Fang Yin and Vera Malkovska
Acute myeloid leukemia (AML) is a heterogeneous group of diseases characterized by uncontrolled proliferation of myeloid progenitor cells that gradually replace normal hematopoiesis in the bone marrow. The genetic changes arising in the neoplastic clone lead to cascades of molecular events that cause abnormal proliferation and aberrant differentiation of the malignant cells and ultimately result in inhibition of normal hematopoiesis.
Characterization of transforming genetic events is becoming increasingly important in establishing diagnosis, defining prognosis, and planning therapy in AML. Aggressive chemotherapy with optimal supportive treatment has improved outcomes in younger patients with AML. Most of these patients achieve a complete remission (CR) but many relapse and their 5-year survival remains below 50% in large studies.1 Patients over the age of 60 have a median survival of less than a year and long-term survival rates below 10%. Both the unfavorable biology of AML and poor tolerance of chemotherapy in the elderly are to blame.2 The current challenge is to improve the understanding of the molecular mechanisms of AML and design leukemia-specific treatments that are effective in chemotherapy-resistant disease and applicable to older patients.
EPIDEMIOLOGY
The age-adjusted incidence rate of AML in the United States is 3.5 per 100,000, accounting for approximately 10,000 deaths per year. AML accounts for about 15% to 20% of acute leukemias in children and adolescents and 80% in adults. The incidence of AML rises rapidly after the age of 60 and the median age at diagnosis is 67 years (Fig. 11.1).1-3
ETIOLOGY
The molecular origins of AML are unknown. The pathophysiologic mechanisms are multiple, act in concert, and are distinct in different types of AML. Inherited genetic predisposition and environmental mutagens such as radiation, drugs, and other toxins all play a role in the development of AML.4 Genetic causes are suggested by the increased incidence of AML in identical twins, as well as the known association of AML with a variety of congenital disorders. Genetic predisposition is probably important in the development of AML in children and young adults while it has not been shown to play a major role in older adults. A large European population-based study found no significant familial aggregation for AML or myelodysplastic syndromes (MDS).4 These findings suggest that in most patients with AML/MDS environmental factors play a more important role than germline mutations. AML arising from pre-existing hematologic disorders, most commonly MDS or myeloproliferative disorders, have inferior prognosis. AML following exposure to chemotherapy and radiation is characterized by resistance to treatment and short survival.
FIGURE 11.1 The age-related incidence of AML in the United States.
Known risks factors for the development of AML are shown in Table 11.1.
PATHOGENESIS
At the molecular level, pathogenesis of AML is a complex multistep process that results from the interaction of two different classes of mutations. The first class of mutations impairs cell differentiation resulting in clonal expansion of myeloid progenitors. The second class causes abnormal cell proliferation by constitutive activation of cellular proto-oncogenes including FLT3 tyrosine kinase, RAS, c-KIT, and others. The silencing of tumor suppressor genes also contributes to the pathogenesis. Some molecular genetic alterations in AML are highlighted by distinct chromosomal changes including translocations, inversions, and deletions while others can be only identified by molecular analysis. The common mutations found in AML that are thought to cooperate in the malignant transformation are shown in Table 11.2.5 The analysis of these cytogenetic and molecular changes is used to predict clinical outcomes and to formulate treatment paradigms in AML. Studies are underway to understand the downstream molecular pathways triggered by these mutations that lead to the unrestricted growth of leukemic cells and suppression of normal hematopoiesis by the malignant clone. Whole genome sequencing of AML cells and their normal counterparts opens the possibility of identifying all pathogenetic mutations and using them for disease classification and therapy.
CLINICAL FEATURES
Patients with AML usually present with bone marrow failure that causes symptoms of anemia, bleeding from thrombocytopenia, and neutropenic infections. Tissue infiltration with leukemic blasts involving gums, skin, meninges, and other organs is most commonly associated with monocytic morphology. Striking bruising and life-threatening hemorrhage should raise suspicion of disseminated intravascular coagulation (DIC) frequently seen in acute promyelocytic leukemia (APL). However, DIC can occur in any type of AML. Leukostasis and hyperviscosity causing organ dysfunction usually occur with blast cell counts over 100,000/µL. Manifested by confusion, visual impairment, and shortness of breath, leukostasis can also lead to hemorrhage in the retina, brain, lungs, and other organs. Rare but striking manifestations of AML include the Sweet syndrome, a skin rash with neutrophilic infiltrates in the dermis, and chloromas, tumors of myeloid blasts. Extramedullary leukemia portends a worse prognosis.
Table 11.1 Risk Factors for Development of Acute Myeloid Leukemia
Environmental Exposures
Benzene
Ionizing radiation
Smoking
Genetic Disorders
Down syndrome
Bloom syndrome
Fanconi anemia
Dyskeratosis congenita
Ataxia telangiectasia
Li-Fraumeni syndrome
Kostmann syndrome
Klinefelter syndrome
Pre-Existing Hematologic Disorders
Myelodysplastic syndromes (MDS)
Myeloproliferative disorders
Aplastic anemia
Treatment-Associated
Alkylating agents: AML usually arises from MDS, after a 3–10-yr latency period and is associated with deletions involving chromosomes 5 or 7.
Topoisomerase II inhibitors: AML lacks preceding myelodysplasia, has a shorter latency (1–3 yr), exhibits monocytic morphology, and is associated with changes involving the long arm of chromosome 11 (11q23).
Radiotherapy alone or in combination with chemotherapy
Symptoms and signs on presentation include the following:
Marrow failure
Fatigue
Shortness of breath
Fever
Focal bacterial infections
Petechiae
Bruising
Bleeding (if severe, suspect promyelocytic leukemia)
Tissue involvement
Bone pain, tenderness
Moderate splenomegaly
Gingival hyperplasia
Central nervous system (CNS) and cranial nerve dysfunction
Visual changes (retinal involvement, hemorrhage, papilledema)
Rare manifestations
Sweet syndrome
Chloromas
LABORATORY FINDINGS
The most common laboratory findings in AML include anemia, thrombocytopenia, neutropenia, and myeloid blasts on the blood smear. The blasts have distinct immunophenotypes detected by flow cytometry. In aleukemic leukemia, blasts are seen only in the bone marrow. Coagulopathy resulting from DIC is common in promyelocytic leukemia. Hyperuricemia from high cell turnover is often seen on presentation and worsens during chemotherapy. Rapidly rising serum levels of uric acid, potassium, and phosphate with decreasing calcium herald a tumor lysis syndrome that can result in acute renal failure. Renal tubular dysfunction caused by muramidase released from leukemic blasts can add to the electrolyte abnormalities commonly seen in AML. Lactic acidosis tends to occur with leukostasis while high lactate dehydrogenase (LDH) is associated with CNS involvement. High numbers of leukemic blasts in blood samples can lead to spurious hypoglycemia, hypoxemia, hypokalemia, and other abnormalities resulting from cellular metabolic activity in vitro. Rapid processing of anticoagulated blood samples avoids this artifact.
In addition to routine chest X-rays, imaging studies including computerized tomography (CT) and magnetic resonance imaging (MRI) scans directed according to symptoms can reveal leukemic infiltrates, hemorrhage, or infection.
Laboratory findings in AML include the following:
Hematologic
Increased white blood cell count with blasts in peripheral blood
Anemia
Granulocytopenia
Thrombocytopenia
DIC
Chemistry
Hyperuricemia
Elevated blood urea nitrogen and creatinine (urate nephropathy)
High LDH
Hypokalemia (tubular dysfunction)
Lactic acidosis (leukostasis)
Hypercalcemia, rarely hypocalcemia
Spurious hypoxemia, hypoglycemia, hyperkalemia or hypokalemia
Imaging studies
Intracranial hemorrhage (often with hyperviscosity) (CT)
Thickened nerve sheaths (MRI)
Lung infiltrates (CT)
CLASSIFICATION
The classification of AML has evolved from the mainly morphology-based French-American-British (FAB) to the more comprehensive World Health Organization (WHO) classification. The older FAB classification based on morphology, cytochemical staining, and immunophenotype of the predominant cells divides AML into eight subtypes (M0–M7) (Tables 11.3 and 11.4).
The WHO classification is based on the concept that different subgroups of AML can be defined as unique diseases through correlation of morphology, cytochemistry, immunophenotype, genetics, and clinical features.6 In this classification, diseases associated with specific genetic abnormalities, namely AML with t(8;21)(q22;q22), AML with inv(16)(p13,1q22) or t(16;16)(p13.1;q22), and APL with t(15;17) (q22;q12), can be diagnosed as AML regardless of the blast cell count. In all other entities within the category of AML with recurrent genetic abnormalities, 20% or more blasts must be present in the blood or bone marrow to establish the diagnosis of AML. The category “AML with myelodysplasia-related changes” includes disease that evolves from previously documented MDS, AML with myelodysplasiaassociated cytogenetic abnormalities (e.g., −7 or −5), or AML that exhibits dysplasia in 50% or more of the cells in two or more myeloid lineages.6
Table 11.4 World Health Organization Classification of Acute Myeloid Leukemias (AMLs)
AML with Recurrent Cytogenetic Translocations
AML with t(8;21)(q22;q22); RUNX1-RUNX1T1
AML with inv(16)(p13.1q22) or t(16;16)(p13.1;q22); CBFB- MYH11
APL with t(15;17)(q22;q12); PML-RARA
AML with t(9;11)(p22;q23); MLLT3-MLL
AML with t(6;9)(p23;q34); DEK-NUP214
AML with inv(3)(q21q26.2) or t(3;3)(q21;q26.2); RPN1-EVI1
AML (megakaryoblastic) with t(1;22)(p13;q13); RBM15-MKL1
AML with mutated NPM1
AML with mutated CEBPA
AML with Myelodysplasia-Related Changes
Therapy-Related Myeloid Neoplasms
AML, not Otherwise Specified
AML with minimal differentiation
AML without maturation
AML with maturation
Acute myelomonocytic leukemia
Acute monoblastic/monocytic leukemia
Acute erythroid leukemia
Acute megakaryoblastic leukemia
Acute basophilic leukemia
Acute panmyelosis with myelofibrosis
Myeloid Sarcoma
Myeloid Proliferations Related to Down syndrome
Transient abnormal myelopoiesis
Myeloid leukemia associated with Down syndrome
Acute Leukemias of Ambiguous Lineage
APL, acute promyelocytic leukemia.
Modified from Vardiman JW,Thiele J, Arber DA, et al.The 2008 revision of the World Health Organization (WHO) classification of myeloid neoplasms and acute leukemia: rationale and important changes. Blood. 2009;114:937-951.
The WHO classification of AML continues to evolve and incorporate newly detected molecular mutations with a defined impact on disease biology. The hope is that continuing refinement of the classification into biologically distinct entities will define prognosis and help guide the type and intensity of therapy.
DIAGNOSTIC EVALUATION
Leukemic myeloblasts are usually seen on the blood smear and are always found in the bone marrow biopsy. According to the WHO consensus, the diagnosis of AML requires at least 20% myeloid blasts in the peripheral blood or bone marrow. The evaluation of a patient with AML at diagnosis includes the following:
Blood count and inspection of blood smear
Bone marrow aspirate and biopsy
Morphology with Wright-Giemsa stain
Immunophenotyping by multiparameter flow cytometry
Cytogenetics
Molecular analysis
Lumbar puncture (after blasts cleared from blood): if CNS symptoms, monocytic morphology or high blast count
The morphologic diagnosis of AML can be supported by the presence of Auer rods in the cytoplasm, positive cytochemistry with Sudan black, and staining for myeloperoxidase and esterases. Immunophenotyping with a panel of monoclonal antibodies is particularly useful for distinguishing AML from acute lymphoblastic leukemia (ALL) and for identification of the subtypes, including AML with minimal differentiation, erythroleukemia, and megakaryoblastic leukemia. Cytogenetic and molecular abnormalities associated with the morphologic subtypes can further support the diagnosis and help predict treatment outcomes (Table 11.5).
PROGNOSTIC FACTORS
The most powerful prognostic factors that have been established over decades including age over 60, cytogenetics, prior MDS, and treatment-related AML have been recently supplemented by molecular genetic abnormalities.6 The independent prognostic variables associated with treatment outcomes are listed below:
Clinical prognostic factors
Age
AML arising from pre-existing MDS
Treatment-related AML
Performance status
Extramedullary disease
Comorbid conditions
Laboratory-based prognostic factors
White blood cell count >20,000/μL at presentation
Cytogenetics
Molecular genetic changes (mutations in nucleophosmin 1 (NPM1), FLT3, CCAAT enhancer binding protein alpha (CEBPA), KIT)
CD34 positive blasts
Multidrug resistance
The best predictors of long-term outcome next to age are chromosomal and molecular genetic findings in leukemic cells (Tables 11.6 and 11.7).6,7 Cytogenetic analysis is the basis of stratification of patients into three risk groups with different responses to chemotherapy (Table 11.3). For example, patients in the better-risk group with core-binding factor mutations are most likely to obtain a long remission after consolidation with high-dose cytosine arabinoside while patients with poor-risk cytogenetics do not benefit from this treatment. Approximately 40% to 50% of all patients have a normal karyotype at the time of diagnosis and most of them fall into the intermediate-risk category (Table 11.3). The molecular genetic changes in cytogenetically normal AML that correlate with prognosis have been used to subdivide this group and create more refined treatment algorithms.7 The frequency and prognostic impact of gene mutations in AML with normal karyotype are listed in Table 11.7.7,8 Some of these mutations including NPM1, FLT3, CEBPA, and KIT already impact therapy while the rest are under investigation. The best-studied mutations, the internal tandem duplications (ITD) of the FLT3 gene which codes for a tyrosine kinase receptor, are found in approximately 20% to 30% of AML patients. Several retrospective studies have demonstrated that patients harboring such mutations respond poorly to treatment and have significantly shorter disease-free survival (DFS) and overall survival (OS).7 By contrast, in the absence of FLT3 mutations patients with NPM1 and CEBPA mutations have improved CR rates and OS (Table 11.6).7 These mutations provide substantial insights into the pathogenesis of AML and help identify molecular targets for treatment. Studies of wholegenome sequences, mircoRNA, and gene expression in AML cells may alter prognostic categories in the near future.7,9
The outcome of younger patients with AML has markedly improved over the last three decades, due to advances in both chemotherapy and supportive care. Over 50% of patients under the age of 60 in the favorable prognostic category can be cured with current chemotherapy. The cure rates of younger patients without better-risk chromosomal and molecular markers are about 30% to 40%. Unfortunately, little progress has been made in the long-term survival of older adults with AML. These patients have more poor-risk disease characteristics, higher frequency of comorbid conditions, and poor tolerance of toxic therapy. Novel treatment strategies are needed for the majority of older patients.2,10
TREATMENT
Because AML is strikingly heterogeneous, its treatments must be individualized. Treatment plans in AML patients depend on age, performance status, and prognostic factors detected by conventional cytogenetics, fluorescence in situ hybridization, and polymerase chain reaction (PCR).7,8 The first example of a targeted treatment directed against a specific mutation is the successful use of all-trans-retinoic acid (ATRA) for APL, which is discussed separately in this chapter.
The treatment for AML is generally divided into two phases: remission induction and postremission therapy. The goal of the former is to achieve a CR defined by the following criteria: less than 5% blasts in a bone marrow that is 20% or more cellular, absent extramedullary leukemia, a neutrophil count greater than 1,000/μL, and a platelet count greater than 100,000/μL. The achievement of CR as defined by these simple criteria translates into improved survival. Disappearance of karyotypic or molecular abnormalities is not required for the definition of CR. It is therefore not surprising that clinical trials confirm an almost 100% risk of relapse when patients receive only induction chemotherapy. Long-term survival requires postremission treatment. Intensive chemotherapy given after achievement of CR (similar to that given during induction) is termed consolidation therapy.
Initial Management
The early management of AML patients should be well organized and implemented by an experienced team.
The initial evaluation should include the following:
History and physical exam
Complete blood count with differential
Examination of the peripheral blood smear
Coagulation studies: prothrombin time (PT), activated partial thromboplastin time (aPTT), fibrinogen, D-dimer
Serum chemistries with uric acid, calcium, and phosphorus
Evaluation of renal and liver functions
Hepatitis B and C, herpes simplex virus (HSV), cytomegalovirus (CMV), varicella, and human immunodeficiency virus (HIV) serologies
Pregnancy test in females of reproductive age
Bone marrow aspirate for morphology, cytochemistry, cytogenetics, molecular genetic studies, and flow cytometry
Bone marrow biopsy
Human leukocyte antigen (HLA) typing of patients who are candidates for hematopoietic stem cell transplant
Lumbar puncture delayed until blasts cleared from blood, in patients at high risk for CNS involvement (CNS symptoms, elevated leukocyte count, extramedullary disease, and monocytic morphology FAB M4 or M5)
Chest radiograph and electrocardiogram
Evaluation of cardiac function by echocardiogram or multigated acquisition scan (MUGA) in selected patients
Central venous-access catheter placement
The workup should be followed by an unhurried discussion with the patient about the diagnosis, prognosis, side effects of therapy, probable impact on life style, and anticipated requirements for support by family and friends. In frail elderly patients, a decision to give only supportive treatment without chemotherapy may be reached jointly by the patient and the attending physician.
The final outcome depends not only on the choice of chemotherapy but also on close monitoring, prevention, and meticulous management of complications. Many events are associated with AML therapy and their timing is predictable. For example, hyperleukocytosis, tumor lysis, and DIC tend to occur early while marrow aplasia with resulting complications can be expected from the second week of chemotherapy. Patients should be monitored for side effects such as cardiotoxicity due to anthracyclines or neurotoxicity from high doses of cytosine arabinoside. Infection prophylaxis includes meticulous care for indwelling central venous catheters and prevention of mucositis. Empirical broad spectrum intravenous antibiotics should be administered immediately if the patient becomes febrile during neutropenia. Cultures should be obtained before administration of antibiotics. Prompt treatment or prophylaxis of oral and perianal herpetic ulcerations prevents discomfort and bacterial superinfection. Prophylaxis against herpes viruses can prevent these complications and decrease the severity of mucositis. Prophylactic antifungal therapy may lead to decrease in fungal infections. Transfusion support with irradiated and leukodepleted blood products should be provided to prevent symptoms of severe anemia and to maintain the platelet count above 10,000/μL. Transfusions of blood products in AML are guided by the same principles as described for ALL (Chapter 12).
Induction Therapy
The most commonly used chemotherapy regimen consists of one or two cycles of a combination of cytarabine 100 to 200 mg/m2 given by continuous intravenous (IV) infusion over 7 days and 3 days of an anthracycline (e.g., daunorubicin 60–90 mg/m2, idarubicin 10–13 mg/m2, or mitoxantrone 10–12 mg/m2 by IV bolus). This “3 + 7” regimen results in CR rates of approximately 70% to 80% in patients younger than 55 to 60 years.10-12 If leukemia persists in the bone marrow at day 14 to 21, a second identical or modified course of chemotherapy is usually given.
Several recent studies demonstrated that high-dose daunorubicin at 90 mg/m2/day for 3 days is well tolerated and results in improved CR rates and a significantly higher OS, especially in younger adults under the age of 50 or 65 years and those with better-risk or intermediate-risk cytogenetics. Therefore, high-dose daunorubicin is now considered a new standard care in the initial treatment of patients <65 years of age with a good performance status and adequate cardiac function.12 Idarubicin could be an alternative to daunorubicin, because a recent study showed no difference in outcome between high-dose daunorubicin and standard-dose idarubicin.13
Elderly patients with no comorbidities have a better outcome with intensive chemotherapy induction than with supportive care.14,15 Clofarabine is active as a single agent and is currently explored as part of induction regimens for elderly patients.16 In older patients unlikely to benefit from chemotherapy, agents such as azacitidine, decitabine, and lenalidomide, may prolong survival even in the absence of CR. Novel agents including tipifarnib, sapacitabine, and lenalidomide are being tested in this population. The goals and end points of therapy for elderly patients are therefore shifting.
Further strategies to intensify induction chemotherapy include the addition of high-dose cytarabine, etoposide, 6-thioguanin, double induction and timed sequential therapy, priming and growth-factor support. Randomized trials demonstrate benefit in DFS but not in OS with the above approaches that may cause additional toxicities.
The anti-CD33 immunoconjugate, gemtuzumab ozogamicin, used in combination therapy shows promising outcome in de novo AML.17,18 In a recent prospective phase III trial, patients were randomized to receive a traditional induction 3 + 7 regimen or to receive the same regimen in combination with fractionated small doses of gemtuzumab. The trial demonstrated no significant difference in fatal adverse events between the two arms, and a highly significant advantage in event-free survival and DFS at 2 years for patients who were treated on the gemtuzumab arm. This translated into a 10-month OS advantage for the combination arm.17 Another recent randomized study from the UK showed similar results in AML patients age greater than 60.18 Though gemtuzumab was withdrawn from the US market in June 2010 due to concerns of toxicity, the above results are encouraging. In the United States, further studies can be conducted with gemtuzumab as an investigational new drug (IND).
Postremission Treatment
Nearly all patients in CR after induction therapy have residual disease that, without further treatment, leads to relapse. The main strategies to prevent relapse involve postremission treatment with high-dose cytarabine and allogeneic or autologous hematopoietic stem cell transplantation (HSCT). Maintenance therapy is not part of current standard treatment for AML. However, maintenance therapies including immunotherapy, demethylating agents, and targeted therapies are currently being tested in this disease with the intention of achieving leukemia cure.
Intensive consolidation treatment improves survival in younger patients with AML. A dose- dependent response to cytarabine has been shown in randomized controlled trials.11 Consolidation with high-dose cytarabine (HDAC)-based therapy using daily doses of 1 to 6 g/m2 (e.g., 2–3 g/m2 twice daily on days 1, 3, and 5 or twice daily for 6 days) is the standard for patients younger than 60 years. HDAC appears to be most beneficial to younger patients with better-risk cytogenetics, especially those with core-binding factor mutations (i.e., t(8;21) and inv (16)). Patients over the age of 60 years do not benefit from HDAC consolidation.19 The optimal number of courses of HDAC-based consolidation has not been determined, but the evidence suggests that 1 to 4 courses are reasonable.20,21 For older patients, there are also studies that suggested no particular value in intensifying postremission therapy beyond a single course of consolidation.21,22 The optimal cumulative dose of cytarabine in the initial therapy of younger patients with AML is still an open issue. The reduction in cytarabine doses (cumulative dose not beyond 12 g/m2 within first consolidation) during conventional chemotherapy did not appear to worsen treatment outcome.14,23 For younger AML patients, the choice between consolidation chemotherapy and allogeneic HSCT should be based on the risk of relapse.
Hematopoietic Stem Cell Transplantation
High doses of marrow-ablative chemotherapy and total body radiation followed by autologous or allogeneic stem cell rescue have been widely used in AML. Autologous HSCT (auto-HSCT) requires stem cell collection from the patient while in CR. Allogeneic stem cells are obtained from HLA-matched siblings, unrelated donors, and cord blood or more recently haploidentical donors. Allogeneic HSCT (allo-HSCT) confers an additional immune-mediated antileukemic activity, the so-called graft-versus-leukemia effect.
Multiple prospective randomized trials comparing standard consolidation chemotherapy with HSCT have demonstrated that allo-HSCT provides the best antileukemic therapy with the lowest risk of recurrence, followed by auto-HSCT, which in some studies is superior to conventional postremission chemotherapy (Table 11.8). Therefore, in patients lacking an available HLA-matched donor, auto-HSCT can offer a benefit over conventional consolidation chemotherapy. However, the OS has not improved in AML patients undergoing auto-HSCT compared to patients receiving intensive consolidation chemotherapy because of more opportunities for salvage with second-line chemotherapy and stem cell transplantation in patients relapsing on the chemotherapy arm.26,27
The powerful antileukemic activity of allo-HSCT has not always translated into better OS, due to the high treatment-related mortality (TRM) rate associated with allo-HSCT (Table 11.9). The availability of a donor has been used as a surrogate for randomization in these studies. These donor–no donor comparisons may be confounded by limited application of the assigned therapy, and the results of comparative studies have not always been consistent. Recent analyses demonstrate superior OS after allo-HSCT for AML patients excluding those with better-risk cytogenetics, and this benefit is most pronounced in younger patients.30 A large meta-analysis of 24 prospective studies including over 6,000 patients with AML in first CR (CR1) compared the role of HSCT and non-HSCT treatments. Overall, 3,638 patients were analyzed by cytogenetic risk category, including 547 better-risk, 2,499 intermediate-risk, and 592 poor-risk patients. Compared with non-allogeneic therapy, the hazard ratio of relapse or death with an allo-HSCT for patients in CR1 was 0.80 (95% confidence interval 0.74–0.86). When the analysis was broken down by risk category and outcome, there was a significant relapse-free survival and OS benefit for allo-HSCT during CR1 in intermediate-risk and poor-risk AML patients, but not for better-risk AML patients.31
The role of myeloablative allo-HSCT is limited in older patients due to lower tolerance of intensive chemotherapy and significant comorbidities. Reduced-intensity conditioning (RIC) regimens have broadened the applicability of allotransplants to elderly patients and to younger patients with comorbidities who would not tolerate standard myeloablative conditioning regimens. Studies reveal that more than one-third of patients can achieve long-term survival with RIC transplantation.32 The best insight into the differences in outcome comes from large registry-based retrospective studies demonstrating that RIC is associated with a reduction in TRM but an increased risk of relapse. Patients undergoing RIC stem cell transplantation in remission have similar DFS and OS compared to myeloablative conditioning.32
In the absence of an HLA-matched donor, alternative donors such as unrelated umbilical cord blood (UCB) or haploidentical donors are increasingly used. Both strategies have important advantages such as shorter time to transplant, which is particularly relevant to patients requiring urgent transplantation, and the availability of mismatched donors for virtually all patients. Recent studies have demonstrated similar leukemia-free survival after UCB transplantation compared to HLA-matched HSCT in patients with acute leukemia.33,34 The European Bone Marrow Transplant Registry (EBMTR) study also indicated that the haplo-HSCT results for poor-risk AML in CR are similar to those reported for unrelated donor transplants.35 As both the transplantation technology and chemotherapy continue to improve, the best treatment approach remains a moving target. Since AML is a heterogeneous malignancy, the most appropriate therapy will be ultimately determined by the cytogenetic and molecular characteristics of the disease. With progress in understanding of the disease biology and identification of new molecular markers, current practice will continue to evolve.
Risk-Based Approach to Acute Myeloid Leukemia Treatment in Younger Patients
Better-Risk Acute Myeloid Leukemia
Patients with a better-risk karyotypes (core-binding factor AML with t(8;21) or inv(16), normal karyotype with an NPM1 mutation but no FLT3-ITD), or (with a CEBPA mutation) do well with either intensive consolidation chemotherapy or autologous HSCT (Table 11.6). Long-term DFS of 60% to 70% can be achieved with these approaches in younger patients. Allo-HSCT, with its higher TRM, is generally not used in this subset of patients outside clinical trials.10,31
Poor-Risk Acute Myeloid Leukemia
Patients with poor-risk cytogenetics or patients with normal karyotype and FLT3-ITD as well as patients with secondary AML have very poor prognosis (Table 11.6). If an HLA-matched donor is available, these patients should be evaluated for HSCT as soon as possible. The outcome remains dismal with conventional consolidation or autologous HSCT in this group of patients. Allo-HSCT from unrelated donors matched by high-resolution HLA-typing has comparable outcomes to transplants from HLA-matched family donors.19 Younger and healthier patients without an HLA-matched donor should be considered for allo-HSCT from a haploidentical family member, or UCB transplant. Patients with significant comorbidity who achieve CR can be offered allo-HSCT with RIC. New therapeutic agents given on clinical trials should be offered to the majority of patients without a donor. Patients with FLT3-ITD should be considered for clinical trials with FLT3 tyrosine kinase inhibitors such as midostaurin, lestaurtinib, sorafenib, and quizartinib.7,36
Intermediate-Risk Acute Myeloid Leukemia
Treatment decisions are particularly complex in the largest and most heterogeneous prognostic group of patients with normal cytogenetics, no molecular markers, or markers of unknown impact on therapy (e.g., IDH1 and IDH2, RUNX1, TET2, and DNMT3A mutations) (Tables 11.6 and 11.7). Allo-HSCT, consolidation chemotherapy, and auto-HSCT are considered of equivalent benefit. Recent meta-analysis revealed allogeneic HSCT has significant relapse-free survival and OS benefit for intermediate-risk AML in CR1 compared with non-allogeneic HSCT therapies.31 However, novel genetic lesions detected by new technologies, such as whole genome analyses, RNA and microRNA profiles continue to refine AML risk. Until outcome data are available for these new subgroups, treatment decisions should be carefully individualized based on patient age, comorbidity, type of transplant, and risk of relapse. If an HLA-matched family donor is available, allo-HSCT should be offered to patients less than 55 to 60 years of age with good performance status, given the superior antileukemic activity of this therapy. Auto-HSCT, following intensive consolidation for in vivo purging, can be offered to patients without an HLA-matched donor. In general, the benefit of allo-HSCT over auto-HSCT or HDAC consolidation will be seen in subgroups of younger healthier patients in whom the improvement in relapse risk outweighs the higher procedural mortality.
Treatment of Acute Promyelocytic Leukemia
APL is a well-defined disease entity with a distinct epidemiology, characteristic morphology, and potentially life-threatening coagulopathy at presentation but overall better prognosis than other AML subgroups. It is the first example of leukemia in which therapy directed against the leukemogenic event, the t(15;17) resulting in the PML-RARα fusion transcript, leads to improved outcome. With better management of the associated coagulopathy and the introduction of the differentiating agents, ATRA and arsenic trioxide (ATO), APL now represents the most curable subtype of AML with cure rates of >90% quoted in large clinical trials.37,38 The most widely tested treatment for APL consisted of a combination of ATRA and anthracycline-based chemotherapy for induction, followed by at least two courses of anthracycline-based chemotherapy and ATRA for consolidation, and maintenance therapy with intermittent ATRA alone (15 days every 3 months) or combined with 6-mercaptopurine and methotrexate. Accumulating data from historically controlled and randomized trials suggest a benefit for ara-C during induction and/or consolidation in both low-risk (WBC count < 10,000/μL) and high-risk patients.37,38
It is now established that ATO is a more effective agent in APL than ATRA. In particular, single-agent ATO can cure APL much more frequently than single-agent ATRA.38 ATO-based initial therapy can be used as an alternative for patients who cannot tolerate anthracycline therapy. The North American Intergroup Trial incorporated ATO into the consolidation schema aiming to decrease toxicity. Patients randomized to receive two courses of 25 days of ATO (5 days a week for 5 weeks) had a significantly better event-free and OS than those who received only two courses of ATRA plus chemotherapy.39
Despite the introduction of ATRA and improvement in the treatment of coagulopathy, bleeding remains an important cause of death during initial management. Therefore, APL should be treated as a medical emergency. ATRA and aggressive supportive measures should be started as soon as the diagnosis is suspected, even before the confirmatory genetic tests are available. The coagulopathy is treated with fresh-frozen plasma, fibrinogen, and platelets with the goal of maintaining fibrinogen level above 150 mg/dL and platelets well above 30,000/μL until the resolution of coagulopathy.38,39 The APL differentiation syndrome emerged as a major toxicity associated with both ATRA and ATO. It is characterized by pleural and pericardial effusions, weight gain, edema, dyspnea, fever, episodic hypotension, and pulmonary infiltrates. It should be promptly treated with intravenous administration of dexamethasone at a dose of 10 mg twice daily. The incidence of this complication is reduced in patients receiving concomitant chemotherapy during induction. Patients receiving ATO should be also closely monitored for prolongation of QT interval and their electrolytes should be maintained within normal range.38,39 The goal of induction and consolidation therapy should be the attainment of PCR negativity for the PML-RARα rearrangement, as the persistence of such minimal residual disease (MRD) predicts relapse. Patients treated in molecular relapse may have better outcome compared with those treated at frank hematologic relapse.40 Novel formulations of the same targeted agents such as the oral formulation of ATO and an oral synthetic retinoid (Tamibarotene) designed to overcome ATRA resistance may be a promising therapeutic option for patients with relapsed disease.
Treatment of Relapsed and Refractory Acute Myeloid Leukemia
Unfortunately, AML relapses in the majority of patients. Moreover, approximately 25% of younger patients are refractory to standard induction chemotherapy. Patients with recurrent AML experience lower CR rates with reinduction chemotherapy compared to initial treatment. If second CR is achieved, it tends to be shorter. Therefore, patients with relapsed and refractory AML are candidates for clinical trials exploring innovative therapeutic strategies. Best supportive care is an option for those who do not wish to pursue intensive treatment.
If a suitable HLA-matched donor is available, the first choice of therapy is an allo-HSCT.19 Patients transplanted at the time of relapse have similar outcomes to those treated in second CR. If a suitable donor is not available, management should be guided by the duration of CR1. For patients with CR greater than 12 months, reinduction with HDAC-containing regimen is reasonable since it can achieve a CR rate of 50% to 60% and a 5% to 10% long-term DFS. For patients with shorter CR durations, the priority is treatment on a clinical trial. Novel agents currently being investigated include clofarabine, farnesyltransferase inhibitors (FTIs, FLT3 inhibitors, and other agents (see Table 11.11). Clinical trials should be considered in relapsed and refractory patients since response rates with currently available agents remain dismal.
Hematopoietic Growth Factors
Colony-stimulating factors (CSFs) can shorten the duration of neutropenia during AML treatment and have the potential to improve outcomes. Both granulocyte-macrophage colony- stimulating factor (GM-CSF) and granulocyte colony-stimulating factor (G-CSF) have been shown to accelerate neutrophil recovery after induction chemotherapy. G-CSF and GM-CSF have also been used to sensitize blasts to chemotherapy by recruiting cells into the cell cycle. However, numerous clinical trials have failed to show a reproducible survival benefit of such approach. Studies to date demonstrate that the benefit of CSFs is limited to reduction of neutropenic and febrile (FTI) days. The addition of CSFs to chemotherapy yielded no difference in all-cause mortality, CR, or relapse rates in patients with AML.41 Growth factors may be considered in the elderly after chemotherapy is complete. Their use may confound interpretation of the bone marrow biopsy. Patient should be off G-CSF and GM-CSF for a minimum of 7 days before obtaining bone marrow to document remission. Myeloid growth factors should not be used in APL.
Acute Myeloid Leukemia in Older Patients
Older adults have a dismal prognosis that has not changed significantly over the last three decades. Care of these patients is challenging due to the higher rates of poor-risk prognostic features including comorbidities, poor-risk cytogenetics, and multidrug resistance of leukemic blasts.2,10 Older age per se, however, should not be a reason to withhold intensive therapy.
Studies suggest that remission-induction chemotherapy provides better quality of life and longer survival than supportive care only.15 Patients over the age of 60 years without significant comorbidities can be treated with standard “3 + 7” induction therapy. Their outcomes worsen with advancing age and decreasing performance status (Table 11.10). Patients who benefit most from standard chemotherapy can be reasonably selected by the following criteria: age 60 to 69 years, no secondary AML or pre-existing MDS, good performance status, no pretreatment infection, and normal bilirubin and creatinine.2 Results from MD Anderson Cancer Center in this small subgroup of patients show a 70% CR rate and a median survival of 14 months.2 Choice of post remission therapy for older patients remains problematic. In contrast to younger patients, no studies with HDAC-based therapy have demonstrated a survival advantage in patients over 60 years of age. As this population represents a majority of patients with AML, new strategies are needed to improve their outcomes. For the majority of elderly patients, clinical trials with novel agents should be considered at the time of diagnosis due to poor results with conventional therapy. Several modestly successful options are available outside clinical trials including azacitidine, decitabine, low-dose cytarabine, and RIC stem cell transplantation. The regimen of azacitidine 75 mg/m2/day for 7 days every 4 weeks used as outpatient therapy in older adult patients with low marrow blast count (20% to 30%) significantly prolongs OS and improves several morbidity measures compared with palliative treatment.42 Data from recently published randomized trials suggest that the efficacy of treatment with hypomethylating agents including decitabine may be comparable to that of intensive chemotherapy and superior to that of other palliative treatment approaches.42,43
Azacitidine has become an upfront option for patients who cannot tolerate intensive induction or for whom clinical trials are unavailable. Low-dose decitabine has also been used in high-risk MDS and elderly AML patients with CR rates of 14% to 28% and is better tolerated than standard induction.42,43 Low-dose cytarabine (20 mg subcutaneously twice daily for 10 days every 4 to 6 weeks) also produced better OS than did supportive care and hydroxyurea in an earlier randomized trial.44 Clofarabine is active as a single agent and is currently explored as part of induction regimens for elderly patients.16 Choice among these agents is based on individual preference because comparisons of efficacy or toxicity are problematic across different studies. In many patients, especially those with poor performance status, abnormal organ function, or active infection, palliative care may be the treatment of choice.
Acute Myeloid Leukemia in Pregnancy
The prevalence of leukemia during pregnancy is low, roughly one in 75,000 to 100,000 pregnancies. AML accounts for two-thirds of cases.45 Acute leukemia is usually reported during the second and third trimesters of pregnancy. This could be a result of a selection bias of unreported cases that occurred early and resulted in termination of pregnancy. The management of pregnant patients with AML poses major challenges. Patients are at high risk for pregnancy-associated complications due to the bleeding and infection.
Acute leukemia needs immediate treatment irrespective of gestational stage because delay or modification of therapy results in inferior maternal outcome. A recent systematic review identified 87 AML patients (88 pregnancies) treated with systemic therapy during the course of pregnancy. With few exceptions, patients were electively treated following the end of the first trimester.46 Nearly 50% of those who were exposed to chemotherapy during the first trimester had poor fetal outcomes. Chemotherapeutic agents and target therapies, such as ATRA, should be avoided during the first trimester. ATRA, a highly effective agent in APL, is associated with substantial toxicity to the fetus when used in the first trimester, including CNS and cardiovascular malformations. Hence, women diagnosed early in gestation should be offered a termination of pregnancy. Administration of chemotherapy and ATRA in the second and third trimesters is less likely to result in teratogenesis, although it increases the risk of intrauterine growth retardation.45 In summary, management of AML in pregnancy should focus on survival of the mother, while minimizing treatment-related fetal toxic effects.
Evaluation of Minimal Residual Disease
Postremission assessments of MRD by using PCR and multiparametric flow cytometry techniques are rapidly becoming part of the standard of care in AML. The recent development of real-time quantitative PCR allows for monitoring of patients with known genetic markers with the sensitivity of detecting one leukemic cell in up to 105 normal cells. Flow cytometry can be also used to detect and quantify low numbers of AML cells with a unique leukemia-associated immunophenotype. It is applicable to more than 85% of AML. Most laboratories can detect the leukemic cells with a sensitivity of between 10−4 and 10−5 by using the combination of 10-color flow cytometry.8
Extensive data support the prognostic value of MRD detection in AML patients after therapy. The quantification of residual AML cells at defined time points permits individualization of postremission therapy. It can also serve as a surrogate end point for the evaluation of new treatments. The goal of MRD detection is to identify patients at higher risk for relapse and assign them to different therapeutic approaches. MRD has been best studied in APL, where the persistence of PML-RARA fusion transcript at the end of consolidation therapy or subsequent recurrence of PCR positivity after a molecular remission precedes overt relapse. In other types of AML, MRD can be an early predictor of clinical relapse, but the optimal timing and frequency of MRD monitoring, and the optimal intervention based on MRD remains to be determined. Caution should be taken in applying MRD detection techniques to clinical practice outside well-designed prospective studies.
Novel Therapeutic Targets in Acute Myeloid Leukemia
Standard treatment for AML is based on aggressive cytotoxic chemotherapy given in repetitive cycles in order to eradicate disease. Only a minority of patients is cured by this approach. The success of ATRA and ASO in APL demonstrates that more selective and less toxic drugs can substantially increase cure rates of AML. The concept of targeting leukemic cells and sparing normal ones from the broad attack of chemotherapy is now under investigation in other types of AML. With improved understanding of the mechanisms underlying leukemogenesis, novel classes of drugs have entered clinical trials (Table 11.11). These include antibody-based therapies and other immunotherapeutic approaches, FLT3 inhibitors and other tyrosine kinase inhibitors, FTIs, clofarabine, and inhibitors of methylation (Table 11.11). Most of these drugs have limited activity as single agents and their full potential may be only realized in combination therapy.
Gemtuzumab ozogamicin (Mylotarg) is the first monoclonal antibody-directed chemotherapy targeting the CD33 epitope. It was withdrawn from the US market in June 2010 due to concerns of toxicity, though it may be of benefit for patients with better-risk cytogenetics or APL. In two recent randomized phase III trials, small doses of gemtuzumab in combination with the traditional 3 + 7 regimen demonstrated no significant difference in fatal adverse events compared to chemotherapy alone, and a highly significant advantage in DFS and OS for the combination arm.17,18 Both trials suggest that gemtuzumab warrants further investigation.
FLT3 kinase inhibitors have generated interest because activating FLT3 mutations in AML confers poor prognosis. Specific inhibitors of FLT3 such as sorafenib, lestaurtinib (CEP-701), and quizartinib (AC220) have been studied in relapsed or refractory AML. A phase III randomized trial of salvage chemotherapy with or without lestaurtinib failed to demonstrate a general benefit.36 The target inhibition of FLT3 was achieved in only 58% of patients receiving lestaurtinib but it did correlate with higher remission rate and longer OS. This suggests that FLT3 inhibition remains a promising approach and inhibitors with improved pharmacokinetic and pharmacodynamic properties should be studied. Sorafenib was demonstrated to slow down disease progression in relapse but did not induce second remission.51 The results of quizartinib large-scale phase III trials in relapsed AML are still pending.36
FTIs represent a new class of small-molecule inhibitors that selectively inhibit farnesylation of a number of intracellular proteins such as Ras. Single-agent tipifarnib has shown antileukemic activity in patients with MDS and refractory/poor-risk AML.52 However, a phase III study comparing single-agent tipifarnib to best supportive care including hydroxyurea in patients 70 years of age or older with untreated AML failed to demonstrate a survival advantage.53 Subsequently a phase I trial combining tipifarnib and etoposide in elderly poor-risk AML patients led to an improved CR rate of 25% across multiple dose levels of both drugs compared with 14% for single-agent tipifarnib.47 High RASGRP1 and low APTX gene expression was found to predict clinical response to tipifarnib and etoposide.47
Efforts are also under way to optimize the delivery of ara-C, for example, via conjugation to the lipid moiety elaidic acid, which allows bypassing the nucleoside transporter protein hENT1. Another novel formulation, CPX-351, fixes a 5:1 ratio of cytarabine to daunorubicin within a liposomal carrier. Novel topoisomerase II inhibitors are also under development. There is a plethora of new agents and drug combinations that awaits testing in AML. All patients with AML including the elderly should be offered therapy on well-designed clinical trials whenever possible.
SUMMARY
AML represents a genetically, morphologically, and clinically heterogeneous group of hematopoietic malignancies characterized by a rapid growth of myeloid blasts and suppression of normal hematopoiesis. The initiating genetic events and the pathways involved in the pathogenesis of AML are the subject of intense investigations. These events determine the type of AML, response to therapy, and to some extent, the final outcome. The well-established prognostic factors including age and cytogenetic changes are continuously expanded by molecular characteristics. Mutation testing is now incorporated into prognostic models and used to identify treatment targets. Unfortunately, less than half of younger patients and less than 10% of patients over 60 years are cured of their disease with current therapies. In younger patients, the longest DFS can be achieved with repeated cycles of intensive chemotherapy containing anthracyclines and cytarabine. Postremission consolidation therapy is crucial while maintenance is not necessary with the exception of APL. HDAC is the consolidation of choice for younger patients with better-risk cytogenetics. Patients with intermediate- or poor-risk disease who have an HLA-matched donor and good performance status benefit from allo-HSCT. Although questions remain about the optimal drugs for induction and about the numbers of consolidation cycles, it is unlikely that further modifications of standard chemotherapy will result in dramatic improvement in survival. Aggressive chemotherapy is not suitable for older patients with comorbidities and decreased performance status. These patients can derive modest benefit from hypomethylating agents or low-dose cytarabine. New targeted treatments including antibody-based therapies, tyrosine kinase inhibitors and other drugs targeting signal transduction, and vaccines have shown some activity and acceptable toxicity in clinical studies. Combinations of these agents based on studies of cell biology should lead to better outcomes than unselective cytotoxic therapy.
References