Pathobiology of Acute Myeloid Leukemia


Acute myeloid leukemia (AML) is a cancer of hematopoietic stem/progenitor cells, characterized by recurrent genetic and epigenetic alterations. Historically, human leukemias were distinguished according to clinical and histological features and subsequently by morphology. Analysis of the AML genome at increasing resolution, from the level of whole chromosomal changes to individual base pairs, together with an appreciation of epigenetic changes and interactions within the bone marrow microenvironment, has furthered the understanding of the biology and clinical behavior of this disease. Characterizing the patient-specific biology of AML is increasingly important both for understanding the origins of disease and also to guide therapeutic decision-making.

Phenotype of Acute Myeloid Leukemia

Normal hematopoiesis is characterized by self-renewal and differentiation of long-term hematopoietic stem cells (HSCs) to short-term HSCs, multipotent progenitors, and common lymphoid and myeloid progenitors (see Chapter 9 ). These lineage-committed progenitors further differentiate to mature lymphoid or myeloid cells, including erythrocytes, granulocytes, macrophages, and platelets. This process is regulated by lineage-specific transcription factors at key points during normal hematopoiesis. Functional analysis of recurrent chromosomal, molecular, and epigenetic alterations in AML has revealed that many of these lesions corrupt hematopoietic differentiation programs, impacting proliferation, survival, and maturation of myeloid progenitor cells. Consequently, a hallmark of the AML phenotype is an accumulation of immature myeloid precursors. A myeloblast count of 20% or greater distinguishes AML from other myeloid malignancies.

Historically, leukemias were classified according to morphologic criteria using the French-American-British (FAB) classification, first proposed in 1976. This system classifies AML by the extent of maturation and lineage specificity, ranging from M0 (undifferentiated) to M3 (promyelocytic), M4eo (myelomonocytic with eosinophilia), M6 (erythroid), or M7 (megakaryocytic). Certain FAB subtypes were subsequently found to correlate with underlying cytogenetic abnormalities, notably M3 with t(15;17), M4eo with inv(16), and M2 with t(8;21). As greater understanding of recurrent cytogenetic and now molecular and epigenetic aberrations has developed, it has become clear that morphology alone is inadequate to fully describe the disease spectrum of AML.

Etiology of Acute Myeloid Leukemia

The majority of patients who develop AML lack any recognized antecedent condition, and for these patients their disease is classified as sporadic or de novo. In these cases, AML appears to result from the accumulation of spontaneously acquired somatic mutations in self-renewing hematopoietic cells. In contrast, a diagnosis of secondary AML (sAML) is preceded by a known predisposing condition, including environmental exposure, antecedent hematologic malignancy, or inherited factors.

Toxins and Exposures

A number of environmental, occupational, and iatrogenic exposures have been identified that contribute to sAML via genotoxic damage to hematopoietic cells. Exposure to benzene, an organic component of many commonly used chemicals including plastics, dyes, pesticides, solvents, and petroleum products, has been linked to the subsequent development of AML. This relationship was identified in the 19th century, when bone marrow aplasia and myeloid leukemia were noted among workers exposed to benzene-containing chemicals. Individuals with occupational benzene exposure have an approximately threefold increased relative risk of developing AML. Workplace benzene exposures have decreased significantly since this discovery, but other sources of benzene exposure remain a concern (e.g., through cigarette smoking). Although cytopenias can occur within months of benzene exposure, there is a latency of several years between benzene exposure and the development of leukemia.

Cases of AML arising after chemotherapy or radiation have been historically designated as therapy-related AML (tAML). Two classes of chemotherapy drugs in particular are associated with an increased risk of tAML ( Table 59.1 ); the first class of drugs with clear links to tAML are the topoisomerase II inhibitors. The most commonly used topoisomerase II inhibitors are anthracyclines, such as doxorubicin, idarubicin, and daunorubicin, and the epipodophyllotoxin etoposide, which are critical components of many treatment regimens for both solid tumors and hematologic malignancies. Topoisomerase II is an adenosine triphosphate (ATP)-dependent enzyme that religates deoxyribonucleic acid (DNA) at sites of double-strand breaks to manage supercoils; inhibition of this enzyme increases the number of double-strand breaks. Resolution of these double-strand breaks may occur via error-prone nonhomologous end joining, resulting in accumulation of DNA damage or apoptotic cell death. tAML arising after exposure to topoisomerase II inhibitors typically occurs with a latency of 1 to 3 years and is often characterized by balanced chromosomal translocations, with the majority involving the KMT2A (mixed-lineage leukemia [ MLL1 ]) locus on chromosome 11q23. Typical lesions are reciprocal translocations such as t(9;11)(p21;q23) and t(11;19)(q23;p13). Other rearrangements that occur in de novo AML also occur in tAML after topoisomerase II inhibitors, including t(15;17), t(8;21), and inv(16). The risk of tAML varies based on the chemotherapy dosing schedule, cumulative dose received, additional cytotoxic agents, and underlying disease characteristics but generally does not exceed 5% of patients treated with topoisomerase II inhibitors.

Table 59.1
Characteristic Features of Therapy-Related Acute Myeloid Leukemia
Class Examples Latency Associated Phenotypes
Topoisomerase II inhibitors Doxorubicin, idarubicin, daunorubicin, etoposide, mitoxantrone 1–3 years after chemotherapy Balanced translocations, often with KMT2A ( MLL1 ), or with t(15;17), inv(16), or t(8;21)
Alkylating agents Cyclophosphamide, ifosfamide, chlorambucil, busulfan, carmustine, dacarbazine ~5–7 years after chemotherapy Monosomal karyotype (−7, del5q, −5, etc.), TP53 mutations are common, underlying dysplastic features
Irradiation Therapeutic radiation typically involving bone marrow, nuclear accidents, radioiodine ~5–7 years Often similar to alkylating agents
Immunosuppressive agents Azathioprine ~8 years Abnormalities in chromosomes 5 and 7, possibly related to diminished immune surveillance
PARP inhibitors Olaparib, rucaparib, niraparib, talozoparib Poorly defined Limited to date. Possible increased risk in patients with germline BRCA1/2 mutations
MLL , Mixed-lineage leukemia; PARP , poly (ADP-ribose) polymerase.

Alkylating agents are the second class of chemotherapy drugs with a clear role in the pathogenesis of tAML. The first leukemogenic agents identified in this category were nitrogen mustards. Frequently implicated drugs in contemporary clinical practice include cyclophosphamide, ifosfamide, and melphalan; weaker associations have been described with other alkylating agents such as busulfan, thiotepa, and cisplatin. Alkylating agents create adducts in DNA bases, which are variably mutagenic or cytotoxic. Cytogenetic lesions in alkylator-associated tAML are typically unbalanced, including loss of the long arms of chromosomes 5 or 7 [del(5q), del(7q)], or complete loss of these chromosomes (−5, −7). The risk of tAML following alkylator exposure is up to 1% per year but typically has a longer latency (5 to 7 years) compared with topoisomerase II-associated tAML. The risk increases with age and cumulative exposure to these agents. In some cases of tAML, small clonal populations harboring TP53 mutations antedate chemotherapy exposure. TP53 deficiency may confer enhanced fitness on these clones, allowing them to expand under the selective pressure of therapy.

Several other therapies have also been implicated as risk factors for tAML, including some immunosuppressive therapies such as azathioprine, while other new associations continue to be investigated—for instance, possible tAML risk associated with poly (ADP-ribose) polymerase (PARP) inhibitors, particularly in patients with germline BRCA1/2 mutations.

Exposure to ionizing radiation also has been identified as a causative mechanism for tAML. This relationship was identified in the context of occupational exposures during the development of radiography and subsequently in the setting of mass exposures such as the atomic bomb detonations or nuclear power plant disasters, where a time-limited spike in leukemia incidence occurred following the event. Outside of these events, therapeutic radiation therapy represents the most common setting for significant radiation exposure, which is associated with a small but significant increase in tAML risk and likely varies depending on the site and dose of directed radiation therapy. Radiation-associated tAML is characterized by an increased frequency of mutations otherwise implicated in de novo AML pathogenesis—for instance, mutations in RUNX1 , as well as balanced translocations such as RUNX1-RUNX1T1 and DEK-NUP214 —suggesting some selectivity in the patterns of DNA damage.

Prior Hematologic Malignancy

Other myeloid malignancies, including myelodysplastic syndromes (MDSs) and myeloproliferative neoplasms (MPNs), carry a risk of disease evolution to sAML. The risk varies depending upon the underlying disease and may be facilitated by certain exposures, including genotoxic chemotherapy.

Patients with MPNs have an approximately 10% risk of evolution to AML (MPN in blast phase [MPN-BP]) over 10 years, which varies according to the underlying disease. The risk is lowest in essential thrombocythemia and may be as high as 20% for myelofibrosis. There is a clear association between chemotherapies used in treating MPNs, specifically alkylating agents and radioactive phosphorus, and leukemic evolution; treatment with these agents results in a threefold to fourfold increase in progression to MPN-BP. Another mechanism that may contribute to clonal evolution and disease progression may be a chronic inflammatory state related to the underlying MPN. Sequencing of sAML cases developing in the background of an MPN has identified recurrent mutations in TET2 , JAK2 , IDH , IKZF1 , and ASXL1 . Moreover, a number of patients with a JAK2 -mutated MPN may progress to MPN-BP that is JAK2 wild-type, thought to arise either from a common pre- JAK2 founding clone, or due to parallel expansion of a distinct hematopoietic clone. MPN-BP with mutated JAK2 typically proceeds through an accelerated myelofibrosis phase, whereas MPN-BP that no longer harbors a JAK2 mutation tends to arise from chronic phase disease and may be associated with the use of cytotoxic therapies (see Chapter 72 ).

Prior to the introduction of tyrosine kinase inhibitors (TKIs) for chronic myeloid leukemia (CML), patients with CML typically progressed from chronic phase to blast phase within 5 years, at a rate of more than 20% per year (see Chapter 69 ). Most cases of blast phase CML have a myeloid phenotype, whereas approximately 30% of patients have a lymphoid phenotype. Additional mutations may occur during transformation of CML, and approximately 80% of patients have additional cytogenetic abnormalities, such as duplication of the Philadelphia chromosome, and other trisomies that are recurrent in de novo AML. Up to one third of patients with CML in myeloid blast phase harbor mutations in the tumor suppressor genes P16 or TP53 . In addition, BCR-ABL signaling upregulates transcription factors implicated in AML pathogenesis, including HOXA9 and EVI1 , which may contribute to leukemic transformation. The rate of transformation to blast phase CML in the TKI era has decreased markedly to approximately 1% per year, which underscores the continued requirement for BCR-ABL1 signaling in CML evolution.

Approximately one-third of patients with MDS progress to sAML, although this varies significantly according to the underlying MDS subtype and disease characteristics, including the percentage of bone marrow blasts, presence of characteristic cytogenetic abnormalities, and cytopenias (see Chapter 61 ). Progression to leukemia is associated with acquisition of additional somatic mutations as well as epigenetic alterations within the MDS clone. Mutations in transcription factors and cytokine signaling genes, including RUNX1 , NRAS , and ETV6 , are more common at progression to sAML, compared with the frequency of these mutations at MDS diagnosis. Epigenetic modifications of the MDS genome appear to also play a significant role in AML progression, particularly through DNA methylation–mediated silencing of tumor suppressor genes. These alterations are also enriched in the subtype of AML with myelodysplasia-related changes, even where a prior MDS diagnosis was not known.

Congenital Bone Marrow Failure Syndromes

A number of inherited bone marrow failure syndromes are associated with an increased risk of developing advanced myeloid malignancies (see Chapter 30 ). This may be due to the proliferative stress imposed by chronic cytopenias or defects in DNA repair that are hallmarks of several of these syndromes. These syndromes are reviewed in detail in Chapter 30 .

Fanconi anemia (FA) is the most common inherited bone marrow failure disorder and is caused by germline mutations in factors involved in DNA repair. These disorders have an autosomal recessive inheritance pattern except for FANCB , which is X-linked. To date, more than 20 genes have been identified as a part of the FANC gene family, and together their protein products are responsible for identifying DNA damage and targeting these sites for repair. The cumulative risk of AML or MDS among FA patients is approximately 10% to 15%, with peak incidence during the teenage years.

Dyskeratosis congenita (DKC) is a bone marrow failure syndrome characterized by inherited mutations in the telomere maintenance pathway. DKC can be inherited in an autosomal dominant (Online Mendelian Inheritance in Man [OMIM] 127550), autosomal recessive (OMIM 224230), or X-linked recessive pattern (OMIM 305000). Mutations in TERT , DKC1 , TERC , or TINF2 account for most cases. These patients will typically develop bone marrow failure by 20 to 30 years of age, although variable penetrance of short telomere syndromes may result in later onset of leukemia—particularly in those with a telomere mutation but without the DKC “triad” phenotype of skin hyperpigmentation, nail dystrophy, and oral leukoplakia. Transformation to AML occurs in approximately 10% of patients and is thought to occur via genomic instability related to shortened telomeres and associated DNA damage, resulting in dysplasia and an increased risk of hematopoietic malignancy.

Shwachman-Diamond syndrome (OMIM 260400) is an autosomal recessive disorder caused by mutations in SBDS . AML or MDS occurs in up to a third of patients by 30 years of age and is thought to relate to chromosomal instability and accelerated rates of apoptosis, which may be due to the role of SBDS in stabilizing the mitotic spindle during mitosis.

Severe congenital neutropenia or Kostmann syndrome is associated with neutropenia at birth and has been associated with a variety of genetic mutations. The pattern of inheritance can be autosomal dominant ( ELANE or GFI1 ), autosomal recessive ( HAX1 , G6PC3 , VPS45 , or JAGN1 ), or X-linked ( WAS ). Nearly a third of patients develop AML or MDS during adolescence. Transformation into AML is frequently characterized by the acquisition of somatic mutations in CSF3R , which encodes the granulocyte colony-stimulating factor (G-CSF) receptor. The causal relationship to chronic G-CSF therapy remains controversial.

Diamond-Blackfan anemia (DBA) (OMIM 105650) is characterized by red cell aplasia and typically spares the leukocyte and platelet lineages. DBA is typically inherited in an autosomal dominant fashion and is associated with mutations in a number of ribosomal proteins. AML can occur in up to 20% of patients and typically occurs after 40 years of age.

Congenital amegakaryocytic thrombocytopenia (CAMT) (OMIM 604498) and thrombocytopenia with absent radii (TAR) (OMIM 274000) syndrome are both characterized by hypoplastic thrombocytopenia. CAMT is inherited in an autosomal recessive manner via mutations in the MPL gene, which encodes the receptor for thrombopoietin (TPO). CAMT is associated with an increased incidence of AML, typically in the second decade of life. TAR syndrome has been associated with mutations in RBM8A , which is involved in messenger RNA (mRNA) splicing. The thrombocytopenia in TAR syndrome often improves over time; both acute lymphoblastic leukemia and AML have been reported among patients with this rare disorder.

Down syndrome, caused by trisomy 21, is associated with an approximately 10- to 20-fold elevated relative risk of AML and MDS compared with the general population and in particular an increased risk for acute megakaryocytic leukemia, FAB M7 (see Chapter 29, Chapter 30 ). Infants with Down syndrome may experience transient abnormal myelopoiesis (TAM), where circulating peripheral blood blasts are seen and may be accompanied by hepatic dysfunction, effusions, and rash; this occurs in approximately 10% of these patients. The majority of TAM cases harbor somatic mutations in GATA1 , resulting in altered function of this transcription factor that plays an important role in hematopoietic cell maturation, particularly in the megakaryocyte lineage. Indeed, up to 30% of persons with TAM will progress to AML, commonly acute megakaryocytic leukemia. The development of AML in patients with Down syndrome likely relates both to acquired somatic mutations, such as GATA1 , and also the presence of additional copies of genes on chromosome 21 that facilitate leukemogenesis, such as the oncogenes RUNX1 , ERG , and ETS2 .

Mendelian Acute Myeloid Leukemia Predisposition Syndromes

A number of genes that are targets of recurrent somatic mutation in AML are also mutated in the germline in families with predisposition to myeloid malignancy without a prodrome of bone marrow failure. These include mutations in RUNX1 , CEBPA , DDX41 , and GATA2 . Predisposition to AML is also associated with germline variants in ANKRD26 , SRP72 , ETV6 , and SAMD9 / SAMD9L . Because gene panel testing (see box on Molecular Diagnostics in Acute Myeloid Leukemia ) of unpaired tumor samples cannot reliably discriminate germline from somatic mutations, detection of a known pathogenic mutation in one of these genes in a bone marrow or peripheral blood sample should prompt referral for genetic counseling and germline testing. Although these familial syndromes are rare, they are important to recognize because affected individuals and asymptomatic carriers require specific clinical management. In particular, asymptomatic carriers require counseling about risks to themselves and their offspring, and they should be deferred as graft donors for allogeneic stem cell transplantation. The prevalence and risk of MDS/AML development has not been defined for most of these syndromes due to ascertainment bias in most of the reported series ( Table 59.2 ).

Molecular Diagnostics In Acute Myeloid Leukemia

The various mutations and dysregulated pathways integral to the pathobiology of acute myeloid leukemia (AML) also act as specific therapeutic targets and offer clear prognostic implications for patients. As such, molecular testing at the time of AML diagnosis has become the standard of care, to assist with subclassification of disease, risk stratification, selection of an induction regimen, and consolidation preferences. Diagnosis of AML requires analysis of a specimen demonstrating excess myeloblasts. For patients with high levels of circulating disease, some studies may be performed with the use of peripheral blood samples; however, assessment of a bone marrow biopsy and aspirate is essential.

Standard Evaluation

The cornerstone of the diagnosis of AML remains morphologic assessment. In the current World Health Organization (WHO) guidelines, at least 20% of the bone marrow cellularity or peripheral blood must be composed of myeloblasts, except in the presence of the t(8;21), t(16;16)/inv(16), or t(15;17) rearrangements, which are sufficient for an AML diagnosis regardless of blast count. In addition, promonocytes in acute monocytic leukemia, megakaryoblasts in acute megakaryocytic leukemia, and abnormal promyelocytes in acute promyelocytic leukemia are added to the blast percentage. Only in pure erythroleukemia are erythroblasts included in the blast count.

Flow cytometry uses multiparametric analysis of single cells to assess cellular granularity and size, cell surface, intracellular antigen expression, and other features. The coexpression of certain cell surface markers may help to confirm myeloid cell origin, identify immature blasts, typically CD34 + and CD117 + , and also distinguish an aberrant phenotype of a leukemic blast population. Flow cytometry can enumerate small populations of leukemic cells, below the limit of detection by morphology. For this reason, flow cytometric analysis has been developed as a platform to monitor minimal residual disease (MRD).

A critical element in the initial laboratory assessment of AML is cytogenetic analysis of a bone marrow aspirate specimen. This provides important prognostic data for risk stratification and informs therapeutic strategies. Cells from the aspirate are cultured, mitosis is interrupted, and the paired chromosomes are arranged to identify missing, translocated, or duplicated segments. Fluorescence in situ hybridization uses fluorescently labeled DNA probes and can identify gains and losses of chromosomal material, as well as rearrangements that may be cryptic using conventional banding techniques, and may complement, but not replace, standard karyotyping.

Detection of somatic mutations that are known drivers of AML biology can aid in the initial risk classification of patients with AML, particularly those with intermediate-risk cytogenetics. Current National Comprehensive Cancer Network guidelines recommend testing for recurrent gene mutations ( KIT , FLT3, NPM1 , IDH1, IDH2, TP53 , and CEBPA ) at diagnosis because their prognostic and therapeutic significance has been validated in large cohorts (level 2 A evidence). Testing for mutations in RUNX1 and ASXL1 is recommended for WHO classification. In addition, use of targeted therapies is increasingly dependent on detection of a specific tumor genotype. Multigene panels that test for recurrently mutated genes in AML and other myeloid malignancies are increasingly used for these purposes (later).

Investigational Testing

Given the rapidly expanding number of genes that are recognized targets of recurrent somatic mutation in AML, more comprehensive mutational profiling is increasingly part of routine clinical practice. With increasing numbers of genes to query, next-generation sequencing approaches offer advantages in sensitivity, cost, and efficiency over traditional testing methods (e.g., polymerase chain reaction, Sanger sequencing). Large panels of genes can be tested simultaneously by preenriching for the targets of interest (by automated amplicon generation or hybridization capture). These platforms have also been adapted to monitor MRD for specific known alterations, including NPM1 and FLT3 , among others. With further improvements in analytical workflow and cost reduction, whole-genome and transcriptome sequencing could displace some existing diagnostic tools because these platforms can provide simultaneous detection of mutations, gene expression, copy number alteration, and structural variation.

Table 59.2
Familial Acute Myeloid Leukemia Syndromes
Gene Chromosome Inheritance Risk of MDS/AML Associated Phenotypes
RUNX1 21q22.12 AD ~40% Thrombocytopenia, platelet dysfunction
ANKRD26 10p12.1 AD Not defined Thrombocytopenia, platelet dysfunction
ETV6 12p13.2 AD Not defined Thrombocytopenia, platelet dysfunction
CEBPA 19q13.1 AD ~100% None
DDX41 5q35.3 AD Not defined None
GATA2 3q21.3 AD ~80% Emberger, MonoMAC
SAMD9 7q21.2 AD Not defined MIRAGE (myelodysplasia, infection, restriction of growth, adrenal insufficiency, genital phenotypes, enteropathy)
SAMD9L 7q21.2 AD Not defined Ataxia, pancytopenia
SRP72 4q12 AD Not defined Cytopenias, sensorineural hearing loss
AD , Autosomal dominant; AML , acute myeloid leukemia; MDS , myelodysplastic syndrome.

Familial platelet disorder with predisposition to acute myelogenous leukemia (OMIM 601399) is associated with autosomal dominant inheritance of germline mutations in RUNX1 . Mutation carriers frequently present with easy bruising/bleeding due to quantitative or qualitative platelet dysfunction and have an approximately 40% lifetime risk of developing AML, typically in the third or fourth decade. The most frequently reported mutant alleles in RUNX1 are loss-of-function nonsense, frameshift, or missense mutations in the DNA-binding domain. The second copy of RUNX1 is frequently mutated in cases that evolve to MDS/AML.

Germline mutations in CEBPA are a rare cause of autosomal dominant familial predisposition to AML (OMIM 116897). The germline mutations are typically truncating at the N-terminus of the protein, whereas somatic acquisition of mutations affecting the C-terminus is a nearly invariant event in the development of AML among these patients. These cases have a relatively favorable prognosis, similar to de novo AML with somatically acquired biallelic CEBPA mutations.

Germline mutations in GATA2 cause a spectrum of disorders with overlapping features, including Emberger syndrome (OMIM 614038) and immunodeficiency 21 (IMD21) (OMIM 614172), also described as monocytopenia with susceptibility to mycobacterial, fungal, and papillomavirus infection and myelodysplasia (MonoMAC syndrome) or dendritic cell, monocyte, B-lymphocyte, and natural killer lymphocyte deficiency (DCML). Patients with IMD21 have decreased monocyte counts and natural killer and B-cell deficiency and are at an increased risk of developing viral and nontuberculous mycobacterial infections. Emberger syndrome patients have a similar presentation but also have deafness and lymphedema and often develop pancytopenia. More than 100 germline GATA2 mutations have been described, most commonly deletions, nonsense, or missense substitution affecting the C-terminal zinc finger of GATA2 , in addition to intronic mutations and substitutions at a 5′ enhancer. AML or MDS arises in approximately 70% of carriers and is associated with cooperating genetic events, such as mutations in ASXL1 , SETBP1 , STAG2 , or hemizygous deletions involving chromosome 7.

SAMD9 and SAMD9L are paralogous genes on chromosome 7q. Inherited missense mutations in both genes appear to confer gain-of-function growth suppressive activity in hematopoietic cells. A variety of compensatory somatic events have been described that allow cells to escape from this antiproliferative pressure, including acquisition of loss-of-function mutations in cis that abrogate the effects of the germline allele, gene correction by homologous recombination, and loss of the chromosome arm harboring the mutant allele—a phenomenon termed “adaptation by aneuploidy.” The latter is associated with increased risk of MDS/AML development, whereas the former mechanisms are protective. Affected individuals may present with MDS/AML without prior sequelae or may have preceding stigmata of ataxia-pancytopenia ( SAMD9L ; OMIM 159550) or the MIRAGE syndrome ( SAMD9 ; OMIM 617053).

Germline mutations in DDX41 on chromosome 5q are associated with autosomal dominant predisposition to MDS/AML with a long latency (OMIM 616871). Affected individuals may present with cytopenias prior to diagnosis of MDS/AML, but often a family history of hematologic malignancies is the only feature that raises suspicion of an inherited syndrome. Pathogenic germline DDX41 alleles include deletions, frameshift, nonsense, or missense mutations. Somatic mutations in the remaining DDX41 allele are acquired in nearly half of individuals who progress to MDS/AML.

Clonal Hematopoiesis

HSCs acquire somatic mutations with age (see Chapter 19 ). Mutations that confer a fitness advantage create a state of clonal hematopoiesis (CH) in which a clonally derived population of cells can be detected in peripheral blood or bone marrow. CH is nearly ubiquitous in the adult population when very sensitive error-corrected sequencing technologies are used. Using more stringent thresholds typically employed in routine clinical sequencing (i.e., variant allele fraction of 2%), CH is detectable in nearly 10% of apparently healthy individuals older than 65 years of age. This is termed clonal hematopoiesis of indeterminate potential (CHIP) if the affected individual has normal blood counts and no antecedent hematologic malignancy. In contrast, CH in patients with idiopathic cytopenias but without a morphologically apparent hematologic malignancy is referred to as clonal cytopenias of undetermined significance (CCUS). The genes most frequently mutated in CHIP and CCUS encode epigenetic regulators (e.g., DNMT3A , TET2 , ASXL1 ) which are also targets of recurrent mutations in MDS/AML.

Patients with CHIP carry a small but significantly increased risk of progression to overt hematologic malignancy, approximately 0.5% to 1% per year, most often to MDS or AML. The risk of progression to a hematologic malignancy is higher in individuals with CCUS, a large clone size, more than one driver mutation, and mutations in specific genes (particularly, spliceosome components). Patients with CH also have a significantly increased risk of cardiovascular events, typically atherosclerosis-related vascular events, possibly related to altered proinflammatory signaling in clonally derived monocytes, among other mechanisms. Additional epidemiologic study will be required to determine how often CH is a precursor to MDS/AML and whether surveillance and/or early intervention are warranted.

Genetic and Epigenetic Alterations in Acute Myeloid Leukemia

AML is characterized by a variety of recurrent cytogenetic, molecular, and epigenetic modifications that illustrate the pathophysiologic role of acquired somatic alterations affecting specific gene products. These mutations have been incorporated into strategies for prognostic stratification and risk-adapted treatment, in addition to providing a framework for targeted therapy. The World Health Organization (WHO) classification now recognizes a number of recurrent genetic abnormalities, including balanced translocations and inversions, as defining features of AML.

Early understanding of the pathogenesis of AML suggested a model in which AML occurred in the setting of acquired mutations in two pathways: mutations that block normal mechanisms of differentiation and mutations that activate cell proliferation and survival pathways. This oversimplified model gained early support from genetically engineered mouse models in which at least two genetic “hits” were required to generate leukemia, such as a mutation activating cytokine signaling and a mutation inactivating a hematopoietic transcription factor. However, this model does not account for the wide spectrum of more recently described somatic alterations, nor do all AML cases carry mutations in these two classes.

Chromosomal Abnormalities

The central role of acquired mutations in AML was first recognized through the identification of recurrent nonrandom cytogenetic alterations in the mid-20th century (see Chapter 57 ). Recurrent karyotypic lesions are frequent events in AML, present in approximately 50% to 60% of patients at diagnosis, and have distinct prognostic significance that are central to treatment decisions, including the identification of patients for whom HSC transplantation should be considered in first remission, as well as patients likely to achieve a favorable outcome with chemotherapy alone ( Fig. 59.1 ).

Figure 59.1, FREQUENCY OF CHARACTERISTIC CYTOGENETIC ABNORMALITIES IN ADULT PATIENTS WITH ACUTE MYELOID LEUKEMIA.

Common chromosomal abnormalities include chromosome translocations or inversions, chromosome deletions, and monosomies or trisomies. Patients can be stratified according to karyotype into favorable-, intermediate-, and poor-risk categories. Typically, favorable risk includes patients with t(15;17), t(8;21), or inv(16)/t(16;16), and adverse risk includes inv(3q)/t(3;3), t(6;9), monosomy 7, monosomy 5, loss of 5q, 7q, or 17p, and complex (three or more) chromosomal abnormalities, as well as most translocations involving the KMT2A ( MLL1 ) locus on chromosome 11q23. In lieu of additional molecular studies, all other karyotypic lesions are generally classified as intermediate risk. This includes patients with a normal karyotype, which comprise approximately 45% of all AML cases.

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