ACUTE LYMPHOBALSTIC LEUKEMIA

Read Also

Affecting about 4,000 patients each year in the United States, acute lymphoblastic leukemia (ALL) is the most common malignancy in children ¹⁰⁸. It is now curable in more than 80%  of children, but the overall cure rate is only 40%  in adult ALL, although much progress has been achieved during the past two decades ¹⁰⁸. Most of the current knowledge of the biology and treatment of ALL originates from studies of children, which is also the emphasis of the following dissuasion unless stated otherwise.

GENETICS

Molecular analysis has contributed greatly to our understanding of the pathogenesis and prognosis of ALL. Although the frequency of particular genetic subtypes differs in children and adults, the general mechanisms underlying the induction of ALL are similar. Chromosomal abnormalities represent a major genetic alteration in leukemia patients. They include the aberrant expression of proto-oncogenes, chromosomal translocations that create fusion genes encoding active kinases and altered transcription factors, and hyperdiploidy involving more than 50 chromosomes.

A number of leukemia subtypes have been defined based on their genetic make up. These subtypes  include B lineage leukemias that contain t(12;21)[TEL-AML1], t(9;22)[BCR-ABL], t(1;19)[E2A-PBX1], rearrangements in the MLL gene on chromosome 11, or a hyperdiploid karyotype and T lineage leukemias (T-ALL)¹⁰⁹. The most common translocation found in childhood B-precursor ALL is the t(12;21)(p13;q22), which causes TEL-AML1 translocations, presenting in approximately 25% of childhood ALL ¹¹⁰. Another common chromosomal aberration found in B-precursor ALL is the presence of more than 46 chromosomes (hyperdiploid ALL) ¹⁰⁹. Mechanism of leukemogenesis in hyperdiploid ALL is unknown while activating mutations in the receptor tyrosine kinase FLT3 were identified in approximately 20% of hyperdiploid ALL ^111,112. The t(1;19)(q23;p13) encoding the E2A-PBX fusion protein is present in about 6% of all B-precursor ALLs and in 25% of cases with a preB immunophenotype ^113,114. E2A contains a bHLH domain through which to bind sequence specific DNA binding and dimerization, and it plays a critical role in lymphocyte development ¹¹⁵. Loss of E2A function and dysregulation of HOX by PBX1 may contribute to leukemogenesis in this subtype of ALL ¹¹⁶. t(9;22)[BCR-ABL] is more common in adult ALL while is only detected in 4% childhood ALL ¹¹⁰.

Transcription factor genes, such as bHLH genes MYC ¹¹⁷, TAL1(SCL) ¹¹⁸,  and LYL1 ¹¹⁹, are the preferred targets of chromosomal translocations in the acute T-cell leukemias ¹¹⁰. When rearranged near enhancers within the TCRβ or α/δ-chain locus, these regulatory genes become active, and their protein products inappropriately enhance target gene transcription. In addition to genes encoding bHLH proteins, additional classes of regulatory genes are rearranged near TCR loci, including those encoding the proteins LMO1 and LMO2 within the cysteine-rich LIM family ^120,121. Mechanisms of leukemia transformation by LIM proteins is unclear, but LMO1 showed transformation capability in the transgenic mice model and LMO2 can bind to the bHLH protein TAL1 in vitro ^122,123. HOX11 and HOX11L2 are the two major HOX genes that are inappropriately placed under the control of TCR loci ¹¹⁰. About 20% of childhood T-ALL patients demonstrated a HOX11L2 gene translocation by fluorescence in situ hybridization ¹²⁴. NOTCH1 is a gene that normally encodes a transmembrane receptor that is involved in the regulation of normal T-cell development and may other tissues during embryologic development. NOTCH-1 activation by truncation had previously been shown in a rare t(7;9) T-cell ALL  ¹²⁵ and the same truncated fragment was shown to induce T-cell ALL in mouse models ^126,127. It’s suggested that specific mutations in sequences encoding both the heterodimerization and PEST domains of NOTCH1  exist in over 50% of primary patient T-cell ALL samples ¹²⁸.

These genetic alterations contribute to the leukemic transformation of hematopoietic stem cells or their committed progenitors by changing cellular functions. They alter key regulatory processes by maintaining or enhancing an unlimited capacity for self-renewal, subverting the controls of normal proliferation, blocking differentiation, and promoting resistance to death signals (apoptosis).

TREATMENT

Since ALL is a heterogeneous disease, its treatment is more often individualized according to genotype, phenotype and risk factors ¹⁰⁸. Mature B-cell ALL is the only subtype that is treated with short-term intensive chemotherapy ^129,130. For all other ALL patients, three stages of treatment are included: the initial remission-induction therapy followed by intensification (or consolidation) therapy and continuation treatment to eliminate residual leukemia ¹⁰⁸. The remission-induction therapy includes the administration of a glucocorticoid (prednisone, prednisolone, or dexamethasone), vincristine, and at least one other agent (usually asparaginase, an anthracycline, or both). This treatment phase eradicates more than 99%  of the initial burden of leukemia cells and restores normal hematopoiesis and a normal performance status. When normal hematopoiesis is restored, patients then receive intensification therapy which includes high-dose methotrexate with mercaptopurine, high-dose asparaginase given for an extended period, and reinduction treatment. Allogeneic transplantation is the ultimate form of treatment intensification and has been indicated to benefit certain very-high-risk pediatric and adult patients, such as those with BCR-ABL+ ALL or those with a poor initial response to treatment ^131,132. A combination of methotrexate administered weekly and mercaptopurine given daily forms the most common continuation regimens. Although as many as two-thirds of childhood cases may be curable with only 12 months of treatment, it is not possible to reliably identify this subgroup prospectively ¹³³. Therefore, ALL patients generally receive prolonged continuation therapy to ensure maximum cure rate ¹³⁴. Therapy directly acting at the central nervous system is critical for treatment success, since it is often the place harboring residual leukemia cells after chemotherapy, thereby contributing to relapse ¹³⁵.

Biological differences in leukemogenesis between adult and childhood ALL contribute to the differential prognosis. Childhood and adult ALL differ markedly in the prevalences of various cytogenetic abnormalities. For example, Philadelphia chromosome (Ph)-positive ALL, a high-risk cytogenetic subset, accounts for one quarter of adult ALL cases but occurs in less than 5% of children. Similarly, ETV6/RUNX1 (TEL-AML1) fusion and hyperdiploidy, both of which are good risk genetic features, together comprise about 50% of childhood ALL, but only about 10% of adult ALL ¹³⁶. Age influences the prognostic effect of the same genetic lesions. Among patients with t (9;22), children one to nine years of age have a better prognosis than adolescents with the same disease ¹³⁷, who in turn fare better than adults ^138,139. Among patients with MLL-AF4 fusion, infants fare considerably worse than older children, and adults have an especially poor outcome ^138,140. In T-cell ALL, the presence of t (11;19) with MLL-ENL fusion and overexpression of the HOX11 gene confer a good prognosis.

GLUCOCORTICOID INDUCED CELL DEATH

Glucocorticoids (GC) such as dexamethasone (Dex) induce cell apoptosis and cell cycle arrest in thymocytes and some leukemia (Fig.1.3). Although the mechanism of GC-induced cell death is still elusive, the integrity of the GC signaling pathway, including DNA binding of the GR and subsequent transcriptional regulation of specific genes, appears to be important for its pro-apoptosis effect. The gene for the GR is located on chromosome 5 (5q31) and it consists of nine exons encoding three characteristic domains of the GC protein ¹⁴¹. The N-terminal domain (NTD) contains a transactivation domain (AF-1) that is involved in transcriptional activation of target genes ^142,143. The DNA binding domain (DBD) is in the middle, and it consists of two highly conserved zinc finger domains essential for binding to the glucocorticoid response element (GRE) sequences of regulated genes. The C-terminal of the GR protein contains the ligand binding domain (HBD) that is also required for binding heat-shock proteins and GR dimerization 

GC-induced apoptosis depends on sufficient levels of GR and subsequent alterations in gene expression. However, basal level GR expression is not enough to mediate GC-induced apoptosis, and positive autoregulation is a necessary component of this process in leukemia cell lines ¹⁴⁴. Although non-genomic actions of the GR are not excluded, most data suggest that GC induced apoptosis is linked to de novo gene expression. The target genes whose transactivation or transrepression initiates apoptosis remain unclear. Expression profiling suggested that GCs modulate the expression of distinct sets of genes, rather than causing generalized transcriptional alterations ^145-149. Expression of many components in the extrinsic and intrinsic death pathways are altered in a pro-apoptotic manner upon treatment of sensitive lymphoid cells with GCs ¹⁴³.  

The extrinsic pathway is initiated by death receptors of the tumor necrosis factor (TNF) receptor superfamily such as CD95 (APO-1/Fas) or TNF-related apoptosis-inducing ligand (TRAIL) ¹⁵⁰. In this pathway, caspase-8 activates downstream effector caspases such as caspase 3 through either mitochondria-dependent or mitochondria-independent pathways ¹⁵¹. In the former, Bid, a pro-apoptotic member of the Bcl-2 family, is activated and triggers cytochrome c release from mitochondria. Cytochrome c binds APAF-1 and then activates caspase-9, which in turn activates effector caspases. In the mitochondria-independent pathway, caspase-8 directly cleaves and activates caspase-3, bypassing mitochondria and cytochrome c release. GC-induced thymocyte apoptosis was unaffected in Bid-deficient mice, indicating that the death receptor pathway doesn’t play an essential role in GC-mediated cell death ¹⁵².

The intrinsic and mitochondria-mediated pathway responds to intracellular signals such as GC. This leads to release of pro-apoptotic molecules upon depolarization of the mitochondrial membrane potential ¹⁵³. The apoptotic response is tightly regulated by the interaction between pro- and anti-apoptotic Bcl-2 family members. Pro-apoptotic factors  include Bim, Bid, Bad and Puma, which activate Bax and Bak,  and anti-apoptotic Bcl-2 family members consist of Bcl-2 and Bcl-x_L, which bind and neutralize their pro-apoptotic counterparts ^153-155. Following formation of pores in the outer mitochondrial membrane by Bax and Bak, cytochrome c and other factors such as Smac/Diabolo are released into the cytosol. Cytochrome c triggers caspase-3 activation through formation of the cytochrome c/Apaf-1/caspase-9-containing apoptosome, which finally leads to apoptosis ¹⁵⁶. Smac/Diabolo promotes caspase activation through neutralizing the inhibitory effects of IAPs ¹⁵⁷. Thus, apoptosis is tightly controlled by the balance between anti- and pro-apoptotic Bcl-2 family members together with the caspase-inhibitory IAP molecules.

The mitochondrial apoptosis pathways have been shown to be important for GC-induced apoptosis ¹⁵⁸. GC-treatment induces loss of mitochondrial membrane potential in thymocytes and leukemic T cells,  which is prevented by caspase-9 deficiency ^159-162. Another essential principle for GC-induced mitochondrial apoptosis may be induction of proapoptotic ^163,164 and repression of antiapoptotic ^165,166 Bcl-2 family proteins, leading to transcriptional deregulation of the Bcl-2 rheostat. For example, Bim-xL, a proapoptotic member of the Bcl-2 family, was noted to be induced by GCs at 20 hours and is thought to precipitate apoptosis ¹⁶⁷. GC-regulated proapoptotic and BH3-only proteins Bim and Puma have been shown to be key initiators of GC-induced apoptosis in vivo ⁵⁴. Bcl-2-deficient or Bcl-XL-deficient mice display lymphoid apoptosis in vivo and enhanced cell death of thymocytes in vitro after GC-treatment ^155,168.

A number of signaling pathways are known to modulate the complex mechanism of GC-induced apoptosis. It is likely that the balance between pro-survival and pro-apoptotic signaling pathways determines the ultimate fate of the cell. These include GC-mediated repression of the proto-oncogene c-myc or induction of IκB, an inhibitor of the survival transcription factor NF-κB ^169,170. NF-κB is a heterodimeric transcription factor for more than 100 genes including cytokines, cytokine receptors, chemotactic proteins, and adhesion molecules ^171,172, and is involved in the regulation of apoptosis ^173,174. In human leukemic T cells, GCs induce synthesis of IκBα, which causes the retention of NF-κB in the cytoplasm and correlates with the induction of apoptosis ¹⁷⁵. GR can also directly repress NF-κB activity ¹⁷⁶. Another example of how GC-induced apoptosis is regulated by crosstalk with other signaling pathways involves protein kinase C (PKC). PKC includes several subfamilies of enzymes including Ca²⁺-dependent PKC and Ca²⁺-independent PKC. It has been shown that the Ca²⁺-independent PKC subfamily is involved in induction of apoptosis, whereas activation of Ca²⁺-dependent PKC is capable of inhibiting GC-induced apoptosis ¹⁷⁷. It has been shown that sustained expression of the protooncogene c-myc provides protection against GC-induced cell death in the human leukemia cell line CEM-C7, and the downregulation of c-myc accompanies induction of apoptosis by GCs ¹⁷⁸. However, recently it was found that repression of c-myc is essential for cell cycle arrest in the G1 phase, but is not required for GC-induced apoptosis ¹⁷⁹. GC-mediated cell death may also involve GR-dependent repression of MAP kinase phosphatase-1 (MKP-1) and subsequent activation of the proapoptotic JNK pathway, which can be inhibited by Rapamycin ^180,181.

GLUCOCORTICOID RESISTANCE

The mechanism of GC resistance in some human leukemia and lymphoma cells is poorly understood. Cancer cells can employ multiple strategies that ultimately evade apoptosis following chemotherapy. General mechanisms of resistance include decreased drug uptake, increased drug efflux, alterations in the drug target, drug metabolism, repair of DNA damage, cell cycle checkpoint mediators, and changes in downstream mediators of the apoptotic pathway ¹⁸².

Since the integrity of the GC signaling pathway is critical for GC induced apoptosis, disruption of any step of this pathway could lead to resistance to GC. With supporting experimental evidence, many underlying reasons for GC-resistance have been suggested and include a low number of GR, mutations of the GR, expression of different GR splice variants, different phosphorylation patterns of GR, multidrug resistance by overexpression of P-glycoprotein, increased levels of glutathione and glutathione S-transferase, abnormal expression of GR binding proteins, dysregulation of transcription factors such as NF-κB, AP-1, dysregulation of GR target genes such as c-myc, autoregulation of GR itself, or dysregulation of members of the apoptosis pathway, such as anti-apoptotic expression of Bcl-2 family proteins ¹⁵⁸. Depending on molecular components affected in the GC induced signaling pathway, the mechanisms of GC resistance may be grouped into ‘upstream’ and ‘downstream’ levels ¹⁸³. The ‘upstream mechanisms’ concern the GC, GR and GR-associated proteins that control its function and the ‘downstream mechanisms’ interfere with individual GC responses, such as transcriptional expression and induction of apoptosis.

Overexpression of the mdr-1 gene-encoded P-glycoproteins, which are ATP-binding cassette (ABC) transporters that pump various drugs out of the cell, has been suggested to account for cross-resistance to GC resulting after exposure to combination chemotherapy ^184-186. GR expression levels have been correlated to GC sensitivity in several experimental systems. Since GR is a target gene of GC signaling, GR auto-induction but not the basal GR expression was shown to be critical for GC-induced apoptosis ^187,188.  As a transcriptional factor, GR recruits a number of cofactors such as SRC-1, TIF2/GRIP1, CBP/p300, NcoR and SMRT required for its gene regulatory activities ^189,190. Mutation or abnormal expression of these cofactors compromises transcriptional activities of GR and therefore reduces sensitivity to GC. Inefficient expression of GC target genes might contribute to GC resistance. A number of such genes, including c-myc ^191,192, IkB ^175,193, and c-jun ^187,194, have been indicated to be required for GC induced apoptosis in some leukemia cell lines. However, their role in GC resistance is still controversial based on reports from different research groups.

GC resistance might also result from deregulation of the apoptotic effector machinery, which leads to an imbalance of pro-apoptosis and anti-apoptosis forces in the cells. Upregulation of cell survival signals inhibits GC-induced apoptosis, resulting in resistance to GC-induced apoptosis in leukemia therapy ¹⁹⁵. Activation of MAPK kinase (MEK) and extracellular signal-regulated kinase (ERK) antagonizes GC-induced apoptosis in CD4+ T cells ¹⁹⁶. Also, other protein tyrosine kinases are deregulated in hematological malignancies ¹⁹⁷, which in turn activate the Ras/Raf/MEK, NF-κB, PI3-K/AKT and β-catenin survival pathways. AKT inhibitors increased the sensitivity of a follicular lymphoma cell line to GC-induced apoptosis by inducing Bad translocation to the mitochondria ¹⁹⁸.

Together, resistance towards GC-induced apoptosis may concern defects in the GR itself, GR binding partners, dysregulation of GR target genes and transcription factors, or it may be due to activity of general resistance mechanisms, defects in the apoptosis pathway, or a shift of the balance of cellular signaling pathways to anti-apoptotic signaling. In line with the latter case, it is a consistent finding that GC sometimes show paradox regulations, such as regulation of some apoptosis genes in a pro-apoptotic manner in situations where GC promote survival and vice versa ¹⁹⁹. Thus, pro- and anti-apoptotic signaling may occur in parallel within a single cell but the balance between survival and suicide may shift to pro-apoptotic signals upon a strong apoptosis signal and vice versa.