL may be associated with inferior outcomes. In B-ALL, MEF2D has multiple fusion partners, but all MEF2D fusions all are characterized by a gene expression profile with overexpression of HDAC9 which is a known MEF2D target. Thus, inhibition of HDACs may represent a viable therapeutic adjuvant for the treatment of affected individuals. Although no significant difference in outcome was found in patients with ZNF384 rearranged B-ALL as a whole, ZNF384 has multiple potential fusion partners ( TCF3, EWS1, CREBBP , and EP300 ) and outcomes may vary depending on the specific fusion. Patients with the TCF3-ZNF384 fusions are relatively resistant to corticosteroids and carry an increased risk of relapse. Prospective studies of larger, uniformly treated patients harboring these novel fusions will be necessary to precisely define their prognostic impact with modern era therapy.

Recent genomic studies of large cohorts of B-ALL patients have identified a number of collaborating genetic lesions with biologic and potentially prognostic and therapeutic significance. A single nucleotide polymorphism (SNP) array demonstrated substantial differences in the frequency of copy number abnormality (CNA) among various ALL subtypes. KMT2A -rearranged cases had less than one CNA per case, suggesting that KMT2A fusions are potent oncogenes that requires very few cooperating lesions to induce leukemia transformation, especially in infants, while other subtypes such as ETV6-RUNX1 and BCR-ABL1 leukemias have over eight lesions per case. Such SNP array studies have identified recurrent deletion of several transcription factors critical to normal B-cell development including PAX5, EBF1 , and IKZF1 . While deletion of each of these genes likely contributes to the development of B-ALL, only IKZF1 deletion appears to impact prognosis, with IKZF1 deletions and mutations being associated with a poor prognosis. Deletions or sequence mutation of the IKZF1 are present in 10% to 20% of pediatric B-ALL patients and are enriched in high-risk subsets including over 80% of Ph-positive ( BCR-ABL1 –positive) B-ALL cases and is common in Ph-like ALL. IKZF1 deletions are also common in DUX4/ERG dysregulated B-ALL, but in this context are not associated with a poor prognosis. Cooperative groups have attempted to incorporate IKZF1 status into risk stratification. The Associazione Italiana Ematologia ed Oncologia Pediatrica-Berlin-Frankfurt-Muenster (AIEOP-BFM) ALL 2000 trial demonstrated that patients with IKZF1 deletion with concomitant deletion of CDKN2A, CKN2B, PAR1 , or PAX5 without deletion of ERG (termed IKZF1 ^plus ) had a significantly worse 5-year event-free survival (EFS) than those with IKZF1 deletions without the IKZF1 ^plus status, and those without IKZF1 deletions (53 ± 6% vs 79 ± 5% vs 87 ± 1%, respectively).

Recent genome-wide association studies have identified germline polymorphic variants in several genes (including ARID5B , CEBPE , GATA3 , CDKN2A , BMI1-PIP4K2A , and IKZF1 ) that are associated with an increased risk of ALL or specific ALL subtypes. Rare germline mutations in PAX5 and ETV6 are linked to familial ALL. In addition to identifying common, low-penetrance susceptibility alleles, these data provide insights into disease causation by identifying risk variants annotating genes involved in transcriptional regulation and differentiation of B-cell progenitors.

T-Acute Lymphoblastic Leukemia Genomics

Unlike B-ALL, genomic classifiers are not widely used to risk stratify patients with T-ALL. However, some consortia are exploring possible integration of clinical features, MRD, and recurrent genomic lesions for refinement of risk-adapted therapy for T-ALL.

T-ALL is characterized by a number of structural chromosomal lesions and genetic mutations. Translocation involving transcription factors and epigenetic regulators (e.g., TAL1 , TLX1 , TLX3, KMT2A, MLLT10 ) are common. Deletions of CDKN2A/B and PTEN and mutations of signaling pathway and epigenetic regulators are also recurrent in T-ALL.

Additionally, over 70% of T-ALL patients have activating mutations of NOTCH1 , which encodes a transmembrane receptor and critical regulator of normal T-cell development and another 20% have loss-of-function mutations of FBXW7 , a ubiquitin ligase that negatively regulates NOTCH1. Given the frequency of NOTCH1 activating lesions in T-ALL, inhibition of NOTCH1 has been explored. Most notably, gamma secretase inhibitors (GSIs) which block the cleavage and release of the active intracellular component of NOTCH1 have been clinically tested, but demonstrated limited efficacy and excess rates of intolerable gastrointestinal toxicity.

Early T-Cell Precursor Acute Lymphoblastic Leukemia Genomics

Early T-cell precursor or early thymic precursor (ETP) ALL is a distinct subtype of T-ALL characterized by a stem-progenitor cell gene expression signature and an immature immunophenotype. While the first descriptions of ETP ALL found and associated with poor outcome, more recent data indicates that with appropriate risk-adapted therapy based on MRD outcomes of patients with ETP ALL are the same as those with non-ETP T-ALL.

Importantly, genomic profiling of ETP-ALL samples has identified several potentially targetable lesions. Most prominently, ETP-ALL is characterized by mutations of activated signaling including activating Ras-pathway and FLT3 mutations as well as epigenetic regulator mutations.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here