Acute Myelogenous Leukemia (AML) Treatment & Management

Updated: Sep 16, 2017
  • Author: Karen Seiter, MD; Chief Editor: Emmanuel C Besa, MD  more...
  • Print

Approach Considerations

Treatment options for acute myelogenous leukemia (AML) comprise a variety of chemotherapy regimens, biologic agents, and stem cell transplantation. [35, 36] Treatment recommendations include general recommendations, which take into account patient age and performance status, as well as recommendations for relapsed or refractory disease and acute promyelocytic leukemia (APL). See also Acute Myeloid Leukemia Treatment Protocols.

Current standard chemotherapy regimens cure only a minority of patients with AML. As a result, all patients should be evaluated for entry into well-designed clinical trials. If a clinical trial is not available, the patient can be treated with standard therapy. For consolidation chemotherapy or for the management of toxic effects of chemotherapy, readmission is required.

When receiving chemotherapy, patients should avoid exposure to crowds and people with contagious illnesses, especially children with viral infections. Any patient with neutropenic fever or infection should immediately be treated with broad-spectrum antibiotics. (See Antimicrobial Agents in Neutropenic Cancer Patients and Neutropenic Fever Empiric Therapy .)

Appropriate transfusion support must be provided to patients with AML. This includes transfusion of platelets and clotting factors (fresh frozen plasma [FFP], cryoprecipitate) as guided by the patient’s blood test results and bleeding history. Blood products must be irradiated to prevent transfusion-associated graft versus host disease (GVHD).

Patients with AML are best treated at a center whose staff has significant experience in the treatment of leukemia. Patients should be transferred to an appropriate (generally tertiary care) hospital if they are admitted to hospitals without appropriate blood product support, leukapheresis capabilities, or physicians and nurses familiar with the treatment of leukemia patients.


Treatment of Acute Myelogenous Leukemia

Induction therapy

Various acceptable induction regimens are available. The most common approach, "3 and 7," consists of 3 days of a 15- to 30-minute infusion of an anthracycline (idarubicin or daunorubicin) or anthracenedione (mitoxantrone), combined with 100 mg/m2 of cytarabine (arabinosylcytosine; ara-C) as a 24-hour infusion daily for 7 days. Traditional dosages have been as follows:

  • Idarubicin: 12 mg/m 2/d for 3 days
  • Daunorubicin: 45-60 mg/m 2/d for 3 days
  • Mitoxantrone: 12 mg/m 2/d for 3 days

These regimens require adequate cardiac, hepatic, and renal function. On these regimens, approximately 50% of patients achieve remission with one course. Another 10-15% of patients enter remission after a second course of therapy.

Improved outcomes have been reported with induction regimens using a higher dose of daunorubicin (90 mg/m2/d for 3 d). In a study by Fernandez et al in 657 patients younger than 60 years with untreated AML, the complete remission rate with high-dose daunorubicin  was 70.6%, versus 57.3% with conventional-dose daunorubicin, and overall survival was a median of 23.7 months versus 15.7 months, respectively. [37]

In a similar study in 813 patients 60 years of age or older by Lowenberg et al, the complete remission rate was 64% in the escalated-dose group compared with 54% in conventional-dose group. In both groups, daunorubicin was administered over 3 hours on days 1-3. Cytarabine was given in a dose of 200 mg/m2/d as a continuous infusion for 7 days, followed by a second cycle at a dose of 1000 mg/m2/12 h for 6 days. [38]

No significant difference was seen between the groups in terms of hematologic toxic effects, 30-day mortality, or other significant adverse events. Although survival endpoints did not differ between the two groups overall, in patients aged 60-65 years the complete remission rate, event-free survival, and overall survival were superior in the escalated-dose group. [38]

A study by Liu et al in 74 patients older than 60 years with newly diagnosed non-M3 AML reported a significantly higher complete remission rate with reduced-intensity idarubicin plus cytarabine than with daunorubicin plus cytarabine (70.4% vs 40%, P = 0.028). The difference was especially marked in patients with white blood cell (WBC) counts >10 × 109/L (P = 0.042) and ECOG (Eastern Cooperative Oncology Group) score <2 (P = 0.021). However, the overall survival of the entire population was poor, with a median survival of 10 months. [39]

Alternatively, high-dose cytarabine combined with idarubicin, daunorubicin, or mitoxantrone can be used as induction therapy in younger patients. The use of high-dose cytarabine outside the setting of a clinical trial is considered controversial. However, 2 studies demonstrated improved disease-free survival rates in younger patients who received high-dose cytarabine during induction.

A study of dosing regimens for cytarabine induction therapy determined that lower doses produce maximal antileukemic effects for all response end points. [40] Thus, high-dose cytarabine results in excessive toxic effects with no therapeutic advantage.


Midostaurin (Rydapt), an orally administered multitargeted kinase inhibitor, was approved by the US Food & Drug Administration (FDA) in April 2017 for adults with newly diagnosed AML who are FLT3 mutation positive. The FLT3 mutation is observed in approximately 30-35% of patients with AML.

Midostaurin is used in combination with standard cytarabine and daunorubicin induction and cytarabine consolidation chemotherapy. Midostaurin and its major active metabolites inhibit the activity of wild type FLT3, FLT3 mutant kinases (ITD and TKD), KIT (wild type and D816V mutant), PDGFR-alpha/beta, VEGFR2, and members of the serine/threonine kinase protein kinase C (PKC) family.

Approval of midostaurin was based on the CALGB 10603 (RATIFY) study, conducted in 717 adults younger than 60 years with newly-diagnosed FLT3-mutated AML. Patients who received midostaurin plus standard induction and consolidation chemotherapy had significantly longer overall survival than patients who received standard treatment plus placebo (hazard ratio [HR] for death, 0.78; one-sided P=0.009), as well as longer event-free survival (HR for event or death, 0.78; one-sided P=0.002). [41]

Consolidation therapy in younger patients

In patients aged 60 years or younger, treatment options for consolidation therapy include high-dose cytarabine, autologous hematopoietic stem cell transplantation (HCT), and allogeneic HCT.

Mayer et al conducted a randomized study of three different doses of cytarabine in patients with AML who achieved remission after standard “3 and 7” induction chemotherapy. [42] Patients received 4 courses of cytarabine at one of the following dosages:

  • 100 mg/m 2/d by continuous infusion for 5 days
  • 400 mg/m 2/d by continuous infusion for 5 day
  • 3 g/m 2 in a 3-hour infusion every 12 hours on days 1, 3, and 5

The probability of remaining in continuous complete remission after 4 years in patients aged 60 years or younger was 24% in the 100-mg group, 29% in the 400-mg group, and 44% in the 3-g group. The outcome in older patients did not differ. On the basis of this study, high-dose cytarabine for four cycles is a standard option for consolidation therapy in younger patients. [42]

As an alternative to high-dose cytarabine, Li et al reported that consolidation with fludarabine plus cytarabine can reduce relapse in patients with t(8;21) AML. In their prospective controlled trial, relapse-free survival (RFS) at 36 months was 81.73% in the fludarabine-cytarabine arm (n=23) and 50.73% in the high-dose cytarabine arm (n=22) (P=0.04), overall survival (OS) was 91.1% and 48.4%, respectively, and cumulative incidence of relapse was 18.27% and 47.39%, respectively (P=0.05). In patients without c-kit mutations, 36-month RFS was 100% in the fludarabine-cytarabine arm and 57.8% in high-dose cytarabine arm (P=0.005). [43]

Patients with good-risk AML (ie, t[8;21] or inversion of chromosome 16[inv16]) have a good prognosis after consolidation with high-dose cytarabine and should be offered such therapy. This is given as cytarabine at 3 g/m2 twice a day on days 1, 3, and 5 of each cycle, repeated monthly (after recovery from the previous cycle) for 4 consolidation cycles. Alternatively, autologous HCT can be given after (typically) 1-2 cycles of consolidation therapy. Allogeneic stem cell transplantation should be reserved for patients who relapse.

Patients with high-risk cytogenetics findings are rarely cured with chemotherapy and should be offered transplantation in first remission. However, these patients also are at high risk for a relapse following transplantation.

The best approach for patients with intermediate-risk cytogenetics findings is controversial. Some refer patients in first remission for transplantation, whereas others give consolidation chemotherapy with high-dose cytarabine for 4 courses and reserve transplantation for patients who relapse. Studies using newer molecular markers, such as FLT3, NPM1, CEBPa, BAALC, and ERG, are helping to define which patients with cytogenetically normal AML should receive standard consolidation therapy versus transplantation.

Hematopoietic stem cell transplantation in younger patients

The American Society for Blood and Marrow Transplantation (ASBMT) considers that in patients with AML who are under age 55, allogeneic HCT offers no survival advantage for those with low-risk cytogenetics who are in first clinical remission, but does offer a survival advantage versus chemotherapy for those with high-risk cytogenetics. [44]

In patients younger than 60 years, National Comprehensive Cancer Network guidelines recommend matched sibling or alternate donor HSCT as an option in the following situations [21] :

  • After induction failure with standard-dose or high-dose cytarabine
  • In patients with significant residual disease without a hypocellular marrow, after high-dose cytarabine induction therapy
  • As post-remission therapy in patients with intermediate-risk or poor-risk cytogenetics and/or molecular abnormalities, or those with treatment-related disease

Before referral for allogeneic transplantation, a suitable donor must be identified. Ideally, this is a fully HLA-matched sibling; however, many patients do not have such a donor. In these patients, alternatives include transplantation using a matched unrelated donor or using cord blood. Newer studies are examining the possibility of transplanting across HLA barriers (ie, with haploidentical-related donors) via intensive conditioning regimens and high doses of infused CD34+ donor cells.

Targeted therapy

In September 2017, the FDA approved gemtuzumab ozogamicin (Mylotarg) for the treatment of adults with newly diagnosed AML whose tumors express the CD33 antigen (CD33-positive AML). The FDA also approved gemtuzumab ozogamicin for the treatment of relapsed or refractory CD33-positive AML in patients aged 2 years and older. [45]

Gemtuzumab ozogamicin originally received accelerated approval in May 2000 as a stand-alone treatment for relapsed CD33-positive AML in older patients, but was voluntarily withdrawn from the market after subsequent confirmatory trials failed to verify clinical benefit and demonstrated safety concerns, including a high number of early deaths. The September 2017 approval includes a lower recommended dose, a different treatment schedule, and a new patient population. [45]

The approval is based on data from several trials, including the ALFA-0701 and AML-19 trials. The multicenter, open-label phase III ALFA-0701 trial randomized 271 patients with newly-diagnosed AML to daunorubicin and cytarabine alone or combined with gemtuzumab ozogamicin. Gemtuzumab at 3 mg/m2 was administered on days 1, 4, and 7 during induction and day 1 of each of the 2 consolidation chemotherapy courses. The primary endpoint was event-free survival (EFS) with a secondary endpoint of overall survival (OS). Gemtuzumab ozogamicin was associated with a statistically significant improvement in EFS of 7.8 months. However, the drug was not associated with a significant improvement in OS. [46]

In the open-label phase III study AML-19, elderly patients who could not tolerate other AML treatments were randomized to gemtuzumab ozogamicin (n=118) or best supportive care (n=119). Gemtuzumab ozogamicin was initially administered at 6 mg/m2 on day 1 and 3 mg/m2 on day 8. Patients without evidence of disease progression then received the drug at 2 mg/m2 on day 1 every 4 weeks. Patients received a single course of gemtuzumab ozogamicin. The trial measured how many patients achieved a complete remission. Following treatment with gemtuzumab ozogamicin, 26% of patients achieved a complete remission that lasted a median 11.6 months. [47]

Therapy in older patients

Overall, the results of treatment of AML in elderly patients (particularly those older than 75 years) remain unsatisfactory. In a Cancer and Leukemia Group B (CALGB) study, patients older than 60 years had a complete remission rate of 47% after standard therapy. There were 31% aplastic deaths, and only 9% of patients were alive at 4 years. It should be noted that patients with antecedent hematologic disorders were excluded; accordingly, these results overestimate the benefit of chemotherapy in elderly patients.

Many patients are never referred for treatment, because of serious comorbid medical conditions and the knowledge that the treatment results are poor in this group of patients. For example, Menzin et al analyzed Medicare claims for treatment of AML. [48] In this study, only 30% of patients received chemotherapy (44% of patients aged 65-74 y, 24% of patients aged 75-84 y, and only 6% of patients 85 y or older).

Despite this, approximately 90% of patients were hospitalized and the patients spent approximately one third of their remaining days in the hospital. Therefore, novel treatments need to be developed for this patient population. [48]

There is evidence that patients who are treated have longer survival than those who are not treated. In the study by Menzin et al, the median survival was 6.1 months for patients who received chemotherapy versus 1.7 months for those who did not. [48]

Similarly, Lowenberg et al reported a median survival of 21 weeks for elderly patients randomized to therapy compared with 11 weeks for patients randomized to a “watch and wait” approach. [49] In a Medical Research Council study, the median survival was significantly improved for patients who received low dose ara-C as opposed to hydroxyurea.

Some older patients do reasonably well with standard therapy. In an analysis of 998 older patients treated at MD Anderson Cancer Center, age greater than 75 years, poor performance status, previous antecedent hematologic disorder, unfavorable karyotype, renal insufficiency, and/or treatment outside of a laminar flow room were associated with an adverse outcome. [50]

Patients with none of these risk factors had a complete remission rate of 72%, 8-week mortality of 10%, and median 2-year survival of 35%, whereas patients with 3 or more risk factors had a complete remission rate of 24%, an 8-week mortality of 57%, and a median 2-year survival of only 3%. [50] Thus, some low-risk elderly patients can benefit from standard intensive chemotherapy.

A study in elderly patients with newly diagnosed AML compared conventional-dose daunorubicin (45 mg/m2/d for 3 d) with high-dose daunorubicin (90 mg/m2/d for 3 d). [38] These regimens were administered with cytarabine 200 mg/m2/d for 7 days for the first cycle. A second cycle of cytarabine alone (1000 mg/m2/d for 6 d) was also administered.

Complete remission occurred in 64% in the high-dose daunorubicin group compared with 54% in the conventional-dose group [38] ; remission after the first cycle was 52% in the high-dose daunorubicin group compared with 35% in the conventional-dose group.

Other therapies are being studied in older patients who are not candidates for intensive chemotherapy. [51] As part of National Cancer Research Institute Acute Myeloid Leukemia 14 Trial, 217 patients who were deemed unfit for intensive chemotherapy were randomized to receive low-dose cytarabine (20 mg twice daily for 10 d) or hydroxyurea with or without all-trans retinoic acid (ATRA). Low-dose cytarabine produced a better remission rate (18% vs 1%; P =0.00006) and better overall survival (OR, 0.60; P =0.0009) Overall survival was 80 weeks for patients achieving a complete remission verus 10 weeks for patients with no remission. [52]

The hypomethylators azacytidine and decitabine are also options for the treatment of AML in elderly patients. In a study of azacytidine treatment in 149 previously untreated AML patients (median age, 74 years) who were considered ineligible for intensive chemotherapy, 2-year OS was 51% in responders to azacytidine and 10% in non-responders (P<0.0001). [53]

A randomized, open-label, phase III trial in 488 patients age ≥65 years with newly diagnosed AML with >30% bone marrow blasts reported a 1-year survival rate of 46.5% with azacitidine versus 34.2% with conventional care regimens. [54] In a study of azacitidine treatment in 130 AML patients older than 50 years (median age, 67 years) who had experienced relapse or induction failure with intensive chemotherapy, the overall response rate was 17% (complete response [CR], 10%, CR with incomplete blood count recovery [CRi] , 7%). [55]

Decitabine is another hypomethylator with activity in AML. Ansstas et al reported a single institution, retrospective study of patients older than 60 years with either de novo AML or AML arising out of myelodysplastic syndrome who were treated with decitabine at 20 mg/m2 for 5 consecutive days of a 4-week cycle. Patients continued to receive decitabine until disease progression or an unacceptable adverse event occurred. The best response to therapy was CR/CRi 29%, stable disease/partial remission 49%, and progressive disease 22%. The median duration of CR/CRi was 393 days (range, 184-748 d). Median overall survival for patients presenting with WBC less than 10,000 cells/µL was 11 months (range, 0.5-59.8 mo) and WBC greater than 10,000 cells/µL was 7.1 months (range, 1.9-32.7 mo). [56]

In a multicenter, randomized trial of decitabine versus physician's treatment choice in elderly patients with AML, decitabine showed a nonsignificant increase in median overall survival (7.7 months vs 5 months, P =0.108) compared with treatment choice. The CR plus CRp rate was 17.8% for decitabine versus 7.8% for treatment choice. [57]

Clofarabine is a purine analogue that is approved by the US Food and Drug Administration (FDA) for the treatment of relapsed pediatric acute lymphocytic leukemia (ALL). A study of clofarabine and cytarabine in newly diagnosed patients with AML who were 50 years or older yielded a complete response rate of 52% and a CRp rate of 8%. Induction deaths occurred in 7% of patients. [58]

No standard consolidation therapy exists for patients older than 60 years. Options include a clinical trial, high-dose ara-C in select patients, or repeat courses of standard-dose anthracycline and cytarabine (2 and 5; ie, 2 d of anthracycline and 5 d of cytarabine). Select patients can be considered for autologous stem cell transplantation or nonmyeloablative allogeneic transplantation.

Although allogeneic HCT is a potentially curative treatment option for patients with AML, all age groups have a significant risk of death from the procedure. The risk of death increases with age, particularly in patients older than 40 years. However, the median age of patients with AML is 65 years; therefore, only a small percentage of patients with AML are candidates for such aggressive therapy.

Following ablative allogeneic transplantation, death occurs due to sepsis, hemorrhage, direct organ toxicity (particularly affecting the liver; ie, veno-occlusive disease [VOD]), and GVHD. In an attempt to reduce these toxicities, investigators have developed less toxic conditioning regimens, including reduced-intensity conditioning and nonmyeloablative transplants. [59, 60, 61]  Although myeloablative conditioning regimens are preferred for HCT for most patients with AML who are age 60 years or younger, reduced-intensity and nonmyeloablative regimens allow many otherwise ineligible patients to undergo HCT. [62]

Reduced-intensity and nonmyeloablative regimens feature the use of the purine analog fludarabine and lower doses of alkylating agents or total body irradiation (TBI). Nonmyeloablative regimens may cause only minimal cytopenias that do not require stem cell support, whereas reduced-intensity regimens do require stem cell support. [63]

A study in 190 patients age 60-70 years with AML in first remission reported lower risk of relapse and longer leukemia-free survival with reduced-intensity allogeneic HCT than with induction and postremission chemotherapy using CALGB protocols. At 3 years, risk of relapse was 32% vs 81%, respectively (P < 0.001) and leukemia-free survival was 32% vs 15% (P = 0.001); however, nonrelapse mortality was higher with transplantation (36% vs 4% at 3 years; P <0.001). [64]

Current guidelines from the ASBMT do not generally recommend autologous HCT for AML. The ASBMT considers allogeneic HCT to be standard therapy for AML in adults with high-risk cytogenetics who are in first remission (CR1) and for those in second or subsequent remission. For patients with intermediate cytogenetics who are in CR1, the ASBMT considers allogeneic HCT standard of care based on sufficiently large single- or multicenter cohort studies. [44]

For patients 60 years of age or older who have residual disease after standard-dose cytarabine, the NCCN recommends reduced-intensity HCT as an option. Reduced-intensity HCT is also an option for post-remission therapy in patients with a complete response to intensive therapy. Allogeneic HCT, preferably in a clinical trial, can be considered in patients with induction failure after previous intensive therapy. [21]

Treatment of therapy-related AML

Patients who develop AML as a complication of conventional chemoradiotherapy for a primary malignancy have traditionally had a poor prognosis, with a median survival of only 6 months. Allogeneic HCT has generally been recommended, because these cases respond poorly to traditional chemotherapy. [65]

In August 2017, the FDA approved a fixed-dose combination of cytarabine and daunorubicin liposomal (Vyxeos) for newly diagnosed therapy-related AML and AML with myelodysplasia-related changes. Approval was based on a phase III trial in 309 patients aged 60-75 years that evaluated the efficacy and safety of the combination product with cytarabine and daunorubicin given separately in the 7+3 regimen (cytarabine, 100 mg/m2/day x 7 days; daunorubicin, 60 mg/m2 on days 1, 2, and 3). [66]

The complete response (CR) or CR with incomplete platelet or neutrophil recovery (CRi) rate was 47.7% compared with 33.3% for the combination and 7+3, respectively. For CR alone, the rates were 37.3% for the combination and 25.6% percent for 7+3. OS rate at 12 and 24 months for the fixed-dose combination was 41.5% and 31.1% compared with 7+3 OS of 27.6% and 12.3%. [66]

In an exploratory analysis of the phase III study for those with secondary, untreated AML, 34 of the 52 patients (65%) in the fixed-dose combination arm who proceeded to transplant remained alive after a median follow-up of 521 days. In the 7+3 arm, after 442 days of follow-up, 13 of 39 patients remained alive (33%). [66]



Treatment of Acute Promyelocytic Leukemia

APL is a special subtype of AML. It differs from other subtypes of AML in that patients are, on average, younger (median age 40 y) and most often present with pancytopenia rather than with elevated white blood cell (WBC) counts. In fact, WBC counts higher than 5000 cells/µL at presentation are associated with a poor prognosis.

APL is the subtype of AML that is most commonly associated with coagulopathy due to disseminated intravascular coagulation (DIC) and fibrinolysis. Therefore, aggressive supportive care is an important component of the treatment of APL. Platelets should be transfused to maintain a platelet count of at least 30,000/µL (preferably 50,000/µL). Administer cryoprecipitate to patients whose fibrinogen level is less than 100 g/dL.

The bone marrow demonstrates the presence of more than 30% blasts resembling promyelocytes. These cells contain large dense cytoplasmic granules along with varying numbers of Auer rods.

Although the initial diagnosis of APL is based on morphology, the diagnosis is confirmed on the basis of cytogenetic and molecular studies. Do not delay treatment pending the results of confirmatory tests.

In more than 95% of cases of APL, cytogenetic testing reveals t(15;17)(q21;q11). Molecular studies reveal the PML/RARa rearrangement. Patients with either t(15;17) or the PML/RARa rearrangement respond well to all-trans-retinoic acid (ATRA) and chemotherapy.

A small percentage of patients have other cytogenetic abnormalities, including t(11;17)(q23;q11), t(11;17)(q13;q11), t(5;17)(q31;q11), or t(17;17). Patients with t(11;17)(q23;q11) are resistant to ATRA. Older studies using standard chemotherapy regimens without ATRA showed that approximately 70% of patients achieved complete response and 30% were disease free at 5 years. Induction failures were due to deaths resulting from hemorrhage caused by DIC, with few actual resistant cases. [67, 68, 69]

In the 1980s, reports from China, France, and the United States demonstrated that most patients with APL could enter remission with ATRA as the single agent. Unfortunately, in the absence of further therapy, these remissions were short-lived.

In addition, a new toxicity, the retinoic acid syndrome, was discovered. [70] The retinoic acid syndrome results from differentiation of leukemic promyelocytic cells into mature polynuclear cells and is characterized by fever, weight gain, pleural and pericardial effusions, and respiratory distress. The syndrome occurs in approximately 25% of patients, and, in the past, was fatal in 9%.

Subsequently, the early addition of chemotherapy resulted in a reduction of deaths caused by retinoic acid syndrome. Studies have also demonstrated that the addition of chemotherapy (idarubicin and cytarabine) to ATRA results in remissions in more than 90% of patients. As many as 70% of these patients are long-term survivors.

Chemotherapy is most effective when added early in induction (ie, day 3) rather than after attainment of a complete response. Initiate chemotherapy on day 1 of therapy for patients with high WBC counts (eg, >5000/µL).

Traditionally, the most commonly employed approach was the combination of ATRA and anthracycline-based chemotherapy. Current National Comprehensive Cancer Network guidelines recommend a variety of induction regimens that combine ATRA with daunorubicin and cytarabine; idarubicin; arsenic trioxide; or idarubicin and arsenic trioxide. Regimen options differ for high-risk APL and low/intermediate risk APL. [21]

Once patients with APL are in remission, the standard approach is consolidation therapy followed by maintenance therapy.

A North American Intergroup study evaluated the addition of 2 cycles of consolidation therapy with arsenic trioxide followed by 2 cycles of chemotherapy with cytarabine and daunorubicin to two cycles of cytarabine and daunorubicin chemotherapy without arsenic trioxide. [71] Event-free survival, the primary endpoint, was 77% at 3 years in the arsenic trioxide arm (median not reached) compared with 59% at 3 years in the standard arm (median, 63 mo).

Overall, 84% of adults were alive at last follow-up. Overall survival was 86% at 3 years in the arsenic trioxide arm compared with 77% at 3 years in the standard arm (medians not reached). Maintenance therapy with ATRA, 6-mercaptopurine (6-MP), and methotrexate is effective in preventing relapses compared with no maintenance therapy; however, the optimal schedule of this therapy is not yet determined.

Subsequent studies eliminated cytarabine from the induction therapy for newly diagnosed patients. [72] For example, the GIMEMA AIDA regimen (ie, ATRA 45 mg/m2 daily combined with idarubicin 12 mg/m2 on days 2, 4, 6, and 8 until remission) yields remissions in 95% of patients.

However, a randomized study from France questioned this approach. Newly diagnosed APL patients younger than 60 years with a WBC count of less than 10,000/µL were randomly assigned to receive either ATRA combined with and followed by 3 daunorubicin plus cytarabine courses and a 2-year maintenance regimen (cytarabine group) or the same treatment but without cytarabine (no-cytarabine group).

Patients older than 60 years and patients with an initial WBC count of greater than 10,000/μL were not randomly assigned but received risk-adapted treatment, with higher dose of cytarabine and central nervous system (CNS) prophylaxis in patients with WBC counts greater than 10,000/μL. Overall, 328 (96.5%) of 340 patients achieved complete remission.

In the cytarabine and the no-cytarabine groups, the complete remission rates were 99% for the ara-C arm and 94% for the no-cytarabine arm, the 2-year cumulative incidence of relapse (CIR) rates were 4.7% in those who received cytarabine and 15.9% in those who did not receive cytarabine, the event-free survival rates were 93.3% in the cytarabine group and 77.2% in the no-cytarabine group, and survival rates were 97.9% in patients who received cytarabine and 89.6% in those who received no cytarabine.

In patients younger than 60 years with WBC counts more than 10,000/μL, the complete response rate was 97.3%, 2-year CIR was 2.9%, event-free survival was 89%, and survival rate was 91.9%.

Another approach is the development of risk-adapted approaches to consolidation therapy. In the Programa de Estudio y Traitmiento de las Hemopatias Malignas (PETHEMA) study, patients with intermediate and high risks of relapse (ie, whose baseline WBC count was >10,000/µL or platelet count was < 40,000/µL) received 3 courses of consolidation therapy with ATRA and increased doses of anthracyclines (idarubicin, month 1; mitoxantrone, month 2; idarubicin, month 3). [73]

More recently, some have advocated a "non-chemotherapy" approach to the treatment of APL.

Lo-Coco et al conducted a phase 3 randomized trial of ATRA plus chemotherapy verus ATRA plus arsenic trioxide in patients with APL classified as low-to-intermediate risk (WBC ≤10 x 109/L). Complete remission was achieved in 100% of patients in the ATRA-arsenic group and 95% of the ATRA-chemotherapy group. Two-year event-free survival rates were 97% for ATRA-arsenic and 86% for ATRA-chemotherapy group. Overall survival was better with ATRA-arsenic trioxide, largely due to a reduction in cytopenic deaths. Longer follow-up is needed to determine if long term cure rates will be the same. [74]

Patients who have a relapse are usually treated with arsenic trioxide. Arsenic trioxide induces complete remission in 85% of patients. Toxicities include the APL differentiation syndrome (similar to that seen with ATRA), leukocytosis, and abnormalities found on electrocardiographs (ECGs). Patients can also be retreated with chemotherapy plus ATRA, depending on the duration of their first remission and cardiac status. Evaluate patients in second remission for allogeneic or autologous stem cell transplantation.


Treatment of Relapsed Acute Myelogenous Leukemia

Patients with relapsed AML have an extremely poor prognosis. Most patients should be referred for investigational therapies. Young patients who have not previously undergone transplantation should be referred for such therapy.

Estey et al reported that the chances of obtaining a second remission with chemotherapy correlate strongly with the duration of the first remission. [75] Patients with an initial complete response duration of longer than 2 years had a 73% complete response rate with initial salvage therapy. Patients with an initial complete response duration of 1-2 years had a complete response rate of 47% with initial salvage therapy.

Patients with an initial complete response duration of less than 1 year or with no initial complete response had a 14% complete response rate with initial salvage therapy. Patients with an initial complete response duration of less than 1 year (or no initial complete response) who had no response to first-salvage therapy and received a second or subsequent salvage therapy had a response rate of 0%. These data underscore the need to develop new treatment options for these patients.

Response to third-line therapy is even worse. Giles et al studied 594 patients with AML undergoing second salvage therapy from 1980 to 2004. [76] The patient median age was 50 years. Salvage therapy included allogeneic stem cell transplantation (SCT), standard-dose cytosine arabinoside (ara-C) combinations, high-dose ara-C combinations, non–ara-C combinations, and phase I-II single agents. Overall, 76 patients (13%) achieved CR. The median CR duration was 7 months. The median survival was 1.5 months, and the 1-year survival rate was 8%. A multivariate analysis identified the following 6 independent adverse prognostic factors:

  • First CR duration < 6 months
  • Second CR duration < 6 months
  • Salvage therapy not including allogeneic SCT
  • Noninversion 16 AML
  • Platelet counts < 50 × 10 9/L
  • Leukocytosis > 50 × 10 9/L.

Owing to the poor outcome with salvage therapy, it is important to refer patients for well-designed clinical trials whenever possible. Agents in late-stage clinical trials include clofarabine, vosaroxin, and elacytarabine.

For patients who are unable to participate in a clinical trial, options include high-dose cytarabine-based regimens, hypomethylators, and supportive care. Standard regimens with high response rates include the following:

  • CLAG-M (cladribine combined with high doses of ara-C, mitoxantrone, and granulocyte–colony-stimulating factor [G-CSF])
  • MEC (mitoxantrone, etoposide, and intermediate-dose cytarabine).

In a study by the Polish Adult Leukemia Group, the CLAG-M regimen yielded complete remissions in 58% of patients. White blood count cell greater than 10 g/L and age older than 34 years were factors associated with increased risk of treatment failure. Hematological toxicity was the most prominent toxicity of this regimen. The probability of overall survival at 4 years was 14%, and the probability of 4-year disease-free survival was 30% for all 66 patients in complete remission. [77]

In an initial study of MEC, 32 patients with refractory acute myeloid leukemia received salvage therapy with mitoxantrone at 6 mg/m2 intravenous (IV) bolus, etoposide at 80 mg/ m2 IV for a period of 1 hour, and cytarabine (Ara-C) at 1 g/ m2 IV for a period of 6 hours daily for 6 days. Eighteen patients were primarily resistant to conventional daunorubicin and Ara-C induction treatment, 8 patients had relapsed within 6 months from initial remission, and 6 patients had relapsed after a bone marrow transplantation (BMT) procedure. Overall, 21 patients (66%) achieved a CR. The median remission duration was 16 weeks; the overall median survival was 36 weeks. Severe myelosuppression was observed in all patients, resulting in fever or documented infections in 91% of patients. Nonhematologic toxicity was minimal. [78]

In a study by Levis et al, lestaurtinib was administered twice daily to patients with FLT3 mutant AML in first relapse. [79] Because such a small proportion of patients in the trial achieved sustained FLT3 inhibition in vivo, conclusions regarding the efficacy of combining FLT3 inhibition with chemotherapy are limited. Overall, lestaurtinib treatment after chemotherapy did not prolong survival or increase response rates in first relapse. Other FLT3 inhibitors used to treat patients with FLT3 ITD mutated AML include sorafenib, sunitinib, and the investigational agent AC-220.

In August 2017, enasidenib—an oral, selective inhibitor of mutant-IDH2 enzymes—was approved by the FDA for treatment of relapsed/refractory IDH2-mutated AML. The approval was based on data from a phase I/II dose-escalation and expansion study (AG221-C-001) in which enasidenib proved safe and well tolerated, and induced hematologic responses. Median overall survival (OS) with enasidenib was 9.3 months. In those achieving a CR, the median OS was 19.7 months and in the non-CR responders the median OS was 13.8 months. In those without a response, the median OS was 7 months. [80]


Supportive Care

Replacement of blood products

Patients with AML have a deficiency in the ability to produce normal blood cells and, therefore, need replacement therapy. The addition of chemotherapy temporarily worsens this deficiency. All blood products should be irradiated to prevent transfusion-related GVHD, which is almost invariably fatal.

Packed red blood cells are given when the hemoglobin concentration is lower than 7-8 g/dL or at a higher level if the patient has significant cardiovascular or respiratory compromise.

Platelets should be transfused if the platelet count is lower than 10,000-20,000/µL. Patients with pulmonary or GI hemorrhage should receive platelet transfusions to maintain a value greater than 50,000/µL. Patients with CNS hemorrhage should be transfused until they achieve a platelet count of 100,000/µL. Patients with APL should have their platelet count maintained above 50,000/µL, at least until evidence of DIC has resolved.

FFP should be given to patients with a significantly prolonged prothrombin time, and cryoprecipitate should be given if the fibrinogen level is less than 100 g/dL.

Antibiotic therapy

Intravenous (IV) antibiotics should be given to all febrile patients. At minimum, antibiotics should include broad-spectrum coverage, such as that provided by a third-generation cephalosporin with or without vancomycin. In addition to this minimum, additional antibiotics should be given to treat specific documented or suspected infections.

Patients with persistent fever after 3-5 days of antibacterial antibiotics should receive antifungal antibiotics. In the past, amphotericin was the standard antifungal antibiotic. Patients with fever but without a focus of infection received amphotericin at a dose of 0.5 mg/kg. Patients with sinopulmonary symptoms received 1 mg/kg.

In the past few years, however, a number of other antifungal agents have become available. These include the lipid-preparation amphotericins (Abelcet and AmBisome), newer azoles (voriconazole and posaconazole), and the echinocandins (caspofungin, anidulafungin, and micafungin). These drugs have varying roles in the treatment of neutropenic patients with either suspected or proven fungal infections.

Prophylactic antibiotics are usually used in nonfebrile patients undergoing intensive chemotherapy. Both the National Comprehensive Cancer Network (NCCN) and Infectious Diseases Society of America (IDSA) guidelines strongly recommend antifungal prophylaxis in this group of patients.

A commonly used regimen is ciprofloxacin, fluconazole or posaconazole, and acyclovir or valacyclovir. A randomized trial of posaconazole versus either fluconazole or itraconazole (center choice) in patients with AML and MDS undergoing intensive chemotherapy demonstrated a significant reduction in all-cause mortality at day 100, as well as a decrease in invasive fungal infections and a decrease in aspergillosis in patients randomized to posaconazole. [81]

Once patients receiving these antibiotics become febrile, the regimen is changed to IV antibiotics, as indicated above.

Treatment of hyperuricemia

Allopurinol at 300 mg should be given 1-3 times a day during induction therapy until the clearance of blasts and resolution of hyperuricemia. For patients who cannot tolerate oral medications, IV drugs such as rasburicase are an option. Rasburicase should also be considered in patients at high risk of severe tumor lysis (very high lactate dehydrogenase [LDH] level, baseline renal insufficiency).

Growth factors

Several randomized studies have been performed that attempted to determine the effect of growth factors on induction therapy.

In an early Japanese study, patients with poor-risk acute leukemia randomly received either granulocyte colony-stimulating factor (G-CSF) derived from Escherichia coli or no drug. Patients in the G-CSF group had a faster neutrophil recovery (20 d) than those receiving no drug (28 d), a smaller number of febrile days (3 d vs 7 d, respectively), and fewer documented infections. [82] No significant difference in response rate or remission duration was observed between the 2 groups.

In a French study of G-CSF, the duration of neutropenia was shorter in the G-CSF arm (21 d) compared with those in the placebo arm (27 d), and the complete response rate was higher in those who received G-CSF (70%) than in those who received placebo (47%); however, the overall survival rate was unaffected. [83]

In a Southwestern Oncology Group (SWOG) study, a decrease was observed in the time to neutrophil recovery and days with fever in those who received G-CSF; however, no difference in complete remission rate and overall survival rate was observed for patients receiving G-CSF versus no drug. [84]

Other groups have studied the effect of granulocyte macrophage colony-stimulating factor (GM-CSF) on induction therapy.

In an Eastern Cooperative Oncology Group (ECOG) study of yeast-derived GM-CSF in elderly patients with AML, no significant increase in response rate was observed; however, a significant decrease in the death rate from pneumonia and fungal infection was observed. [85] Neutrophil recovery rate was increased in the GM-CSF group (14 d vs 21 d, respectively), and overall survival was significantly improved (323 d vs 145 d, respectively). [85]

In a CALGB study of GM-CSF derived from E coli, no difference was observed in response rates between the GM-CSF group and the placebo group. [86] The risk of severe infection and resistant leukemia was similar in the 2 groups. However, in an EORTC study using GM-CSF derived from E coli, patients who randomly received GM-CSF after induction had a significantly lower complete rate (48%) than those who did not receive GM-CSF (77%). [87]

These data suggest that G-CSF and yeast-derived GM-CSF accelerate neutrophil recovery and decrease the risk of infection in patients who are undergoing induction therapy. [87] For this reason, most clinicians use either of these growth factors in patients who are at high risk for complications from infection.

Venous catheter placement

Placement of a central venous catheter (eg, triple lumen, Broviac, Hickman) is necessary. Caution should be used in APL and presence of bleeding risk and should withold invasive procedures.


Diet and Activity

Patients with AML should follow a neutropenic diet (ie, no fresh fruits or vegetables). All foods should be cooked. Meats should be cooked completely (ie, well done).

Patients should limit their activity to what is tolerable. They should refrain from strenuous activities (eg, lifting, exercise).



Long-Term Monitoring

Patients should come to the office for monitoring of disease status and chemotherapy effects. Guidelines from the National Comprehensive Cancer Network (NCCN) recommend that after completion of consolidation therapy for AML, patients should undergo surveillance with a complete blood cell count every 1–3 months for 2 years, then every 3–6 months up to 5 years. The NCCN recommends a bone marrow aspirate and biopsy only if a peripheral smear is abnormal or cytopenias develop. [21]